UNIVERSITY IN NOVI SAD MEDICAL SCHOOL DOSE ASSESSMENT AND OPTIMIZATION OF THE - StudyAce – Custom Writing & Research Support for All Levels

UNIVERSITY IN NOVI SAD

MEDICAL SCHOOL

DOSE ASSESSMENT AND OPTIMIZATION OF THE PROTOCOL IN

Standard Review multilayer computed tomography

– Ph.D. –

mentors:

Prof. MD, PhD Sanja Stojanovic

Prof. PhD Olivera Ciraj Bjelac

candidate:

Darko Hadnađev Šimonji

Novi Sad, 2015.

UNIVERSITY IN NOVI SAD

MEDICAL SCHOOL

KEY INFORMATION DOCUMENTATION

Serial number:

RBR

Identification number:

IBR

Type of document:

monographic documentation

Record Type:

Text printed material

Type of work (BSc., MSc., Doc.):

doctoral thesis

Name and surname:

Darko Hadnađev Šimonji

Mentor (title, name, surname,

Prof. Dr. med sci Sanja Stojanovic

position):

Prof Dr sci Olivera Ciraj Bjelac

Title of work:

“Assessment of the dose and in the optimization of the protocol

standard examinations multilayer

Computed Tomography “

Language Publications

Serbian Serbian

Language excerpts:

Jul. / Eng.

Country of publication:

Serbia

Immediate geographical area:

Vojvodina

UGP

Year:

2015

Publisher:

author reprint

FROM

City and address:

21 000 Novi Sad, Serbia, Hajduk Veljkova 3

The physical description of the work:

the number of Chapter 7/126 pages / 35 images / 10

charts / tables 125, 103 references / 5 Contributions

Scientific area:

medicine

Science discipline:

Radiology

Subject headings, keywords:

Multilayer computed tomography; dose

AS PER

radiation, radiation protection; Medical Physics

UDC

616-073: 004.9

Stores:

In the library of the Medical Faculty in Novi

Sad, Serbia, Hajduk Veljkova 3

Important Notice:

Excerpt:

FROM

Computed tomography (CT) is one of the most important diagnostic modality, the use of which is growing from decade to decade. The total number of radiological review of CT is represented by 5-10%, but its contribution to the total dose for the population of greater than 50%. Basic indicators of image quality and dose, and accompanying radiation risk depend on the applied imaging techniques and CT examination. In terms of good radiological practice, and in accordance with the basic principles of radiation protection, preferred is the use of the lowest possible dose to the patient while maintaining the image quality and the diagnostic information. The aim of the research is to define the optimal protocol review multilayer CT for the diagnosis of certain regions of the body, as well as to determine the dose and radiation risk to patients before and after optimization protocol. The analysis included a total of 437 patients, which are divided into groups according to the regions of the body that have been recorded: I – non-contrast CT head, II – a head CT of contrast, III – thoracic CT scan, IV – CT of the abdomen and pelvis, and V – CT angiography (CT angiography aortoiliac segment and the lower extremities). The study was performed in 2 stages: in the first stage using a standardized protocol for the region of the body to be recorded, and in the second phase of the CT examinations were performed by a modified protocol (by changing the parameter value of mAs), with the minimum requirements in terms of quality indicators slike.Na basis of dosimetric estimated effective dose and a radiation risk for the patients in both phases. The study used the guidelines of the Guide EUR 16262 EN, in which the parameters for assessing the quality of images to analyze different anatomical sections of certain regions of the body that were recorded. The image quality for each patient was assessed by a three-point scale for each parameter visualization of anatomic regions 0 – details are visible, 1 – details appear, 2 – details are clearly displayed. We used subjective methods where two experienced radiologists performed the interpretation of images. The final image quality score of each examination corresponds to the sum of all the parameters estimated by the three-level scale visualization. Then, for the purpose of calculating the size of FOM (figure of merit) calculated index value assessment of image quality (the sum of all evaluation parameters / number of parameters). FOM value is calculated as the ratio of the index estimating image quality and the effective dose per patient. The average value of the FOM for each group of respondents has served us as a relative indicator for comparison between non-optimized and optimized groups of patients for the same type of review. The comparison of the effective dose in the first and second phase of the research quantified the reduction in radiation burden to patients after optimization

Protocol. The results showed that the optimal choice of protocol parameters in terms of exposure (mAs by decreasing the value) it is possible to significantly reduce the dose of radiation in examining the head of 7.5%, in examining the head of angiography by 7%, while viewing of the chest by 40%, while viewing abdomen and pelvis of 25%. Group CT angiography could not be optimized because the machine did not accept the change in picture quality when given parameters optimization. Using standard protocols it is better picture quality than is necessary, and therefore higher radiation dose than necessary. Optimal choice of protocols in terms of exposure parameters it is possible to significantly reduce radiation dose while maintaining image quality that is sufficient for an adequate interpretation of radiological images.

Date Accepted by Scientific

NN higher:

Date of Defense:

TO THE

Members of the Commission:

President:

(Name / title / title /

member:

organization name / status)

member:

UNIVERSITY OF NOVI SAD

MEDICAL FACULTY

KEY WORD DOCUMENTATION

Accession number:

ANO

Identification number:

INO

Document type:

Monograph documentation

Type of record:

Textual printed material

Contents code:

Ph.D.Thesis

Author:

Darko Hadnađev Šimonji

mentor:

Prof. Dr. med sci Sanja Stojanovic

Prof Dr sci Olivera Ciraj Bjelac

Title:

“Assessment and optimization of dosage of

YOU

protocol at standard examinations with

multislice computerized tomography “

Language of text:

Serbian (Roman)

Language of abstract:

Serbian (Roman) / Angleščina

Country of publication:

Serbia

Locality of publication:

Vojvodina

Publication year:

2015

Publisher:

Author reprint

Publication place:

21 000 Novi Sad, Serbia, Hajduk Veljkova 3

Physical description:

number of chapters 7/126 pages / 35 figures

/ Graphs 10 / tables 125, 103 references /

5 appendices

Scientific field

Medicine

Scientific discipline

Radiology

Subject, Keywords

Multislice Computed Tomography; Radiaton

SKW

Dosage; Radiation Protection; Health Physics

616-073: 004.9

Holding data:

Medical Library Library of Novi Sad, 21 000

Novi Sad, Serbia, Hajduk Veljkova 3

Note:

Abstract:

Computed tomography (CT) is one of the most significant diagnostic methods whose application has been increasing from decade to decade. Among the total number of radiological examinations CT accounts for 5-10%, however, its contribution in the whole dosage for the population is greater than 50%. Main indicator of the quality of images and dosages as well as the accompanying risk irradiation depend on applied radiographic technique that is CT examination. In the sense of good radiological practice and in accordance with basic principles of protection from irradiation, the application of the lowest possible dosage for a patient together with preserving the quality of image and diagnostic information are mostly welcomed. The goal of the research is to define the optimum examination protocol by multislice CT and diagnostics of certain body regions as well as to determine dosages and irradiation risk for patients both before and after protocol optimization. The analysis has included 437 patients divided into groups according to body regions which have been scanned: I-unenhanced head CT, II contrast enhanced head CT, III – chest CT, IV – abdomen and pelvis CT and V – angio CT (CT angiography of aortoiliac segment and lower extremities). The study has been conducted in two phases: in the first phase standard protocol for the scanned body region has been applied, and in the second phase CT examinations have been carried out according to the modified protocol (by change of parameters of values mAs) with minimum requirements regarding the image quality. On the basis of dosimetric indicators the efficient dosage and irradiation risk for the patients in both phases have been assessed. In the study the guidelines form the Guide EUR 16262 EN have been observed where parameters for the assessment of image quality have been defined in order to analyze different anatomic cross sections of certain body regions. Image quality for each patient was assessed by three-level scale visualization for each anatomic region of parameter: 0 – details are visible, 1 – details are presented, 2 – details are clearly presented. A subjective method was applied where two experienced radiologists performed the image interpretation. Final assessment of image quality of every examination corresponds to the sum of all parameters according to three-level scale visualization. further,

been calculated. The value of FOM has been calculated as a quotient of the image quality assessment index and effective dosage per patient. The average value of FOM for every group of patients has offered us a relative indicator for comparison of non-optimum and optimum group of patients for the same type of examinations. By comparison of values of effective dosage in the first and second phase of the research a decrease in load irradiation for patients after protocol optimization was quantified. The results have shown that by optimum protocol selection in the sense of exposition parameters (by reduction of values of mAs) it is possible to reduce significantly the irradiation dosage at unenhanced CT head examination for 7.5%, at contrast enhanced CT examination for head 7%, at chest CT examination for 40% at the abdomen and pelvis CT examination for 25%. The group CT angiography could not be optimized since the device did not accept the change in image quality at a set of optimization parameters. By application of standard protocols the image quality better than required was achieved and along with this, a higher irradiation dosage occurred than required. By selection of protocol in the sense of exposition parameters it is possible to reduce irradiation dosage significantly along with preserving image quality which is sufficient for adequate radiological image interpretation. a higher irradiation dosage occurred than required. By selection of protocol in the sense of exposition parameters it is possible to reduce irradiation dosage significantly along with preserving image quality which is sufficient for adequate radiological image interpretation. a higher irradiation dosage occurred than required. By selection of protocol in the sense of exposition parameters it is possible to reduce irradiation dosage significantly along with preserving image quality which is sufficient for adequate radiological image interpretation.

Accepted on Scientific Board on:

Defended:

Thesis Defend Board:

president:

member:

I want to thank:

Mentor Prof. MD, PhD and Prof. Sanja Stojanovic PhD Oliveri Ciraj white man, for guidance during the drafting of the thesis;

The members of the committee for evaluation of doctoral dissertations, and especially Prof. MD, PhD Viktor Tilu for motivation and useful suggestions;

Danijeli Aranđić, from the Laboratory for protection from radiation and protection of the environment of the Institute for Nuclear Sciences, for assistance in the embodiment on dosimetric measurements and interpret the results;

Radiological technicians of the Center of Radiology, Clinical Center of Vojvodina, especially Svetlana Brujić, Jelena and Ivana Kovač Cazic that their efforts have enabled the implementation of the optimization of protocols for standard CT examinations support and understanding;

All fellow radiologists Center of Radiology, Clinical Center of Vojvodina, on the understanding and support;

Radovan parents and Mili, Igor and his wife, for the generous support, motivation and encouragement to this work to an end.

THE CONTENT

1. INTRODUCTION 2

1.1. Application of ionizing radiation in medicine 2

1.1. CT and the basic principles of radiation protection 3

1.2. The biological effects of ionizing radiation 4

1.3. Fundamentals of radiation physics 5

1.4. X-ray source and the basic parts CT scanner 8

1.5. Tomography and CT – general principles 12

1.6. Generation CT scanner 14

1.7. The principle of the formation of images in CT 15

1.8. The reconstruction of CT images on the device 16

1.9. Dosimetry in diagnostic radiology 19

1.10. Dosimetric quantities and units 20

1.11. Phantoms in diagnostic radiology 26

1.12. Factors affecting the dose of radiation in CT 27

1.13.Kvalitet images in CT 30

2. OBJECTIVE AND WORKING HYPOTHESIS 41

2.1. The goal 41

2.2. The working hypothesis 41

3. METHODOLOGY 42

4. THE RESULTS 49

4.1. Inspection of the head without contrast 49

4.2. Viewing head with contrast 60

4.3. Review chest 71

4.4. Examination of the abdomen and pelvis 83

4.5. image quality 94

4.6. Only data in the group CT angiography without modification protocol 100

5. DISCUSSION 102

5.1. Viewing head 102

5.2. Viewing head with contrast 106

5.3. Review chest 107

5.4. Abdominopelvični review 109

5.5. image quality 112

6. CONCLUSIONS 116

7. REFERENCES 119

Darko Hadnađev Šimonji doctoral thesis

1. INTRODUCTION

1.1. Application of ionizing radiation in medicine

Ionizing radiation, which is used for daily examinations of patients around the world are on the rise (1). Is administered in more than 10 million diagnostic radiological procedures, approximate 100 000 in diagnostic nuclear medicine procedures, and 20 000 of radiotherapy and therapeutic procedures in nuclear medicine (2).

Medical use of radiation makes up more than 99.9% of artificial sources of radiation which is exposed to the world population. The number of medical procedures that use ionizing radiation has increased from about 1.7 billion in 1980 to nearly 4 billion in 2007. The largest distribution is in a well-developed countries where is carried out about 75% of these procedures (1).

Categories of procedures that use radiation as follows: diagnostic radiology (radiographic, fluoroscopic procedures, diagnostic computed tomography (CT), fluoroscopy, or CT-guided interventional procedures), nuclear medicine (use of radiopharmaceuticals for diagnostic or therapeutic purposes), radiation therapy (brachytherapy and tele-) (1).

Of the total number of diagnostic medical examination in the health care system and countries categories makes CT 7.9%, in the countries of category II slightly more than 2%, and somewhat lower than the 14% in countries III / IV category. However, the contribution to the total collective CT to the effective dose of the diagnostic medical examinations is approximately 47% in countries of category I, 15% in countries category II and of 65% in countries III / IV class (uncertainty information for countries III / IV). According to a global survey of the medical application of radiation exposure and Scientific Committee of the United Nations effects of atomic radiation (United Nations Scientific Committee on the Effects of Atomic Radiation – UNSCEAR), CT accounted for 43% of the total collective effective dose in diagnostic radiology (v. Annex 1 5) (3).

In Table 1 it is evident that CT and angiography have the highest value of the average of the effective dose in a period, and from 1970 to 2007, as compared to other selected radiographic diagnostic exams in the countries and health care category (3).

Page 1 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 1. The trend of the average of effective doses of the different selected medical diagnostic

review of the country and category of health care (3)

overview

The average effective dose per-view (mS)

1970-1979

1980-1990

1991-1996

1997-2007

chest radiography

0.25

0.14

0.07

X-ray of the abdomen

1.9

1.1

0.53

0.82

mammography

1.8

0.51

0.26

1.3

4.4

8.8

7.4

angiography

9.2

6.8

9.3

The growing trend of frequency CT examinations per annum and a significant dose per examination have an important impact on the total dose to the population from medical exposure. The introduction of spiral and multi-layered scanning is reduced scan time. As a result, it is possible to perform multiple views at the time, to expand the region of interest in scanning, and introduce new techniques and examinations. Easily produce images could result in unnecessary exposure of patients to radiation (3).

1.1. CT and the basic principles of radiation protection

The system of protection against ionizing radiation has different forms, depending on the type of radiation source and nature of human activities that lead to exposure to ionizing radiation. Of particular importance is reducing the unnecessary exposure, which is accomplished by the application of three basic principles of radiation protection:

justification views:

referring patients (based on different criteria)

assessment of feasibility

optimization:

equipment

daily monitoring

limitation of individual doses and the risk (not applicable for medical exposure) (4)

Justification review

The practice is justified if the benefit to the exposed individuals or society as a whole is greater than the damage caused by exposure to ionizing radiation. The publication of the International Commission on Radiological Protection (International Commission on Radiological Protection-ICRP) 60 Principle

Page 2 of 126

Darko Hadnađev Šimonji doctoral thesis

justification of practices is exposed through a two-stage model. The first level defines the general principles of justification of practices, while at another level considering the individual procedure. In the case of radiological diagnostics and benefits and claims are directed to individuals-individual patients. Proper selection of the radiological procedure type is based on an estimate of the available clinical evidence of, the benefits obtained from the proper diagnosis and treatment of a patient (4).

A special review of medical radiation exposure is required in sensitive group (children, pregnant women), whereby is to be guided the choice in the case of unavoidable indication (vital indications eg polytrauma), and to the use of optimized modified programs (CT “Image Gently” program American pediatric radiology company-v.dalje) (1).

optimization

As the CT classified as high- procedure, it is necessary to observe the principles ALARA (to maintain the radiation dose as low as Reasonably Achiavable), and not to the loss of diagnostic information that is not to endanger the quality of the image (5). In the literature there are many strategies that can be primenti to optimize MDCT examination, a first strategy you can use is a choice of protocols and exposure parameters (for more details about the strategies in. Chapter Factors affecting dose in CT).

In June 2009, the American College of Radiology (ACR) and the Radiological Society of North America (RSNA) set up a joint working group for Radiation Protection adults. The mission of this working group is to raise awareness about the possibilities of high-elimination of unnecessary examinations, and to reduce the amount of radiation to the one that is sufficient to obtain adequate images for interpretation. At the first meeting of a joint working group in November 2009 launched a campaign “Image Wisely”, which provides educational resources to optimize the dose of adult patients, a connection with the manufacturers of imaging equipment available the latest information on techniques for dose reduction for certain equipment (6 ).

1.2. The biological effects of ionizing radiation

Radiation protection of patients involves, inter alia the prevention of deterministic effects (tissue reaction) and minimizing the probability of the stochastic effects.

Page 3 of 126

Darko Hadnađev Šimonji doctoral thesis

Tissue reactions (deterministic (direct) effect) are the adverse effects of ionizing radiation and are formed when the recommended limit by exposure of the patient to provide a reduction or loss of function of certain organs or tissues (sunburn, cataracts, sterility). Limit values, the dose ranges from 0.5 Gy to several Gy, depending on the tissue or organ radiosenzibilnosti. In diagnostic radiology doses are on the order mGy, which is below the threshold for the occurrence of deterministic effects. If the dose values higher than expected levels, but not high enough to produce visible signs of radiation damage (except in case of accidents), the problem may be unnoticed and unreported, putting patients at increased risk for stochastic effects, delayed harmful effect. Stochastic effects, proportional to the dose, are divided into two groups: somatic and genetic. The somatic effects of radiation include cancer that may have a long latent period incurred. Genetic effects are a result of damage of the reproductive cells and could be expressed in the progeny through several generations (4,7).

Ionizing radiation is a strong carcinogen, a cancer is one of the stochastic effects. How can regular CT examinations performed below the threshold of deterministic effects of interest are stochastic effects. We should not forget that patients are often referred for repeat control CT examinations, while receiving a total dose of about 100 mSv. This dose is sufficient we can talk about the increased likelihood of stochastic effects.

Risk between the dose and the probability of the stochastic effects is a continuous function, no minimum threshold. Therefore, it can be concluded that the lower the dose, the better for the patient. Lower doses also means lower quality of the reconstructed image, which is a bigger challenge for an accurate diagnosis. All this obliges the legislature to establish guidelines and diagnostic reference levels of dose exposure (8).

1.3. Fundamentals of radiation physics

Radiation is a process in which the particle energy, or the energy waves traveling through a vacuum, or through the matter which is not necessary for their propagation (8). Thus, the radiation is directed through the transfer of energy particles or waves. If the radiation is composed of particles or corpuscles called corpuscular (particle) radiation, and if energy is transmitted through the waves (quantum-energy photons) is called electromagnetic radiation (9).

ionizingradiation is further divided into direct and indirect, depending on the nature of ionizing particles. Charged particles (electrons, protons, alpha particles) are in direct ionizing

Page 4 of 126

Darko Hadnađev Šimonji doctoral thesis

radiation, they carry enough energy to ionize atoms or molecules initiatives. Uncharged particles (neutrons, photons), however, initiate direct ionizing radiation, though themselves not directly ionizing.

The origin and properties of X-radiation, photon interaction with matter

Background of ionizing radiation discovery dates from 1858, when the observed flicker Julius plucker rarefied gas which was between the electrodes at high voltage. He spotted rays that went from near cathode current towards the anode, and called them cathode rays. Wilhelm Hittorf, William Crookes and other researchers have shown that the cathode radiation spreads in a straight line, high speed and are negatively charged particles. Have been known since research cathode Eugene Goldstein radiation of 1881 and 1897. J. Thompson (10). Cathode rays are used as the basis for obtaining X-rays.

Wilhelm Conrad Röntgen was 1895. In a statement presented all the characteristics of X-rays, and the discovery in 1901 received the Nobel Prize. X-rays are named in honor of the X-rays, and the unit of ionizing radiation before the X-ray [R] which is now equivalent to 2.58 x 10⁴ C / kg.

As is known, the principle of the generation of X-rays comprising an electron bombardment of the anode to the so-called accelerated. a vacuum x-ray tube, in which the approximation of the vacuum. For the production of X-rays are needed free electrons, a high voltage anode and, with that, due to the slowing down of the electron emitted X-ray radiation (11).

The elements of the X-ray tube and the characteristics of the X-ray to see the next chapter.

There are three possible mechanisms of energy loss in the interaction of photons and absorber of which the first two are important in diagnostic radiology:

photoelectric effect

Compton effect

The pair

photoelectric effect It means complete absorption of photons in matter. Namely, when a photon of X-rays from electron collisions with a deeper layer, closer to the core, ejects it from its orbit, whereby a carbon ionization. Then it is necessary reorganization within the atom itself, that

Page 5 of 126

Darko Hadnađev Šimonji doctoral thesis

So what is takes an electron from the outer orbit and thereby light is emitted photon. The ejected electrons received the whole energy of the incident X-ray photons, which is actually completely absorbed (8.11). This is important for low-energy and high atomic numbers.

Compton effect occurs when the X-photon strikes the free electron absorber and remove it from its orbit, with the photon loses some of its energy and changes direction of propagation under different angles. Due to the loss of energy X-photon changes its wavelength, which increases and decreases the pervasiveness of X-photons (the air becomes softer). The ejected electron from its orbit, along its path, further collisions throws new electrons from atomic orbits and in this way to create pairs of ions. That created electrons are called “Compton electrons” (8:11). The imaging Compton effect is important because it spoils the image and radiating staff.

The probability of interaction for the performance of each of these mechanisms depends on the photon energy and atomic number of the absorber and is shown in Graph 1.

The atomic number of the absorber [Z]

dominant

dominantly

photoelectric effect

creation

pairs

dominant

Compton effect

Photon energy hIn MeV

Figure 1. Three mechanisms of energy loss in the absorber of photons, x-axis with logarithmic values (12)

On the graph above the green line represents the photon energy and atomic number that have an equal probability of occurrence of two adjacent mechanism. As can be seen, for very low atomic numbers dominant Compton effect for all energy and with increasing atomic number of the absorber, the photoelectric effect begins to dominate for low power while making pairs begins to dominate the high-energy (12).

Page 6 of 126

Darko Hadnađev Šimonji doctoral thesis

The dominant mechanism of interaction in the absorption of X-ray in the soft tissue of a Compton effect, as evidenced by the atomic numbers of elements that make up the human body: oxygen (Z = 8), carbon (Z = 6), hydrogen (Z = 1) and the nitrogen ( Z = 7). These four elements together constitute about 96% of the weight of the human body (19), and all are below the red line in the graph. Although the X-rays may have energy as low 120 eV range of diagnostic imaging is close to 100 keV, the labeled red circle in the image, wherein the dominant Compton effect (13).

Thus, the X-rays after passing through the material (human body) of different structure, the density and atomic weight due to poor absorption and scattering. Hard-rays, higher frequencies and shorter wavelengths and have more energy and less absorb and disseminate more or less remain in the body. The reverse is true for softer rays that are less frequency, longer wavelengths, lower energy, more absorbing and less disseminate, more is retained in the body of the stray rays. Scattered radiation is greater if larger volume irradiated object, if it is higher density if the applied voltage to the X-ray tube higher (11).

1.4. X-ray source and the basic parts CT scanner

The main parts of CT scanners are:

gantry with X-ray tubes, detectors and stand

moving bed for the patient

high voltage generator

evaluation and control console with memory, monitor, system for transmitting images on film and printer.

Gentry represents a base unit each CT unit and contains the X-ray tube with detectors. This is a rotating ring with a significant radius of the cooling system and the system of transmission of signals from the detectors to the Analog-Digital (A / D) converter (Fig 1). In most CT device has a diameter of 70 cm. However, the surface of which is used to measure the attenuation coefficient is lower, and ranges from 50 to 55 cm. Because of these differences, it is possible to create an artifact that makes the image useless, if the dimensions of the patient is greater than the dimension which measures the attenuation coefficients. Specially designed code CT, such as PET-CT, the opening and gantry amounts to 90 cm. On the gantry there are laser systems for precise patient positioning. Gantry with their content (X-ray tube and detectors) can be achieved at angles of 0 ° to + 30 °, and 0 ° to -30 ° (14, 15).

Page 7 of 126

Darko Hadnađev Šimonji doctoral thesis

Picture 1. Pipe-system with a continuous rotation of the detector in the scanner of the spiral (14)

X-ray tube is the most important part of the CT system to a good and useful images (14). The X-ray tubes that are used in modern CT scanners are based on the same principle of the model used by Roentgen in the end of the 19th century (See Figure 2).

The target of

tungsten

Figure 2. Schematic representation of the X-ray tube (8)

Within the x-ray tube there are conditions close to a vacuum; cathode emits a steady stream of electrons whose orbits are controlled by focusing cup. The anode is placed opposite the cathode, and bears firmly attached metal target, usually made of copper and / or tungsten. Large potential difference across the anode and the cathode, up to 150 kV, causing the movement of the electrons directed from the cathode toward the anode. These electrons lead atoms of metal targets at the anode in the excited state, which results in the emission of X-ray diffraction as mentioned above. These X-rays leaving the x-ray tube through a dialog glass x-ray tube, and to the so-called. characteristic X-rays, which depend on the material of the anode (8).

There is also another type of X-ray, which is also transmitted from the X-ray tube, together with the characteristic X-rays. When the electrons reach the metal targets, they stray from their path as

Page 8 of 126

Darko Hadnađev Šimonji doctoral thesis

the result of the fact much greater positive charge of the atomic nucleus, which in addition to closely pass. In this turn electrons reduce his speed, losing kinetic energy, which is consecutively emitted as a photon. This emission of X-rays by bremsstrahlung is referred to as ( “bremsstrahlung”) and in contrast to the characteristic X-ray diffraction provides a continuous heterogeneous spectrum of X-rays of different wavelengths, which does not depend on the material of the anode (8,11).

The first filter through which the X-rays is a glass wall of the bulb x-ray tube, the other is a layer of oil between the shell and the metal tube, ie. window through which X-rays are output, a third filter is a material of which is made the material window (11).

Unlike other diagnostic methods that use X-rays, CT is the most demanding due to higher thermal loads. In earlier generations, the recording lasted much longer, and one of the reasons it was cooling and anode. In recent generations, this break is reduced to a minimum, the development of materials that make up the anode and the cooling mode. By X-pipe with conventional CT apparatus could be used 1,000 hours, and today, with the development of technology from 10 000 to 40 000 operating hours (15).

Common values of power exercised by the X-ray tube from 20kW to 150kW with voltages from 80kV to 140kV, current up to 800 mA with a heat capacity of 1 000 000 to 2 000 000 units of heat .. Therefore, special attention is paid to the development of the anode plate. The maximum power of the device can be realized over a longer period of time. Tube for X-rays is mounted on a round base, which rotates around the patient’s body and supplied with electricity. A number of scanners used cables that wrap around the base until it rotates. This method allows for several rotations. The stand must be stopped and the rotation begins in the opposite direction to the cables unrolled. Another way is through the use of a sliding electrical contact, no cables, allowing continuous, fast rotation – slip rings (Figure 1) (11,15).

In a computerized tomographic technique is applied a narrow x-ray beam and the normal of the imaging methods are different, inter alia, and in that the X-ray film detectors replaced. The receiver is comprised of images detectors and associated electronic circuits (preamplifier weak currents that come out from the detector) (11). The detector system has a special role in the construction of the CT component, because that turns on X-rays of different intensity into an electrical signal. In ST, as well as other digital technique, the resulting image is the product of multiple detection, measurement and calculation of digital information (details about the origin of the image with the CT v. Section 2.5).

Page 9 of 126

Darko Hadnađev Šimonji doctoral thesis

Depending on the apparatus, the detector has from 300 to 1000. The essential elements of the detector are absorbed by a substance which the energy, and is designed on the basis of a crystal and a photodiode. Some materials have the ability to a much greater extent detected by X-rays, and some of them are based on ultra fast ceramic (UFC). This feature new types of materials significantly improves image quality, and therefore faster data acquisition (14.16). There are three types of detector are based on different physical principles (11):

scintillation detectors

gas detectors

semiconductor detectors

Most of the new CT apparatus used or xenon detectors (3rd generation CT) scintillation detectors or solid-state (of tungstatata (CWO crystals) gadolinijumskih or ceramic material) (17).

Due to the limited thickness that can be achieved by using a silicon crystal, the detector system is made by the principle of stacking the small blocks of the detector (Figure 3). One detector block consists of 24 · 64 = 1536 elements, while an entire detector system consists of 38 · 4 = 152 blocks (14).

Figure 3. The working principle of the detector (14)

Multi-row detectors more efficiently use X-rays which are delivered to them, as opposed to the single row of detectors (14).

TABLE FOR PATIENT is designed so that the patient can withstand up to about 200 kg, and the special devices and to the CT 270 kg, with the accuracy of the displacement of 1 to 2 mm. They are made of

Page 10 of 126

Darko Hadnađev Šimonji doctoral thesis

carbon fiber, with minimal attenuation coefficient, so as not to affect the measurement.

Table length is up to 200 cm (15).

The main parts of the generator are: a high voltage transformer, numerous smaller transformers and switches, are necessary for the smooth operation of the appliance. High voltage generator receiving AC electrical power from the city power grid and transform it into direct current high voltage that is required to operate the X-ray tube, usually from 80 to 140 kV. Electricity is transmitted through the slip – ring connections, in order to reduce the possibility of electrical arcing between high gear in the gantry device (14).

1.5. Tomography and CT – general principles

Computed tomography (CT) is a digital imaging method that has been introduced into clinical practice in 1972, although the basic settings tomography detected early 20th century. This is certainly led to a revolution in diagnostic radiology, because the former planar radiological methods did not provide a three-dimensional display morphological changes (18,19).

The first clinically applicable CT scanner developed in 1972 by English electrical engineer Sir Godfrey Hounsfield, who along with Allan McLeod Cormack, South African physicist, creator of the idea of this use of X-rays, shared the Nobel Prize in Medicine for the development of computed tomography (18,19).

The essence of the computerized tomography is in parallel and a number of measurements of the absorption coefficient of a certain cross-section area of one or more organs through which the X-rays in a very narrow beam and from various angles. After passing through the authorities receive x-rays attenuated that accept and measures properly spaced detectors. What will be the absorption of X-rays depends on the thickness of the layer, density of the tissue and its ordinal number in the periodic table of elements (11).

Photodetectors include crystals that are more sensitive than crystals of silver bromide the X-ray film, even if there are minor differences in the absorption (from water, normal blood and of blood clot), which reflects the different opacity of the Contrast Ratio, as opposed to x-ray film, where they are not detected . The difference in the detectors is not crucial at planar imaging and CT. In CT there is no superposition of tissue. Computerized tomography is representative of the difference in the density of the tissue, and the absorption of X-rays that exists within the organ tissue, and between the sick and healthy tissue (11).

Page 11 of 126

Darko Hadnađev Šimonji doctoral thesis

During the review tomographic X-ray tube moves axially along the axis of the patient, precise and synchronized with the detector (film for conventional radiography, level detector for digital radiography). The obtained coronal cross-section, giving the axial view of the region of interest. In CT, X-ray tube and the detector move transaxial normal to the patient’s longitudinal Z-axis, so as to give the transverse section of the region of interest (18). Schematic representations given in figures 4,5 and 6.

Figure 4. Schematic representation of the classic tomografije- axial coronal section (18)

Figure 5. Schematic representation of the principle of computed tomography (18)

Figure 6. Schematic representations of the axial tomographic sections and CT transaksijalnog cross-section (18)

Page 12 of 126

Darko Hadnađev Šimonji doctoral thesis

1.6. Generation CT scanner

Technological developments scanner is reflected in its many generations. The first EMI-generation CT scanner is contained only one detector. The movement system was translational-rotational and required a lot of time (about 5 min.) For each record, and the patient had to be fixed for that time, it was uncomfortable for him (11).

To shorten the scan time introduced a system of multiple detectors that are mutually shifted by small angle x-ray beam in a plane layer. This is a characteristic CT scanner of the second generation (See Figure 7). Still, the system moved rotationally-translationally (11).

1) 2) 3)

Figure 7. 1) Scanners 2nd generation. A) Transmission of multiple narrow beams by a number of detectors during each translation. BD) Small angle between the narrow beam allows each detector to obtain a completely separate data at different angles; 2) The geometry of the 3rd generation.

Detectors in a wide range. The tube and detector are firmly connected to the interface; 3) Geometry 4th generation scanners. Fixed detector ring. In the beginning of the X-ray tube was inside the ring and rotated, the subsequent design of X-ray tube is moved outside of the ring. (20)

Third-generation CT scanner (See Figure 7) is based only on the rotary motion of the system. X-ray tube is placed in a circular frame, gantry, opposite the detector array (8). Canceling the translational movement, and the x-ray beam is extended and covers the whole layer (cross-section) of the object (11,18).

The fourth generation CT device has a full circle of evenly spaced detectors that are stationary. Moving only the X-ray tube and detector inside the circle (See Figure 7). First, they were CT devices with a rotating X-ray tube and stationary detectors (11,18).

Page 13 of 126

Darko Hadnađev Šimonji doctoral thesis

The spiral, subsekundni scanners have improved clinical applications, eg. volume scanning in one breaths without missed lesions. Also increased patient comfort and decreased artifacts due to the movement, so that the basic characteristics of the CT scanner 5th generation were subsekundno during continuous rotation and volume of acquisitions (18).

Multidetector CT (MDCT) is used more detector rows along the Z-axis or. along the patient (longitudinal direction of right angles to the axial plane of the CT). A large number of detector rows collect data that can be combined as if they were collected with a detector, which is a great advantage MDCT. The term “channel” is the smallest unit of the detector in the direction of the Z-axis, from which they may independently collect data. Commercial MDCT can simultaneously collect data from up to 64 independent channels. Each channel has collected enough data to create an axial cross-section (21).

Manufacturer GE (General Electric) was used detector rows 16 of the same size along the Z-axis, while the producers of SIEMENS, PHILIPS and TOSHIBA used detector rows along the Z-axis of different sizes. The use of detector rows of different sizes can be achieved by a number of different section width. If necessary sections of smaller width, use only the central detector elements of detector rows (19).

MDCT scanners enable rapid collection of data from the recording volume with less load on the X-ray tube, and thus shorten the exposure time. Their advantage lies in the fact that can capture specific anatomical volume with a thinner cross-sections, which significantly improves the spatial resolution in the longitudinal direction (21).

1.7. The principle of the formation of images in CT

Basically emergence radiography images is based on the phenomenon of attenuation of X-rays in their passage through the body. In CT scanner measurement of intensity attenuation is performed by means of sensitive detectors, which under the influence of the received x-rays generate electrical signals proportional to the intensity of the radiation (11).

X-ray tube and the detector are mechanically joined together so as to form a firm coupling, which is always oriented in space in such a way that the central air falls exactly on the detector and lying in the plane layer that is being recorded. During radiation coupling moves first translational example. from left to right, so that

Page 14 of 126

Darko Hadnađev Šimonji doctoral thesis

sweeps out the entire layer. Then the team moves to a certain angle (1⁰) And re-translational sweeps the whole layer far from right to left. This is repeated until the team does not turn 180⁰, Which results in that the selected layer is recorded from all the projection (11).

Electrical signals that give the detectors are converted to digital values and lead to the computer where they are processed and organized in a matrix or so. digital “numerical” picture. The matrix also has the measured values of attenuation of each point of layers arranged so that their physical position in the layer has a corresponding place in the array. Thus created matrix or latent image in a digital form can be read by a special procedure, and for the reconstruction of the images and the intensity of the 0% and 100% are added to the Process control electronics different levels of gray (white = 0%, black = 100%) (11) .

Prior to the image reconstruction should perform correction data obtained from the detector. This is important because the beam of X-rays formed in an X-ray tube is not mono energy (19).

1.8. The reconstruction of CT images on the device

Image reconstruction involves forming an image from the measurement data obtained during one scan and then display it on the monitor screen image (11). In fact, it is an electronic record of attenuation of X-rays in the form of a matrix of intensity (18). Attenuation of X-rays passing through the material-tissue is defined as the linear coefficient of attenuation, and denoted by the symbol μ. It is influenced by the radiation energy, the energy spectrum of x-ray radiation, the composition of the tissue through which the x-rays and subsequent interaction (absorption, scattering, …) (11).

It is necessary to determine μ at each point of the subject. For this reason, the building is divided into a number of picture elements (pixels) which are two-dimensional representation of the corresponding tissue volume. Pixels are arranged in a corresponding number of rows and columns in a given matrix (256 x 256, 512 x 512, etc.). Each measured value μ in the matrix has its place corresponds exactly to its position in the house. The numerical value attached to each pixel is called CT number or Hounsefield Unit (HU) (11,18). Voxel represents the volume of a three-dimensional view of tissue caused by pixel and the thickness of the CT layer (11). Only rekonstrukcionog matrix is shown in Figure 8.

Page 15 of 126

Darko Hadnađev Šimonji doctoral thesis

Figure 8. The reconstruction matrix. Haunsfildova vision of the scanned cross-section consisting of matrix of small cubes of tissue – voxels, each with its own attenuation coefficient μ. Mere transmission of the X-ray (Ni) may be expressed as the sum of the values of attenuation which occurs in voxels along the path

There is no air. The aim of the reconstruction of CT images to determine how much attenuation narrow beam of X-rays produced at each voxel reconstruction matrix. These computed values can then represent the attenuation of the level of gray as a two-dimensional image cross-section (plane X and Y) and is referred to as a pixel. (20)

Each pixel is represented on the screen as a shade of gray suits and CT number in the range of –

1,000 to about 1,000, wherein the CT number of the water is 0, and the air-1000. Medical CT scanners operate with a typical range of -1024 to +3071 HU (18). As a reference value of the absorption in CT selected is water and it has been given a value of the relative linear attenuation coefficient is equal to zero. It is determined according to other values of the tissues and organs (V.tabelu 2) (11).

Table 2. CT numbers for some of the tissues and organs (19)

The tissue / organ

CT number (HU)

bones

1000 +

hemorrhage

60 to 110

liver

50 to 80

Muscles

44 to 59

blood

42 to 58

gray matter

32 to 44

white matter

24 to 36

Heart

24

cerebrospinal fluid

0 to 22

Water

Fat tissue

-20 to -100

Lungs

-300

Air

-1000

Page 16 of 126

Darko Hadnađev Šimonji doctoral thesis

Trained human eye can distinguish only about 30-40 shades of black and white rocks. For this reason it is necessary to choose only one band absorption spectrum of the total value of the image, so that only one spread over a certain number of degrees of gray that the human eye can distinguish (method “window”) (11). Specifying the CT numbers (for window-window width) and the center of the window CT numbers (window) receives a greater contrast resolution, especially in soft tissue, because his contrasting the differences are small (22). This setting can be manually or automatically (See Figure 9).

Figure 9. The width of the center window CT numbers (19)

Gore noted that the obtained numerical (digital) image of the subject, which is electronically transferred to the computer. Image reconstruction, in fact, the mathematical algorithm by which receives two-dimensional distribution of attenuation coefficients calculated from a sufficient number of projections. In modern CT devices used algorithm based on Fourier transformations. For avoidance of interaction between the surrounding elements (pixels), the special mathematical filters, which removes unsharpness. Upon completion of calculating the approach to determining the degree of opacity of electron for each element of the matrix, in proportion to the calculated values (11). The computer is necessary to simultaneously solve a large number of equations underlying the image reconstruction (18).

Data processing (data processing) is much more complicated systems in CT imaging compared with conventional radiography because the individual images to be reconstructed from sinograma individual sections. Sinogram contains data collected during discrete intervals

Page 17 of 126

Darko Hadnađev Šimonji doctoral thesis

rotation, so the process of image reconstruction thing reversion rotation. This concept is known as back projection (back projection) (8).

The image acquisition code refers to a term describing the process in which a sinogram is first converted into an initial image using the filtered projection backwards (FBP). This figure is defined as original reconstruction, and as such can use two methods of iterative reconstruction (IR): classic methods where the iteration (repetition) between the picture space and space data, and newer methods limited to iterations (repetitions) within the picture space. Both methods can be combined (23,24).

1.9. Dosimetry in diagnostic radiology

Evaluation of dose values is the most important aspect of the feasibility review, and in addition the optimization process is based on the knowledge of dose values, depending on its exposure parameters and the choice of equipment.

Doses of tissue or organs are not directly measurable. In practice, the applicable measurable specific size, which are modified by a coefficient of a conversion of the absorbed dose to provide a total dose value of an organ or tissue (25). Determination of the dose should be done calibrated instruments.

In a CT-specific application size is recommended by the International Commission on Radiation Units and Measures (International Commission on Radiation Units and Measurements-ICRU) (26).

Computed tomography is a unique modality because of its geometry and application, with a continuous exposure around the patient, and have their own specific parameters for calculation of the radiation dose. This mode of use typical of thin cross-sections in the range of 0.5 to 20 mm nominal collimation of air. However, it also uses multiple exposures specified length of the patient to cover the anatomical volume. This exposure can be performed in the sequence of the scans (such as, pre- and postkonstrastna series of scans).

CT scans of tomographic exposure with full rotation of 360 ° results in a radially symmetrical gradient dosage levels within the patient. To magnitude of a dosage gradient (difference

Page 18 of 126

Darko Hadnađev Šimonji doctoral thesis

between the center and the periphery) due to several factors, including the size of object, range of an x-ray attenuation of the tissues and (27).

It is important that the beam of radiation used in CT narrow (kolimisan), and that a significant part of the energy goes outside the nominal beam width. It should also be noted that CT examination includes a series of single-section until the cover observed anatomical area, with the possibility of overlap of adjacent sections, and it is not known whether the listed dosage relates to a single-section or the whole series section. For this reason, it is necessary to define specific CT dosimetry measures (19).

1.10. Dosimetric quantities and units

In general the dose rate of radiation include exposure, the absorbed dose and the effective dose, while specific CT dosimetric quantities CT dose index (Computed Tomography Dose Index-CTDI), the product of the length and the dose (Dose Length Product-DLP) and effective dose (27).

The mean value of kerma in the air at one intersection in the literature referred to as CT index kerma in the air, CTDIair or CT dose index (28,29).

Extreme kolimisan beam of X-radiation which is used in the CT produces an extremely non-uniform dose distribution in the patient’s body. Assessment methods Patient doses in CT are based on the basic dozimetric size – CTDI, which is a measure of the local dose and protocol review (19), which is given by:

∫+ ∞ ()

(30)

-∞

where T is the thickness of the tomographic layer and D (z) of the absorbed dose distribution along the axis parallel to the axis of rotation of gantry that is labeled with (4).

CTDI is the mean absorbed dose along the Z-axis of the patient of a series of adjacent exposures.

Page 19 of 126

CTDI dozaRelativna

Position-section Figure 2. Only CTDI (31)

Therefore, the measure of the CTDI irradiation within the observed cross-section and is dependent on the primary collimation, distance collimator tube, and an amount of leakage polusenke radiation. Measurements can be performed only for one rotation of the X-ray tube (an axial cross-section), and is calculated as the quotient of the integral of the absorbed dose and the nominal total collimation of the beam. Therefore, the quotient of the CTDI is equal to the area under the curve of the dosage and the total width of the profile sections (31) (See Figure 2).

CTDI unit is J / kg, and is called the gray (Gy).

CTDI100

CTDI100represents the integral of the radiation dose profile for a single axial section and for defined limits of the integral that is ± 50 mm. Dosimetry size CTDI100 is defined as:

Darko Hadnađev Šimonji doctoral thesis

100 =

∫+50

() .

(21)

∙

-50

Normalized CTDI (nCTDI) is the ratio of CTDI and products amperage and time

exposure (mAs), respectively:

100

(21)

wherein Q is the product of tube current intensity and the exposure time (the amount of the charge tubes).

Page 20 of 126

Darko Hadnađev Šimonji doctoral thesis

Normalized CTDI is a characteristic of the observed model scanner. Indicates scanner capacity in terms of radiation output, used for easy comparison of different devices and has no meaning when it comes Patient doses in CT examinations (19).

The weighted CTDI (CTDIw)

CTDIw represents the dose mean for the observed cross-section and is obtained by combining the value of CTDI100 in the center (c) and on the periphery (p) of the phantom in the following way:

∙

∙ (

Σ4

)

(21)

100

= 1

100,

CTDIwcan be measured separately in the phantoms of the head and body. In the case of using a phantom head (made from polymethylmethacrylate PMMA-16cm) value in the center (CTDI100, c) And on the periphery (CTDI100, p) Are almost identical, which is not the case when using the phantom body (32 cm PMMA), where the value was measured in the center is approximately one-half the value in the periphery, as a result of the phantom larger in diameter and consequently a large attenuation of the radiation. CTDIw could also be normalized, analogous to (19).

CTDI is measured in air or in phantom ionization chamber and TLD. Based on the recommendations of the Food and Drug Administration (FDA), the measurements are performed in the center and on the periphery of the cylindrical phantom made of PMMA diameter of 16 cm and 32 cm. (4)

CTDI is measured cylindrical ionization chamber length of 10 cm. Size CTDI10cm is defined by the expression:

∫+50 () .

(32)

-50

For measuring the difference between the phantom CTDIFDA i CTDI10cmcan be significant. During the measurement of the CTDI in the air scattered radiation contribution is negligible (4).

CT pitch factor (p)

In helical scanning patient that moves simultaneously with the rotation system CEV X-ray detector, which describes a helical path around the patient’s chair. Therefore, one of the most important factors in the so-called helical scan. PIC (pitch) (factor of thinning), which is defined by the scan speed. By definition is the quotient of the displacement of the patient table speed (in mm) per one rotation of the X-ray tube and the layer thickness, ie. collimation (in mm) (18) (See Figure 10).

Page 21 of 126

Darko Hadnađev Šimonji doctoral thesis

Infinitely

The distance per rotation

beamwidth

Figure 10. Schematic representation of the budget pic factors (33)

If PIC factor is exactly 1, it means that adjacent sections of touch; if PIC factor less than 1, the adjacent sections overlap, and if it is greater than 1 there are gaps between adjacent sections. According to the International Electrotechnical Committee (International Electrotechnical Commission IEC) (34) of the definition of CT PIC factor is:

ℎ =

(34)

∙

where Δd is the displacement of the patient bed according to the rotation, T is the nominal thickness of the cross section in mm, N is the number of simultaneous tomographic cross section for one rotation of the tube (N = 1 to N and SDCT> 1 for MDCT) and N · T is the nominal beam collimation (34).

If the value of the PIC factor increases, the MDCT scanner software can automatically increase the volume of tube current (mA) and noise so that the dose received by the patient is relatively constant with a change of the PIC factors (35).

Volume CTDI (CTDIox)

By definition, the mean dose of the scanned volume:

(21)

ℎ

Page 22 of 126

Darko Hadnađev Šimonji doctoral thesis

Essentially, CTDIox adjusted value of CTDIw the value Pic factors, containing information for a specific recording protocols, taking into account gaps or overlaps of neighboring dosage profile (19).

A product dose and duration (DLP)

All the above dosimetric quantities based on CTDI measures are local dose in the reporting section. The confusion arises when it is asked whether the patient exposure in patient 10 with the cross-section of 10 times the dose of hits with a single cross-section. Although at first glance does not seem logical, in both cases, the dose is the same. This contradiction is derived from the definition of the dose absorbed as the amount of energy absorbed per unit mass of dE dm, and with an increase in cross-section, and automatically increases the weight, and at the same doses remained constant (19).

DLP is a measure of overall risk (dose) to the patients. The DLP computed tomography dose index is the product of the length of the spiral scan:

∙ =

(31)

DLP is the additive value. It is calculated as the integral under the curve. SI unit for the DLP mGy ∙ cm.

If radiological examination consists of several stages, then the DLP for overall review is given by the sum of GLP-s of each individual phase (V. Figure 3).

Figure 3. Only DLP (31)

Newer scanners on display show the value of CTDIoxand DLP (Figure 11). DLP as a measure of the total energy absorbed for the given protocol scanning is an integral measure of the risk of exposure to ionizing radiation. For example. CT examination of the abdomen can have the same CTDIox and CT

Page 23 of 126

Darko Hadnađev Šimonji doctoral thesis

pelvic or abdominal DLP will be greater in proportion to the anatomical area that is scanned (31).

Figure 11. An example presented dosimetric parameters of CT examination region wide. (Center of Radiology, Clinical Center Vojvodine- Novi Sad)

effective dose

The effective dose of (E) is, as mentioned earlier, the parameter indicating the risk of exposure to radiation and to normalization of the partial exposure or the exposure part of the body in relation to the exposure of the entire body. It shows the ratio of possible damage to health due to stochastic effects, and depends on the sex and age of the patient at the exposition. For its assessment of the need to include radio knowledge of certain organs of the body, which is usually obtained by using Monte Carlo simulations and by using special boys. SI unit for effective dose is the sievert (Sv). Utilizing E can be compared exposure to different sources of ionizing radiation, as well as natural ventilation (19,36,37).

Practical approach for estimating the value of E is to use DLP, year and regional specific conversion factor k according to the equation (37):

∙ =

(21)

where k is a conversion factor, which is a unit of mSv ∙ mGy-1∙ cm-1 and depends on the observation of organs and the patient’s age (V. tabulation of coefficients for the various parts of the body – Table 3).

Page 24 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 3. Normalized effective dose for adult patients DLP standard material for pediatric patients of different ages and for different parts of the body (38)

anatomical region

k (mSv  mGy-1  cm-1)

Age range

0 years

1 year

5 years

10 years

Adults

Head and neck

0,013

0.0085

0,0057

0.0042

0,0031

Head

0,011

0,0067

0.0040

0.0032

0.0021

neck

0,017

0,012

0,011

0,0079

0,0059

Chest

0,039

0,026

0,018

0,013

0,014

Abdomen and Pelvis

0,049

0,030

0,020

0,015

hull

0,044

0,028

0,019

0,014

0,015

The concept of effective dose is primarily developed for the control of occupational exposure,

and in radiation protection used for the purposes of optimization, and should not be applied in any

epidemiological studijamailizaprocenurizikazapojedince.Zatakvunamenuse use

absorbed dose. In pacijentnoj dosimetry biggest advantage of the concept of effective dose is that

There is a possibility of comparing different doses of diagnostic and therapeutic procedures and the possibility

comparisons of the same imaging technique, or in a variety of medical institutions (in. Table 4).

Table 4. Typical values of the effective dose for a particular diagnostic radiology procedures (38)

type of review

effective dose

CT head

– 2 mSv

CT chest

– 7 mSv

CT of the abdomen

– 7 mSv

pelvic CT

– 4 mSv

CT of the abdomen and pelvis

– 11 mSv

CT angiography

– 12 mSv

radiography of the head

 0.1 mSv

Radiographs of teeth

 0.1 mSv

Radiographs of the chest

0.1 – 0.2 mSv

mammography

0.3 – 0.6 mSv

Radiography profile spine

– 6 mSv

IUD

– 6 mSv

Diagnostic angiography

– 10 mSv

1.11. Phantoms in diagnostic radiology

Fantom as a general term means an object with no content, and in diagnostic radiology and

dosimetry represents a physical object or a mathematical abstraction. Its function is to simulate

transport of X-rays through the patient’s body in terms of absorption and scattering, and to enable measurement and

Budget doses in conditions that a satisfactory level of confidence reflect clinical practice.

In addition, the phantoms can be used to assess the image quality in diagnostic radiology (4). For

Page 25 of 126

Darko Hadnađev Šimonji doctoral thesis

determine the dose in CT are used in standard cylindrical phantoms diameter of 16 cm to simulate a pediatric examination and head adult patients, and 32 cm in diameter for determining the examination of the body of adult patients.

1.12. Factors affecting the dose of radiation in CT

The factors which directly influence on the dose of radiation are as follows: energy X-ray (voltage), current (mA), time of rotation or the exposure, the total length of the scanned region, thickness of the sectional shape, the thickness of the object or the attenuation, the PIC factor reduction techniques dosage forms such as variation or modulation current, the distance from the x-ray tube to the isocenter. Factors that indirectly affect the radiation dose to which they have a direct impact on the quality of the image. reconstruction filter. Automatic exposure control (automatic exposure control -AEC) and iterative reconstruction are more qualitative than quantitative impact on the dose and image quality. The selection of these parameters can affect the operator to change settings, which directly influence on the dose of radiation (8.27).

Amperage (MA)

Act to affect the number of photons coming out of the x-ray tube, as determined by the number of electrons emitted from cathode ray. For practical reasons, shall be used mAs, current multiplied by the scan time in seconds (8). Reducing the strength of current is the most practical way to reduce a dose of the CT, since the dose of radiation is directly proportional to the amount of current. Thus, for example. If we reduce the amount of current by 50% will reduce the radiation dose in half. It must be addressed without increasing the noise image, which can affect the outcome of a diagnostic review (39) (See Figure 12).

Figure 12. Expiratory MDCT in the rms value of 140 kV and 80 mAs. simulated scans lower effective mAs values at 60, 40 and 20 mAs. Zone “air trapping” are alike

detectable, at each dose. (40)

Page 26 of 126

Darko Hadnađev Šimonji doctoral thesis

Tube voltage (kV)

Represents the difference in potential between the cathode and anode in the x-ray tube, which determines the acceleration of electrons. This fact affects the kinetic energy of electrons when they reach the anode, and therefore the number of interactions before they are absorbed. As a consequence, an increase in voltage (kV) pipe will increase the dose, or if other factors remain constant. Increasing the voltage increases exponentially radiation dose, while lower voltage means better contrast. Rarely is changing common value of 120 kV. Some reviews may use different voltages, but rarely outside the range of 80 to 140 kV (41). The effect of stress on the pipe image quality is more complex, because it affects the image noise and tissue contrast. Further research on the application of a lower tube voltage in terms of dosage strengths, because the complex relationship between tissue contrast,

Rotation time (s)

Modern scanners have gantry rotation time of 0.4 seconds in the field. The main consequence of reducing the time of rotation of the increase of forests and reduction of the absorbed dose. To avoid noise, usually increases the strength of the current tube (8).

The total length of the scanned region (cm)

The absorbed dose is higher if most of the patients exposed to radiation / scan. For this reason, it is necessary to clearly limit the length of the scanned region which is diagnostically relevant to the patient, as it will otherwise appear unnecessarily increasing the dose (overscan).

Sectional thickness (mm)

When MSCT scanner width of each individual detector element in the longitudinal direction is determined by the minimum thickness of the intersection, a merging of multiple neighboring detector elements during detection can increase the thickness of the section. This significantly affects the image quality, as Thin sections have better spatial resolution compared with thicker sections, or have a lower signal-to-noise ratio. In order to reduce the signal-to-noise ratio, it is necessary to increase such. the amount of current which significantly increases the dose of the patient (42).

Pic factor

If we increase the value of the PIC factor, the greater the rate table for a given collimation, and the shorter the exposure of the patient, and therefore a lower dose of radiation (especially if the other scan parameters, including the amount of current was constant). Although scanning with high value Pic factors generally

Page 27 of 126

Darko Hadnađev Šimonji doctoral thesis

more effective dose, it tends to cause a spiral artifacts, degradation of sensitivity profile section and reducing the spatial resolution (39).

In order to reduce the dose were developed many technical innovations, some of which are still in the experimental stage. Some of these innovations are prepacijentna air beam collimation, the use of better filters and algorithms of processing images, automatic tube current modulation strength, efficient configuration of detectors (39).

Application of bismuth protect the region radiosenzitivnih organs, such as breast, eye lens and gonads, is of particular importance in pediatric patients and adults younger patients. There is data from various studies that prove that the radiation dose can be reduced in the region of the breast, thyroid, and, on average, from 45% to 76% bismuth by using the protection (39) (See Figure 13).

A B

Figure 13. A. Photograph boys with bismuth protection for the eyes with a B. Scan bismuth protection in the same region. Successful reduction of the doses of 26.4%. (43)

Role in the reduction of patient dose are largely qualified medical professionals who should determine the need for CT scans as indicated, a justification of the review is the responsibility of the radiologist, whose duty is to eliminate inappropriate CT examinations (39). It should be considered and the application of procedures which do not use ionizing radiation, such as ultrasound and magnetic resonance (MR), if appropriate clinical indications for the preparation of the same or greater diagnostic information (7).

Page 28 of 126

Darko Hadnađev Šimonji doctoral thesis

1.13. Image quality in CT

Radiological picture is a representation of the spatial variation of some physical size: the flux of X-rays, optical density (radiography) or levels of gray scale (image on the monitor). At the same time, it is a representation of the spatial distribution of tissue from the patient. Visualization of important details required separation “structure of interest” in the background. This difference between the major structure and the background is called the signal (44,45).

Monochrome (black and white) images the spatial distribution of the intensity of certain shades of gray (for minimum intensity is taken white and black color for maximum). In a mathematical sense the image is a continuous function of two parameters: the position of the plane of points in the image and intensity of the gray color of that point. Computers do not work with continuous values, and it is necessary to carry out the digitization of images on the spot, whose position is described with two values, and the intensity of gray. Digitizing the image obtained “two-dimensional network” (image matrix), which is the basic element square pixel (v.raniji text image reconstruction in CT). Each pixel is described in the position of the matrix of the digital image and containing a predetermined amount of intensity shades of gray, so that the matrix contains all the information about the image.

The human eye can distinguish shades of gray if they differ by more than 2%. For digital images commonly used 256 shades of gray, because then 1 pixel occupies 8 bits or 1 byte of memory (19).

Figure 14. Example of discretization of gray scales: a) 16, b) 8, c) 4 d) 2 shades of gray (46)

Page 29 of 126

Darko Hadnađev Šimonji doctoral thesis

The quality of digital images represents credibility presentation of the subject. It can be defined by physical parameters and visual images (inter- observer, usually radiologists). The physical parameters of the image can be calculated or measured for the observed image, and that usually include spatial resolution, contrast and noise. Spatial resolution refers to the sharpness or the ability to display the details in the picture. The contrast difference in the intensity (density) between the surfaces (sections) in the image. Noise is the result of random fluctuations in the detection and radiation or broadcasting. the result of a finite number of photons absorbed by the receiver image, and a negative effect on the detection of low-contrast structures. These three parameters can not be considered separately, because improving one of them usually means a worsening of other parameters (4,19,47).

In CT scanning, image quality has many components, and it is influenced by a number of technical parameters. Clinically acceptable image quality has become more a question of strategy for the reduction of radiation dose, particularly in pediatric patients. Several components of the quality of the CT images are to noise ratio, cross-section thickness (Z-axis resolution), the low-resolution and high contrast, as well as the dose of radiation (48).

image noise

Most common definition, is measured as the standard deviation of voxel values in a homogeneous (usually aqueous) to the phantom. There is a more complex definition of image noise that takes into account the contrasting scale scanner. Factors affecting the image noise are: kVp, mA, exposure time, collimation / thickness reconstructional section, reconstruction algorithm or filter, Helical PIC / speed table, helical interpolation algorithm, and others (48). Image noise is an important indicator of the quality of CT images, because it covers the diagnostic information. For homogeneous structures and tissues can be recognized by a fluctuation in the optical density or brightness. The most significant contribution of the diagnostic system of the forest is attributed to variations in signal due to a random fluctuation in the number of photons which form the signal (quantum noise). Quantum noise is due to the stochastic nature of the following processes:

the formation of X-radiation in X-ray tubes; (B) the interaction of X-radiation with the subject; (C) interaction of missed photons of a detector system (4).

By reducing the current intensity and the voltage of the barrel decreases the radiation dose, or increasing the image noise, which can lead to degradation of image quality (See Figure 15). However, increasing the dose of radiation above a certain level, we can further improve image quality.

Page 30 of 126

Darko Hadnađev Šimonji doctoral thesis

Figure 15. Low dose display acute appendicitis (arrow), with elements of increased noise Pictures. Augmented with tape appendix infiltrates periapendikularnog adipose tissue. Hair

Reformation of 3 mm. The acquisition of the MDCT (4×2.5 mm, 120 kV) and 30 mAs eff, AEC and without natively. (40)

The presence of artifacts in the image, ie, the occurrence of some forms of objects in the image that do not exist in the picture, is also a parameter of the quality of digital images. It is important to know which blows that could be avoided or identified, otherwise it can make it difficult to set accurate diagnosis for the patient (19).

In the image quality affects the patient habitus. For obese patients, the dose is increased, but the larger picture noise, while other patients administered dose ALARA principle (49).

sectional thickness

Factors which affect the width of the reconstructed cross-section in a helical scan are: collimation of the beam of X-rays (especially singlslajsnih CT scanners), the width of the detector (in particular for multidetector CT scanners), the helical Pic / speed table, algorithm helical interpolation. For some producers multidetector scanner reconstructed thickness cross section is independent of the speed of the table, because they use interpolation algorithm (48).

High-contrast resolution (spatial resolution)

The resolution, high contrast and spatial resolution is often determined using objects that have a high signal to noise ratio. On the spatial resolution is affected by several factors:

Page 31 of 126

Darko Hadnađev Šimonji doctoral thesis

The limits of the geometric resolution of the system-size focal spot (focal spot size), the width of detector, the air-sampling (sampling ray)

pixel

The properties of the convolution kernel / filter mathematical reconstruction

These filters increase the reconstructive or stored visokoprostorne frequencies (smaller object) at the expense of increased noise in the image (48).

Low contrast resolution

System of low-contrast resolution is often determined by using the facilities that are slightly different from the background (typically a difference of 4 to 10 HU). In this case, where the signal is so small (the difference between the subject and the background) noise is a significant factor. Although there are several purely quantitative methods, the most accepted method is still one that is based on a subjective detection of the object by the observer (48).

The dose of radiation

See previous chapter Factors affecting the dose in CT

Compromises between radiation dose and image quality

Some examples of compromise:

Reduction mAs: the radiation dose decreases in proportion to the reduction of the value of mAs, but it increases the image noise in a proportion of √ () / ().

It follows that if the value of mAs is reduced to half, is expected to increase forest of 1.41 (41%), which degrades the low contrast resolution.

Increasing the speed of the table or the PIC: reducing the dose of radiation in proportion to an increase in the value of Pic (unless the scanner does not automatically increase the value of mAs with the increase of the PIC, as is the case for multidetector scanner or Philips, Siemens). However, it can increase sensitivity by creating a cross-section larger effective thickness section and reducing the resolution of the Z-axis.

Reducing the kVp: reduces the radiation dose (non-linear); can increase the contrast of signals for some of the tissue and iodine (high Z); due to the increased photoelectric effect, may significantly increase the number of artefacts if the beam hardening energy beam becomes too low (e.g. 80kVp).

Page 32 of 126

Darko Hadnađev Šimonji doctoral thesis

Indirect effects: when using thinner sections or filters visokoprostorne frequency, and other factors are held constant, increases the noise. To compensate for increasing noise images the user can increase the value of mAs, which will consequently increase the dose of radiation (48).

Image quality can be quantified through the signal-to-noise ratio (SNR). SNR enables flexible optimization of the process of forming the diagnostic image in terms of its quality and dose values (4). An important and often used measure of quality assessment of digital images is the contrast ratio – noise ratio (CNR – contrast to noise ratio). Due to a possible later processing of digital images, CNR is more indicative measure of the description of the digital image contrast. Another advantage of the introduction of the concept of CNR is relatively simple possibility of measurement using the ROI (Region Of Interest) method, which is available on all CT devices and a large number of computer applications to display and analyze DICOM image formats (19).

The impact of the clinical task to determine the acceptable level of image quality

To reduce the radiation dose should be considered the effects of image quality and the ability to fulfill the task of clinical imaging.

Various clinical tasks have different requirements for image quality. Here are a few examples:

A higher SNR for detection of solid nodules in the lung (except the “ground glass” nodes), the detection of calcifications in the coronary arteries, the identification of pulmonary emphysema.

Low SNR: abdominal sections (detection of lesions in the liver and kidneys), diffuse pulmonary disease

Medium SNR: brain, abdomen / lungs (in the pediatric population) -isključiti bleeding, diseases of the lungs (48)

All of the above physical parameters are defined in the protocol and can be changed to better suit the individual patient. Protocols for MDCT views can be changed directly and the application of the automatic exposure control (AEC) (8,50).

Automatic Exposure Control (AEC)

Automatic exposure control (AEC) is an important component of modern diagnostic systems. The purpose of the AEC is to form diagnostic images constant mean value of the optical density or brightness, independent of the anatomical structures of the patient and exposure of selected factors.

Views of different producers and their AEC are given in Table 5.

Page 33 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 5. Relevant manufacturers and their types of AEC (51)

manufacturer

AEC

ravan-section

The longitudinal plane

combined

smart Scan

Auto mA

smart mA

Philips

DOM

Z-DOM

Siemens

Care Dose

CareDose 4D

Toshiba

Real EC

Sure Exposure

Most of the multi-layer CT system, that have been recently produced, containing the function CT-AEC in a standard configuration (52).

The main objective of the AEC-CT system is to significantly reduce the dose or to eliminate variation in image quality between the different images. In this way also reduces the variation in the radiation dose at the intersection of patients of various sizes (dimensions). In the current system, this is achieved volume control current through the X-ray tube in order to achieve the required level of image noise (53).

CT-AEC system works on three levels:

AEC in relation to the size of the patient

AEC system adjusts the tube current, depending on the overall size (dimension) of the patient. Same mA is maintained throughout the examination or series sections.

AEC in relation to the Z-axis

The strength of tube current is adjusted for each rotation of the tube, taking into account variations in the attenuation of the patient along the Z-axis. The aim is to reduce the variation in image quality obtained images of the same series. This is particularly useful for the review of the chest and abdomen. On examination of the chest mA value is relatively low due to less attenuation due to the presence of air in the lungs, and in the abdomen, which is denser, the greater the value of mA.

Rotary AEC

Current consumption decreases and increases rapidly (modulates) during each rotation to compensate for differences in attenuation between the lateral (left-right) and AP (anterior-posterior) projection. Generally, lateral projections provide more attenuation than the AP, especially in asymmetric regions of the body, such as hips and shoulders.

Page 34 of 126

Darko Hadnađev Šimonji doctoral thesis

Amplitude modulation mA while rotating AEC reflects asymmetry to patient, with the lower modulation in circular regions of the patient’s body, such as the head. Greater modulation occurs in asymmetric regions. Through phantoms shoulder optimal modulation of AP projection is less than 10% of mA used for the lateral projection (53).

Rotary AEC requires X-ray tube and generator whose “output” can be rapidly and precisely vary, especially in subsekundne rotation that is now common use of multislice scanner.

AEC spinning affects the noise on the image in various ways in relation to the previously mentioned two types of the AEC system. Noise in each CT image is a function of the measurement unreliability all which contribute to each pixel. Measurements of attenuation with the greatest effect on image noise are the ones with the greatest uncertainty, these are the ones where the least number of X-rays due to the detector. AEC spinning working to reduce variations in measurement unreliability attenuation by increasing the amount of current through the corners of the projections having the highest attenuation, and reduces the value of the attenuation mA where available. The effect on the image is to equalize variations observed along the field (field of view). This also reduces the presence of artifacts through asymmetrical body region (53).

Only three types of AEC system is shown in Figure 16.

. Figure 16. Three-level automatic control: a) AEC in relation to the size of the patient, higher mA value is used for the larger size of the patient b) AEC in relation to the Z-axis: the higher the value mA is used along the position of Z-axis which attenuates more c ) rotating AEC: degree of modulation depends on asymmetry at each position of Z-axis d) applying the combined effect of the three levels of AEC (53).

Page 35 of 126

Darko Hadnađev Šimonji doctoral thesis

In practice, two or all three levels of AEC system can be combined to obtain the optimal exposure at any point during the scan acquisition.

AEC system introduction into clinical practice should be careful with. Scanners should have a “default” AEC protocols to achieve a compromise between the required image quality and the radiation dose. Today’s CT-AEC systems control the amount of current, and there is potential for the future development of the ability to control the tube voltage and time of rotation (53).

Although the use of CT-AEC generally pretty simple, there are significant differences between these systems. For example. tube voltage change will not affect the amount of current in the Siemens CARE Dose systems, but will with other AEC system. Similarly, changes reconstructional kernel will change the amount of current used by the system Toshiba SureExposure, but not in other systems. Manufacturers should educate users on how best to use this system.

The application of CT-AEC system may allow better communication between imaging centers practice, especially when it comes to scanners and more manufacturers. This would lead to greater uniformity of protocols and consistently good quality of diagnostic images obtained at the optimal dose of radiation (53).

Iterative reconstruction (IR)

The standard method used for the reconstruction of the CT image and the noise reduction is filtered back projection (FBP), which has an analytical approach to the reconstruction of the image. The advantage of this method is that it is less demanding mathematically compared to iterative methods. Image reconstruction with FBP is fast, which is essential for clinical efficacy, however, is not suitable for use in niskodoznih protocol where data are limited (54,55).

IR process is the basis for reducing the dose, and preliminary data on its application are positive. Manufacturers claim that it reduces the dose by 80%, but independent studies have shown a significant reduction in the range of from 36 to 65% (23,24).

Since 2009, manufacturers have adopted a number of types of algorithms for use in IR CT: iterative reconstruction in image space (Iterative Reconstruction in Image Space IRIS; Siemens), the iterative adaptive reduction of the dosage (Adaptive Iterative Dose Reduction Aidra; Toshiba, Tochigi, Japan), adaptive statistical IR (Adaptive Statistical iterative reconstruction ASIR; GE Healthcare), and affirmative sinogram IR (Sinogram affirmed iterative reconstruction SAFIRE-iterative reconstruction for the specific scanner Siemens) and idos (Philips) (50,55). common

Page 36 of 126

Darko Hadnađev Šimonji doctoral thesis

applications for each of these algorithms IR is a predictor model of statistical noise, which supports the propagation of noise in the image domain. These algorithms allow a dose reduction of 29 to 66% at an abdominal CT multidetektorskog are acceptable as standard IR algorithms in clinical practice. It is now used most complete iterative algorithm-mbira (model-based iterative reconstruction), which uses a complex system of predictive models, which includes modeling of optical factors (55).

Mbira incorporates a physical model of CT systems in the reconstruction process, and perform characterization of the process of data acquisition, including noise, beam hardening and scattering. This method leads to dramatic improvements in image quality, especially in niskodoznih scans. Its use results in less image noise of up to 47% compared to ASIR-alkyl and 58% in comparison to FBP. Although the Mbiri reliable and safe method that assists in reducing the dose in CT, the time for reconstruction is longer by 15 to 30 minutes per cross-section, and can be applied in imaging non urgent cases (55).

Figure 17 shows a comparative display image quality with three reconstructive methods: FBP and two types of IR (Assyrian, mbira).

Figure 17. CT images show a subjective image quality and artifacts. Below is a picture quality with three Reconstructive methods: A) ASIR, B) and C FBP) Mbiri. Of note is the subtle stepped artifacts bone contour in the pelvic ring C) (55)

Practical assessment of the quality of diagnostic images

Rating of the quality of diagnostic images is the parallel process of the measurement of patient doses (56) and can be carried out using the following methods: a subjective – on the basis of defined criteria, by using the test object and the phantom, and objectively ‘ROC analysis.

Page 37 of 126

Darko Hadnađev Šimonji doctoral thesis

The observer interprets subjective image in a way that depends on its ability to extract individual structures important for making the diagnosis. The image quality can be examined analytically, via the signal-to-noise for different spatial frequencies, which determines the size of the detail that can be distinguished clearly in the diagnostic image (4).

In order to find the correct balance between image quality and dose, it is necessary to define techniques for evaluating the quality of images that reflect the characteristics of good clinical practice (v. The influence of pre-clinical task to determine an acceptable level of image quality). The aim of optimizing the image quality that successfully meets the clinical task with minimum exposure of patients. This means that before any assessment of image quality requirements should be defined in terms of whether the image contains the information to the viewer needed regardless of its visual impact (4).

Based on the results of the evaluation of a large number of recordings and multidisciplinary approach, defines the criteria of eligibility diagnostic images that meet two fundamental requirements:

satisfactory and consistent quality diagnostic images;

reasonably low value Patient doses

Diagnostic requirements are often subject to personal affinity radiologists, available equipment and current clinical situation. In general, can be divided into criteria that define anatomical details, whose visualization is necessary for making a diagnosis, and the criteria that define the minimum dimensions of elements in the picture that should be visible. The manner in which these criteria are defined by a visual assessment, which is a prerequisite for their extensive practical implementation. Determination of eligibility criteria, diagnostic imaging is a reliable method for qualitative assessment of the characteristics of the diagnostic system. For different types of radiological examinations in conventional and pediatric radiology and CT, acceptance criteria and diagnostic images are systematized in the publication EUR 6260 EN, EUR 16261 EN 16262 and EN EUR (57, 58,59). The criteria relating to the characteristics of the anatomical details of the three-stage system of visualization:

details are just visible;

details are displayed;

details are clearly visible.

Some of these criteria are dependent on positioning, while others reflect the technical characteristics of the diagnostic system. Patient doses criterion refers to the concept

Page 38 of 126

Darko Hadnađev Šimonji doctoral thesis

diagnostic reference levels. Example of good radiographic techniques contains a set of parameters to ensure compliance with the criteria of the image quality and dose values (4).

Page 39 of 126

Darko Hadnađev Šimonji doctoral thesis

2. OBJECTIVE AND WORKING HYPOTHESIS

2.1. The goal

To determine the optimal protocol for CT scan to diagnose the disease regions.

Determination of the dose and radiation risk to patients before and after optimization practices.

2.2. The working hypothesis

Using standard protocols without adhering to adapt to an individual patient parameters is achieved by a better picture quality than is necessary for reliable diagnosis, and therefore the higher dosage of radiation than is necessary.

Optimal choice of protocols in terms of exposure parameters, reducing the voltage or current, in some views it is possible to significantly reduce the dose, and to preserve the image quality required for an adequate interpretation of radiological images.

Page 40 of 126

Darko Hadnađev Šimonji doctoral thesis

3. METHODOLOGY

respondents

The study was conducted at the Center of Radiology, Clinical Center of Vojvodina in Novi Sad, in the period from January 2011 to March 2015. The study, which is divided into 2 stages (stage without optimization and optimization phase of hits) included adults on, of both sexes, who were referred for recording a multi-layer body region computed tomography (MDCT).

The requirement to perform the study were informed of the respondents, and their signed informed consent. Factors for excluding patients from the study were patients who for a certain type of examination did not succumb to the standard protocol dose of radiation (the dynamic recording mode), as well as patients who were unable to give their consent to this test (comatose, delirious-a lack of data on body weight and height) . For each type of CT examinations have been pre-defined minimum requirements in terms of image quality.

Views

Depending on which region is recorded, the participants were divided into 5 groups: group I – CT head (Head routine), a group II-CT with contrast of the head (Head routine contrast iv), Group III – thoracic CT scan (Thorax routine) , IV group CT abdomen and pelvis (abdomen routine multiphase), a group of V-CT angiography (abdominal aorta, of the lower extremities).

X-ray apparatus for CT

MDCT recording was done in the two devices makes “CT SOMATOM SENSATION SIEMENS 64” (64 detector rows) of the first four groups of subjects and CT SOMATOM Emotion SIEMENS 16 “(16 detector rows) for the fifth group of subjects.

Table 6. Characteristics of CT devices

Number

years

automatic

manufacturer

model

detector

control

installations

rows

exposure

Siemens

Somatom

2007.

YES

Emotion 16

Siemens

Somatom

2006.

YES

sensation

Page 41 of 126

Darko Hadnađev Šimonji doctoral thesis

Protocol (optimized / non-optimized)

In the phase I trials using standard protocols for the review of a particular type of review, after which the calculated dose to the patient based on the value of CTDI and DLP image quality and relevant quality criteria for each type of inspection. In the second phase testing subjects within each group were examined using an optimized and individualized protocol, and parallel to this is estimated dose and image quality by using the methodology as in the first phase. All subjects were measured values of body weight and height to be able to calculate body mass index

(BMI- body mass index) of the formula its definition: = (2) (). The values for BMI

each patient were necessary to be able to properly execute the protocol individualization and optimization of inspection.

Table 7. Tabulation standard protocol for screening the head, chest and abdomen pelvis

Head

Header with iv

Chest

The abdomen + pelvis

contrast

parameter

Siemens / Somatom

sensation

scout

lat

mod

gantry angle

collimation

64×0,6

64×1,2 (0.6)

pitch

0.8

1.4

1.2 / 1.4 / 1

U [kV]

120

Have]*

380

trot [S]

0.5

* – mean rounded to a whole number

Tabla 8. Tabulation optimized protocol for screening the head, thorax and abdomen

pelvis.

Head

Header with iv

thorax

The abdomen + pelvis

contrast

parameter

Siemens / Somatom

sensation

scout

lat

mod

gantry angle

collimation

64×0,6

64×1,2 (0.6)

pitch

0.8

1.4

1.2

U [kV]

120

Have]*

340

335

trot [S]

0.5

* – mean rounded to a whole number

Page 42 of 126

Darko Hadnađev Šimonji doctoral thesis

The dose for CT

The data were obtained by recording the value of the display console CT apparatus. CTDIoxand DLP, as previously noted, are well accepted dosage descriptors in CT (1,2,4,5). Data acquisition was performed in line with the recommendations in the Report of the Institute of Physics, 88 & Engineering in Medicine (Institute of Physics and Engineering in Medicine – IPEM Report 88) (17). Data were collected in the form of the value of CTDIoxand GLP in said group of patients and the type of view. Dosage levels are calculated as the mean value from at least 10 patients by type of inspection, or the median of the CTDIoxDLP and the optimized group. Exposure parameters (tube voltage, tube current potency values are expressed in percentages by weight, the time of rotation) are also recorded for each patient. MAs value for each patient was calculated as the average of the region viewed if the system of automatic exposure control was included.

image quality

Before any assessment of image quality requirements are defined in terms of whether the image contains the necessary information to the observer, regardless of its visual impact.

Categorization quality CT images in our study for all four groups of examinations (CT head natively, CT head with contrast chest CT scan and CT of the abdomen to the pelvis) is determined on the basis of valid publications EUR 16262 EN ‘European guides to criteria of image quality CT “, which clearly defined the parameters for assessing the picture quality depending on the region of the body that was shot. For each region of the body which is recorded, there were two groups of parameters: the visualization of anatomical structures and critical reproduction of anatomical structures (94,95,96).

After the optimization phase II trials, is given practical / final evaluation of the quality of diagnostic images, which is actually a parallel process of measuring the patient dose. Used is subjective method wherein two experienced radiologists performed interpretation image, which is depended on their individual ability to extract significant structure for making a diagnosis.

The system of the three-point rating scale for each parameter: 1-details are visible, 2-details display, 3- details are clearly displayed. Based on the sum of all parameters, determined by the final score of image quality, which is also represented by the aforementioned three-stage scale. Then, for the purpose of calculating the size of FOM (figure of merit) calculated index value assessment of image quality (the sum of all evaluation parameters / number of parameters). FOM value for each patient is calculated according to the formula (which is given below) the image quality index estimates / effective dose per patient.

Page 43 of 126

Darko Hadnađev Šimonji doctoral thesis

The average value of the FOM for each group of respondents has served us as a relative indicator for comparison between non-optimized and optimized groups of patients for the same type of review. In the non-optimized groups received the maximum score value of all the parameters, and in the optimized values for different groups of the same parameters vary sporadically.

= Σ ,

where: – index evaluation of image quality; n – the number of parameters; – an index of the final evaluation of image quality; dose – dose per patient.

Diagnostic tasks (methods):

The brain

The criteria for image quality (according to the European guide for quality criteria in CT):

visualization

Cerebrum, cerebellum, the base of the skull, blood vessel (after the administration of contrast iv)

critical reproductions

The sharp boundary between white and gray brain mass

Clearly shows the basal ganglia, a chamber system, the cerebrospinal fluid space around the midbrain, the cerebrospinal fluid space extracerebral, large blood vessels and choroid plexus after the administration of contrast iv

Chest

visualization: Of the thoracic wall, and the thoracic aorta v.cava-e, the heart, the lung parenchyma, vascular disease (after the administration of contrast iv).

Critical Play: Clearly shown in the thoracic aorta, structure anterior mediastinum, including the residue of the thymus (if present), the trachea and main bronchi, paratracheal tissue and Karin and

Page 44 of 126

Darko Hadnađev Šimonji doctoral thesis

region of lymph nodes, esophagus, pleuromedijastinalna limit, and a large mean pulmonary blood vessels, segmental bronchi, pulmonary parenchyma, the boundary between the pleura and the thoracic wall.

abdomen

visualization: Diaphragm, liver and spleen, retroperitoneal parehnimatozni organs (pancreas, kidney), the abdominal aorta and the proximal parts of common iliac arteries, the abdominal wall (including all herniation), blood vessels (after application thereof iv contrast).

Critical Play: Shown clearly in the parenchyma of the liver and intrahepatic vessels, parehnim spleen, intestine, perivascular, retroperitoneal space, the contours of the pancreas, duodenum, kidney, and proximal portions of the urethra, aorta, a bifurcation and the proximal parts of common iliac arteries, lymph nodes diameter of less than 15 mm, the aortic branch , vena cava and its tributaries (especially the renal vein).

pelvis

visualization: Of the iliac bone, ischial bone, the pubic symphysis, bladder, all peripelvičnih muscle, blood vessels (after the administration of contrast iv).

critical reproductions: A clear demonstration of the walls of the bladder, of the distal portion of the urethra, the rectum, perirektalnog distance, uterus, parametrijalnog tissues or seminal vesicle, prostate gland.

statistical analysis

Data were processed by the appropriate mathematical-statistical methods. Analysis was carried out in three steps as follows: testing of working hypotheses as to whether there is a difference or similarity between the subjects unoptimized and optimized protocol review with respect to image quality, or whether optimization protocol has a negative effect on the image quality, whether using neoptimizovanog Protocol review gets much higher doses of radiation received by the patient, but not better picture quality than those optimized protocol and the statistical significance of these differences. Also, certain measures laid down differences with defining their characteristics and graphic representation.

Page 45 of 126

Darko Hadnađev Šimonji doctoral thesis

In this paper, the descriptive parameters, the mean value, standard deviation, minimum and maximum of all values, and the variation coefficient of the confidence interval, measures the asymmetry Skjunis, measurements spoljoštenosti Kurtozis value and Kolmogorov-Smirnov test.

Procedures were used multivariate discriminant analysis and MANOVA. From univariate methods were applied ANOVA t-test and Roy’s test.

To avoid losing information, finding the finest links and information on nonparametric sizes, carried the scaling of data in tables of contingency. This procedure is based on the frequency, each class is associated with a real number. Based on the above, it is evident that the scaled data possible application of multivariate analysis of variance (MANOVA), discriminant analysis and other parametric procedures and methods. Of univariate method was Roy’s test, Pearson correlation of contingency coefficient (), Multiple correlation coefficient (R).

Calculating the quotient of discrimination stand out characteristics that determine the specificity of the two examined groups, which should be excluded from further processing. Also, see the evaluation of the homogeneity of the group of respondents defined characteristics, the distance between them and cluster analysis, aims to better insight into the characteristics of each group of respondents. Statistical analysis was performed as part of the software package agency “Smartline”.

Graphic displays

The most significant results are shown graphically. Use of an ellipse in a graphical representation is of great significance. Size ellipse indicates the homogeneity of the sample. Already ellipse means that the sample is homogeneous. Length of major axis shows the correlation between the two studied parameters. Forms an angle that the major axis of the ellipse with the abscissa represents the direction of the relationship (increasing or decreasing). Center the ellipse is the mean value of the sample in relation to both parameters.

When plotting the qualitative properties of the ellipse showing the frequency modalities. What is the ellipse extends beyond means that there are more modalities. The scale on the axis represents the modalities, which are not evenly spaced on the axis on the same spacing between the two modes, it is possible and their fit. Also, it is possible to not suspend the order on modalities axis if no application procedures that preserve the order. In the case of two or more groups, the center of the ellipse that is closest to the maximum value skalnoj axis is connected with a longer with this value. Along also connects the nearest center of the ellipse with a minimum scaled value on the axis. It demonstrates that subsamples that was presented this ellipse has over other modalities of the frequency with which it is connected. If it is

Page 46 of 126

Darko Hadnađev Šimonji doctoral thesis

subsample presented a line instead of the ellipse, which means that after one topic represented only one modality. In case the point of confusing the ellipse means that the characteristics of both sub-sample is represented with a single modality.

In the case of two or more subsamples, visually indicating the presence of the similarity or difference between them. If the two ellipses coincide difference does not exist, when the ellipse are separated, that is, have no points in common, there is a significant difference between the subsamples for the observed parameters, and when it is an ellipse, partially overlap, must be always concluded only on the basis of the analysis.

Page 47 of 126

Darko Hadnađev Šimonji doctoral thesis

4. RESULTS

This section presents the results of our study. The first chapter refers to the CT examination of the head without contrast, followed by chapters CT examination of the head with a contrast CT scan of the chest, abdominopelvični CT, CT angiography and abdominal aorta and the arteries of the lower extremities.

Results for the subjects in the first four types of CT are designed to determine the existence and significance of statistical differences among the observed groups of subjects (group A and group B) with respect to the measured parameters.

The first section will be shown the protocols before and after optimization of CT, then the mean value of the parameters: anthropometric characteristics of subjects (height, weight and BMI), age of the patients and dosimetry indicator (CTDIoxi DLP) Groups A and B for the given type of CT examinations. Will be determined by the minimum and maximum value of the monitored parameter for both groups, as well as deviations (SD) of the mean value; the existence of differences between the groups A and B for the monitored parameter, and any statistically significant differences. There will also be presented and full distribution by groups.

On statistical analysis, it is envisaged groups determine the homogeneity of each group, measures of asymmetry, determine the distance between the groups according to the conditions. In the end, it will be displayed in image quality before and after optimization of CT examinations in relation to the FOM, significant differences parametric data quality assessment images (summation image quality parameters and the quality of the final score.

4.1. Inspection of the head without contrast

In this type of CT in the present study included a total of 100 subjects: 50 patients in Group A and 50 patients in the group B. Patients in group A are recorded in the first phase of the research by the standard protocol (See Table 9), while the group of group B recorded in the second phase of the study according to the modified protocol (See Table 10). The aim of the modification of the standard protocol was to reduce the value of mAs as possible without loss of diagnostic information for the patient. For this type of examination in our study, the maximum value for which we performed reduction is -40 mAs.

Page 48 of 126

Darko Hadnađev Šimonji

doctoral thesis

Table 9. Protocol without optimization

Table 10. A modified protocol

parameter

value

parameter

value

120

(KV)

I · t

380

I · t

to 340

(MAs)

Time

rotation (s)

mod

Helical

mod

Helical

pitch

0.8

pitch

0.8

factor

collimation

64×0,6

collimation

64×0,6

(Mm)

Analysis of the anthropometric characteristics of subjects in Group A and B after the body height, body weight and body mass index (BMI) is given in Table 11 and 12.

Table 11. Central and dispersion parameters and measures of skewness and kurtosis anthropometric the characteristics of the group A

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

Body

height

1.71

0.09

1.5

1.9

5.24

1.68

1.74

-0.34

-0.93

0,942

(M)

Body

73.1

mass

77.31

13.22

50.0

120.0

17,10

81.43

0.54

1.23

0,965

(Kg)

BMI

26,53

4.25

17.3

38.7

16,01

25.2

27.85

0.31

0.40

0,896

(Kg / m2)

Note: The value of skewness and kurtosis in the range of -0.04 to 0.04 are not discussed

minimum and maximum values of the anthropometric characteristics of the group A indicates that the values fall within the expected range. The values of the coefficient of variation indicates the homogeneity of the characteristics of body height, body weight and BMI. Increased value Skjunisa (sk) indicate that the distribution of negatively skewed, it means that the distribution curve results tends towards higher values, or that there are more large value compared to the normal distribution, with the body weight and BMI. Reduced sk value indicates that the distribution is positively skewed, it means that the distribution curve tends to result manjii values, or to have several smaller value compared to the normal distribution of body height. Larger values Kurtozisa (ku) indicate that the curve

Page 49 of 126

Darko Hadnađev Šimonji doctoral thesis

with the elongated body weight and BMI. Negative values indicate that the house was flattened curve, with the body height. The distribution of values generally remain within a normal distribution (p) at the body height, body weight and BMI.

Table 12. Central and dispersion parameters and measures of asymmetry and kurtosis anthropometric the characteristics of those of group B

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

Body

height

1.69

0.12

1.5

1.9

6.83

1.65

1.74

0.11

-0.73

0,973

(M)

Body

mass

77.63

13,49

52.0

102.0

17,38

72.59

82.67

0.09

-0.93

0,836

(Kg)

BMI

27.26

5.12

18.0

38.3

18.79

25,35

29.18

0.53

-0.56

0,299

(Kg / m2)

minimum and maximum values of the anthropometric characteristics of the group B indicates that the values fall within the expected range. The values of the coefficient of variation indicates the homogeneity of the characteristics of body height, body weight and BMI. Sk increased values indicate that the distribution is negatively skewed at body height, body weight and BMI. Negative values indicate that the house was flattened curve, with the body height, body weight and BMI. The distribution of values generally remain within a normal distribution (p) at the body height, body weight and BMI.

Table 13 shows the significant differences by group of respondents anthropometric characteristics.

Table 13. Significance of differences between the group A and the group B subjects in relation to the anthropometric characteristics

analysis

MANOVA

0,532

0,663

discriminative

0,372

0,690

Based on the values of p = 0.663 (result MANOVA) and p = 0.690 (discriminant analysis), no significant difference is neither a clearly defined border between the two groups of subjects. Even after reduction of the starting whole, or of a 3-in features of the system characteristics from 2 does not exist any difference between groups of the borders.

Page 50 of 126

Darko Hadnađev Šimonji

doctoral thesis

Table 14. Distance (Mahalanobis) between Group A and Group B patients compared to

anthropometric characteristics

Group A

Group B

Group A

0.00

0.21

Group B

0.21

0.00

By calculating the Mahalanobis distance between both groups obtained another

indication of the similarities or differences. Distance different areas can be compared. Distance from Table

14 indicates that the distance between both groups less.

Based on the graphic display ellipses (confidence interval) it is possible to observe the relative positions

and characteristic of each of the two groups, with respect to the most discriminating characteristics of the two

anthropometric characteristics, as follows: BMI and body weight (See Figure 4).

BMI_

Figure 4. Ellipses (confidence interval) in both groups of subjects by the BMI and the body weight Legend: BMI (BMI); body weight (k)

In Chart 4 the axis of abscissas is the BMI, and the ordinate is the body weight (k). The minimum value of BMI and body weight observed in group A, and the maximum value of the BMI and body weight in group B.

Page 51 of 126

Darko Hadnađev Šimonji doctoral thesis

Comparison age of patients in comparison to the group

The table 15 and 16 are shown in the mean value of age, the maximum and minimum values, SD, coefficient of variation, a confidence interval, dkjunis, kurtozis and deviation from the normal distribution of values.

Table 15. Central and dispersion parameters and measures of skewness and kurtosis of the group A

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

Age

respondents

55.90

15,61

24.0

86.0

27.92

51,04

60,77

-0.21

-0.61

0998

(years)

Note: The value of skewness and kurtosis in the range of -0.04 to 0.04 are not discussed

Higher coefficients of variation indicate the heterogeneity of this group. Reduced sk value indicates that the distribution is positively skewed, it means that the distribution curve tends to result manjii values, or to have several smaller value compared to the normal distribution. Negative values indicate that the house was flattened curve. The distribution of values remain within the normal distribution (p = 0.998).

Age of respondents (years)

Table 16. Central and dispersion parameters and measures of skewness and kurtosis of respondents optimized group

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

63.27

18.28

19.0

85.0

28,89

56.44

70.09

-0.97

-0.06

0,455

Higher coefficients of variation indicate the heterogeneity of the group B. The reduced value sk indicate that the distribution is positively skewed. Negative values indicate that the house was flattened curve. The distribution of values generally remain within a normal distribution (p = 0.46).

Table 17. Significance of differences among groups of patients as compared to age

analysis

MANOVA

3,375

0,070

discriminative

3,375

0,070

Based on the values of p = 0.070 (result MANOVA) and p = 0.070 (discriminant analysis) in the Table 17 it is seen that there is a difference, and is clearly defined boundaries between the groups of patients.

Page 52 of 126

Darko Hadnađev Šimonji

doctoral thesis

Table 18. Distance (Mahalanobis) of the obtained data in relation to age of

Group A

Group B

Group A

0.00

0.44

Group B

0.44

0.00

Distance from Table 18 indicates that the distance between the groups of respondents moderately.

Average value of dose indicators for both groups are shown in Tables 19, 20 and 21.

Table 19. The average value of dose indicator in group A

The mean value ± SD

(Min-max)

CTDIvol 59.4

(MGy)

DLP 1057.4 ± 175.6

(mGy

cm-1) (936-2030)

I · t 380

(MAs)

Table 20. The median dose indicator in the group B, required to reduce by weight

median

reduction

CTDIox

DLP

(MAs)

(MGy)

(MGy cm-1)

55.52

980

-25

56.3

980

-20

57.08

1042

-15

53,96

1012

-35

53.17

1042

-40

Table 21.CTDI value and DLP in group B

The mean value ± SD

(Min-max)

CTDIvol

55.7 ± 1.3

(MGy)

(53.9 to 57.1)

DLP

1009 ± 55.32

(MGy cm-1)

(873-1085)

Page 53 of 126

Darko Hadnađev Šimonji doctoral thesis

Comparing a group of subjects compared to the dose indicators

Table 22. Central and dispersion parameters, measures of asymmetry and kurtosis characteristics dosimetry indicators in group A patients

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

DLP

1,057.36

175.64

936,0

2030.0

16.61

1007.4

1107.2

4.34

19.65

0.00

-1

)

(mGycm

Note: The value of skewness and kurtosis in the range of -0.04 to 0.04 are not discussed

minimum and the maximum value of the group A indicates that the values fall within the expected range. The values of the coefficient of variation (16.61) indicate the homogeneity characteristics of DLP. Increased values sk (4.34) indicate that the distribution of negatively skewed. Larger values ku (19,65) indicate that the elongated curve. The distribution of values departs from the normal distribution (p = 0.00).

Table 23. Central and dispersion parameters, measures of asymmetry and kurtosis characteristics dosimetry indicators in group B patients

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

DLP

1,009.00

55.32

873.0

1085.0

5.53

975.4

1024.5

-0.62

-0.23

0.99

(mGycm-1)

minimum and the maximum value of the group B indicates that the values fall within the expected range. The values of the coefficient of variation (5.53) point to the homogeneity of the characteristics of DLP. Reduced values sk (-0.62) indicates that the distribution is positively skewed. Negative values ku (-0.23) indicating that the curve flattened. The distribution of values ranging substantially within the normal distribution (p = 1.00).

Table 24. Analysis Significance of differences among groups of patients according to dose indicators

analysis

MANOVA

154.819

0,000

discriminative

154.818

0,000

Based on the values of p = 0.000 (result MANOVA) and p = 0.000 (discriminant analysis) in Table 24, it is established that there is a difference and clearly-defined boundary between the groups of patients.

Page 54 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 25. Significance of differences among groups of patients according to dose indicators

analysis

todsk

CTDIox

472.216

0,000

6,805

DLP

1,745

0,191

0,009

Legend: todsk the coefficient of discrimination

As the p <0.1 for parameter CTDIox, It means that there is a significant difference between the groups of patients (p = 0.000) in this parameter. Discrimination coefficient indicates that the difference is greatest for CTDIox(6,805), followed by DLP (0,009). (Table 25)

Table 26. Distance (Mahalanobis) between the groups of patients according to dose indicators

Group A

Group B

Group A

0.00

5.23

Group B

5.23

0.00

Distance from Table 26 indicates that the distance between the group A and group B larger.

1120

doz2

1080

1040

1000 2

dose

960

54 55 56 57 58 59 60

Figure 5. Ellipses (confidence interval) for the groups of patients dosage indicators CTDIox and DLP Legend: CTDIox (Dose), DLP (doz2)

In Chart 5 abscissa is CTDIoxAnd the ordinate the DLP. Group B has the lowest value of CTDIox and DLP, and the group A has a maximum value of CTDIox and DLP.

Page 55 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 27. Assessment of the quality of images in the group A and group B

The number of parameters

The average value

collects parameters

quality assessment

The average value

index reviews

quality

The average value

effective dose

percentage

reduction

effective dose

The average value

FOM

Group A

Group B

7 7

16 15

2 1.91

2.22 2.10

7.5% CTDIvol

0.90 0.91

Table 27 shows all the variables that are taken into consideration for the assessment of image quality

Group A and Group B. It can be seen that it is possible percentage reduction in the effective dose in our study for

about 7.5%.

The analysis of differences between groups of patients relative to indicator FOM

Table 28. Significance of differences among groups of patients relative to FOM

analysis

MANOVA

0,229

0.634

discriminative

0,229

0.634

Based on the values of p = 0.634 (result MANOVA) and p = 0.634 (discriminant analysis) in Table 28, no significant difference is neither a clearly defined boundary between the groups of patients.

Table 29. Distance (Mahalanobis) between the groups of patients relative to FOM

Group A

Group B

Group A

Group B

0.00 0.12

0.12 0.00

Table 29 shows the Mahalanobis distance, which is smaller of the obtained data.

Page 56 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 30. The significance of differences between groups of respondents in relation to the overall assessment of image quality

analysis

MANOVA

0,003

0,997

discriminative

34,258

0,000

Based on the values of p = 0.997 (result MANOVA) and p = 0.000 (discriminant analysis) in Table 30, it can be seen that there is no difference between group participants, and in addition there is no clearly defined boundary between the groups of patients. This fact suggests that there are probably latent characteristics that in conjunction with other characteristics (synthesized) contribute to discrimination groups.

Table 31. Significance of differences among groups of patients according the sum parameter estimation and the final assessment of image quality

analysis



todsk

MANOVA

0,578

0,708

71.538

0,000

2,304

discriminative

0,578

0,708

71.538

0,000

0,516

Legend: todsk the coefficient of discrimination

As the p <0.1, this means that there is a significant difference between the groups of subjects with: the sum of image quality parameters (0.000) and the final assessment of image quality (0.000). Discrimination coefficient indicates that the difference between the groups with the largest sum of image quality parameters (2,304), and then in the final assessment of image quality (0.516). (Table 31)

Table 32. Distance (Mahalanobis) between the groups of patients relative to a quality assessment

Group A

Group B

Pictures

Group A

Group B

0.00

2.01

0.00

Distance from Table 32 indicates that the distance between the groups of patients increased.

Page 57 of 126

Darko Hadnađev Šimonji doctoral thesis

full distribution

Table 33. Number (n) and the percentage (%) of participants in the presence of the pole in relation to the two groups

Group A

Group B

gender male

female gender

26.0

52.0

24.0

48.0

21.0

40.9

29.0

59.1

Inspection shown in Table 33 it can be seen that the group of A a maximum male subjects (26 subjects out of 50 (52.0%). The group B is represented more female terminal (29 out of 50 patients (59.1%)) .

Since p = 0.386 2 – test, one can say that there is no correlation between the groups by gender.

Since the = 0.102 correlation is very low.

Table 34. The significance of differences between groups of respondents with respect to gender

analysis

MANOVA

0,739

0.393

discriminative

0,739

0.393

Based on the values of p = 0.393 (result MANOVA) and p = 0.393 (discriminant analysis) shown in Table 34, no significant difference, there is no clearly defined boundary between the groups of patients.

Coefficient of the study and their characteristics can be logically arranged in a hierarchical structure which, is determined by their contribution, which is defined by the order of the importance of characteristics.

Page 58 of 126

Darko Hadnađev Šimonji

doctoral thesis

Table 35. Contribution feature of the continent research

contribution%

between

72.549

Dose indicator of respondents

the group

21,669

Estimates of the image quality of the respondents

the group

5,782

Age of respondents

the group

0,000

Anthropometric characteristics of subjects

the group

0,000

FOM

the group

0,000

Half of respondents

the group

Table 35 shows that the dosage indications have the greatest contribution to the whole study (72.55%). This means that these characteristics of the group clearly expressed, and the distance between the groups is higher compared to other distance. Followed by contributions of the following: assessment of image quality for both groups of patients (21,669%), the age of the respondents (5,782%). Contribution to the anthropometric characteristics of subjects, FOM and half of the respondents as a whole is 0.000%.

4.2. Viewing head with contrast

In this type of CT in the present study included a total of 100 subjects: 50 patients in Group A and 50 patients in the group B. Patients in group A are recorded in the first phase of the research by the standard protocol (See Table 36), while the group of group B recorded in the second phase of the study according to the modified protocol (See Table 37). The aim of the modification of the standard protocol was to reduce the value of mAs as possible without loss of diagnostic information for the patient. For this type of examination in our study, the maximum value for which we performed reduction is -40 mAs.

Table 36. Protocol without optimization

Table 37. A modified protocol

parameter

value

parameter

value

120

(KV)

I · t

380

I · t

to 342.5

(MAs)

Time

rotation (s)

mod

Helical

mod

Helical

pitch

0.8

pitch

0.8

factor

collimation

64 x0,6

collimation

64 x0,6

(Mm)

Page 59 of 126

Darko Hadnađev Šimonji doctoral thesis

Analysis of the anthropometric characteristics of subjects in Group A and B after the body height, body weight and body mass index (BMI) is given in Table 36 and 37.

Table 38. Central and dispersion parameters and measures of skewness and kurtosis anthropometric characteristics and age of the subjects of group A

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

Age

56,06

17,50

20.0

84.0

31.21

51.09

61.03

-0.87

-0.06

0.33

(years)

Body

height

1.71

0.13

1.5

2.0

7.74

1.67

1.75

-1,62

7.39

0,942

(M)

Body

mass

75.58

15.69

45.0

110.0

20.75

71.12

80.04

0.00

-0.51

0.865

(Kg)

BMI

25.81

4.16

17.3

35.1

16.14

24.63

27,00

0.25

-0.73

0.996

(Kg / m2)

Note: The value of skewness and kurtosis in the range of -0.04 to 0.04 are not discussed

minimum and the maximum value of the age and the anthropometric characteristics of the subjects of group A indicates that the values fall within the expected range. Higher values indicate coefficient of variation of the heterogeneity of different age groups (31,21) and body weight (20,75). The values of the coefficient of variation indicates the homogeneity of the characteristics of body height (7.74) and BMI (16.14). Sk increased values indicate that the distribution is negatively skewed at BMI (0.25). Reduced sk value indicates that the distribution is positively skewed with the characteristics of age (-0.87) and height (-1.62). Sk value indicates that the distribution is skewed with the characteristics of body weight (0.00). Higher values indicate that the house is guilty of an elongated features at body height (70.39). Negative values indicate that the house was flattened curve with the characteristics of age (-0.06), body weight (-0.51), and BMI (-0.73).

Page 60 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 39. Central and dispersion parameters and measures of skewness and kurtosis anthropometric characteristics and age of the subjects of group B

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

Age

63,00

22.0

30.67

53,52

65,48

-0.97

0.57

0.81

(years)

Body

height

1.73

0.10

1.5

1.9

6.74

1.66

1.78

-0.39

-0.86

0.91

(M)

Body

mass

78,00

13.0

52.0

110.0

19,80

69.89

82.31

-0.25

-1.33

1.00

(Kg)

BMI

26.15

3.07

18.6

31.46

15,12

24,64

26.94

-0.61

-0.57

0.99

(Kg / m2)

minimum and the maximum value of the age and the anthropometric characteristics of the subjects of group B indicates that the values fall within the expected range. Higher coefficients of variation indicate the heterogeneity of the groups of characteristics of age (30,67). The values of the coefficient of variation indicates the homogeneity of the characteristics of body height (6.74), body weight (19,80), and BMI (15.12). Reduced values sk indicates that the distribution is positively skewed at Features: Age (-, 97), the body height (-0.39), body weight (-0.25), and BMI (-0.61). Higher values indicate that the house is guilty of an elongated features at the age of (0.57). Negative values indicate that the house was flattened curve characteristics at body height (-0.86), body weight (-1.33) and BMI (-0.57). The distribution of values generally remain within a normal distribution (p) at the age of the features of (0.81), height (0.91),

Table 40. Significance of differences between the groups in relation to the anthropometric characteristics of the respondents

analysis

MANOVA

0,435

0.783

discriminative

0,875

0,422

Based on the values of p = 0.783 (result MANOVA) and p = 0.422 (discriminant analysis) in Table 40, but the significant difference was detected, nor is it clearly defined boundaries between the groups. Even after reduction of the starting whole, or system, of the features in the system 4 of the features 2, there is no difference between the groups and no limits.

Page 61 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 41. Distance (Mahalanobis) between the groups in relation to the anthropometric characteristics of the respondents

Group A

Group B

Group A

Group B

0.00

0.39

0.00

Distance from Table 41 indicates that the distance between the groups less.

BMI_

TDCA

Figure 6. Ellipses (confidence interval) groups of patients according to age and BMI

Legend: A group (1); Group B (2); age (TDCA); BMI (BMI)

The graph abscissa is 6 years old, and the ordinate the BMI. It is noted that the minimum value

age of the parameter in the group A (1), and the highest in the group B (2). In relation to the BMI group B (2) has the

the minimum value, a maximum value of the group A (1).

Average value of dose indicators for both groups are given in Tables 42, 43 and 44.

Page 62 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 42. The average value of dose indicator in group A

The mean value ± SD

(Min-max)

CTDIvol 59.43 ± 0.002

(MGy) (59.42 to 59.44)

DLP 2147.4 ± 267.5

(mGy

cm-1) (1836-2030)

I · t 380

(MAs)

Table 43. The median dose indicator in the group B, required to reduce by weight

median

reduction

CTDIox

DLP

(MAs)

(MGy)

(MGy cm-1)

57.86

2081.5

-20

55.8

1998

-25

57.08

2072

-15

54.74

2009

-30

-35

53.86

2043

-40

Table 44.CTDI value and DLP in group B

The mean value ± SD

(Min-max)

CTDIvol

55.19 ± 1.98

(MGy)

(53.86 to 58.25)

DLP

2069.69 ± 147.8

(MGy cm-1)

(From 1759 to 2764.0)

Comparing a group of subjects compared to the dose ratios are given in Tables 45 and 46.

Table 45. Central and dispersion parameters, measures of asymmetry and kurtosis characteristics dosimetry indicators in group A patients

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

DLP

2,147.40

267.47

1836.0

3302.0

12,45

2,071.37

2,223.43

2.86

9.45

0,003

(mGycm-1)

Note: The value of skewness and kurtosis in the range of -0.04 to 0.04 are not discussed

Page 63 of 126

Darko Hadnađev Šimonji doctoral thesis

minimum and the maximum value of the group A indicates that the values fall within the expected range. The values of the coefficient of variation (12,45) indicate the homogeneity characteristics of DLP. Increased values sk (2.86) indicate that the distribution of negatively skewed. Larger values ku (9.45) indicate that the elongated curve. The distribution of values departs from the normal distribution (p =

0.00).

Table 46. Central and dispersion parameters, measures of asymmetry and kurtosis characteristics dosimetry indicators in group B patients

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

DLP

2,069.69

147.8

1759.0

2764.0

10.81

1,967.74

2,195.33

2.55

7.25

0,065

(mGycm-1)

minimum and the maximum value of the group B indicates that the values fall within the expected range. The values of the coefficient of variation (10.81) indicate the homogeneity characteristics of DLP. Increased values sk (2.55) indicate that the distribution of negatively skewed. Larger values ku (7.25) indicate that the elongated curve. The distribution of values departs from the normal distribution (p = 0.065).

Table 47. Significance of differences among groups of patients according to dose indicators

analysis

MANOVA

105.142

0,000

discriminative

105.167

0,000

Based on the values of p = 0.000 (result MANOVA) and p = 0.000 (discriminant analysis) in Table 47, and there is a difference clearly defined boundaries between the groups of patients.

Table 48. Significance of differences among groups of patients according to dose indicators

analysis

todsk

CTDIox

320.427

0,000

5,069

DLP

0.939

0,336

0,080

Legend: kdsk the coefficient of discrimination

As the p <0.1,, This means that there is a significant difference between the groups of subjects with characteristics CTDIox(0,000). Discrimination coefficient indicates that the difference is greatest in: CTDIox(5,069) followed by the DLP (0,080). (Table 48)

Page 64 of 126

Darko Hadnađev Šimonji

doctoral thesis

Table 491. Features and homogeneity of the group of respondents with respect to the dose indicators

Group A

Group B

dpr

(%)

CTDIox

the evening* 1

less

96,940

DLP

the evening

less

1,530

hMG

100.00

93.75

–

(%)

Legend: HMG – homogeneity; dpr – contributing traits characteristics

The status of each subsample group most defines CTDIoxBecause the contribution features characteristics 96.94%, followed by: DLP (1.53%) and effective dose (1.53%). The homogeneity of the group A is 100.00% and in group B was 93.75%.

Based on the dose indicator can be said that:

Group A value of CTDIox DLP and larger,

in group B values for CTDIox DLP and less.

Table 50. Distance (Mahalanobis) between the groups of patients according to dose indicators

Group A

Group B

Group A

Group B

5.18

0.00

5.18

0.00

Distance from Table 50 indicates that the distance between the groups larger.

Page 65 of 126

Darko Hadnađev Šimonji doctoral thesis

2300

doz2

2200

2100

2000

dose

1900

55 56 57 58 59 60

Figure 7. Ellipses (confidence interval) a group of subjects with characteristics of the CTDIox and DLP Legend: A group (1); Group B (2); CTDIox(The dose); DLP (doz2)

The graph abscissa 7 is CTDIox(Dose), and the ordinate is the DLP (doz2). It may be noted that in group B (2) The minimum value of CTDIox and DLP, and in group A (1) their greatest value.

Page 66 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 51. Assessment of the quality of images in the group A and group B

The number of parameters

The average value

collects parameters

quality assessment

The average value

index reviews

quality

The average value

effective dose

percentage

reduction

effective dose

The average value

FOM

Group A

Group B

1.96

4.51

4.36

7% CTDIvol

0.44 0.45

Table 51 shows all the variables that are taken into consideration for assessing the quality of images in the group A and group B. It can be seen that it is possible percentage reduction in the effective dose in our study of about 4%.

Since p = 0.002 2 – test, one can say that there is a correlation between the groups of characteristics summation of image quality parameters, given that = 0.360 correlation is low.

Table 52. The significance of differences between groups of respondents in relation to the assessment of image quality

analysis

MANOVA

0,003

0,997

discriminative

5,420

0,007

Based on the values of p = 0.997 (result MANOVA) and p = 0.007 (discriminant analysis) from Table 52, it can be seen that there is no difference between group participants, and in addition there is no clearly defined boundary between the groups. This fact suggests that there are probably latent characteristics that in conjunction with other characteristics (synthesized) contribute to discrimination against the group.

Page 67 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 53. The significance of differences between groups of respondents in relation to the assessment of image quality

analysis



todsk

The sum of image quality parameters

0,360

0,386

11,364

0,001

0.616

The final score of image quality

0,360

0,386

11,364

0,001

0,153

Legend: kdsk the coefficient of discrimination

As the p <0.1, this means that there is a significant difference between the sum of characteristics of image quality parameters (0.001) and the final image quality score (0.001). Discrimination coefficient indicates that the difference is greatest in the summation of the characteristics of image quality parameters (0.616) and then with a final assessment of image quality (0.153). (Table 53)

Table 54. Distance (Mahalanobis) between the groups of patients relative to a quality assessment Pictures

Group A

Group B

Group A

Group B

0.96

0.00

0.96

0.00

Distance from Table 54 indicates that the distance between the groups of respondents moderately.

Table 55. Central and dispersion parameters and measures of skewness and kurtosis FOM Group A

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

FOM

0.45

0.04

0.3

0.5

9.72

0.44

0.46

-1.69

4.50

0.00

Note: The value of skewness and kurtosis in the range of -0.04 to 0.04 are not discussed

minimum FOM and maximum values in Table 55 indicate that the values are in the expected range. The values of the coefficient of variation indicates the homogeneity of the characteristics of FOM (9.72). Reduced values sk (-1.69) indicates that the distribution is positively skewed. Larger values ku (4.50) indicate that the elongated curve. Distribution value deviates from the normal distribution (p) FOM (0.00).

Page 68 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 56. Central and dispersion parameters and measures of skewness and kurtosis FOM in group B

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

FOM

0.45

0.04

0.3

0.5

8.67

0.43

0.47

-1.42

2.10

0,954

minimum FOM and maximum values in Table 56 indicate that the values are in the expected range. The values of the coefficient of variation indicates the homogeneity of the characteristics of FOM (8.67). Reduced values sk (-1.42) indicating that the distribution of positive values asimetrična.Veće ku (2.10) indicate that the fault with the elongated FOM. The distribution of values generally remain within a normal distribution (p) FOM (0,95).

Table 57. Significance of differences among groups of patients relative to FOM

analysis

MANOVA

0,018

0,893

discriminative

0,018

0,893

Based on the values of p = 0.893 (result MANOVA) and p = 0.893 (discriminant analysis) from Table 57, no significant difference nor clearly defined boundaries between the groups.

Table 58. Distance (Mahalanobis) between the groups of patients relative to FOM

Group A

Group B

Group A

Group B

0.00

0.04

0.00

Distance from Table 58 indicates that the distance between the groups less.

Table 59. Number (n) and the percentage (%) representation of half the subjects in relation to the group

Group A

Group B

gender male

female gender

24.0

48.0

26.0

52.0

31.0

62.5

19.0

37.5

Inspection shown in Table 59 it can be seen that in group A, the most frequent female half (26 of 50 patients (52.0%)). In Group B is the most represented male half (31 of 50 patients (62.5%)).

Page 69 of 126

Darko Hadnađev Šimonji doctoral thesis

Since p = 0.312 2 – of the test, it can be said that there is no correlation between the sex groups, since it is = 0.123 correlation is very low.

Table 60. The significance of differences between groups of respondents with respect to gender

analysis

MANOVA

1,005

0,320

discriminative

1,005

0,320

Based on the values of p = 0.320 (result MANOVA) and p = 0.320 (discriminant analysis) in the table 60, but the significant difference was detected, nor clearly defined boundaries between the groups.

Table 61. Distance (Mahalanobis) between groups of subjects by sex

Group A

Group B

Group A

Group B

0.00

0.29

0.00

Distance from Table 61 indicates that the distance between the groups less.

Table 62. Contribution feature of the continent research

contribution%

between

84,338

dosage indicator

the group

15,662

Assessment of image quality

the group

0,000

Anthropometric characteristics and age of the respondents

the group

0,000

FOM

the group

0,000

Half

the group

Table 62 reveals that the dosage indications have the greatest contribution to the whole study (84.34%), followed by assessment of image quality (15.67%).

4.3. Review chest

In this type of CT examinations in our study included 114 subjects: 52 patients in group A and 62 patients in group B. Patients in group A were recorded in the first phase of research at

Page 70 of 126

Darko Hadnađev Šimonji doctoral thesis

protocol without optimization (See Table 63), while the group of group B recorded in the second phase of the study according to the modified protocol (See Table 64). The aim of the modification of the standard protocol was to reduce the value of mAs as possible without loss of diagnostic information for the patient. For this type of examination in our study, the maximum value for which we performed reduction is -70 mAs.

Page 71 of 126

Darko Hadnađev Šimonji

doctoral thesis

Table 63. Protocol without optimization

Table 64. A modified protocol

parameter

value

parameter

value

120

(KV)

I · t

44,02

I · t

26.95

(MAs)

Time

0.5

Time

0.5

rotation (s)

mod

Helical

mod

Helical

pitch

1.4

pitch

1.2

factor

collimation

64 x0,6

collimation

X0,6 64 (1.2)

(Mm)

Analysis of the anthropometric characteristics of subjects in Group A and B after the body height, body weight and body mass index (BMI) is given in Table 65 and 66 thereof.

Table 65. Central and dispersion parameters and measures of skewness and kurtosis anthropometric characteristics of the patients in the group A

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

Body

height

1.72

0.17

0.7

2.0

9.67

1.67

1.76

-4.41

25.39

0,068

(M)

Body

mass

77.67

18,62

0.7

118.0

23,97

72.48

82.85

-0.92

4.33

0,576

(Kg)

BMI

25.71

5.93

0.7

42.7

23,06

24,06

27,36

-0.79

5.72

0,655

(Kg / m2)

Note: The value of skewness and kurtosis in the range of -0.04 to 0.04 are not discussed

minimum and maximum values of the anthropometric characteristics of those in Table 65 demonstrate that the values fall within the expected range. Higher values indicate coefficient of variation of the heterogeneity of the groups according to body weight (23,97), and BMI (23,06). The values of the coefficient of variation indicates the homogeneity of the characteristics of body height (9.67). Reduced values sk indicates that the distribution is positively skewed at Features body height (-4.41), body weight (-0.92), and BMI (-0.79). Higher values indicate that the house is guilty of an elongated features at body height (25.39), body weight (4.33) and BMI (5.72). The distribution of values generally remain within a normal distribution (p) in the characteristics of the body weight (0.58), and BMI (0.65). The distribution of values departs from the normal distribution (p) in the characteristics of a body height (0.07).

Page 72 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 66. Central and dispersion parameters and measures of skewness and kurtosis anthropometric characteristics of the patients in the group A

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

Body

height

1.69

0.10

1.5

1.9

5.93

1.67

1.72

0.07

-0.68

0.814

(M)

Body

mass

77.95

16.85

49.0

130.0

21.62

73.67

82.23

0.73

0.52

0.542

(Kg)

BMI

27,18

5.28

18.3

42.5

19.43

25.83

28.52

0.70

0.06

0.518

(Kg / m2)

minimum and maximum values of the anthropometric characteristics of subjects in Table 66 indicate that the values are in the expected range. Higher coefficients of variation suggests heterogeneity optimized groups of body weight (21,62). The values of the coefficient of variation indicates the homogeneity of the characteristics of body height (5.93) and BMI (19.43). Sk increased values indicate that the distribution is negatively skewed with all three characteristics of body height (0.07), body weight (0.73) and BMI (0.70). Higher values indicate that the house is guilty of an elongated features with the weight (0.52) and BMI (0.06). Negative values indicate that the house was flattened curve at body height (-0.68). The distribution of values generally remain within a normal distribution (p) at all the three characteristics: a body height (0.81), weight (0.54) and BMI (0.52).

Table 67. The significance of differences between groups respondents in relation to the anthropometric characteristics

analysis

MANOVA

2,542

0,060

discriminative

2,542

0,060

Based on the values of p = 0.060 (result MANOVA) and p = 0.060 (discriminant analysis) from Table 67, it is seen that there is a difference and clearly-defined boundary between the two groups of subjects.

Table 68. Distance (Mahalanobis) between groups of subjects in relation to the anthropometric

Group A

Group B

characteristics

Group A

Group B

0.00

0.52

0.00

Distance from Table 68 indicates that the distance between the moderate group.

Page 73 of 126

Darko Hadnađev Šimonji doctoral thesis

1.78

vstl

1.76

1.74

1.72

1.7

1.68

1.66

Figure 8. Ellipses (confidence interval) a group of subjects with body weight and body height Legend: A group (1); Group B (2); body weight (k); Body height; (Vstl)

The graph 8 abscissa the weight (k) and the ordinate is the height (vstl). It is possible to

note that in relation to the body weight of the group A (1) has a minimum value, a maximum value of the group

B (2). In relation to the body height, a group B (2) has a minimum value, a maximum value of the group A

(1).

Average values of indicators of dosage for both groups are shown in Tables 69, 70 and 71.

Table 69. The average value of dose indicator in group A

The mean value ± SD

(Min-max)

CTDIvol 6.84 ± 2.08

(MGy) (3.17 to 12.45)

DLP 490.9 ± 213.7

(mGy

cm-1) (144-907)

I · t 44.0 ± 13.8

(MAs) (20.5 to 81.0)

Page 74 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 70. The median dose indicator in the group B, required to reduce by weight

median

reduction

CTDIox

DLP

(MAs)

(MGy)

(MGy cm-1)

5.44

430

-15

4,005

275.5

-20

4.04

295

-25

4.53

353.5

-30

5.69

412

-35

3,505

233.5

-40

3.31

240

-45

3,555

306.5

-50

2,325

192

-60

2.36

113.5

-70

Table 71.CTDI value and DLP in group B

The mean value ± SD

(Min-max)

CTDIvol

3.9 ± 1.1

(MGy)

(2.3 to 5.7)

DLP

285.2 ± 97.3

(MGy cm-1)

(113.5 to 430)

Comparing a group of subjects compared to the dose indicators

Table 72. Central and dispersion parameters and measures of skewness and kurtosis dose indicator patients in group A

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

INTENZA

t current

43.19

14,94

0.7

81.0

34,58

39,03

47,35

0.30

0.70

0.637

(MAs)

CTDIox

6.73

2.24

0.7

12.4

33,25

6.10

7.35

0.36

0.55

0,710

(MGy)

DLP (m

481.46

222,23

0.7

907.0

46,16

419.57

543.34

0.24

-0.73

0,439

Gycm-1)

Note: The value of skewness and kurtosis in the range of -0.04 to 0.04 are not discussed

minimum and the maximum value of the indicators of the respondents dosage in Table 72 demonstrate that the values fall within the expected range. Higher coefficients of variation suggests heterogeneity in the group: the intensity of electricity (34.58), CTDIox(33,25) and DLP (46,16). increased

Page 75 of 126

Darko Hadnađev Šimonji doctoral thesis

sk value indicates that the distribution is negatively skewed with: the intensity of current (0,30), of the CTDIox(0.36) and DLP (0.24). Higher values indicate that ku is elongated curve at the current intensity (0.70), and CTDIox(0.55). Negative values indicate that the house was flattened at fault: DLP (-0.73). The distribution of values generally remain within a normal distribution (p) with: the intensity of current (0,64), of the CTDIox (0.71) and DLP (0.44).

Table 73. Central and dispersion parameters and measures of skewness and kurtosis dose indicator patients in group B

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

intensity

currents

26.95

11.65

8.0

71.0

43.25

23,99

29.91

1.16

2.37

0,579

(MAs)

CTDIox

4.18

1.75

1.3

10.9

41.86

3.73

4.62

1.25

2.59

0.556

(MGy)

DLP (mG

311,90

136.49

63.0

608.0

43,76

277.23

346.57

0.17

-0.75

0,741

YCM-1)

minimum and the maximum value of the indicators of the respondents dosage in Table 73 demonstrate that the values fall within the expected range. Higher coefficients of variation suggests heterogeneity in the group: the intensity of electricity (43.25), CTDIox(41,86) and DLP (43,76). Higher values of sk indicates that the distribution is negatively skewed in the current intensity (1,16), of the CTDIox(1.25) and DLP (0.17). Higher values indicate that ku is elongated curve at the current intensity (2.37) and CTDIox(2.59). Negative values indicate that the house was flattened curve DLP (-0.75). The distribution of values generally remain within a normal distribution (p) at the current intensity (0,58), of the CTDIox (0.56) and DLP (0.74).

Table 74. Significance of differences among groups of patients according to dose indicators

analysis

MANOVA

14,064

0,000

discriminative

13,935

0,000

Based on the values of p = 0.000 (result MANOVA) and p = 0.000 (discriminant analysis) in Table 74, and there is a difference clearly defined boundaries between the groups of patients.

Table 75. Significance of differences among groups of patients relative to dosimetric indicators

analysis

todsk

Airflow

42,499

0,000

0,005

CTDIox

46,579

0,000

0,032

DLP

24,912

0,000

0,086

Legend: k.dsk the coefficient of discrimination

Page 76 of 126

Darko Hadnađev Šimonji doctoral thesis

As the p <0.1,, this means that there is a significant difference between the group of subjects with the current intensity (0,000), CTDIox(0,000) and DLP (0,000). Discrimination coefficient indicates that the difference is greatest in DLP (0,086), followed by CTDIox(0.032) and the current intensity (0,005). (Table 75)

Table 76. Features and homogeneity of the group of respondents with respect to dosimetric parameters

Group A

Group B

dpr

(%)

DLP

the evening* 1

less

41,148

CTDIox

the evening* 1

less

15,311

Airflow

the evening* 1

less

2,392

hMG

69,23

82.26

(%)

Legend: HMG – homogeneity; dpr – contributing traits characteristics

The status of each subsample most defines DLP, because the contribution features characteristics

41.15%, followed by CTDIox(15.1%) and current density (2.39%). Homogeneity in group A was 69.23%, and in group B was 82.26%. (Table 76)

Based on dosimetry indicators, respondents can say that they are:

group A the value of DLP, CTDIox and the intensity of current more.

group B the value of DLP, CTDIox and the intensity of the currents are less.

Table 77. Distance (Mahalanobis) between the groups of patients according to dose indicators

Group A

Group B

Group A

Group B

0.00

1.43

0.00

Distance from Table 77 indicates that the distance between the groups larger.

Page 77 of 126

Darko Hadnađev Šimonji doctoral thesis

dose

doz2

200 300 400 500 600

Figure 9. Ellipses (confidence interval) between the groups and DLP CTDIox

Legend: A group (1); Group B (2); DLP (doz2); CTDIvol (dose)

Figure 9 on the abscissa is the DLP (doz2), and the ordinate is CTDIvol (dose). It is noted that in comparison to DLP and CTDIox Group B (2) has a minimum value, a maximum value of the group A (1).

Table 78. Assessment of the quality of images in the group A and group B

The number of parameters

The average value

collects parameters

quality assessment

The average value

index reviews

quality

The average value

effective dose

percentage

reduction

effective dose

The average value

FOM

Group A

Group B

17 17

41 40

2.41 2.39

6.87 4.37

40% CTDIvol

0.35 0.55

Table 78 shows all the variables that are taken into consideration for the assessment of image quality

in group A and group B. It can be seen that it is possible percentage reduction in the effective dose in our study for

about 40%.

Page 78 of 126

Darko Hadnađev Šimonji

doctoral thesis

Table 79. The significance of differences between groups of respondents in relation to the assessment of image quality

analysis

MANOVA

2,472

0,087

discriminative

2,477

0,089

Based on the values of p = 0.087 (result MANOVA) and p = 0.089 (discriminant analysis) in Table 79, and there is a distinction has been clearly defined boundaries between the groups of patients.

Table 80. The significance of differences between groups of respondents in relation to the assessment of image quality

analysis



todsk

The sum of image quality parameters

0,199

0,203

4,847

0,030

0,026

The final score of image quality

0,136

0,137

2,161

0,144

0,002

Legend: kdsk the coefficient of discrimination

As the p 0.1,This means that no significant differences between groups of respondents celebrate the final image quality score (0.144). Discrimination coefficient indicates that the difference is greatest in the summation of the characteristics of image quality parameters (0.026), and then at the final assessment of image quality (0.002). (Table 80)

Table 81. Distance (Mahalanobis) between the groups of patients relative to a quality assessment

Group A

Group B

Pictures

Group A

Group B

0.00

0.42

0.00

Distance from Table 81 indicates that the distance between the moderate group.

Table 82. Central and dispersion parameters and measures of skewness and kurtosis of respondents according to FOM group A

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

FOM

0.44

0.23

0.2

1.2

52.67

0.38

0.51

1.37

1.58

0,111

Note: The value of skewness and kurtosis in the range of -0.04 to 0.04 are not discussed

minimum FOM and maximum values in Table 82 indicate that the values are in the expected range. Higher coefficients of variation indicate the heterogeneity of the group at FOM (52.67). Sk increased values indicate that the distribution is negatively skewed (1.37). Higher values indicate that the house is guilty of an elongated (1.58). The distribution of values generally remain within a normal distribution (p) FOM (0.11).

Page 79 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 83. Central and dispersion parameters and measures of skewness and kurtosis of respondents according to FOM group B

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

FOM

0.72

0.49

0.3

2.7

68.06

0.59

0.84

2.33

5.94

0,003

minimum and maximum values of respondents by FOM in Table 83 indicate that the values are in the expected range. Higher coefficients of variation suggests heterogeneity with at FOM (68.06). Sk increased values indicate that the distribution is negatively skewed (2.33). Higher values indicate that the house is guilty of an elongated at FOM (5.94). The distribution of values departs from the normal distribution (p) in FOM (0,00).

Table 84. The significance of differences between groups of respondents by FOM

analysis

MANOVA

13,782

0,000

discriminative

13,782

0,000

Based on the values of p = 0.000 (result MANOVA) and p = 0.000 (discriminant analysis) in Table 84, and there is a difference clearly defined boundaries between the groups of patients.

Table 85. Distance (Mahalanobis) between the groups of respondents FOM

Group A

Group B

Group A

0.00

0.70

Group B

0.70

0.00

Distance from Table 85 indicates that the distance between the groups of respondents moderately.

Table 86. Central and dispersion parameters and measures of skewness and kurtosis age respondents group A

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

Age

54.36

14,94

0.7

83.0

27,48

50,20

58.52

-0.95

2.11

0,989

(years)

Note: The value of skewness and kurtosis in the range of -0.04 to 0.04 are not discussed

Higher coefficients of variation indicate the heterogeneity of the groups by age (27,48). Reduced values sk (-0.95) indicates that the distribution is positively skewed. Higher values indicate that the house is guilty of an elongated (2.11). Distribution of values generally remain within the normal distribution (p) (0.99). (Table 86)

Page 80 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 87. Central and dispersion parameters and measures of skewness and kurtosis age respondents group B

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

Age

62.94

10,80

23.0

85.0

17,16

60.19

65.68

-0.66

1.88

0.607

(years)

The values of the coefficient of variation indicates the homogeneity of the characteristics of age (17,16). Reduced values sk (0.66) indicate that the distribution is positively skewed. Higher values indicate that the house is guilty of an elongated (1.88). The distribution of values generally remain within a normal distribution (p) (0.61). (Table 87)

Table 88. The significance of differences between groups of respondents by age

analysis

MANOVA

12,600

0,001

discriminative

12,600

0,001

Based on the values of p = 0.001 (result MANOVA) and p = 0.001 (discriminant analysis) in Table 88, and there is a difference clearly defined boundaries between the groups of patients.

Table 89. Distance (Mahalanobis) of the obtained data in relation to age of

Group A

Group B

Group A

0.00

0.67

Group B

0.67

0.00

Distance from Table 89 indicates that the distance between the moderate group.

Table 90. Number (n) and the percentage (%) of participants in the presence of the pole in relation to the group

gender male

female gender

Group A

30.0

57.7

22.0

42.3

Group B

35.0

56.5

27.0

43.5

Inspection shown in Table 90 it is possible to observe that in group A, the most represented male half (30 of 52 patients (57.7%)). In group B the representation of males 56.5% (35 of 62 patients).

Since p = 0.894 2 – of the test, it can be said that there is no correlation between the sex groups, since it is = 0.012, correlation is very low.

Page 81 of 126

Darko Hadnađev Šimonji

doctoral thesis

Table 91. Significance of differences among groups of subjects by sex

analysis

MANOVA

0,017

0,895

discriminative

0,017

0,895

Based on the values of p = 0.895 (result MANOVA) and p = 0.895 (discriminant analysis) from Table 91, no significant difference nor clearly defined boundaries between the groups of patients.

Table 92. Distance (Mahalanobis) of the obtained data with respect to gender

Group A

Group B

Group A

0.00

0.02

Group B

0.02

0.00

Distance from Table 92 indicates that the distance between the groups less.

Contribution of characteristics and traits characteristic whole experiment are shown in Table 93.

Table 93. Contribution feature of the continent research

contribution%

between

38,242

dosimetry indicators

the group

18,653

FOM

the group

17,825

Age

the group

14,003

anthropometric characteristics

the group

11,277

Assessment of image quality

the group

0,000

half

the group

On the basis of Table 93 reveals that the dosimetric parameters have the greatest contribution to the whole study (38.24%). Then follows: FOM contribution of 18,653%, 17,825% of age, anthropometric characteristics of 14,003%, estimates the image quality 11,277%, while the contribution of half the 0.000%.

4.4. Examination of the abdomen and pelvis

In this type of CT in the present study included a total of 82 subjects: 50 subjects in Group A and 50 patients in the group B. Patients in group A are recorded in the I phase of the study per protocol with no optimization (See Table 94), while the group the group B recorded in the second phase of the study according to the modified protocol (See Table 95). The aim of the modification of the standard protocol was to reduce the value of mAs as possible without loss of diagnostic information for the patient. For

Page 82 of 126

Darko Hadnađev Šimonji doctoral thesis

This type of examination in our study, the maximum value for which we performed reduction is -60 mAs.

Table 94. Protocol without optimization

Table 95. A modified protocol

parameter

value

parameter

value

120

(KV)

I · t

56.29

I · t

42,48

(MAs)

Time

0.5

Time

0.5

rotation (s)

mod

Helical

mod

Helical

pitch factor

1,0; 1,2; 1,4

pitch

1.2

factor

collimation

X0,6 64 (1.2)

collimation

X0,6 64 (1.2)

(Mm)

Analysis of the anthropometric characteristics of subjects in Group A and B after the body height, body weight and body mass index (BMI) is given in Table 96 and 97.

Table 96. Central and dispersion parameters and measures of skewness and kurtosis anthropometric characteristics of the patients in the group A

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

Body

height

1.70

0.10

1.5

1.9

5.66

1.67

1.72

0.35

-0.26

0,078

(M)

Body

mass

73.70

15.69

45.0

111.0

21,28

69.24

78.16

0.32

-0.59

0,155

(Kg)

BMI

25.60

5.13

16.9

39.3

20.02

24,15

27,06

0.44

-0.34

0,710

(Kg / m2)

Note: The value of skewness and kurtosis in the range of -0.04 to 0.04 are not discussed

minimum and maximum values of the anthropometric characteristics of the group A indicates that the values fall within the expected range. Higher coefficients of variation indicate the heterogeneity of the groups of: body weight (21,28) and BMI (20.02). The values of the coefficient of variation indicates the homogeneity of the characteristics of body height (5.66). Sk increased values indicate that the distribution is negatively skewed, at body height (0.35), body weight (0.32) and BMI (0.44). Negative values indicate that the house was flattened curve at body height (-0.26), body weight (-

Page 83 of 126

Darko Hadnađev Šimonji doctoral thesis

0.59) and BMI (-0.34). The distribution of values generally remain within a normal distribution (p) in body weight (0.16) and BMI (0.71). Distribution value deviates from the normal distribution (p) at body height (0.08). (Table 96)

Table 97. Central and dispersion parameters and measures of skewness and kurtosis anthropometric characteristics of the patients in the group B

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

Body

height

1.68

0.11

1.5

1.9

6.55

1.64

1.72

0.46

-0.87

0,502

(M)

Body

mass

71.16

12.88

50.0

100.0

18.10

66,51

75.80

0.42

-0.60

0.789

(Kg)

BMI

25,14

3.67

19.2

34.0

14,60

23.82

26.47

0.29

-0.52

0,980

(Kg / m2)

minimum and maximum values of the anthropometric characteristics of the group B indicates that the values fall within the expected range. The values of the coefficient of variation indicates the homogeneity of the characteristics of body height (6.55), body weight (18.10) and BMI (14.60). Sk increased values indicate that the distribution is negatively skewed at body height (0.46), body weight (0.42) and BMI (0.29). Negative values indicate that the house was flattened curve at body height (-0.87), body weight (-0.60) and BMI (-0.52). The distribution of values generally remain within a normal distribution (p) at the height of the body (0,50), of body weight (0.79) and BMI (0.98). (Table 97)

Table 98. Significance of differences among groups of patients according to the anthropometric characteristics of the

analysis

MANOVA

0,270

0.847

discriminative

0,369

0,693

Based on the values of p = 0.847 (result MANOVA) and p = 0.693 (discriminant analysis) in Table 98, no significant difference is neither clearly defined boundaries between the groups. Even after reduction of the starting continent 3 features in the system of the characteristics of 2, there is no difference or boundary between groups.

Page 84 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 99. Distance (Mahalanobis) between groups of subjects in relation to the anthropometric

Group A

Group B

characteristics

Group A

Group B

0.00

0.20

0.00

Distance from Table 99 indicates that the distance between the groups of less.

The average values of the dose indicator to the group A and B are shown in Tables 100, 101 and

102.

Table 100. The values of dose indicator in group A

The mean value ± SD

(Min-max)

CTDIvol 8.50 ± 2.38

(MGy) (3.93 to 14.19)

DLP 1368.2 ± 572.8

(mGy

cm-1) (305-3132)

I · t 56.3 ± 15.5

(MAs) (28.0 to 92.5)

Table 101. The median dose indicator in the group B, required to reduce by weight

median

reduction

CTDIox

DLP

(MAs)

(MGy)

(MGy cm-1)

7.53

1472

-20

5.77

946

-30

6,265

1213

-40

5.21

1036

-50

5.59

1232.5

-60

Table 102. CTDI value and DLP in group B

The mean value ± SD

(Min-max)

CTDIvol

6.1 ± 0.9

(MGy)

(5.2 to 7.5)

DLP

1179.9 ± 202.9

(MGy cm-1)

(946-1472)

Page 85 of 126

Darko Hadnađev Šimonji doctoral thesis

Only a comparison group compared to the dose indicators

Table 103. Central and dispersion parameters and measures of skewness and kurtosis dosimetry

indicators in group A

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

intensity

currents

56.28

15.52

28.0

92.5

27.57

51.87

60.70

0.49

-0.60

0.08

(MAs)

CTDIox

8.50

2.38

3.9

14.2

27,98

7.82

9.18

0.47

-0.55

0.16

(MGy)

DLP

1,368.22

572.81

305.0

3132.0

41.87

1205.4

1531.1

0.56

0.46

0.84

(mGycm-1)

Note: The value of skewness and kurtosis in the range of -0.04 to 0.04 are not discussed

minimum and maximum values of dosimetric indicators of patients in the group indicate that the values are in the expected range. Higher values indicate coefficient of variation of the heterogeneity of the groups of: the intensity of the current (27,57), of the CTDIox(27,98) and DLP (41,87). Sk increased values indicate that the distribution is negatively skewed at: current intensity (0.49), CTDIox(0.47) and DLP (0.56). Higher values indicate that the house is guilty of an elongated DLP (0.46). Negative values indicate that the house was flattened curve at current intensity (-0.60) and CTDIox(-0.55). The distribution of values generally remain within a normal distribution (p) at the CTDIox(0.16) and DLP (0.84). The distribution of values departs from the normal distribution (p) at the current intensity (0,08). (Table 103)

Table 104. Central and dispersion parameters and measures skewness and kurtosis dosimetry indicator patients in the group B

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

intensity

currents

42,48

15,09

22.0

86.5

35,53

37,03

47,92

1.36

1.86

0.33

(MAs)

CTDIox

6.40

2.25

3.2

12.7

35.18

5.59

7.21

1.24

1.63

0.33

(MGy)

DLP

1,254.72

568.07

296.0

2862.0

45.28

1049.9

1459.6

1.15

1.74

0.19

(mGycm-1)

Page 86 of 126

Darko Hadnađev Šimonji doctoral thesis

CTDIox(1.63) and DLP (1.74). The distribution of values generally remain within a normal distribution (p) at the current intensity (0,32), of the CTDIox(0.33) and DLP (0.19). (Table 104)

Table 105. Significance of differences among groups of patients relative to dosimetric indicators

analysis

MANOVA

7,199

0,000

discriminative

2,555

0,046

Based on the values of p = 0.000 (result MANOVA) and p = 0.046 (discriminant analysis) of Table 105, there is a difference and clearly-defined boundary between the groups of patients.

Table 106. Significance of differences among groups of patients relative to dosimetric indicators

analysis

todsk

Airflow

15,777

0,000

0,001

CTDIox

15,852

0,000

0,006

DLP

0.771

0,383

0,030

Legend: kdsk the coefficient of discrimination

As the p <0.1, this means that there is a significant difference between the group of subjects with the current intensity (0,000) and the CTDIox(0,000). Discrimination coefficient that makes the largest contribution to discrimination between the groups of patients relative to dosimetric indicators, or that the difference in the largest DLP (0.030), then with CTDIox(0.006) and the lowest in the current intensity (0,001). (Table 106)

Table 107. Features and homogeneity of the group of respondents with respect to dosimetric parameters

Group A

Group B

dpr

(%)

DLP

the evening

less

44,776

CTDIox

the evening* 1

less

8,955

Airflow

the evening* 1

less

1,493

hMG

70,00

78.13

(%)

Legend: HMG – homogeneity; dpr – contributing traits characteristics

The status of each subsample most defines DLP, because the contribution features characteristics of 44.78%, followed by CTDIox(8.96%) and current density (1.49%). Homogeneity in group A was 70.00%, and in group B was 78.13%. (Table 107)

Based on dosimetry indicators, we can say that in

Page 87 of 126

Darko Hadnađev Šimonji doctoral thesis

group A DLP value, current intensity and CTDIox the evening,

group B DLP value, current intensity and CTDIox less.

Table 108. Distance (Mahalanobis) between the groups of patients relative to dosimetric

Group A

Group B

indicators

Group A

Group B

1.17

0.00

1.17

0.00

Distance from Table 108 indicates that the distance between the groups larger.

dose

doz2

1000

1100

1200

1300

1400

1500

1600

Figure 10. Ellipses (confidence interval) a group of subjects with DLP and CTDIox

Legend: A group (1); Group B (2); DLP (doz2); CTDIox (Dose)

In Chart 10 the abscissa is DLP (doz2) and the ordinate the CTDIox(Dose). It is noted that in comparison to DLP and CTDIox Group B (2) has a minimum value, a maximum value of the group A (1).

Page 88 of 126

Darko Hadnađev Šimonji

doctoral thesis

Table 109. Assessment of the quality of images in the group A and group B

Group A

Group B

The number of parameters

The average value

collects parameters

quality assessment

The average value

index reviews

2.7

2.62

quality

The average value

20,52

18,82

effective dose

percentage

reduction

25% CTDIvol

effective dose

The average value

0.13

0.14

FOM

Table 109 shows all the variables that are taken into consideration for the assessment of image quality

in group A and group B. It can be seen that it is possible percentage reduction in the effective dose in our study for

about 25%.

Since p = 0.001 2 – test, one can say that there is a link between the group and the sum of image quality parameters, given that = 0.331 correlation is low.

Table 110. The significance of differences between groups of respondents in relation to the assessment of image quality

analysis

MANOVA

0,004

0.996

discriminative

5,488

0,006

Based on the values of p = 0.996 (result MANOVA) and p = 0.006 (discriminant analysis)

Table 110, there is no difference between group participants, and in addition there is a clearly defined

the boundary between groups of patients. This fact suggests that there are probably latent characteristics that

in conjunction with other characteristics (synthesized) contribute to the discrimination of the group.

Table 111. The significance of differences between groups of respondents in relation to the characteristics of assessment collection image quality parameters

analysis



todsk

The sum of image quality parameters

0,331

0,351

11,398

0,001

0.507

The final score of image quality

0,331

0,351

11,398

0,001

0,126

Legend: kdsk the coefficient of discrimination

Page 89 of 126

Darko Hadnađev Šimonji doctoral thesis

As the p <0.1, this means that there is a significant difference between the groups of respondents in both features summation of image quality parameters (0.001) and the final image quality score (0.001). Discrimination coefficient indicates that the difference is primarily in the largest sum of characteristics of image quality parameters (0.507) and then with a final assessment of image quality (0.126). (Table 111)

Table 112. Distance (Mahalanobis) between the groups of patients relative to a quality assessment

Group A

Group B

Pictures

Group A

Group B

0.00

0.76

0.00

Distance from Table 112 indicates that the distance between the groups of respondents moderately.

Table 113. Central and dispersion parameters and measures of skewness and kurtosis characteristics of the FOM group A

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

FOM

0.16

0.10

0.1

0.6

61.25

0.14

0.19

2.41

6.51

0,019

Note: The value of skewness and kurtosis in the range of -0.04 to 0.04 are not discussed

minimum and maximum value characteristics FOM in group A indicates that the values are in the expected range. Higher coefficients of variation suggests heterogeneity in the group at FOM (61,25). Increased values sk (2.41) indicate that the distribution of negatively skewed. Higher values indicate that the house is guilty of an elongated at FOM (6.51). The distribution of values departs from the normal distribution (p) in FOM (0.02). (Table 113)

Table 114. Central and dispersion parameters and measures of skewness and kurtosis characteristics FOM in group B

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

FOM

0.17

0.10

0.1

0.5

60,68

0.14

0.21

2.29

5.11

0,029

minimum and maximum value characteristics FOM in group B indicate that the values are in the expected range. Higher coefficients of variation indicate the heterogeneity of the group at FOM (60,68). Increased values sk (2.29) indicate that the distribution of negatively skewed. Higher values indicate that the house is guilty of an elongated at FOM (5.11). The distribution of values departs from the normal distribution (p) in FOM (0,03). (Table 114)

Page 90 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 115. Significance of differences among groups of patients in comparison to the characteristic of FOM

analysis

MANOVA

0,177

0,675

discriminative

0,177

0,675

Based on the values of p = 0.675 (result MANOVA) and p = 0.675 (discriminant analysis) in Table 115, no significant difference nor clearly defined boundaries between the groups.

Table 116. Distance (Mahalanobis) between groups of subjects in relation to the feature of FOM

Group A

Group B

Group A

Group B

0.00

0.10

0.00

Distance from Table 116 indicates that the distance between the groups less.

Table 117. Central and dispersion parameters and measures of skewness and kurtosis characteristics age patients in group A

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

Age

56.28

12,30

30.0

80.0

21.86

52.78

59.78

-0.32

-0.52

0,155

(years)

Note: The value of skewness and kurtosis in the range of -0.04 to 0.04 are not discussed

Higher coefficients of variation indicate the heterogeneity of the groups celebrate age (21,86). Reduced sk value indicates that the distribution is positively skewed (-0.32). Negative values indicate that the house was flattened curve with the characteristics of age (-0.52). The distribution of values generally remain within a normal distribution (p) at the age of the features of (0.16). (Table 117)

Table 118. Central and dispersion parameters and measures of skewness and kurtosis characteristics of age patients in group B

median

Standard

Minimum

maximum

coefficient

interval

value

deviation

value

variations

trust

Age

65.38

11.63

43.0

91.0

17,80

61.18

69,57

0.05

-0.31

0,791

(years)

The values of the coefficient of variation indicates the homogeneity of the characteristics of age (17,80). Increased values sk (0.05) indicate that the distribution of negatively skewed. Negative values indicate that the house was flattened curve with the characteristics of age (-0.31). The distribution of values generally remain within a normal distribution (p) at the age of the features of (0.79). (Table 118)

Page 91 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 119. Significance of differences among groups of patients as compared to age

analysis

MANOVA

11,120

0,001

discriminative

11,120

0,001

Based on the values of p = 0.001 (result MANOVA) and p = 0.001 (discriminant analysis) in Table 119, there is a difference and clearly-defined boundary between the groups of patients.

Table 120. Number (n) and the percentage (%) characteristics of the representation of a half of patients according to the groups

Group A

Group B

gender male

female gender

19.0

38.0

31.0

62.0

17.0

34.4

33.0

65.6

Inspection shown in Table 120 can be noticed that in the group represented by A most female terminal (31 out of 50 patients (62.0%)), and it is significantly higher than the incidence of male patients (19 patients 38.0%, p = 0.018). In the group B of female patients (33 out of 50 patients, 65.6%), is significantly higher than the incidence of male patients (17 subjects 34.4% p = 0.015). Analysis of the obtained characteristic of the male half maximum is represented by the group A (38.00%), and the feature of the female half maximum is represented in the group B (65.63%).

Since p = 0.740 2 – test, one can say that there is no correlation between groups and genders, given that = 0.037 correlation is very low.

Table 121. The significance of differences between groups of respondents with respect to gender

analysis

MANOVA

0,108

0.743

discriminative

0,108

0.743

Based on the values of p = 0.743 (result MANOVA) and p = 0.743 (discriminant analysis) in the table 110, no significant difference is not clearly defined boundaries between the groups.

The dosimetric parameters have the greatest contribution to the whole study (43.66%) (Table 122). Then follows: contribution to image quality assessment 28.263%, 28.077% of age, anthropometric characteristics of the subjects, a contribution to FOM and features half the 0.000%.

Page 92 of 126

Darko Hadnađev Šimonji doctoral thesis

Table 122. Contribution feature of the continent research

contribution%

between

43,659

dosimetry indicators

the group

28,263

Assessment of image quality

the group

28,077

Age of respondents

the group

0,000

Anthropometric characteristics of subjects

the group

0,000

FOM

the group

0,000

Landmarks half of respondents

the group

4.5. image quality

For each type of CT examinations are examples of specific anatomical regions of the body cross-section which is recorded before and after modification protocol.

Inspection of the head without contrast

An overview of the anatomy of the same cross section for the head, the standard protocol (Figure 18) and modified by the protocol for the value of the weight 40 (19), which is the maximum value of mAs reduction in our study. By comparing these images we can see the boundaries blurred corticomedullar gray and white brain mass and the presence of a small forest in Figure 19, but still sufficient quality for interpretation.

Figure 18. The intersection of the basal ganglia in

Figure 19. The intersection at the same anatomical level

standard protocol

reducing the value of mAs to 40 mAs

Page 93 of 126

Darko Hadnađev Šimonji doctoral thesis

Similarly, in Figure 20 and 21 provides an overview of a certain anatomical cross-section of the head by using the standard protocol and the modified protocol by reducing the value of the current intensity to 35 wt. The patient is younger age (23 years) as compared to previous patient (55 years), and there is a physiological anatomic differences in the appearance of the brain parenchyma (the older the patient are obvious signs of cortical cerebral atrophy). It is notable that the younger the patient due to preserved volume of brain parenchyma, more pronounced boundaries of ambiguity corticomedullary gray and white brain tissue of a given cross-section, although the decrease in volume of tube current less than in the older patient scrutiny.

Figure 20. The intersection of the basal ganglia in

Figure 21. The intersection at the same anatomical level

standard protocol

reduction value of 35 mAs mAs

Although our research is completed on the reduction of the volume of tube current 40 mAs, we tried to reduce the value of the current strength of 60 mAs (Figure 23) and 190 mAs -50% (Figure 25) in elderly patients (83 years and 75 years) that we noticed still possible dose reduction. In comparison with the complementary first obtained using a standard protocol (Figures 22 and 24), it can be seen quite satisfactory image quality obtained by applying the modified protocol.

Page 94 of 126

Darko Hadnađev Šimonji doctoral thesis

Figure 22. The intersection of the basal ganglia in

Figure 23. The intersection at the same anatomical level

standard protocol

reducing the value of mAs to 60 mAs

Figure 24. The intersection of the basal ganglia in

Figure 25. The intersection at the same anatomical level

standard protocol

reducing the value of mAs to 190 mAs

Viewing head with contrast

Figures 26 and 27 provides an axial view of contrast CT certain anatomical cross-section of the head. Figure 26 gives an overview of the protocol is formed by a standard, and Figure 27 an overview of the reduction carried out at a 40 percent by weight (maximum value of the reduction in this study). By comparing these images did not show significant difference in image quality.

Page 95 of 126

Darko Hadnađev Šimonji doctoral thesis

Figure 26. The intersection of the basal ganglia at

Figure 27. The same level of intersection with the reduction of mAs

contrast examination of the head in the standard

40 mAs

Protocol

Although our research is completed on the reduction of the volume of tube current 40 mAs, we tried to

reduce the value of the current strength of 100 mAs in patients older (83 years), in

What also did not observe any significant difference in picture quality (Figure 28 and 29).

Figure 28. The intersection of the basal ganglia at

Figure 29. The same level of intersection with the reduction of mAs

contrast examination of the head in the standard

100 mAs

Protocol

Page 96 of 126

Darko Hadnađev Šimonji doctoral thesis

Review chest

Figures 30 and 31 provides an axial cross-sectional view of anatomical thorax (pulmonary and mediastinal window). In Figure 30, the review made by standard protocol, a review in Figure 31 is performed for reduction of the weight 70 (the maximum value of mAs reduction in our study). By comparing these images did not show significant difference in image quality. Figure 31 is a somewhat stronger noise, without loss of diagnostic information.

Figure 30. The axial cross-section of the lung parenchyma

Figure 31. The axial cross-section of the lung parenchyma

in the standard protocol (pulmonary window and

to reduce by 70 mAs mAs

mediastinal window)

Page 97 of 126

Darko Hadnađev Šimonji doctoral thesis

Examination of the abdomen and pelvis

In Figures 32 and 34 gives an axial view of certain anatomical cross-section of the abdomen and pelvis that were obtained using a standard protocol, and in Figures 33 and 35 are shown in the same anatomical level of the resulting modified protocol (reduction of 50 mAs, the maximum value of the reduction in our study is 60 mAs). To display the selected images of the modified screening for the weight 50 is for the reason that the patient has a suspicious diagnosis of malignancy of the operated in the pelvis, while the patients are recorded the maximal reduction by weight (60 wt) had no pathological referral diagnosis substrate. By comparing these images, except some prominent noise, did not show significant difference in image quality.

Figure 32. The axial cross-section of the abdomen at the level of

Figure 33. The axial cross-section of the abdomen at the level of

pancreatic cancer in a standard protocol

pancreas to reduce by weight to 50 percent by weight

Page 98 of 126

Darko Hadnađev Šimonji doctoral thesis

Figure 34. The axial cross-section of the pelvis level

Figure 35. The axial cross-section of the pelvis level

urinary bladder in the standard protocol

bladder to reduce by weight to 50 percent by weight

4.6. Only data in the group CT angiography without modification protocol

Table 123 shows the average value and standard deviation of age, body weight and body.

Table 123. Data on patients

Age

Body

height

mass

(years)

(M)

(Kg)

Mean ± SD (min-max)

65 ± 8

79 ± 18

1.74 ± 0.10

(49-77)

(45-120)

(1.55 to 2.00)

Table 124. CTDI value and DLP without optimization

The mean value ± SD

(Min-max)

CTDIvol

5.8 ± 2.2

(MGy)

(3.2 to 12.9)

DLP

659 ± 389

(MGy cm-1)

(322-1906)

Page 99 of 126

Darko Hadnađev Šimonji doctoral thesis

In Table 124 are given the average value of dosimetry indicator performed inspection, and Table 125 shown in protocol without modification.

Table 125. Protocol without optimization

parameter

value

110

(KV)

I · t

19,33

(MAs)

mod

Helical

pitch

1.5

factor

collimation

16 x0,6

(Mm)

Page 100 of 126

Darko Hadnađev Šimonji doctoral thesis

5. DISCUSSION

Computed tomography is one of the most important diagnostic methods for the examination of patients. In the last few decades is widely used, and the reason for this is that is easily and quickly performed. It is simple in application and can be combined with other diagnostic visualization techniques. Today there are constructed several devices that incorporate imaging methods and then talk about the new “hybrid machines” (60). Shortening the duration of the review is the result of technological innovation and ease of application of CT. the result of the technological innovation of the short duration of the review). These technological innovations lead to a substantial increase in the number of hits in adult patients and children, which is on the one hand facilitate the diagnosis, but significantly increased individual and collective dose to the population in these views.

These considerations suggest that it is necessary to reduce the number of unjustified CT examinations, and this can be achieved by careful analysis of both the feasibility and the clinical dosimetry aspects. There is an obvious and large variations in the dose of radiation in CT scans of the same anatomical region of the razlličitim hospitals. It is these variations in dose are due to several factors: hardware differences CT apparatus, non-standardized protocols for performing CT examinations and the diversity of anatomical constitution patients (BMI). The goal of optimization is to provide a good enough image quality with minimal dose to the patient. The control is achieved by optimizing the parameters that directly or indirectly affect the dose of CT (4,27,39,64, 65).

The aim of this study was to assess the standard protocols for certain types of CT examinations, as well as to modify the terms of reducing patient radiation dose while maintaining diagnostic image quality, which this paper provides important practical significance.

5.1. Viewing head

The subject of this study were non-contrast CT examinations of the head, which are today among the most commonly for testing the existence of intracranial pathology. At the head CT examinations were sent to patients in neurological intensive unit and neurosurgical clinics. These patients with an initial inspection and control are made in the course of a few days or weeks. Because of the potential damage to the ocular lens precisely these patients target group, it is very important to determine whether it is possible to reduce exposure to radiation, and have optimized the image for diagnosis.

Page 101 of 126

Darko Hadnađev Šimonji doctoral thesis

As noted in chapter Degree “Methodology” stated, optimization CT dose for examination of the head is carried out so that the value of mAs is reduced, which is linearly reduced dose to the patient.

The study included patients that were selected, randomized, with different working diagnosis, which was made up CT scan, and the current intensity is reduced by a maximum of 40 percent by weight in relation to the initial value of 380 by weight.

We used the guidelines of the Guide EUR 16262 EN, in which the parameters for assessing the quality of images to analyze different anatomical sections head: corticomedullar boundaries brain parenchyma, basal ganglia, chamber system, a cerebrospinal fluid around the midbrain, a cerebrospinal fluid extracerebral in case of application of contrast large blood vessels and the choroid plexus. These parameters are well defined for evaluating the quality of the image producing relatively reliable content from the standpoint of diagnostic information. The method is subjective, time-consuming and for its implementation are necessary trained radiologists.

Mullins et al (2004) in his study in a similar way optimizivao protocol review with the CT examination of the head. This study was carried out on a 4 channel multi-layer where CT is a standard protocol for the unenhanced head checks involved the measurement of the current of 170 mAs, and CTDIvol 65 mGy (66). The maximum value of optimizing current strength in their study was 90 mAs, and CTDIvol 34 mGy. In comparison with this study, the standard value of the current intensity was 340 mAs, and CTDIvol mGy 59.4 (64-channel multi-layer CT) and the maximum optimized to 340 mAs, and CTDIvol 53.17 mGy. Mullins et al in their study also track the image quality, or using an objective method which is based on determining a physical parameter of the ratio of the contrast to noise ratio (CNR). Analyzed and various anatomical sections of the head, in the meaning of the calculation of a smaller volume (ROI) of a white and gray matter, as well as the ratio of the contrast to noise ratio (CNR) in order to define the significance of the difference between the standard and the modified protocol of image quality. Their results suggest that the increase in image noise modifikovnom Protocol by 22% compared to standard protocol. For the next intracranial pathology in young patients are recommended a modified protocol (not) traumtaska intracranial hemorrhage, rupture of aneurysm, cerebrovascular accident, hydrocephalus. However, modification of the protocol is not recommended for determining discrete pathology (i mikrolakunarne lacunar ischemic lesions). In our study, the quality of images in a modified protocol was assessed qualitatively, based on three-tier scale visualization (v.poglavlje doctorate Methodology).

Page 102 of 126

Darko Hadnađev Šimonji doctoral thesis

corticomedullar border, basal ganglia and in the end the level of the brain stem), and image noise. From a total of 50 subjects who were screened according to the modified protocol in 11 patients the CT image quality head was weaker (starting values for a current reduction of 30 mAs), but sufficient to obtain valid diagnostic information.

Our study has several limitations, and Mullins’s study. First, patients were randomized, and not explicitly defined image quality for a particular intracranial pathology, for example. for some small lesions and hypo ((micro), lacunar or subcortical infarcts, petechial hemorrhage or subarachnoid discrete), which might be overlooked in the modified protocol. For the same reason, Mullins does not recommend using the modified protocol in diagnosing emergencies. This can be a significant proposal for further study of image quality in accordance with the diagnosis.

Gundogdu et al in the 2005 study (67) analyze the qualitative and quantitative results of the measurements in patients who recorded a head CT, three reference levels (posterior fossa on the level of petrous bone, the basal ganglia and the centrum semiovale). Scanning is done in a non-helical mode. The first is a CT scan of the head shot by standard protocol where the scan parameters were 140 kVp, 280 mAs for the intersection of the posterior fossa and 120 kVp and 320 mAs for supratentorial level (protocol 1). Then, after that was surveyed by a modified protocol for targeted anatomical region (Protocol 2: 140kVp and 140 mAs for a first level, of 120 kVp and 160 mAs for second and third level; Protocol 3 140kVp and 120 mAs for a first level and 120 kVp and 120 by weight for the second and third level; protocol 4 140 kVp and 90 mAs to the first level and 120kVp and 100 mAs for second and third level). They also used measures head circumference and maximum AP diameter of the head of the patient who sniman.Rezultati their research are as follows: 1 for the standard protocol CTDI (mGy) for the first level of 58.21, 47.93 for second and third level; Protocol 2 for a CTDI (mGy) for a first level 29,11, 23,96 of the second and third level; Protocol 3 for a CTDI (mGy) for a first level 24,95, 17,97 of the second and third level; Protocol 4 for a CTDI (mGy) for a first level 18,71, 14,98 of the second and third level. Evaluation of image quality in their study was carried out in part quantitatively defining the ROI in four anatomical cross-section that we used in the analysis. The qualitative assessment was based on a five-point scale visualization. Their results show that it is possible to dose reduction up to 50% without significant loss of image quality. In a bid to reduce the dose by 60% received the degradation of CT images at the level of the cerebellum. In our study we used the spiral recording mode, has not changed the value of x-ray tube voltage at Gundogdu and associates, but not the value of mAs when shooting different anatomical levels. In our study, the value of mAs reduction is done gradually, while in the preceding study

Page 103 of 126

Darko Hadnađev Šimonji doctoral thesis

advance reduced by 50%. Although the final value of reducing the volume of electricity with 380 mAs to 345 mAs, and a dose reduction to around 7.5% in our study that was conducted only qualitatively, not to reach a threshold level of optimization.

Britten et al in 2004 (68) conducted judge how the addition of using the method of spatially correlated statistical noise to the image, which is applied to the transaxial CT images of the head of healthy subjects with a view to reducing to simulate the exposure to 50%, and in view of the that the increase in the noise reduction of a direct consequence of the dose. Routine protocol is based on the amount of current of 420 mAs. The spatial correlation of the noise is determined in the image obtained by the exposure routine to simulate the exposure to 340mAs, 300mAs and 260 mAs and 210 mAs. Reference parameters to determine the image quality were periventricul hipodenznih presence of lesions (such as simulated effect of reduction of dose of the diagnostic accuracy) and visualization of the internal capsule. The results showed that reducing the dose parameter of diagnostic accuracy without compromising on the quality, while vizulizacija internal capsule was worse at lower exposures. Method was applied in this study, as well as the results obtained is similar to the method used in the paper Britten et al, and is important in the study prospectively limit optimize the relation between the image quality-dose. However, in this way, noise is added artificially to non-randomized images. In this paper, the optimization of the confession on a randomized sample, using real clinical picture.

Results cadaveric studies Cohn and associates in 2000 showed that the quality of the image changes only if you reduce the dose by 50% (by changing the value of kVp and mAs). Although it is not appropriate to be compared with other studies in the clinical practice, it is apparent that the radiation dose can be reduced to a certain level without compromise in the diagnostic image (69).

In our study possible dose reduction for this review is about 7.5%. The reason for a small percentage of the reduction in dose may be that we have used only the subjective assessment of image quality (evaluation of a radiologist), compared with the above studies that have used and quantitative methods. Also used are randomized images, in dynamic conditions, which must impair the effect of reducing the dose in relation to the studies conducted using the static cadaveric images of patients or the image which has been artificially increased noise level. In all three above mentioned studies (66,67,68), in which the results showed that it is possible to reduce the dose by 50%, for the evaluation of the image quality taken only particular anatomical levels, whereas in the present study evaluated for all levels are defined in an anatomical guidelines of the Guide EUR 16262 EN, which increases the sensitivity of the assessment of the quality of the image.

Page 104 of 126

Darko Hadnađev Šimonji doctoral thesis

5.2. Viewing head with contrast

The unoptimized group CTDIvol average value was 59.43 mGy and DLP 2147.4 mGycm-1While the reduction in the maximum amperage of 40 mAs value CTDIvol 53.86 mGy and DLP 2043 mGycm-1. From a total of 50 patients were recorded according to the modified protocol, in 3 subjects was observed lower picture quality according to defined parameters (starting from the value of the reduction current of 25 mAs). Similarly, as for group I, CT head without contrast the results showed that the value of mAs gradual reduction increases the uncertainty limit of the observed anatomical cross-section of the head (starting from the level of the corticomedullary limits, then the basal ganglia and in the end-level of the brain stem), as well as image noise. How is the contrast CT scan of the head requires higher spatial resolution and application of thin-section for evaluation of blood vessels, it would be expected that the smaller dose reduction compared to non-contrast examinations of the head. However, in our study were monitored all the parameters of the Guide EUR 16262 EN, a view not only evaluation of blood vessels,

In recent literature, a small number of works that deal with the optimization of contrast CT examination of the head. Basically you can find work in this field for screening CT angiography of the head, which means significantly different protocol views, but our results are hardly comparable results difficult.

The study conducted by Smith et al from 2008 (70) showed that in neuroradiological CT protocols possible to significantly reduce the dose to the patient and that it does not affect the quality of images by applying a modulation to the dosage z direction (in CT with detector 16 channel) and the xyz -axis (in CT with detector array 64 rows). Modulation of dose in the z-axis is performed so that the same speed is maintained tube current, or a computer algorithm changes the amount of current applied to each CT section based on the set index of the noise (Ni noise index, only a property of a manufacturer of the scanner, GE). NI is the indication of the noise level of the image that is acceptable to the radiologist for a particular CT scan. NI defines the manufacturer as well as the image quality parameter according to which automation functions. Radiologist rarely has an impact on NI. The dosage modulation by xyz-axis implies that the current intensity can be varied from slice to slice, as well as within a cross-section which depends on the attenuation in the path of X-rays. Their results for patients who recorded CT angiography have shown that the head of the applying dosage modulation by xyz- axis can be reduced to a dose of about 38.5% (and DLP values CTDIvol). Also shown is lower dose reduction in CT with detector array 64 row in comparison with the CT of 16 detector rows, for the reason that with the first CT scanner done

Page 105 of 126

Darko Hadnađev Šimonji doctoral thesis

more sections and thin sections, was present phenomenon “overranging”. Image quality in their study quantitatively evaluated based on indicators of image quality compared to the modulation function for a current which in this case is modified compared to the standard protocol review.

When the scanner that was used in our study (manufacturer Siemens) algorithm by which it is the modulation current mAs eff. or mAs / pitch. Standard protocols are used so that the modulation current mAs eff. is constant. By reducing the value of mAs in this study introduced the deviation compared to the standard protocol in such circumstances, is estimated picture quality. The idea is to optimize the protocol individualized relative to the standard protocol that incorporate equipment manufacturers.

In our study also applied the principle of dose modulation by xyz-axis, and the image quality is based on a subjective assessment of radiologists as previously stated, this is partly for this reason discrepancy in the results.

5.3. Review chest

The thorax consists of a structure which is characterized by a large difference in the value of attenuation of the high natural contrast, it is possible to significantly decrease the MDCT dose and image quality that are not compromised. Naidich et al 1990 (71) has first found that a decrease in the strength of current with 140 by weight to 10 by weight of the quality of the image that is obtained is sufficient for an adequate assessment of the lung parenchyma. Since then, the derivative have been numerous studies on the optimization of the views of the chest, in particular of importance to the pediatric population (Henschke et al 1999 (72), Itoh et al, 2000 (73), Swensen et al 2002 (74), National Lung Screening Trial Team 2010 (75)). There was no significant difference in the structure of the lung parenchyma between niskodoznih views (npr.40 by weight) and of a high (e.g., 400 mAs; Zwirewich et al.1991 (76), Lee et al, 1994 (77)). However, changes such as. ground-glass pulmonary parenchyma is difficult to evaluate at niskodoznih CT examinations due to increased noise. It is recommended that 200 mAs to be the initial value for obtaining the thin-CT-section and a lower dose (e.g., 40-100 mAs) for follow-up examinations. Recommendations given in European guides are based on the balance between theoretical radiation risk and expected medical benefits. The values of diagnostic reference levels are around 14 mGy for CTDIvol, and a range of 450 to 650 mGycm Recommendations given in European guides are based on the balance between theoretical radiation risk and expected medical benefits. The values of diagnostic reference levels are around 14 mGy for CTDIvol, and a range of 450 to 650 mGycm Recommendations given in European guides are based on the balance between theoretical radiation risk and expected medical benefits. The values of diagnostic reference levels are around 14 mGy for CTDIvol, and a range of 450 to 650 mGycm-1GLP (78). Lowering these values is required and depending on the weight of the patient, CTDIvol can be reduced to 2 mGy. By applying modern MDCT and the AEC system, as well as iterative reconstruction technique files, it is possible to obtain high-quality CT images

Page 106 of 126

Darko Hadnađev Šimonji doctoral thesis

with CTDIvol 1-4 mGy and DLP <40 mGycm-1(Singh et al 2011 (79), Pontana et al 2011 (80)). Effective doses less than 0.8 mSv is very close to that of chest radiography.

In our study included patients that were selected, randomized, with different working diagnosis, which was made up CT scan of the chest, and the current intensity is reduced by a maximum of 70 wt. The diagnosis were related to pulmonary and extrapulmonary pathology, which is probably the reason for reduction in power only to the value of 70 mAs. We took into account only the diagnosis which were related to pulmonary pathology, likely to increase the possibility for further optimization taking into account the clinical condition of the patient.

The study used the guidelines of the Guide EUR 16262 EN, in which the parameters for assessing the quality of images to analyze different anatomical sections of the chest: chest wall, thoracic aorta and the superior vena cava, heart, lung parenchyma, blood vessels (after iv administration of contrast medium), the structure including the anterior mediastinum residual thymus (if present), the trachea and main bronchi, paratracheal tissue, Karin, regions of lymph nodes, esophagus, pleuromedijastinalna limit, segmental bronchi, the boundary between the pleura and the thoracic wall.

The results of our study showed that the unoptimized group at an average value for a current of 44 mAs and constant voltage of 120 kV obtained average value CTDIvol 6.84 mGy and DLP 490.88 mGycm-1. In an optimized group at the maximum reduction in volume of tube current of 70 mAs, the average value of current strength of the pipe is about 26.95 by weight, and a constant voltage of 120 kV, the obtained average values were CTDIvol 2.36 mGy and GLP 113.5 mGycm-1. Possible reduction of dose in this review in our study was 40%. From a total of 62 patients who recorded according to the modified protocol, the 8 patients were partially compromised image quality at the reduction of the value for the strength of current of 70 mAs, and then quenched further optimization.

Similar results were obtained in a study conducted by Prasad et al 2002 (81). Compared with our study, they examined the image quality in patients who had more than 65 years, and the diagnosis of lung cancer. Each of their patient is recorded according to a standard protocol (220-280 mAs), and the amended protocol (reduced by 50% of the dose, 110-140 mAs), wherein the observed anatomical level 4 (the lung parenchyma, of the trachea and the principal bronchus, mediastinum and the chest wall hair). Image quality was assessed subjective method based on a five-point scale visualization. Their results showed that in CT scans of the chest at a dose reduction of 50% of the image quality is acceptable for the visualization of the structure of the normal, but that the effect of the accuracy of diagnostic information need to be further examined.

Page 107 of 126

Darko Hadnađev Šimonji doctoral thesis

Similar results were obtained in a study conducted by Ravanel et al, 2001 (82) in a patient in a biopsy guided by the chest CT scan. Their results showed that it is justifiable decrease the tube current by 50% (from 280 mA to 140 mAs) the CT thorax in adult patients. However, Ravana is said that during the optimization necessary to pay attention to the patient’s clinical interest, because low-dose CT scan is applied primarily in the general screening, and the use of CT scans visokodoznog adequate to detect discrete signs of diseases of the chest.

In screening studies carried out by Takahashi et al 1998 (83) and Itoh et al

2000 (73), it was confirmed that image quality at the reduction of the dose (mAs from 250 to 50 by weight) may be perceptually worse, in particular with the increase of the noise, but if it is easy to recognize normal and pathological structures, pictures, and still are of diagnostic quality.

Mayo et al in 1995 (84) in their work showed that the reduction in volume of tube current of 400 mAs to 140 mAs subjective image quality was not significantly changed for the detection of lung and mediastinal structures. Their research has been conducted on non-contrast CT scans of the chest, where the emphasis is emphasized that further research is needed in contrast examination, taking into account the clinical interest.

Although these studies have differently defined patient samples and compared to the methodology of our study, the aim was the same, and our results are in agreement with the results from the above studies.

5.4. Abdominopelvični review

CT of the abdomen and pelvis is often done today because of the wide range of indications including acute abdominal pain. With MDCT scanners is now possible rapid volumetric acquisition, and often an overview of the entire abdomen and pelvic means as a screening test in patients with suspected abdominal disease. There is a growing concern for the radiation dose to which the patient receives in this view, especially in young patients, as well as in those cases where it is screened and control of malignancy (O’Malley et al.2010 (85), la Fougere et al. 2008 ( 86) Rodriguez-Vigil et al. 2006 (87), Yamamura et al. 2010 (88)).

In our study included patients that were selected, randomized, with different working diagnosis, which was made up CT scan of the abdomen and pelvis, and the current intensity is reduced by a maximum of 60 weight (about 25%).

Page 108 of 126

Darko Hadnađev Šimonji doctoral thesis

The study used the guidelines of the Guide EUR 16262 EN, in which the parameters for assessing the quality of images to analyze different anatomical sections of the abdomen and pelvis: diaphragm, liver, spleen, pancreas, kidneys, abdominal aorta and proximal parts of the common iliac artery, abdominal wall, blood vessels (after the administration of contrast iv).

The results showed that in the unoptimized group at an average value for a current tube

56.29 mAs and constant voltage of 120kv obtained average value CTDIvol 8.50 mGy and DLP 1368.22 mGycm-1. In an optimized group at the maximum reduction in volume of tube current of 60 mAs, the average value of current strength of the pipe is about 42.48 by weight, and a constant voltage of 120 kV, the obtained average values were CTDIvol 5.59 mGy and GLP 1232.5 mGycm-1. Possible reduction of dose in this review in our study was 25%. Out of 50 patients who recorded according to the modified protocol, the 6 patients had partially disrupted image quality at the reduction of the value for the strength of current of 60 mAs, and then quenched further optimization. Compared with the previous CT examinations of other regions of the body, in the optimization of the group views can be seen a great impact on the patient’s BMI obtaining adequate image quality. Our results show that patients with normal weight and underweight patients was a difficult process optimization and evaluation of image quality (reduction of more than 60 mAs), while in obese patients could run further optimization review.

Earlier studies have shown that, similar to other types of CT, it is possible a significant reduction in the radiation dose and at abdominopelvičnog views using an automatic modulation of the volume of tube current (Automatic Tube Current Modulation -ATCM). This study Smith and associates in 2008 (70), in the section of this chapter doctorate, which refers to the optimization of CT examination of the head with contrast, deals with the issue of dose modulation using NI, which is actually basically the ATCM. However, the most important application of ATCM at abdomen, because the differences in the habitus of patients most pronounced here. Control dose ATCM using techniques, which is incorporated in the scanner, is the obligation of the manufacturer. ATCM principle is based on maintaining constant image quality at low exposure because ATCM respond quickly to large variations in the attenuation of the beam. The principle of attenuation of X-ray and image noise depends on the size of the object and tissue attenuation. Kalra et al in their research in 2004 (89) showed that it is possible to decrease by weight to about 33% to give the acceptable diagnostic information in CT scans of the abdomen and pelvis using the modulation current intensity along the z-axis.

A study by Rizzo et al, 2006 (90) compared the techniques and the combined angular ATCM with constant current electrolysis at examination of the abdomen and pelvis revealed that the

Page 109 of 126

Darko Hadnađev Šimonji doctoral thesis

possible to reduce the dose of about 42-44% with an acceptable noise level without compromise in the diagnostic information.

Lee et al in 2011 (91) performed a similar study in which they compared the image quality and radiation dose combined ATCM and fixed techniques for CT examination of the abdomen and pelvis CT apparatus 16 detektorkih lines. Their results showed that the possible reduction in current by about 45% using ATCM. Also, dosage levels were lower in patients with a lower BMI. Optimization, which is presented in this paper is to control the dose to routine clinical level, in addition to the ATCM.

In our study was not included ATCM, but is obtained in significant reduction in the dose of about 25%.

For spiral CT scanners, the pitch value is defined as the ratio of velocity of the table movement per rotation of the gantry according to the nominal width of the x-ray beam (92). Recommendations for the value of the pitch depends on the type of review. For the types of inspections carried out in this study, the value of the pitch factor should not be less than 0.9 (38.93). Increasing the value of the pitch factor reducing the length of exposure to radiation anatomical region which is scanned. Higher speed of the chair for a given collimation results in greater values of pitch factor and is associated with a reduced dose of radiation due to the shorter exposure time, particularly if the other scan parameters, including the volume of tube current, constant. There was no significant difference in the quality of the image obtained on the value of the pitch in 1.5 compared with those on the pitch value of 0.75, wherein the dose of radiation smaller by 50% for the examination of abdomen and pelvis (94). However, this is not the case with the scanner using the effective mAs setting, which is defined as the ratio of mAs and pitch factors. In such a scanner, the effective level is maintained constant by weight regardless of the value of the pitch, so that the irradiation dose does not change with a change of pitch (39). This is a characteristic of Siemens scanner, which was used in this study.

Published results of studies the trend of increase dose values to the development of technology-intensive MDCT (95,96,97). This increase in dosage is partly connected with the need to scan something larger volume than planned to obtain enough data to reconstruct the first and last sections. Generally, there are additional polurotacija at the beginning and at the end of the planned length of the scan, which may be counted to increase the dose by 10-20%, for the study of the head-neck, and the study of the spine, and in the study of the thorax and abdomen and pelvis can be achieved and an increase of 30-35% (95,96,97).

Page 110 of 126

Darko Hadnađev Šimonji doctoral thesis

The need to revise the scan parameters is also highlighted in a recent study of CT dose, conducted by the International Atomic Energy Agency (IAEA) in developing countries. The study showed that the values of the DLP and CTDI varied up to a factor 13 -16 for the examination of the abdomen with the pelvis (98).

5.5. image quality

As noted, the parameters of the scan protocol “standard” review of CT are usually implemented by the manufacturer, and are oriented towards providing the best image quality in order to achieve the highest diagnostic criteria. Their technologists do not want to show a routine operation at the minimum dose, because the images with more noise presented by the product manufacturer inferior to the competition (99).

In CT, the effect of changing the dose (npr.promenom value mAs) image quality is sometimes difficult to assess, because the CT technique in which digital image acquisition and display are not connected. Therefore, the over-exposure will not affect the over-exposure image and the degradation of image quality.

Changing doses plug the intensity noise of the image that continues to affect the ability to visualize niskokontrastnih details, which can be improved by reducing the value of kV. Patient size, shape and anatomy of the patient also have a significant effect on image noise. The image noise will dominate in those regions, which have a high asymmetry of attenuation, such as eg. shoulders or pelvis of the patient. Adapting the protocol is based on the attenuation is, therefore, a more accurate relative to the external characteristics of the patient such as age, body weight, BMI or outer dimensions (100).

Image noise and dose that suits him to be done to give clinically acceptable diagnostic task with confidence is very difficult to measure. A next dose of noise and, in addition to affecting image quality and artifacts.

Artefacts of the image system are erroneous interpretation of ideal characteristics of the image, and is generally not associated with colicky dosage form that is administered. There are numerous artifacts: Water beam hardening, bone beam hardening and / or scatter, Partial Volume, Bone Beam Hardening and / or scatter, 3D artefacts, Helical and Cone beam, Patient Motion, Low Signal Streaks, Aliasing Artifacts, Off Focal Radiation, Tube arcing (Spit), Vibration, Electro-Magnetic Interference, Detection-Calibration artefacts.

Page 111 of 126

Darko Hadnađev Šimonji doctoral thesis

In our study we observed in several patients optimized views the presence of artifacts. In the group of head CT without a contrast with the reduction of the strength of current of 35 mAs, artefacts according to the type of the partial volume effect occurring in two subjects, who have the same admission diagnosis (cervikokranijalni syndrome). In the group of head CT with contrast registers the presence of artifacts in three patients, two were due to the shifting of the patient for review (first patient with the diagnosis of the operated malignant tumors other sites, to reduce current intensity of 35 mAs, and the other with the diagnosis of organic psychosyndrome, with reducing the strength of current of 35 mAs), and one is by the type of the partial volume effect in reducing the strength of current of 25 mAs (this patient had a headache admission diagnosis).

In the group of CT scan of the chest was recorded in the presence of artifacts in four patients. In the two patients artifacts were due to the shifting of patients (bad general state of the patient, the patients breathing very rapidly during the examination; one had admission diagnosis, and for the second it is appropriate diagnosis was pneumonia; decrease the current in the first was for 40 percent by weight, and for the second 20 mAs). The third patient had an artifact of the appointment of bismuth care, by type scattering-glare. Reducing the current intensity in this patient was to 70mAs, admission diagnosis and connective tissue diseases. In the fourth particular patient to reduce the value for the strength of current of 45 mAs, it was noted that structures of the front and upper mediastinum weight of the running due to the effect of the contrast bling in a broken blood.

In our study, the presence of artifacts in said patients has been evident only in optimized groups. Although the artifacts are not associated with a dose that is applied, they significantly affect the reduction in image quality. They can occur in patients unoptimized views (which are commonly found in practice), but only random samples neptimizovanih group of patients the study, we recorded them. Artefacts due to the movement characteristic of poor general condition of patients and the diagnosis.

In an optimized group CT scan of the chest three patients we registered an additional parameter which arguably may affect the reduction in image quality. In fact, all three patients had the diagnosis with pathological substrate in the lung parenchyma, whose contours are vaguely demarcated from the surrounding related structures. Although in all three patients the final score of image quality the highest per standard parameters that are defined in the European guidelines for evaluating the image quality of the thorax, we believe that this additional parameter can be an important indicator of the quality of images when defining an optimized protocol for patients with pathological substrate in the lung parenchyma (pneumonic infiltrate, malignant tumor of lung).

Page 112 of 126

Darko Hadnađev Šimonji doctoral thesis

This should also be taken into account that there were no artifacts, and that this parameter is entirely dependent on the reduction in the volume of electricity (for the first patient 45 mAs, then 50 mAs and 70 mAs, respectively). The significance of this parameter in relation to image quality and dose reduction should in future research to be established, because our sample is not representative in this parameter (and other patients have the diagnosis).

In recent literature, a small number of relevant data on the impact of artifacts on the interpretation of the image quality of optimized CT examination. Comparative controlled study conducted by Freeport et al in 2000 (101) analyzes the occurrence of so-called “stripe” artifacts (Low dose streak artifact) in the lung parenchyma in children (age from 15 days to 16 years) which have been sent to the non-contrast low dose CT review chest high resolution. The authors took into account the cooperation of children in the review as one of the important criteria for assessment. Patients were divided into 3 groups (a total of 44 patients, it was), according to the mAs value, which have been optimized: 180 by weight (non-optimized), 50 mAs and 34 mAs, the constant value of the voltage of 120 kV. The results of their study showed that most of these artifacts observed in children with whom to not establish cooperation during the examination at the value for a current of 34 mAs (60%). At least artifact is observed in children whose cooperation has been established in the check, which is optimized at 50 by weight (7%). Although in the sample there was no children, it should be mentioned the fact that the Freeport and collaborators have found significant reduction in the dose during the optimization of their standard protocol -70% (for review at 50 mAs) and 80% (for review by weight at 34).

CT angiography (CTA) of the abdominal aorta carried out in patients with diseases of the aorta after endovascular procedures (EVAR- endovascular aneurysm repair), and in the screening of patients with peripheral arterial occlusive disease (PAOD), which is in our study was the most common admission diagnosis. Patients with PAOD working CTA segment aortoiliac and lower extremity arteries are at risk of contrast-induced nephropathy, as many come with impaired renal function (78).

CTA there are numerous strategies for the reduction of radiation dose: reduction in the voltage, the application of the AEC system with the modulation strength of tube current, an increased value of the pitch and the most recent application of the algorithm is the iterative reconstruction, and virtual views of the non-contrast CT, dual energy (78).

Page 113 of 126

Darko Hadnađev Šimonji doctoral thesis

The literature cites a number of studies that have examined the reduction of the dose of radiation using the AEC system to modulate the strength of current (Mulkens 2005 (102), Fraioli, 2006 (103)), which was our purpose. However, for a group of 5 (CTA angiography of the abdominal aorta and lower extremities) in our study was not performed evaluation of image quality, because for technical reasons it was not possible to optimize a protocol review (tried it with a reduction in the value of mAs or CT device did not achieve reproducible image quality at setpoint). The data of this group from the first phase of the research are presented in chapter doctorates results.

Page 114 of 126

Darko Hadnađev Šimonji doctoral thesis

CONCLUSIONS

The study, which is divided into 2 stages (stage without optimization and optimization phase of hits) included adults, instruments which are to be taken for particular multi-layer region of the body by CT. In accordance with the previously defined objectives, the methodological approach, and the hypotheses, in this study we analyzed the difference between the groups of patients (non-optimized and optimized) for CT checks of the same anatomical region, according to the anthropometric characteristics of the sample (body height, body weight, BMI), age , dose indicators, FOM, assessing the quality of the image and gender. Based on the results derived the following conclusions:

Inspection of the head without contrast

1) There was no difference between the groups of patients according to the anthropometric characteristics, FOM, the assessment of image quality and sex.

Difference There is the obtained data in relation to age, and dosage ratios

CTDI valuesoxDLP and effective doses were higher in unoptimized group. Homogeneity in both groups was 100.00%.

Based on the evaluation of image quality respondents can say that in the unoptimized group homogeneity of 100.00%, and the optimized group is 50.00% homogeneity.

The unoptimized group values for age are less homogenous is 56%, while the value of greater age in the optimized group, homogeneity is 63.64%.

Viewing head with contrast

There was no difference between the groups according to anthropometric characteristics of the subjects, assessing the quality of the image, FOM and gender.

There was no difference between groups with respect to dose indicators, in CTDIox.

Review chest

There was a difference between the groups in relation to the anthropometric characteristics of the respondents. No statistically significant differences between groups with respect to the characteristics

Page 115 of 126

Darko Hadnađev Šimonji doctoral thesis

height, weight and BMI. There is an underlying difference in body height, body weight and BMI.

Establishes the difference between the groups in relation to the dosimetric indicators at the current intensity, of the CTDIox and DLP.

There is a difference between the groups in relation to the assessment of image quality, according to a feature of the sum of image quality parameters, but not in the final assessment of image quality. There is an underlying difference in the final assessment of image quality.

Establishes the difference between the groups with respect to age of patients and FOM, but not in the half of the respondents.

The unoptimized group values for DLP, CTDIox and the intensity of the currents are higher than the optimized group, and homogeneity is 69.23%, while the optimized homogeneity of the group of 82.26%.

Examination of the abdomen and pelvis

Establishes the difference between the groups according to the dosimetric indicators, at the current intensity and the CTDIoxAnd at age characteristics.

There was no difference between groups in relation to the anthropometric characteristics, image quality assessment, FOM and a half.

Based on the above it can be concluded the following:

Using standard protocols it is better picture quality than is necessary for reliable diagnosis, and therefore higher radiation dose than necessary.

The optimum choice of protocol parameters in terms of exposure (mAs value reduction), in examining the head (with / without contrast), chest and abdomen to the pelvis, it is possible to significantly reduce the dose of radiation (7.5% for native CT scan of the head, 7% for

CT head with contrast, 40% for CT scan of the chest, 25% for CT examination of the abdomen to the pelvis), held with picture quality that is sufficient for an adequate interpretation of radiological images.

Identified doses and radiation risks for patients unoptimized and optimized views. Values with optimized group were significantly less than the level of unoptimized groups (especially pronounced in the views of the chest and abdominopelvičnog views), and the recommendation of this research optimizing standard protocols for examinations of the chest and of the multistep abdominopelvičnog

Page 116 of 126

Darko Hadnađev Šimonji doctoral thesis

Radiological examinations in everyday practice to limit this survey, with few exceptions (caution in thinner patients abdominopelvične recording region is getting worse, the general condition of the patients, who were operated after radiotherapy). For examinations of the head without contrast and the contrast is necessary to pay particular attention to the clinical condition of the patient and potential pathologic substrate of the brain parenchyma, because both aspects can play a key role in the decision radiologist to efficiently optimize protocols.

The significance of the results of this study is that the patients were randomized, and what was done clinical evaluation of image quality. It was found that there is a need for individualization CT protocol review by referral diagnosis to obtain valid diagnostic information (especially for repeated checks), and adjustment of scanning parameters according to patients BMI values (highest value in CT examinations of the abdomen and pelvis). However, the values obtained by optimizing the dose CT protocol in this study do not represent the lowest possible dose for the checks of certain anatomical regions of the body. For this reason, the recommendations of this paper is to further modify the CT protocol, taking into account both the subjective and objective measurements (quantification of observed parameters) but also applying the ATCM (CARE4Dose program).

Page 117 of 126

Darko Hadnađev Šimonji doctoral thesis

REFERENCES

Holmberg O, R Czarwinski, Mettler f.The importance and unique aspects of radiation protection and medicine. Eur J Radiol 2010; 76: 6-10.

Holmberg A., Malone J M et al Rehani. Current issues and actions in radiation protection of patients. Eur J Radiol 2010; 76: 15-19.

3. UNSCEAR Report 2008 Annex A available on the:

http://www.unscear.org/docs/reports/2008/09-86753_Report_2008_Annex_A.pdf Downloads: 10.02.2014.

Ciraj-Bjelac A. Evaluation of exposure and the possibility to reduce the patient dose in diagnostic radiology; PHD thesis. University of Novi Sad, Association of Centers for Interdisciplinary and Multidisciplinary Studies and Research, Medical Physics, 2005.183 pages.

Council of the European Union (1997) Council Directive 97/43 / Euratom of 30 June 1997 on health protection of individuals against the dangers of ionizing radiation in relation to medical exposure, and repealing Directive 84/466 / EURATOM.OJEU 1997 / L 180 / 22

Brink JA and Amis SJ, “Image Wisely: A Campaign to Increase Awareness about Adult Radiation Protection”, Radiology, 257 (2010), 601-602.

Hadnađev DR, O Nikolić, Stojanović S. Justification of computerized tomography examinations and radiation risks in everyday radiological practice; The First International Conference on Radiation and Dosimetry in Various Fields of Research (RAD 2012) -realized as a part of the FP7 project: joint research on various types of radiation dosimeters (RADDOS), Niš, Proceedings 2012: 157-159.

Oberg M: Patient Doses for CT Examinations and Denmark; Doctoral thesis. Submitted to the Department of Electrical Engineering at the Technical University of Denmark, University of Copenhagen, 2011.

FER Ergonomics available at: http://web.zpr.fer.hr/ergonomija/2003/klemencic/str02.html Downloads: 1.03.2014.

Karasman V. Radioactivity and radiation detection. Web presentation of 2006/2007.

Available at: https://www.google.rs/url?sa=t&rct=j&q=&esrc=s&source=web&cd=6&cad=rja&ved= 0CFoQFjAF & url = http% 3A% 2F% 2F% 2Frgn.hr ~ dkuhinek 2Fnids_daliborkuhinek% 2F% 1% 2520OEE-

Page 118 of 126

Darko Hadnađev Šimonji doctoral thesis

RN% 2F5Seminari% 2F2006_2007% 2F35% 2520Radioaktivnost% 2520% 2520detekcija % 2520zracenja.ppt & ei = 0L0RU9i2EsK_ygOH34LIBA & USG = AFQjCNG4n4Ny0kmy8S BcRMnTMXr7n7qANg & SIG 2 = Zwi_uw45nN352xP2J3u9Rw & bvm = bv.62286460, d.Ym with Downloads: 1.03.2014.

Zivkovic M, Živković Z, Vukašinović Z. Clinical Radiology 1.Univerzitetski udžbenik.Izdavač IGP “Sports Book” Belgrade 1992, 400 pages.

Coderre, J.22.01 Introduction to Ionizing Radiation. MIT OpenCourseWare: Nuclear Science and Engineering. 2006. Available online athttp://ocw.mit.edu/courses/nuclear-engineering / 22-01-introduction-to-ionizing-radiation-fall-2006 / Retrieved 09/21/2013.

H Petterson et al (Ed). The Encyclopedia of Medical Imaging. Physics, tecniques and Procedures: The NICER Institute, Oslo in 1998.

Design of biomedical devices and devices available in 2012/2013: http://bmi.mas.bg.ac.rs/fajlovi/diplomske/BAU10.pdf Downloads 11.03.2013.

Genzel I: Design measures radiation safety and security of the room with a CT device; Master work. University of Novi Sad, Faculty of Science, Medical Physics, 2012, 90 pages.

http://025.awardspace.com/wap/studije/rtg/CT/ Downloads: 11/08/2014.

Banghart D. CTs- A brief talk on everything! Available at:http://hpschapters.org/ncchps/Sept-08-Presentation.pdf Downloads 15.08.2014.

Pejnović P: Basic principles and technical solutions in computed tomography

(CT) Computerized tomography and central nervous systemStojanovic S et al.Novi Sad, 2007; 165: 1-11.

Lagumdzija A: Comparison of image quality parameters of different types of devices in computed tomography; Master’s thesis. University of Sarajevo, Faculty of Science, 2012.

Goldman LW. Principles of CT and CT technology. J. Nucl. Med. Technol 2007: 35

(3), 115-128.

AAPM 96, 2008, The Measurement, Reporting, and Management of Radiation Dose in CT, American Association of Physicists in Medicine AAPM Report 96, New York,

AAPM

22. Mr. Dougherty Digital Image Processing for Medical Application, Cambridge University Press, Cambridge, UK, 2009.

Page 119 of 126

Darko Hadnađev Šimonji doctoral thesis

Hara AK, Padva RG, Silva AC, Kujak JL, Lawder HJ, Pavlicek W. Iterative Reconstruction Technique for Reducing Body Radiation Dose at CT. Scottsdale: American Journal of Roentgenology 2009; 193: 764-771.

O Gosling et al. A comparison of radiation doses between state-of-the-art multislice CT coronary angiography with iterative reconstruction, multislice CT coronary angiography with standard filtered back-projection and invasive diagnostic coronary angiography. Exeter: Heart 2010; 96: 922-926.

A Meghzifene, Dance DR, D McLean et al. Dosimetry in diagnostic radiology.Eur J Radiol 2010; 76: 11-14.

International Commission on Radiation Units and Measurements. Patient dosimetry for X-rays used in medical imaging.74. Bethesda, MD: ICRU; 2006.

McNitt-Gray MF. AAPM / RSNA physics tutorial for residents: topics in CT. Radiation dose in CT. Radiographics 2002 (6): 1541-53.

International Electrotechnical Commission. Medical electrical equipment. Part 2: Particular requirements for the safety of X-ray equipment for computed tomography. IEC 601-2-44, IEC, Geneva, 1997.

DG Jones and Shrimpton PC, Noramlised organ doses for X-ray computer tomography calculated using the Monte Carlo technique, NRPB-SR250, National Radiological Protection Board, Chilton, 1993.

C. Carlson Integral Absorned Dose and Diagnostic Procedures Roentgen, Acta Radiologica, 1965; (3): 311-326.

Tack D, Gevenois PA et al, Radiation Dose from Adult and Pediatric Multidetector Computed Tomography, Springer Verlag, Berlin Heidelberg New York, Printed in Germany, 2007, 275 p.

H. Gfirtner and Wild J. A new Diamentor for measuring kerma-area product and airkerma simultaneously. Med. Phys.1997; 24: 1954-1959.

Computed Tomography Dose and Image Quality Management available http://www.sprawls.org/resources/CTIQ/module.htm Downloads 11.08.2014.

IEC, Medical Electrical Equipment, IEC Geneva, Switzerland, in 2002, publication No. 60601-2-44. Ed. 2.1.

Mahesh M, Scatarige JC, Cooper J, Fishman EK. Dose and pitch relationship for a particular multislice CT scanner, Am J Radiol 2001; 177: 1273-5

Martin CJ. Effective dose: how should it be applied to medical exposures? Br J Radiol 2007; 80 (956): 639-47.

Page 120 of 126

Darko Hadnađev Šimonji doctoral thesis

Huda W, Ogden KM, Khorasani MR. Converting dose-length product to effective dose at CT. Radiology 2008; 248 (3): 995-1003.

Bongartz G, SJ Golding, Juřík AG et al. European guidelines for multislice computed tomography; March 2004th

kalra M, MM Maher, TL Toth et al. Strategies for CT Radiation Dose Optimization. Radiology, 2004.230: 619-628.

Baert AL, Knauth M, Sartor K.Radiation Dose from Adult and Pediatric Multidetector Computed Tomography (2007).

MK Kalra and Saini S. Standardized Nomenclature and Description of CT Scanning Techniques. Radiology 2006; 241: 657-660.

Kalender WA. Computed tomography: fundamentals, system technology, image quality,

applications. Chichester: Wiley, 2005.

Wang J et al. bismuth Shielding, Organ-based Tube Current Modulation, and Global Reduction of Tube Current for Dose Reduction to the Eye at Head CT. Radiology 2012; 262: 191-198

Bacher K, Smeets P, K Bonnarens, Hauwere AD, Verstaete K, Thierenes H. Dose reduction in patients Undrgoing Chest Imaging: Digital Amorphous Silicon Flatpanel Deteact Versus Film-Screen Conventional Radiography and Computed Radiography Phosphor Based, AJR 2003; 181: 923- 9.

Strotzer M, Wolf M, R Frund, Hamer O, N Zorger, Feurerbach S. Routine Chest Radiography Using a Flat Panel Detector: Image Quality Detector at Standard Dose and Dose 33% Readuction, AJR, 2002; 178: 169-71.

Jahn B, Digital Image Processing, Springer Verlag, Berlin Heidelberg 2005.639 p, 6th edition.

Suetens P Fundamentals of Medical Imaging, Second Edition, Cambridge University Press, Cambridge, UK, 2009, second edition.

McNitt-Gray MF Tradeoffs in CT Image Quality and Dose available at: http://www.oucom.ohiou.edu/ou-microCT / Downloads / Tradeoffs_in_CT_Image_Quality_and_Dose_9794-13379.pdf Retrieved 17/09/2014.

Hadnadjev D, Stojanovic S. Patient dose reduction strategies and multi-detector computed tomography (MDCT). European Congress of Radiology 2013- Electronic Presentation Online System – EPOS C-0356th Educational Exhibit.

Page 121 of 126

Darko Hadnađev Šimonji doctoral thesis

Tamm EP, Rong XJ, Cody DD, Ernst RD, Fitzgerald NE, Kundra V. Quality Initiatives: CT Radiation Dose Reduction: How to Implement Change without Sacrificing Quality Diagnostic. Radiographics 2011; 31.1823 to 1823.

Silkoset, RD Analyze av variasjon and representative doser ved Undersokelser between computer tomographs (CT). Østerås: Statens stråleværn, 2008: 0804-4910.

Oosawa K, Nakahiro S, S Terakawa et al.Evaluation of effectiveness of the CT-auto exposure control (AEC-CT) and Ensuring image quality and reducing exposure dose based on a low-contrast resolution using 3 CT systems. ECR 2009 C-798 Available at:http://posterng.netkey.at/esr/viewing/index.php?module=viewing_poster&task=viewsect ion & pi = 40859 & t = 122403 & searchkey = 3d3380b9c18269ecd6d1037129e70e9c # poster2 Downloads 12.08.2014.

CT scanner automatic control system Report 05016 available www.mhra.gov.uk Downloads 5.09.2014.

Cochrane Miller J, Iterative Image Reconstruction Method (ASIR): Lowering CT

RadiationDoseandImprovingImageQualitydostupnona

http://www.mghradrounds.org/index.php?src=gendocs&link=2009_october Downloads 9.9.2014.

Deak et al Zs. Filtered Back Projection, Adaptive Filtered Back Projection, Adaptive Statistical Iterative Reconstruction, and a Model-based Iterative Reconstruction and Abdominal CT: An Experimental Clinical Study. Radiology 2013; 266 (1): 197-206.

Wall B F. Implementation of DRLs and UK; Rad Prot Dosim 2005; 114: 183-188.

Commission of the European Communities (CEC). European guidelines and quality criteria for diagnostic radiographic images. EUR 16260 EN eds. Carmicheael J, et al. CEC, Brussels, 1996.

Commission of the European Communities (CEC). European guidelines and quality criteria for diagnostic radiographic images in pediatrics. EUR 16261 EN eds. Carmicheael J, et al. CEC, Brussels, 1996.

Commission of the European Communities (CEC). European guidelines and quality criteria for computed tomography. EUR 16262 EN eds. Bongertz G, et al. CEC, Brussels, 1998.

Johnson DA. CT imaging and related risk for cancer from Medscape Gastroenterology-

Viewpoints available http://www.medscape.com/viewarticle/714400?src=mp&spon=35, Downloads 9.9.2014.

Page 122 of 126

Darko Hadnađev Šimonji doctoral thesis

Brenner DJ. Estimating cancer risks from the pediatric CT: going from the qualitative to quantitative. Pediatr.Radiol 2002 32: 228-231.

Mettler FA, West PW, Locken JA et al. CT scanning: Pattern of use and dose. J Radiol Prot 2000; 20: 353-359.

Brenner DJ, Elliston CD, Hall EJ, Berdon WE. Estimates of the cancer risks from pediatric CT radiation are not merely theoretical. Comment on “point / counterpoint: in

X-ray computed tomography, technique factors should be selected appropriate to patient size against the proposition “(letter). Med Phys 2001; 28-2387-2388.

Boone JM. What parameters are most accurate in predicting factors appropriate technique for CT scanning? Radiology 2005; 236 (2): 377-378.

National Radiological Protection Board. Guidelines on Patient Dose to Promote Optimization for Protection for Diagnostic Medical Exposure. Documents of the NRPB, Vol 10, No 1, NRPB, Didcot, in 1999.

Mullins ME, Leu MH, Bove, et al. Comparison of Image Quality Between Conventional and Low-Dose CT Nonenhanced Head. Am J Neuroradiol 2004; 25: 533-538.

Gundogdu S et al. Assessment of image quality and standard of a three-dose protocols in reducing adult cranial CT. Eur Radiol 2005; 15 (19): 1959-68. Epub 2004 Dec 2.

Britten AJ et al. The addition of computer simulated noise to investigate radiation dose and image quality in images with spatial correlation of statistical noise: an example application to X-ray CT of the brain. Br J Radiol 2004; 77 (916): 323-8.

AL Spielman et al. Increasing the radiation dose for renal stone CT: a feasibility study of single- and multidetector CT. AJR Am J Radiol 2002; 178: 1058-1062.

AB Smith et al. Radiation dose reduction strategy for CT protocols: successful implementation and neuroradiology section. Radiology 2008; 247 (2): 499-506. Epub 2008 Mar 27.

Naidich DP et al. Low-dose CT of the lungs: preliminary observations. Radiology 1990;

175: 729-731.

Henschke CI et al. Early lung cancer action project: overall design and findings from baseline screening. Lancet 1999; 354: 99-101.

Itoh S, Ikeda M, Arahata S et al. Lung cancer screening: minimum tube current required for helical CT. Radiology 2000; 215: 175-183

Swensen SJ. CT Screening for Lung Cancer. American Journal of Roentgenology. 2002; 179: 833-836.

Page 123 of 126

Darko Hadnađev Šimonji doctoral thesis

National Lung Screening Trial Team. The national lung screening trial: overview and study design. Radiology 2010; 258: 243-253.

Zwirewich CV, Mayo JR, Müller NL. Low-dose high resolution CT of lung parenchyma. Radiology 1991; 180: 413-17.

Lee HP, Primack SL, CA Staples et al. Chronic infiltrative lung desease: comparison of diagnostic accuracies of radiography and low- and conventional-dose CT-thin -section. Radiology 1994; 191: 669-673.

Tack D et al (eds). Radiation Dose from Multidetector CT, Medical Radiology. Diagnostic Imaging Springer-Verlag Berlin Heidelberg 2012, 312 p.

Singh S, Kalra MC, MD Gilman et al. Adaptive statistical iterative reconstruction technique for radiation dose reduction in chest CT: a pilot study. Radiology 2011; 259: 565-573.

Pontana F, Duhamel et al. Chest computed tomography reconstruction using iterative filtered back projection versus (part 2): image quality of low-dose CT examination and 80 patients. Eur Radiol 2011; 21: 636-643.

Prasad et al. Standard Dose-and 50% Reduced Dose Chest CT: Comparing the Effect on Image Quality, AJR 2002, 179: 2 pp 461-465.

Ravanel et al J. Radiation Exposure and Image Quality in Chest CT Examinations, AJR 2001; 177.2, pp 279-284.

Takahashi M, WM Maguire, Ashtari M, et al. Low-dose spiral computed tomography of the thorax: comparison with the standard-dose technique. Invest Radiol 1998; 33: 68-73.

Mayo JR, Hartman TE, Lee KS, et al. CT of the chest tube minimal current required for good image quality with the least radiation dose. AJR 1995; 164: 603-607.

O’Malley ME, Chung M, Haider et al. Comparison of low dose with standard dose abdominal / pelvic multidetector CT in patients with stage 1 testicular cancer under surveillance. Eur Radiol. 2010 Jul; 20 (7): 1624-30. Epub 2010 Jan 30.

La Fougere C, Pfluger T, Schneider, et al. Restaging of patients with lymphoma. Comparison of low dose CT (20 mAs) with contrast enhanced diagnostic CT and combined [18F] -FDG PET / CT. Nuklearmedizin 2008; 47 (1): 37-42.

Rodríguez-B Vigil, Gómez-León N, Pinilla, et al. PET / CT in lymphoma: a prospective study of enhanced full-dose PET / CT versus unenhanced low-dose PET / CT. J Nucl Med. 2006; 47 (10): 1643-8.

Page 124 of 126

Darko Hadnađev Šimonji doctoral thesis

Yamamura J, K Tornquist, Buchert R, et al. Simulated low-dose computed tomography and oncological patients: a feasibility study. J Comput Assist Tomography 2010; 34 (2): 302-. 308

Kalra MK, MM Maher, TL Toth et al. Comparison of Z-axis automatic tube current modulation technique with fixed tube current CT scanning of the abdomen and pelvis. Radiology 2004; 232 (2): 347-53.

Rizzo S, Kalra M, Schmidt B. et al. Comparison of angular and combined automatic tube current modulation techniques with constant tube current CT of the abdomen and pelvis. AJR Am J Radiol 2006; 186 (3): 673-9.

S Lee, Yoon SW, Yoo SM et al. Comparison of image quality and radiation dose between combined automatic tube current modulation and fixed tube current technique and CT of the abdomen and pelvis. Acta Radiol. 2011; 52 (10): 1101-6.

International Electrotechnical Commission. Medical electrical equipment. Part 2-44: Particular requirements for the safety of X-ray equipment for computed tomography. IEC 60601-2-44, Geneva, IEC 2001.

European Guidelines on Quality Criteria for Computed Tomography EUR 16262, European Commission, Brussels, 1999.

Sahani D et al. Comparison between Low (3: 1) and High (6: 1) Pitch for Routine Abdominal / Pelvic Imaging with Multislice Computed Tomography. J Comput. Assist. Tomogram. 2003; 27 (7): pp 105-109.

Dawson P. Patient Dose and Multislice CT: Why is it Increasing and Does it Matter? Br J Radiol 2004; 77 (1), pp S10-3.

Kalender W. Computed Tomography Computed tomography: fundamentals, system technology, image quality, applications. Publicis MCD Verlag, Erlangen, Munich, Germany, 2000.

Tzedakis A et al. The Effect of the Overscanning on Patient Effective Dose from Multidetector Helical Computed Tomography Examinations.Med. Phys 2005; 32 (6): pp 1621-1629.

Muhogo W et al. Patient Doses and CT Examinations in 18 Countries: Initial Results from the International Atomic Energy Agency Projects. Radiation Protection Dosimetry 2009; 136 (2): pp. 118-26.

AJ Fox. Use of the lowest radiation dose necessary (editorial). Am J Neurosurg 2004; 25: 519.

Page 125 of 126

Darko Hadnađev Šimonji doctoral thesis

Menke J. Comparison of different body size parameters for individual dose adaptation in body CT of adults. Radiology 2005; 236 (2): 565-71.

Freeport J et al. Low-Dose High-Resolution CT of the Chest and Young Adults Dose, Cooperation, Artfact, Incidence, and Image Quality. American Journal of Roentgenology, 2000: 175: 4.

Mulkens TH, Bellnick P, Baeyaert M et al. Use of an automatic exposure control mechanism for dose optimization and multidetector row CT examinations: clinical evaluation. Radiology 2005; 237: 213-223.

F Fraioli et al. Low-dose multidetector-row CT angiography of the infrarenal aorta and lower extremity vessels: image quality and diagnostic accuracy in comparison with standard DSA. Eur Radiol 2006; 16: 137-14.

Page 126 of 126

Darko Hadnađev Šimonji doctoral thesis

CONTRIBUTIONS

Attachment 1. Percentage distribution of average European by age and gender for some X-ray and CT examinations. (UNSCEAR 2013)

Years

Men

women

Men

women

Men

women

Men

women

Men

women

CT examinations

CT head

CT door

CT chest

CT of the spine

CT of the abdomen

0-4

0.75

0.53

0.25

0.29

0.32

0.16

0.10

0.06

0.09

0.07

5-9

0.76

0.46

0.49

0.20

0.28

0.21

0.22

0.16

0.12

0.08

10-14

0.84

0.70

0.34

0.26

0.22

0.21

0.30

0.21

0.18

15-19

1.29

1.24

1.08

0.71

0.75

0.47

1.15

0.90

0.63

0.50

20-24

1.62

1.55

1.43

1.09

1.13

0.75

1.76

1.49

0.82

0.83

25-29

1.75

1.94

2.04

1.34

1.45

0.91

2.57

2.10

1.48

1.24

30-34

2.48

2.61

2.68

2.50

2.09

1.31

4.12

2.90

1.94

1.61

35-39

3.01

2.77

4.06

3.41

2.44

1.99

5.26

4.36

2.37

2.06

40-44

2.84

2.61

3.89

4.09

2.84

2.11

5.55

5.13

2.84

2.56

45-49

3.06

3.25

4.64

4.50

3.45

3.07

5.25

5.48

3.27

3.29

50-54

3.79

3.68

6.08

4.70

5.07

3.97

5.43

5.17

4.45

4.32

55-59

4.15

3.45

6.17

4.67

6.02

4.73

4.74

4.59

5.65

4.76

60-64

4.08

3.64

5.77

3.77

7.25

4.88

3.65

3.96

6.03

4.87

66-69

4.27

3.99

4.66

3.38

7.53

4.78

3.08

3.93

7.20

5.07

70-74

4.80

4.53

4.10

3.29

7.42

5.25

2.72

3.72

6.95

5.55

75-79

4.45

4.96

3.03

2.85

5.52

4.01

2.02

3.27

5.49

4.57

80-84

1.62

3.02

1.10

1.63

0.95

1.08

0.45

0.80

0.97

1.42

85-89

1.62

3.02

1.10

1.63

0.95

1.08

0.45

0.80

0.97

1.42

> 90

0.56

1.07

0.31

0.67

0.28

0.31

0.06

0.21

0.26

0.61

In total

49.11

50.88

54.11

45.89

57.48

42.52

49,53

50.44

53.24

46.75

Appendix 2. The collective contribution to the effective dose of various types of CT (UNSCEAR

2008)

Views

Contribution (%)

new I

Level II

Level III-IV

Other

CT head

5.0

1.7

6.6

4.6

CT chest

9.7

1.9

8.7

CT of the abdomen

7.0

CT of the spine

2.9

0.51

7.5

2.7

pelvic CT

9.3

2.8

9.4

8.4

CT intervention

0.19

0.00

0.93

0.18

CT Others

0.95

1.2

0.00

0.64

Page 1 of 3

Darko Hadnađev Šimonji doctoral thesis

Appendix 3. Trend annual frequency of diagnostic medical radiololoških views expressed as the number per 1000 population (UNSCEAR 2008)

Level

1970-1979

1980-1984

1985-1990

1991-1996

1997-2007

820

810

890

920

1332

140

120

154

332

III

8.8

Appendix 4. Trend in the average effective dose from diagnostic medical radiological review of the health system of the country and category (UNSCEAR 2008)

overview

Mean effective dose per examination

1970-1979

1980-1990

1991-1996

1997-2007

chest radiography

0.25

0.14

0.7

Fotofluoroskopija lung

0.52

0.65

0.78

fluoroscopy lung

0.72

0.98

1.1

2.1

Limbs and joints

0.02

0.06

0.05

Pelvis and hips

2.2

1.7

1.8

1.1

Head

2.1

1.2

0.83

0.08

abdomen

1.9

1.1

0.53

0.82

The upper abdomen

8.9

7.2

3.6

3.4

lower abdomen

9.8

4.1

6.4

7.4

cholecystography

1.9

1.5

2.3

2.0

urography

3.1

3.7

2.6

mammography

1.8

0.51

0.26

1.3

4.4

8.8

7.4

PTCA

11.9

Appendix 5a. The global estimated number of procedures, collective effective dose and effective dose per patient for different categories of radiological procedures of nuclear medicine using ionizing radiation in the United States (UNSCEAR 2008)

type of procedure

number of procedures

collective effective

Effective dose

(Million)

Dose (manSv)

head (mSv)

conventional

radiography and

293

100000

0.3

fluoroscopy

Intervention

128000

0.4

410000

1.5

nuclear medicine

231000

0.8

In total

395

899000

3.0

Page 2 of 3

Darko Hadnađev Šimonji

doctoral thesis

Appendix 5b. The frequency,

effective dose in the population and the collective dose is assumed in the

global model for diagnostic practice (1997-2007) (UNSCEAR 2008)

Number of visits per 1,000 population

The effective dose per examination

annual collective

dose

(M Sv)

(Man SV)

Views

On the

Level I

Level II

Level III-IV

world

Level I

Level II

Level III-IV

world

Level I

Level II

Level III-IV

world

level

CT head

2.3

0.4

2.4

150000

17000

3800

170000

CT chest

0.8

0.7

6.3

7.8

290000

19000

9000

310000

Kodjo

1.8

0.7

8.2

12.4

570000

70000

14000

650000

abdomen

CT of the spine

0.3

0.5

3.0

5.0

87000

5100

4300

96000

pelvic CT

1.0

0.3

5.1

9.4

270000

28000

5400

310000

1.0

0.0

0.1

0.3

3.8

5700

0.0

530

6200

intervention

CT Others

2.8

1.0

0.0

1.2

3.8

16000

12000

0.0

29000

References used for Annex:

1. UNSCEAR report for 2008 (Annexes 2,3,4,5 and 5b) is available at: http://www.unscear.org/docs/reports/2008/11-80076_Report_2008_Annex_D.pdf Downloads 12.04.2014.

2. UNSCEAR The report 2013 (Annex 1) available on the:

http://www.unscear.org/unscear/en/publications/2013_2.html Downloads 12.04.2014.

Page 3 of 3

The post UNIVERSITY IN NOVI SAD MEDICAL SCHOOL DOSE ASSESSMENT AND OPTIMIZATION OF THE appeared first on essayfab.

UNIVERSITY IN NOVI SAD MEDICAL SCHOOL DOSE ASSESSMENT AND OPTIMIZATION OF THE