Institutional Patient Doses in Computed Tomography Angiography

Transcription

1 UNIVERSITY OF PATRAS INTERDEPARTMENTAL PROGRAMME OF POSTGRADUATE STUDIES IN MEDICAL PHYSICS Institutional Patient Doses in Computed Tomography Angiography Anastasia G. Aravantinou Msc Thesis October 2017 Patras

2 ΠΑΝΕΠΙΣΤΗΜΙΟ ΠΑΤΡΩΝ ΔΙΑΤΜΗΜΑΤΙΚΟ ΠΡΟΓΡΑΜΜΑ ΜΕΤΑΠΤΥΧΙΑΚΩΝ ΣΠΟΥΔΩΝ ΤΗΝ ΙΑΤΡΙΚΗ ΦΥΣΙΚΗ Δόση ασθενών σε εξετάσεις Αξονικής Αγγειογραφίας Aναστασία Γ. Aραβαντινού Mεταπτυχιακή Διπλωματική Εργασία Oκτώβριος 2017 Πάτρα

3 THREE MEMBER EXAMINATION COMMITTEE Panayiotakis George, Professor of Medical Physics (Supervisor) Costaridou Lena, Professor of Medical Physics Kalogeropoulou Christina, Associate Professor of Radiology

4 ACKNOWLEDGEMENTS I would like to thank my thesis supervisor, Professor George Panayiotakis for giving me the opportunity to realize this project and for his continuous support and guidance during this Master Thesis as well as Professor Lena Costaridou from the Department of Medical Physics. I am also indebted to the Medical Radiation Physicist from the University Hospital of Patras, Dr. Gerasimos Messaris for his valuable scientific support and co-operation during this research. Without his knowledge and his advices this thesis would not have been possible. My sincerest appreciation to the Associate Professor Christina Kalogeropoulou and the Assistant Professor P. Zampakis from the Department of Radiology, for giving me the opportunity not only to have access to the patient s data regarding the specific Computed Tomography Angiography exams but also for their valuable discussions regarding the medical aspects of the non-invasive angiographic procedures. Last but not least I would like to thank the Radiation Technologists, Ms Vasiliki Vavatsikou and Mr Epameinondas Ntzanis for their valuable help during the collection of the specific exams.

5 ABSTRACT.. 1 ΠΕΡΙΛΗΨΗ 2 GENERAL PART 1. INTRODUCTION DΕVELOPMENT OF CT SCANNER TECHNOLOGY Design of CT Scanners Helical CT Multi Slice CT MSCT Detector CT scanner collimation and filtration Scan Parameter Tube current Time product (Q) Tube potential (kvp) Slice collimation (hcol) and slice thickness (hrec) Gantry rotation time Examination Parameters Scan Length (L) Number of scan series (nser) Number of rotation in dynamic CT studies (n) Definition of helical pitch CT image and display Image Reconstruction Image reconstruction filter Automatic Tube Current Modulation Principles of ATCM system for different CT manufacturers Toshiba scanner General Electric (GE) scanner COMPUTED TOMOGRAPHY ANGIOGRAPHY (CTA) Introduction Helical CT Angiography: Principles and Techniques Technique of Multi detector CT Angiography Scan Speed Contrast Material Injection DOSIMETRIC QUANTITIES Introduction CT Dose Descriptors Computed Tomography Dose Index (CTDI) CTDIFDA CTDI Weighted CTDI (CTDIw) Volume CTDI (CTDIvol) Dose length product (DLP) Organ and Tissue Dose Equivalent Dose Effective Dose Diagnostic Reference Levels (DRLs) [i]

6 SPECIFIC PART 5. PATIENTS & METHOD Systems description Toshiba (Aquilion Prime) GE (LightSpeed 16) Protocols Dose verification reports RESULTS & DISCUSSION Patient data collection Carotids Toshiba Chest Toshiba Abdomen Toshiba Carotids GE Chest GE Abdomen GE Comparison between systems Carotids - Toshiba vs. GE Chest - Toshiba vs. GE Abdomen Toshiba vs. GE Comparison with bibliography Carotids Toshiba & GE Chest Toshiba & GE Abdomen Toshiba & GE CONCLUSIONS AND FUTURE WORKS 89 APPENDIX I 90 REFERENCES. 91 [ii]

7 ABSTRACT Computed Tomography Angiography (CTA) is an imagig modality that combines a CT scan with an injection of a special dye to produce pictures of blood vessels in a part of the body. During the last years CTA examinations have increased a lot, so there is need for optimization of the patient dose. In this study the institutional doses for CTA examinations for carotids, chest and abdomen, in the University Hospital of Patras are presented. The systems that have been used were Toshiba Aquilion Prime (80 slices) and General Electric Lightspeed 16 (16 slices). Totally 175 patients participated in this study, 30 of each examination for the Toshiba scanner and 85 for the GE scanner. The patient characteristics age, sex, body mass index (BMI) for the Toshiba scanner and sex for the GE scanner, were recorded. To calculate the patient dose, the dose indexes utilized were Computed Tomography Dose Index (CTDIvol), Dose Length Product (DLP) and we calculate the Effective Dose (ED) using the DLP values. The 75 th percentile of DLP distribution was calculated for comparison purposes. For Toshiba CT scanner the 75 th percentile of CTDIvol values in carotids, chest, abdomen examinations were 5.45 mgy, 5.88 mgy, 7.05 mgy, whilst the DLP values were mgy cm, mgy cm, mgy cm, respectively. For the GE scanner the corresponding CTDIvol values were mgy, mgy, mgy, whilst the DLP values were mgy cm, mgy cm, mgy cm, respectively. The dosimetric indexes (CTDIvol, DLP) for the three types of examinations (carotids, chest and abdomen) carried out in our hospital utilizing an 80- slice CT system were lower than that of the 16-slice CT system by (85%, 85%), (54%, 50%) and (69%, 69%) respectively for the three types of examinations. Specifically, for the specific 80-slice CT system we have in particular: a) for the carotid examination, the DLP value of mgy cm (which is related with the evaluation of ED) was found to be lower than that of the comparing studies except one which was 37% lower using an 320-slice CT system, b) for the chest examination, the DLP value of mgy cm was found to be lower than that of the comparing studies except one which was 16% lower using a 64-slice CT system mainly to the smaller SL, c) for the abdomen examination, the DLP value was found to be lower than that of the comparing studies except one study which was 19% lower using an 128-slice CT system. [1]

8 ΠΕΡΙΛΗΨΗ Η αξονική αγγειογραφία είναι ένας τύπος ιατρικής εξέτασης που συνδυάζει CT σάρωση με σκιαγραφικό για την παραγωγή εικόνων από αιμοφόρα αγγεία σε ένα μέρος του σώματος. Τα τελευταία χρόνια έχει αυξηθεί σημαντικά ο αριθμός των εξετάσεων αξονικής αγγειογραφίας, έτσι κρίνεται αναγκαίος ο περιορισμός των δόσεων. Σε αυτή τη μελέτη εξετάστηκαν οι δόσεις των ασθενών για καρωτίδες, θώρακα και κοιλιά σε δύο διαφορετικούς αξονικούς τομογράφους, στο Πανεπιστημιακό Γενικό Νοσοκομείο Πατρών. Τα συστήματα που χρησιμοποιήθηκαν ήταν ο Toshiba Aquilion Prime (80 τομές) και ο General Electric Lightspeed 16 (16 τομές). Οι ασθενείς που έλαβαν μέρος σε αυτή τη μελέτη ήταν 175. Έγινε καταγραφή του ύψους, βάρους, φύλου και της ηλικίας του κάθε εξεταζομένου για τον Toshiba και μόνο το φύλο γι αυτούς του GE. Για τον υπολογισμό της δόσης των ασθενών οι δοσιμετρικοί δείκτες που χρησιμοποιήθηκαν ήταν το CTDIvol, το DLP και υπολογίσαμε την ενεργό δόση (ED). Για τον υπολογισμό της δόσης και τη σύγκριση μεταξύ τους αλλά και με άλλες μελέτες ορίστηκε ως η τιμή αυτή που αντιστοιχεί στα τρία τέταρτα των κατανομών των μετρήσεων. Τα αποτελέσματα έδειξαν ότι τα τρία τέταρτα των κατανομών της τιμής του CTDIvol στις εξετάσεις καρωτίδων, θώρακα και κοιλίας στον αξονικό τομογράφο 80 τομών ήταν: 5.45 mgy, 5.88 mgy, 7.05 mgy, ενώ οι αντίστοιχες τιμές DLP ήταν mgy cm, mgy cm, mgy cm. Για τον αξονικό τομογράφο 16 τομών οι αντίστοιχες τιμές του CTDIvol ήταν : mgy, mgy, mgy, ενώ οι τιμές του DLP ήταν mgy cm, mgy cm, mgy cm, αντίστοιχα. Οι δοσιμετρικοί δείκτες (CTDIvol, DLP) για του τρεις τύπους εξετάσεων(καρωτίδες, θώρακας, κοιλιά) που πραγματοποιήθηκαν στο νοσοκομείο μας χρησιμοποιώντας τον αξονικό τομογράφο 80 τομών ήταν χαμηλότεροι από αυτούς του αξονικού τομογράφου 16 τομών κατά (85%, 85%), (54%, 50%) και (69%, 69%) αντίστοιχα για τους τρεις τύπους εξετάσεων. Ειδικότερα για τον αξονικό τομογράφο 80 τομών : α) για τις καρωτίδες η τιμή του DLP ήταν χαμηλότερη από αυτή των αντίστοιχων μελετών, εκτός μιας η οποία ήταν 37% μικρότερη χρησιμοποιώντας σύστημα 320 τομών. β) για το θώρακα η τιμή του DLP ήταν χαμηλότερη από άλλες μελέτες, εκτός μιας η οποία ήταν 16% χαμηλότερη χρησιμοποιώντας ένα σύστημα 64 τομών με μικρότερο μήκος σάρωσης γ) για την κοιλιά η τιμή του DLP ήταν μικρότερη από άλλες μελέτες, εκτός μιας η οποία ήταν 19% χαμηλότερη χρησιμοποιώντας ένα σύστημα 128 τομών. [2]

9 GENERAL PART 1. INTRODUCTION Computed Tomography (CT) scanner technology has been developed significantly since the first CT scanner was constructed in early 1970s (Kalender 2006). Initial CT scanners were single slice axial, but technological development has seen the introduction of helical and multi-slice models. Modern scanners are capable of imaging simultaneously 64, 128 or even 320 parallel slices in one rotation (Geleijns et al 2009). Beam width has increased significantly from a standard of 10 mm to current beam widths of up to 160 mm. The use of CT has been increasing rapidly; there have been 12-fold and 20-fold increases in CT in European countries and the United States over the last 20 years (Hall and Brenner 2008). Moreover, CT is a high radiation dose examination and makes the largest contribution to the patient radiation dose from medical exposures. CT now accounts for 50%, 68% and 70 % of the collective dose in European countries, the United Kingdom and the United States, respectively (Martin 2008, Hart et al 2010). Because the use of CT has been increasing rapidly, there has been growing concern about potential health effects from the high doses that can be delivered (Amis et al 2007, Berrington de Gonzalez et al 2009, Smith-Bindman et al 2009), and patient dose from CT examinations has become a cause for concern among radiological professionals. The standard method for CT dosimetry measurement has been the CT dose index (CTDI). This is designed to measure the output for a single CT slice or a limited number of slices, but is also used for measurements inside phantoms simulating parts of the body for the purpose of patient dose assessment. Scans with an axial slice include most of the radiation within the length of the standard 100 mm pencil chamber used for the measurement. The purpose of the present study is the estimation of Institutional (Dose) Levels using the CTDI vol and DLP values from the most frequent CTA examinations at the University Hospital of Patras by two different scanners of 16 and 80 slices respectively. [3]

10 2. DΕVELOPMENT OF CT SCANNER TECHNOLOGY 2.1. Design of CT Scanners The first CT scanner was developed in the early 1970s by Hounsfield, a computer engineer in England (Kalender 2006). The first generation used a pencil beam having only one detector acquired image data by a translate-rotate method. The combination of the x-ray tube and detector moved in a linear motion across the patient (translate) and this was followed by a one-degree rotation and this procedure repeated for 180 degrees. The total scan time was more than 24 hours in the first generation of CT. The second generation of the CT was introduced in The x-ray source was changed from the pencil beam to a narrow fan shaped beam, together with multiple detectors. The principle of the second-generation CT was still translate and rotate. The beam irradiated a line of detectors, so the number of translation step could be reduced and this gave a significant decrease in total scan time. Instead of sampling a transmission profile from the pencil and the narrow fan beam, a larger fan beam coupled with a large array of detector arc in the third generation was installed in These are able to measure a complete projection, the translation motion becomes obsolete and the systems are operated with only a rotation. The x-ray tube and detector array rotates as one through 360 degrees. In the fourth generation of CT, the detector configuration was changed into a stationary circular array of fixed detectors completely surrounding the patient. With the third and fourth generations of CT scanners, data accumulation times as short as 1 second are achievable. However, there is a disadvantage in the fourth generation CT scanner, since there is a need for many more detector elements, and also the x-ray tube is closer to the patient than the detectors, a geometric magnification and also scatter artefact are large (Hendee and Ritenour 2003). After the year 1975 CT scanners have been developed from the concept of the third CT generation and were single slice axial with the detector array containing long elements along z-axis. Both of the x-ray tube and the detectors which are opposite to each other rotate around the patient, the scan is taken slice by slice and after each slice the scan stops and moves to the next slice and information is obtained. This is called conventional or sequential CT. [4]

11 Figure 1. CT scanners in the (a) first, (b) second, (c) third and (d) fourth generations Helical CT In 1989, technological development has seen the introduction of helical CT, with slip-ring technology invention in Slip rings are electro-mechanical devices. The design consists of sets of parallel conductive rings concentric to the gantry axis which connect to the x-ray tube, detectors and control circuit by sliding contactors. The x- ray source detector assembly is allowed to rotate continuously and a continuous scan is taken. The patient is moved continuously through the scan field in the z-direction while the gantry performs multiple rotations in the same direction in a spiral fashion. The helical CT is referred to as volume scanning, a potential advantage of [5]

12 the helical CT technique is a reduction of patient motion as it is a much quicker process. In a conventional CT scanner, as discussed earlier, the slice would be moved into a particular z position, and the gantry rotated through 360 degrees to acquire all projections. However, for spiral scanning, new projections are interpolated from those available at z-positions different from that of the reconstructed slice. The simplest approach is to estimate a value at a certain position using known data from nearby points of linear interpolation algorithm to derive an interpolated. A slightly more sophisticated approach is to recognise that points repeat every 180 degrees, half the fan angle, and interpolate new rays from projections in opposite directions (Keat 2005a, Keat 2005b) Multi Slice CT Multi-slice CT (MSCT) has been introduced by Elscint since 1992 (Kalender 2006). The MSCT can be called in other terms such as multi row CT, multi detector row CT (MDCT). Because a helical CT scan covering the patient body is a high workload for an x-ray tube, a limitation is imposed by x-ray tube heating. Solutions to this heat issue are to develop x-ray tubes with higher heat capacities or more effectively widen the x-ray beam in the z-direction. When multiple detector rows can be used, CT can collect data from several slices at the same time. Therefore, the scan time and the heat requirement of x-ray tube is reduced. Under these conditions, the projections are not collected on a slice by slice basis. However, virtual projections can be constructed for each required reconstructed slice by suitable interpolation from the adjacent projections MSCT Detector MSCT is different from single slice CT (SSCT) in terms of the detector configuration as illustrated in figure 2-2a, the detectors of MSCT are in an array segmented in the z axis which means there are more rows of detectors next to each other allowing for simultaneous acquisition of multiple images in the scan plane with one rotation (Goldman 2008, Bongartz et al 2004). An early detector design of the first modern MSCT from GE scanner consists of 16 rows of equal 1.25 mm elements in z axis (figure 2-2b), for the acquisition of 4 slices, [6]

13 Figure 2. (a) Single and eight-slice detector scanners and (b) multi detector of 16 rows, 4-slice from GE scanner. the combinations of slice widths that can be acquired simultaneously are: mm, mm, mm and 4 5 mm. These can acquire up to 4 images per rotation with the maximum beam width of 20 mm. The results from thin slices can be combined to get thicker slices (ImPACT 2002). Another detector configuration used at the beginning of the MSCT era from a Siemens scanner is called an adaptive array detector which consists of detector elements of different sizes. Individual elements of the adaptive array can be linked to acquire four slices simultaneously.16-slice GE CT scanners were introduced in 2002, detector element arrays are in the form of an adaptive array which consists of 16 x millimetre inner detector elements, and eight outer elements which are double the size of the inner elements. This allows the simultaneous recording of 16 thin slices. Alternatively, the inner 16 elements can be linked to get thicker slices. By 2005, 64 slice CT scanners were announced. Detector array configurations of some manufacturers are shown in figure 2-3. The design of Siemens scanner is different. They use a periodic motion of the focal spot in the longitudinal direction (zflying focal spot) to double the number of simultaneously acquired slices. Each of the 32 detectors collects two measurements separated by 0.3 mm, therefore the net result gives a total of 64 slices (Goldman 2008). At the present time, modern CT scanners are capable of imaging simultaneously 128 or even 320 parallel slices in one rotation (Geleijns et al 2009). Beam width has increased significantly from a standard of 10 mm to current beam widths of up to 160 mm. [7]

14 2.5. CT scanner collimation and filtration Collimation reduces unnecessary radiation to patients and improves the quality of the images. A distinction can be made between two types of collimator in CT scanners, pre-patient and post-patient collimators. The pre-patient collimator effectively limits the amount of x-radiation that reaches the patient. The post patient collimator which is positioned directly in front of the detectors is used to block scattered photons from reaching the detectors, thus preventing image artefacts. There are two major types of filtration utilized in CT scanner; mathematical and physical filters. The physical filter is discussed in this section. Beside an inherent filtration provided by the x-ray tube, there are two filters inserted for CT scanners. The common filter is a flat filter which is made from aluminium or copper. This filter is used to remove low energy photons that will be absorbed in the patient or reduce the beam-hardening artefact (Bongartz et al 2004). The other type is a bow-tie filter or beam shaping filter which is made from polytetrafluoroethylene (PTFE), aluminium or Teflon. Because the human body shape is elliptical, the radiation passing through such a shape is not uniform. The bow tie filter is used for shaping the beam in the x-y plane; it absorbs off axis the radiation at the edge of field of view. The advantages of using the bow-tie filter are maintaining a more uniform x-ray field at the detector. Modern CT scanners have two or three different bow-tie filters which are implemented automatically for clinical scan protocols. Some scanner manufacturers provide bow-tie filters based on field of view. Therefore, choosing the appropriate protocol or scan field of view (FOV) that matches the body part being scanned can introduce the correct bow tie filter. The GE scanners have two bow-tie filters which are used for head and body scans (ImPACT 2009). For the Toshiba Aquilion scanner, there are two bow-tie filters, small and large filters, depending on the scanning field size. The small filter is used for extra small (SS), small (S) and medium (M) FOVs. The large filter is used for large (L) and extra large (LL) FOVs. The filter movement motor is a stepping motor. This motor rotates to set the appropriate filter position (ImPACT 2009). [8]

15 Figure 3. Different sizes of bow-tie filter or beam shaping filter Scan Parameter Tube current Time product (Q) As in conventional radiology, a linear relationship exists between the tube current time product and dose. The mas product Q for a single sequential scan is obtain by multiplying the tube current I and exposure time t; in spiral scan mode, Q is the product of the tube current I and rotation time trot. The tube current time product is often used as a surrogate for the patient dose. With the advent of multi-slice scanners, additional confusion arose due to the introduction of a different pitch correct mas notation 'effective mas', or mas per slice. Tube current time product should be adapted to characteristic of the scanner, the size of the patient and the dose requirement (Hausleiter et al 2009) Tube potential (kvp) In conventional projection radiology, when the tube potential is increased, both the tube output and the penetration power of the beam are improved, while the image contrast is adversely indicated. Increased tube potential applied in order to ensure reduce patient dose. In CT, increased tube voltages are used preferentially for improvement in tube loading and image quality, the consequence of variation in kvp cannot easily be assessed. The relationship between dose and tube potential U is not linear, but rather of on exponential nature which varies according to the specific circumstances. [9]

16 Decreasing tube current by 50% will essentially decrease radiation dose by 50%. 80 kvp instead of 120 kvp would allow reduction of the patient dose by almost a factor of two without sacrificing image quality. Tube potential other that 120 kvp should be considered only in the case of: Obese patients in whom mas cannot be further increased: use higher kvp setting. Slim patients and pediatric CT, when mas cannot be further reduced: uses lower kvp setting. CT angiography with iodine: use lower kvp setting. Variation in tube potential should not be considered for pure dose reduction purposes except in the case of CT angiography (Hausleiter et al 2009) Slice collimation (hcol) and slice thickness (hrec) With single slice CT (SSCT), the slice collimation hcol used for data acquisition and the reconstructed slice thickness hrec used for viewing purposes were identical. So, there was no need to distinguish between them. With MSCT, the slice collimation (e.g. 0.75mm) and the reconstructed slice thickness hrec (e.g. 5 mm) are usually different. Frequently, the selection of hrec is made with aspect to Multi Planar Reformation (MPR) purposes. As reduced slice thickness is associated with increased image noise, this may have a significant impact on patient dose. Viewing should be preferentially be made with thicker slabs thereby reducing image noise and other artifacts, thinner skips should only be used if partial volume effect is of importance. Except for very narrow slices, there should be no need for any increase in dose settings on reduction of slice thickness. (Hausleiter et al 2009) Gantry rotation time Decreasing gantry rotation time decreases radiation dose in linear fashion. The faster the gantry rotation, the lower the dose increasing the cycle speed of rotation from seconds per 360 o rotation reduced the dose essentially by 50%. Pitch (P): With SSCT scanners, scanning at increased pitch settings primarily severe to increase the speed of data acquisition. MSCT scanners make use of a spiral interpolation scheme that is different from SSCT. Scanners that make use of effective mas (mas per slice) concept not only keep slice profile width, but also image noise constant when pitch change. Pitch setting with MSCT scanners should be made exclusively with respect to scan speed, spiral artifacts and tube power. [10]

17 2.7. Examination Parameters Scan Length (L) CTDI is almost independent of length of the scanned body section. DLP, and effective dose, both increase in proportional to the length of body section. Therefore, limiting the scan length according to the clinical needs is essential. For each patient, the scan length should be selected individually, based on the scan projection radiography that is generally made prior to scanning for the purpose of localization, as should be kept as short as necessary. Moreover, a reduction in the scan range should be considered in multi-phase examination and follow-up studies (Hausleiter et al 2009) Number of scan series (nser) In CT terminology, a scan series is usually referred to as a series of consecutive sequential scan or one complete spiral scan. The number of scan series (Phase) should be kept as low as necessary. This holds true particularly for liver examination, where studies with up to six different phases are sometimes recommended in literature (Hausleiter et al 2009) Number of rotation in dynamic CT studies (n) In dynamic CT studies, e.g. in CT fluoroscopy or in perfusion studies, a multiple number of scans is made at the same position. Therefore, it is meaning full to sum up the local doses. The doses applied in dynamic studies depend on two factors, i.e. the CTDI per rotation and number of rotations. As perfusion studies are regularly made with administration of contrast agents, the benefits of reduced kvp setting should be used to reduce the dose settings. The number of rotation can be kept low by limiting the total length of the study and by reducing the image acquisition rate. Dynamic CT studies should be made with the lowest dose setting, the most narrow beam width, the shortest length and the smallest images rate that is compatible with clinical needs of examination (Hausleiter et al 2009) Definition of helical pitch Currently, manufacturers of multi-slice systems employ two different definitions of pitch; detector pitch and beam pitch. The beam pitch is determined solely by the x-ray [11]

18 collimation and table speed, whereas the detector pitch will also depend on the number of slices acquired per rotation. The detector pitch is similar to the beam pitch on a single slice scanner. Beam Table movement per rotation pitch X ray beam width For the beam pitch, a pitch of less than 1 and a pitch of greater than 1 imply overlaps and gaps between the x-ray beams from adjacent rotations. Therefore, Detector pitch = Table movement per rotation Detector acquisition width Detector pitch = Beam pitch No. of slice acquired 2.9. CT image and display Each CT slice represents a specific plane in the patient s body. The thickness of the slice is referred to the z-axis. The data that form the CT slice are divided into elements, each one of these two-dimensional squares is a pixel (picture element) with the width indicated by x and height by y. The CT image is displayed on the monitor composed of pixels. If the z-axis is taken into account, this element is referred to as a voxel (volume element). Image displayed from CT scanners is normally 512x512 in size at 16 bits per pixel therefore they contain 65,536 shades of grey scale or brightness. The level of brightness can be called various names such as pixel value, grey scale value, digital number or Hounsfield unit. The CT image does not show the linear attenuation coefficient (μ) values directly but values used are called CT numbers in Housefield units. A viewer can adjust how grey levels are to be allocated by specifying a window width or a range of CT numbers (maximum - minimum) that are distributed over the viewable grey scale, for example 100 to 200, and a window level or the CT number in the centre of the viewable grey scale. CT number contains the μ-value of the underlying tissue in every volume element with respect to the μ-value of water. For elements having μ-values less than that of water, CT number is negative, air is a good example. Conversely, for substances having μ-values greater than that of water, CT number is positive, for example the bone (Dendy and Heatson 2012, Kalender 2006). [12]

19 CT number is calculated by the equation CT number = (μ μ water ) 1000 μ water Some typical values for CT numbers are given in table 1. Table 1. Typical CT number for different biological tissues Tissue Hounsfield Units Range Air Lung -200 to -500 Fat -50 to -200 Water 0 Muscle +25 to +40 Bone +200 to Source: Dendy and Heaton Image Reconstruction There are two major steps for image processing. The first step is data acquisition or record of projections and the second step is image reconstruction from projection. There are two major groups of reconstruction methods; analytic reconstruction, filtered back projection (FBP) and iterative reconstruction (IR) (Dendy and Heaton 2012). CT reconstruction has traditionally been performed by FBP. It is fast but dose reduction is difficult with this technique, as reduction results in a readily perceived increase in noise. It is a modification of an older technique, called back projection or simple back projection in which an individual sample is back projected along the ray pointing to the sample to the same value. A back projected image is very blurred. Filtered back projection is a technique to correct the blurring; each view is filtered before the back projection. The procedure is to first convolve each projection with a selected filter function before back projecting convolution result to form an image. The selection of the proper filter is the key to obtaining a good reconstruction from filter (convolution) back projection (Smith 1997). Iterative reconstruction is more versatile but is a slower process. Commercial names for individual CT scanners are iterative Dose (idose) in Philips, Iterative Reconstruction in Image Space (IRIS) and Sinogram Affirmed Iterative Reconstruction (SAFIRE) in Siemens, Adaptive Iterative Dose Reduction (AIDR) in Toshiba, and Adaptive Statistical Iterative Reconstruction (ASiR) in GE scanners. In principle, iterative reconstruction is an algorithm whereby image data are modified through the use of advanced mathematical models. The principle of the iterative [13]

20 technique algorithm is to find a solution by successive estimates. In each cycle, the projections of the current estimate are compared with the measured projections. It can either subtract or divide the corresponding projection in order to obtain correction factors. The result of the comparison is used to modify the current estimate, thereby creating a new estimate and is used to update the original image Image reconstruction filter The produced images by direct back-projection of attenuation profiles are unsharp, and to counteract this the profiles are convolved with a high pass filter. The choice of convolution filter kernel affects the image characteristics. A sharper filter will increase spatial resolution or edge enhancement, but also increased image noise. There are five different types of kernels for basic protocols of the Siemens scanner; H, B, C, S and T which refer to Head, Body, Child Head, Special Application and Topogram. The image sharpness is defined by numbers, the higher the number the sharper the image, while the lower the number, the smoother the image. The endings s f and h indicate standard, fast and high-resolution modes. For the body scans, standard kernels B30s or B40s are recommended. Smoother images are obtained with B20s. The Filter convolution (FC) for the Toshiba scanners can be split into two major groups, with and without beam hardening correction (BHC). For the body scan, they can also divided into body and soft tissue filters; FC01-FC05 (Body filter with BHC), FC11-FC15 (Body filter without BHC), FC07-FC09 (Soft tissue filter with BHC) and FC17-FC19 (Soft tissue filter without BHC), the lower the number of FC the smoother image and the higher the number of FC the sharper image. The first number after FC indicates whether the BHC is used or not, e.g. FC01 and FC11 are the same reconstruction algorithm and the difference is whether BHC is used or not. The filter for the Philips scanner has two major groups; for the body scan and cardiac scan. There are various kernels for the body scan, for example filter A, B, C, D refer to very smooth, smooth, sharp and very sharp. The filter A is recommended for large patients, while filters B and C are recommended for the routine abdomen and pelvis. Filter D is for edge enhancement and the bone. They also have other filters, for example F and L which are sharp and very sharp for scans of the lungs, knee and shoulder. Selections of resolution which are standard, high and ultra-high can be made [14]

21 with each filter. The GE algorithm consists of soft, standard, detail, lung (chest), Bone, edge and bone plus Automatic Tube Current Modulation The Automatic Tube Current Modulation (ATCM) or automatic exposure control (AEC) systems have been another recent development in modern CT scanners. Tube current can be reduced while scanning regions of lower attenuation and increased with those of higher attenuation; the attenuation level depends on patient body size, body shape, and anatomic location. Results from many studies have shown that use of ATCM systems reduced patient dose by about 35%-60% for the body and 18% for the neck, across all sizes of patient, compared with the traditional fixed tube current techniques. These dose reductions vary between different studies and depend on the tube current being used for the fixed technique and the size of the patient. (Lee et al 2011, Soderberg Gunnarsson 2010, Rizzo et al 2006, Lee et al 2009, Peng et al 2009, Papadakis et al 2008) Principles of ATCM system for different CT manufacturers With the ATCM system, the traditional tube current selection is replaced by an input value which is on the basis of the required image quality. For ATCM systems, the tube current is automatically adjusted to the X-ray attenuation of the patient cross section being scanned leading to a potential for a reduction in radiation dose, while obtaining images with a consistent level of image quality (Kalender 2006). The principles of ATCM systems for different CT scanner manufacturers and differences in translation of terms used for different manufacturers are shown in table 2. ATCM can be divided into angular and longitudinal modulations which adjust the tube current as the X-ray tube rotates around the patient and along the longitudinal axis, respectively. [15]

22 Table 2. CT scan parameters from 2 CT scanners Toshiba Aquilion GE Lightspeed User interface exam Plan Exam Rx CT localizer scan projection Scanogram Scout radiograph Tube Current Tube current ma ma Effective Tube current time Effective mas (mas/pitch) - product Pitch CT Pitch Factor Pitch Automatic Tube Current Modulation (ATCM) Sure Exposure Auto ma/smart ma Principle of ATCM Constant target noise by varying the tube currents within the minimum and maximum limits, according to patient attenuation Angular tube current modulation Not as a separate item - Longitudinal tube current Sure Exposure Auto ma modulation Angular and Longitudinal tube Sure Exposure 3D Smart ma current modulation Image quality reference Standard deviation Noise Index parameter Image reconstruction property Filter Convolution (FC) Algorithm Toshiba scanner The Toshiba ATCM system SureExposure 3D modulates the tube current by patient size, shape and attenuation and the tube current is modulated along the longitudinal (z-direction) and axial (x,y) plane (figure 4) (Kalra et al 2004). If a single scanned projection radiograph (SPR) is used, SureExposure will modulate the tube current in the z-direction but if a dual AP and lateral SPR is used SureExposure will modulate the tube current in all three dimensions as the tube rotates and transverses the patient. [16]

23 Figure 4. Tube current (ma) modulation pattern in x-y plane and z-axis shown on the scanner monitor prior to the scan. SureExposure determines the relative attenuation of a patient from SPRs and converts them into a water equivalent thickness. For z-axis modulation, the water equivalent diameter at each level of the patient is calculated and compared to the maximum attenuation. The tube current required to achieve the selected standard deviation for the maximum water equivalent diameter is applied. Tube current is then modulated to maintain the target standard deviation throughout the examination. The image quality level can be automatically set for the clinical examination. Different target image standard deviation modes are available on the Toshiba Aquilion. These correspond to the selection of different preselected image noise levels; a) high quality (standard deviation (SD) =7.5 HU), b) quality (SD=10 HU), c) standard (SD=12.5 HU), d) low dose (SD =15 HU), e) ultra-low dose (SD=17.5 HU), or f) low dose ++ (SD=20.0 HU). The system also allows the user to set any standard deviation of pixel value (in Hounsfield unit) and a minimum and maximum (range) of the tube current. SureExposure aims to maintain the image quality. It is not a standalone tube current modulation algorithm. It incorporates the selected imaging and reconstructing parameters. The primary acquisition parameters affecting image noise are pitch factor, rotation time and kvp. For example, a higher pitch can reduce the scan time but increase image noise. SureExposure counteracts this effect by adjusting the tube current to achieve the target image quality regardless of the selected pitch value. Sure Exposure can also be incorporated with noise reduction tools. One of the noise reduction tools available on the Aquillion scanner is Quantum Denoising Software (QDS). QDS applies a combination of smoothing and enhancing filters for lower ma imaging, the image [17]

24 areas of soft tissue or with little edge are smoothed while sharper image areas are processed with edge enhancing filters. When QDS is used in a protocol, SureExposure decreases the tube current to account for the benefits gained from QDS General Electric (GE) scanner There are two elements of the GE ATCM system; Auto ma and Smart ma. The Auto ma provides longitudinal (z) axis modulation, whereas SmartmA enables both longitudinal and angular modulations. The quality of image depends on a selected noise index (NI). The reference NI which is a default or baseline NI for a given protocol is provided. The NI is defined as the standard deviation of pixel values in the central region of an image of a uniform water phantom (General Electric Company 2008). The system allows the user to set the new NI value by changing the NI value itself or adjusting dose steps. The dose step value of 0 indicates that the prescribed NI is equal to the reference NI for the protocol. When the dose step is decreased by 1, the ma decreased by 10% and the NI increased by 5%. The NI value is used for estimating the tube current. Only a SPR image is required for Auto ma. A table of tube current values can be previewed before scanning. For Smart ma, the system estimates the attenuation level and the oval ratio from SPR images. The attenuation level reflects the density and size of the patient. The oval ratio reflects how circular or elliptical the patient is at that level and is estimated from brightness and width information in the scout image. To determine the appropriate tube current, the system interpolates between the targeted NI relating noise to attenuation level and oval ratio. Using the relationships between image noise and mas, slice thickness, and pitch factor, the ma required to achieve the prescribed noise index is calculated. For this, the tube-current is modulated four times for each rotation along the angular and longitudinal directions (Bruesewitz et al 2008). There are two tube current values in the ma table information for the Smart ma, one for the y-axis (AP) and the other for the x-axis (lateral) directions. For Auto ma, the tube current is kept constant during each rotation and only changes along the longitudinal direction. Since Smart ma reduces the ma along the axis with less attenuation (typically the AP direction), the radiation dose is reduced by an additional amount relative to Auto ma. [18]

25 3. COMPUTED TOMOGRAPHY ANGIOGRAPHY (CTA) 3.1. Introduction CTA is a test that combines the technology of a conventional CT scan with that of traditional angiography to create detailed images of the blood vessels in the body. In a CT scan, x rays and computers create images that show cross-sections, or slices, of the body. Angiography involves the injection of contrast dye into a large blood vessel, usually in the leg, to help visualize the blood vessels and the blood flow within them. When the contrast dye is used to visualize the veins, the study is called a venogram, and when it is used to visualize the arteries, it is known as an arteriogram. CT angiography is similar to a CT scan, but the contrast dye is injected into one of the veins shortly before the x ray image is performed. Because the dye is injected into a vein rather than into an artery, as in traditional angiography, CT angiography could be considered less invasive Helical CT Angiography: Principles and Techniques CTA has been improved significantly with the introduction of four- to 64-section spiral CT scanners, which offer rapid acquisition of isotropic data sets. A variety of techniques have been proposed for postprocessing of the resulting images. The most widely used techniques are MPR, thin-slab maximum intensity projection (MIP), and volume rendering (VR). Sophisticated segmentation algorithms, vessel analysis tools based on a centerline approach, and automatic lumen boundary definition are emerging techniques; bone removal with thresholding or subtraction algorithms has been introduced. These techniques increasingly provide a quality of vessel analysis comparable to that achieved with intra-arterial three-dimensional rotational angiography. Neurovascular applications for these various image post processing methods include steno-occlusive disease, dural sinus thrombosis, vascular malformations, and cerebral aneurysms. However, one should keep in mind the potential pitfalls of these techniques and always double-check the final results with source or MPR imaging. In recent years, rapid advances in computed topographic (CT) technology and image post processing software have been made. CT angiography was improved substantially by increasing scan speed and decreasing section thickness and emerged as a powerful tool in neurovascular imaging. Nowadays, spiral CT systems with acquisition capabilities of up to 64 sections per gantry rotation are introduced in [19]

26 clinical practice. Gantry rotation times decreased to 0.33 second, and section widths of mm are available. Assessment of vascular studies based on axial images alone is not straightforward; two-dimensional (2D) and three dimensional (3D) visualization methods are routinely employed to create images comparable to those acquired with catheter angiography. In the emergency situation (stroke or subarachnoid hemorrhage), a robust and fast imaging technique capable of answering all vital clinical questions and allowing clear therapeutic decisions is mandatory. Optimal image quality depends on two factors: CT angiography technique (scan protocol, contrast material injection protocol, image reconstruction methods) and data visualization technique Technique of Multi detector CT Angiography An essential prerequisite for successful post processing is good quality of the acquired imaging data. While the short arteriovenous transit time in neurovascular applications makes short scan times preferable, the small caliber of cervical and intracranial vessels requires the highest spatial resolution in all three dimensions Scan Speed For evaluation of the basal intracranial arteries, a scan range of approximately 100 mm needs to be covered. With four detector row CT at a collimated section width of 1 mm, a pitch of 1.5, and a gantry rotation time of 0.5 second, this volume can be covered in about 9 seconds. Assuming a cerebral transit time of about 5 seconds, this is not fast enough to avoid venous overlay. With 16 detector row CT at a collimated section width of 0.75 mm, a pitch of 1.5, and a rotation time of 0.5 second, the same range can be covered in 3 seconds, well beyond the arteriovenous transit time. Examination of the whole length of the carotid arteries from the aortic arch to the circle of Willis requires a scan range of approximately 250 mm. With the abovementioned scan parameters, the scan time would be 21 seconds for four detector row CT, 7 seconds for 16 detector row CT, and 4 seconds for 64 detector row CT ( mm, pitch of 1.3, 0.33-second rotation time). The latter protocol allows contrast phase-resolved imaging. [20]

27 3.5. Contrast Material Injection Short scan times require short contrast material injection. The injection protocols need to be simple and standardized to guarantee excellent and reproducible results on a 24-hour basis. To deliver an appropriate amount of iodine, injection rates of 4 5 ml/sec and highly concentrated contrast medium (iodine, mmol/ml) are preferable. The utility of the contrast material bolus can be increased if a saline bolus is appended. Flushing of the veins reduces streak artifacts due to beam hardening, especially at the thoracic inlet. Individual timing of contrast material injection (bolus tracking or test bolus injection) is mandatory to take advantage of phase-resolved image acquisition. To individualize the timing of contrast material injection, automatic bolus tracking techniques (Smart Prep, CARE Bolus, and Sure Start) can be employed. These techniques are fast and easy to use and require only a single contrast material injection. The disadvantage is that a large target vessel for monitoring the contrast material arrival is required, and an additional delay for table movement and patient instruction is necessary. Test bolus injection is the alternative to assess the individual circulation time. Its major advantage is that it provides information about the timing of both arterial and venous enhancement in the vessels of interest. The individual start delay can be optimized by placing the scan between the arterial peak and venous contrast material upslope. Table movement and patient instructions can be performed prior to the optimal image acquisition window. The disadvantage is the necessity for an additional injection of about 10 ml of contrast agent (10% 20% increase of total amount). [21]

28 4. DOSIMETRIC QUANTITIES 4.1. Introduction Medical ionizing radiation sources provide by far the largest contribution to the population dose from artificial sources and most of this contribution comes from diagnostic X rays (above 90%). CT procedures result in organ doses in the range of mgy, generally below the level required to produce deterministic effects. Figure 3.1 shows the mean effective doses for some examination types, and Figure 3.2 the relative contributions to the total collective dose derived from all diagnostic X ray examinations, as reported in the UNSCEAR 2000 report 17 for the first health care level in Figure 5. Effective doses from diagnostic X ray examinations and interventional. [22]

29 Figure 6. Contribution to collective dose from various types of diagnostic X ray. The effective doses are the highest in the interventional procedures, angiography and CT. The differences between Figures 3.1 and 3.2 result from the different examination frequencies employed. For example, CT accounts for about 6% of the total frequency of all X ray examinations and for about 40% of the collective effective dose. In interventional procedures and angiography, the average effective doses per examination are higher than in CT but, owing to their low frequencies, the percentage contributions to the collective dose are only about 5 7%. As an example of a low effective dose, chest radiography accounts for about 30% of the total frequency but for only about 3% of the collective effective dose from all X ray examinations. It is generally recognized that even a 10% reduction in patient dose is a worthwhile objective for optimization. Numerous surveys have shown wide variations in the magnitude of absorbed dose to the patient for the same type of x-ray procedure performed at different facilities or even within the same facility. This has focused attention on the possibility of using reference values as guidance for the levels that can be achieved using good radiographic technique and appropriate x-ray equipment. As an aid to the optimization of absorbed dose to the patient, reference values (ICRP 1996) can be specified. The ICRP introduced the concept of diagnostic reference [23]

30 levels (DRLs) for patients as a tool for optimization and quality assurance, requiring regular assessment of patient doses and comparison with the reference dose levels CT Dose Descriptors Computed Tomography Dose Index (CTDI) The CTDI is the fundamental CT dose descriptor. By making use of this quantity, the first two peculiarities of CT scanning are taken into account: The CTDI [unit: milli- Gray (mgy)] is derived from the dose distribution along a line that is parallel to the axis of rotation for the scanner (= z-axis) and is recorded for a single rotation of the X-ray source. Figure 7 illustrates the meaning of this term: CTDI is the equivalent of the dose value inside the irradiated slice (beam) that would result if the absorbed radiation dose profile were entirely concentrated to a rectangular profile of width equal to the nominal beam width NxT, with N being the number of independent (i.e., non-overlapping) slices that are acquired simultaneously. Accordingly, all dose contributions from outside the nominal beam width, i.e., the areas under the tails of the dose profile, are added to the area inside the slice The corresponding mathematical definition of CTDI therefore describes the summation of all dose contributions along the z-axis 1 CTDI D( z) dz NT where D(z) is the value of the dose at a given location, z, and NxT is the nominal value of the total collimation (beam width) that is used for data acquisition. The value N is equal to the number of data channels used in a particular scan and may be less than or equal to the maximum number of data channels available on the system. The value T is equal to the width of the tomographic section along the z-axis imaged by one data channel. In multiple-detector-row (multislice) CT scanners, several detector elements may be grouped together to form one data channel. In single-detector-row (single-slice) CT, the z-axis collimation (T) is the nominal scan width. [24]

31 Figure 7. Illustration of the term Computed Tomography Dose Index: CTDI is the equivalent of the dose value inside the irradiated slice (beam) that would result if the absorbed radiation dose profile were entirely concentrated to a rectangular profile of width equal to the nominal width NxT. The CTDI is always measured in the axial scan mode for a single rotation of the x- ray source and is calculated by dividing the integrated absorbed dose by the nominal total beam collimation. It theoretically estimates the average dose within the central region of a scan volume consisting of multiple, contiguous CT scans [Multiple Scan Average dose (MSAD)] for the scan length is sufficient for the central dose to approach its asymptotic upper limit CTDIFDA It is a particular definition of CTDI given by FDA. It envolves the integration of D(z) over a distance of 14 times the slice thickness, where D(z) is the dose at a point z on any line parallel to the z (rotational) axis for a single slice of nominal thickness T. To account for this the CTDI value must be normalized to 1/NT. CTDI FDA 1 NT 7T 7T D( z) dz The scattering media for CTDI measurements were also standarized by the FDA. These consist of two polymethacrylate (PMMA, e.g., acrylic or Lucite) cylinders of 14-cm length. Two different standard phantoms exist, i.e., a head phantom with a diameter of 16 cm and a body phantom with a diameter of 32 cm. [25]

32 CTDI100 CTDI100 represents the accumulated multiple scan dose at a 100-mm scan and underestimates the accumulated dose for longer scan lenghts. It is thus smaller than the equlibrium dose or the MSAD. The CTDI100, like the CTDIFDA requires integration of the radiation dose profile from a single axial scan over specific integration limits. In the case of CTDI100, the integration limits are +-50 mm, which corresponds to the 100-mm length of the comercially available pencil ionization chamber. 50mm 1 CTDI 100 D( z) dz NT 50mm The use of a single, consistent integration limit avoided the problem of dose overestimation for narrow slice widths (e.g., <3mm). CTDI 100 is acquired using a 100-mm long, 3-cc active volume CT pensil ionization chamber and the two standard CTDI acrylic phantoms[ head (16-cm diameter) and body (32-cm diameter)]. The measurement must be performed with a stationary patient table Weighted CTDI (CTDIw) The CTDI varies across the field of view (FOV). Typically, the dose distribution within the body cross section imparted by a CT scan is much more homogeneous than that imparted by radiography, but is still somewhat larger near the skin than in the body centre. Therefore, a third measure, the weighted CTDI is introduced. CTDI 100 at the centre and periphery of standard PMMA phantoms either a 16 cm (head) or 32 cm (body) diameters are combined to give a measure relating to patient dose. The CTDI w is employed as a standard measure relating to patient dose (IEC 2003) and this is given by 1 3 CTDI w CTDI 100, center CTDI 100, periphery 2 3 [26]

33 Figure 8. Cylindrical standard CT dosimetry phantoms (16 and 32 cm in diameter) made from Perspex for representative measurements of CTDI in regions of the head and the trunk and a pencil-like detector for measurements of the dose profile integral Volume CTDI (CTDIvol) To represent dose for a specific scan protocol, which almost always involves a series of scans, it is essential to take into account any gaps or overlaps between the x- ray beams from consecutive rotations of the x-ray source. This is accomplished with use of a dose descriptor known as the Volume CTDIw (CTDIvol), where I is the table increment per axial scan (mm). CTDI vol CTDI According to the definition of pitch that is the ratio of the table travel per rotation (I) to the total nominal beam width (N x T) w NT I [27]

34 the Volume CTDI can be expressed as Pitch I N T CTDI vol CTDI w 1 pitch The difference between CTDIw and CTDIvol is that CTDIw represents the average radiation dose over the x and y directions at the center of the scan from a series of axial scans where the scatter tails are negligible beyond the 100-mm integration limit, whereas CTDIvol represents the average absorbed radiation dose over the x, y and z directions. The CTDIvol provides a single CT dose parameter, based on a directly and easily measured quantity, which represents the average dose within the scan volume for a standardized (CTDI) phantom. The SI units are milligray (mgy). CTDIvol is a useful indicator of the dose to a standarized phantom for a specific exam protocol, because it takes into account protocol-specific information such as pitch. Whereas CTDIvol estimates the average radiation dose within the irradiated volume for an object of similar attenuation to the CTDI phantom, it does not represent the average dose for objects of substantially different size, shape, or attenuation or when the 100-mm integration limits omit a considerable fraction of the scatter tails. In addition, it does not indicate the total energy deposited into the scan volume because it is independent of the length of the scan. Therefore, the value remains the same whether the scan coverage is 10 and 100 cm, it estimates the dose for a 100-mm scan length only Dose length product (DLP) To represent the overall energy delivered by a given scan protocol, the air kerma can be integrated along the scan length and Dose-Length Product (DLP) is computed. The DLP reflects the total energy (and thus the potential biological effect) attributable to the complete scan acquisition. The DLP is the product of the CTDIvol and the length of scan (slice thickness number of slices) in centimetres. DLP can be linked to the effective dose for different parts of the body (Huda et al 2008). It should be noted that the DLP is independent of patient size and age. In other words, the reported DLP is the same whether a child or an adult is scanned if the scan length and other scan parameters are the same. The relationship between patient size and effective dose is also a topic of interest. Recently, AAPM published a report on size-specific dose [28]

35 estimates (SSDE) it is not specific organ dose and effective dose but is a size dependent conversion factor to allow estimation of patient dose based on CTDIvol and patient size. For the same CTDIvol, a smaller patient will tend to have a higher patient dose than a larger patient (AAPM, 2011) since the mean dose in the center of the scanned volume is higher. There is a difference between the DLP and the product of the CTDIvol and the total imaged length. Typically, a scanner will need 1 or 2 extra rotations beyond the nominal imaged volume to gather sufficient data to reconstruct all images. An exposure time can be converted to a Scan Length (SL) with knowledge of a total collimation, spiral pitch and rotation time. collimatio n pitch exposure time Irradiated length rotation time DLP can be calculated by multiplying this by the CTDIvol. This will be slightly higher than the CTDIvol x the total imaged length since this includes overranging which is an extended scan length beyond the planed image boundaries to reconstruct the first and last sections of a helical CT scan, as mention earlier Organ and Tissue Dose The ICRP has recommended that the appropriate dosimetric indicator for the probability of stochastic radiation effects is the average absorbed dose in a tissue or organ. The mean absorbed dose in a specified organ or tissue T has been given the symbol D T and is defined either as the integral of the absorbed dose D t over the mass of the tissue divided by its mass m T D T 1 m T m t D dm t where D t is the absorbed dose at a point in tissue material t. The mean absorbed dose in a specified organ or tissue is further simply referred to as organ dose. The subscript T can be replaced by a specific organ, for example, stomach, D stomach Equivalent Dose The biological effects of an absorbed dose of a given magnitude are dependent on the type of radiation delivering the energy (i.e., whether the radiation is from x rays, [29]

36 gamma rays, electrons (beta rays), alpha particles, neutrons, or other particulate radiation) and the amount of radiation absorbed. This variation in effect is due to the differences in the manner in which the different types of radiation interact with tissue. The variation in the magnitude of the biological effects due to different types of radiation is described by the "radiation weighting factor" for the specific radiation type. The radiation weighting factor is a dimensionless constant, the value of which depends on the type of radiation. Thus, the absorbed dose (in Gy) averaged over an entire organ and multiplied by a dimensionless factor, the radiation weighting factor, gives the equivalent dose. The unit for the quantity equivalent dose is the sievert (Sv). Thus, the relation is Equivalent dose (in Sv) = absorbed dose (in Gy) x radiation weighting factor. In the older system of units, equivalent dose was described by the unit rem and 1 Svequals 100 rem or 1 msv equals 0.1 rem. For x rays of the energy encountered in CT, the radiation weighting factor is equal to 1.0. Thus, for CT, the absorbed dose in a tissue, in Gy, is equal to the equivalent dose in Sv Effective Dose The relationship between the probability of stochastic effects and equivalent dose also depends on the organ or tissue irradiated. In medical x-ray imaging, more than one organ is often irradiated. It might therefore be useful to combine the doses to different tissues in such a way that the combined value is likely to correlate well with the total of the stochastic effects. For the radiological protection of workers and the whole population, the ICRP has defined the factor by which the equivalent dose in a tissue or organ has to be weighted, called the tissue weighting factor, w T. Effective dose, E, is a dose descriptor that reflects this difference in biologic sensitivity. It is a single dose parameter that reflects the risk of a non-uniform exposure in terms of an equivalent whole-body exposure. The effective dose is defined as the sum of the weighted equivalent doses in all the tissues and organs of the body. It is the sum over all the organs and tissues of the body of the product of the equivalent dose, H T, to the organ or tissue and a tissue weighting factor, w T, for that organ or tissue, thus: E Unit: J/kg. The special name for the unit of effective dose is sievert (Sv). T w T H T [30]

37 As it is desirable that a uniform equivalent dose to the whole body should give an effective dose numerically equal to that uniform equivalent dose (ICRP 1996), the sum of the tissue weighting factors is normalized to unity T w T The values of the tissue weighting factor proposed by the ICRP are independent of the type and energy of the radiation incident on the body. The consequences following an absorbed dose also depend on the distribution of the dose in time. The effect of all exposure conditions other than those dealt with by the radiation and tissue weighting factors is covered by using different values of the coefficients relating equivalent dose and effective dose to the probability of stochastic effects, rather than by using additional weighting factors in the definitions of the quantities. The values of both the radiation and tissue weighting factors depend on current knowledge of radiobiology and may change from time to time. Another way to calculate the effective dose for a particular scanning protocol from a measurement of CTDI air is by utilizing scanner-specific normalized organ dose data determined for a mathematical anthropomorphic phantom using Monte Carlo techniques. Alternatively, broad estimates of effective dose (ED) may be derived from values of DLP for an examination using appropriately normalized coefficients: ED E 1 DLP DLP where DLP (mgy cm) is the dose-length product and EDLP is the region-specific normalized effective dose (msv mgy -1 cm -1 ). General values of EDLP appropriate to different anatomical regions of the patient (head, neck, chest, abdomen or pelvis) have been published by the European Commission, Table 3. Table 3. Normalized values of effective dose length product (DLP) over various body regions Region of body Normalized effective dose, EDLP (msv mgy -1 cm -1 ) Head Neck Chest Abdomen Pelvis [31]

38 4.6. Diagnostic Reference Levels (DRLs) The International Commission of Radiation Protection introduced diagnostic reference levels (DRLs) and defined them as: Dose levels in medical radiodiagnostic practices, for typical examinations for groups of standard-sized patients or standard phantoms for broadly defined types of equipment. These levels are expected not to be exceeded for standard procedures when good and normal practice regarding diagnostic and technical performance is applied (IAEA 2007). For diagnostic radiology, these levels, which are a form of investigation level, apply to an easily measured quantity at the surface of a simple standard phantom or a representative patient. The ICRP recommends that DRL values should be selected by professional medical bodies and reviewed at intervals that represent a compromise between the necessary stability and the long-term changes in the observed distributions. DRLs are intended to act as thresholds to trigger internal investigations by departments where typical practice is likely to be well away from the optimum and where improvements in dose-reduction are probably most urgently required. Typical levels of dose in excess of a reference dose value should either be thoroughly justified or reduced. In general, patient doses should always be reduced to the lowest levels that are reasonably practicable and consistent with the clinical purpose of the examination. DRLs are aimed at the management of patient doses consistent with the clinical imaging information that is required. This means that in individual cases, the exceeding of DRLs may be justified in terms of a clinical requirement, for example, need for additional diagnostic information, or the unexpected difficulty of a procedure. Diagnostic reference dose values should not be applied locally on an individual patient basis, but rather to mean doses observed for representative groups of patients. (European Communities 1999). [32]

39 SPECIFIC PART 5. PATIENTS & METHOD Currently, for the establishment of national DRLs, the Greek Atomic Energy Commission (GAEC) receives from the hospitals dosimetric data regarding the following CTA examinations: coronary, head, carotids, lung arteries, thoracic aorta, abdomen aorta, liver and kidneys. At the University Hospital of Patras from the requested CTA examinations are performed the following CTA examinations: coronary, head, carotids, lung arteries, thoracic aorta, abdomen aorta, liver and kidneys. For the purposes of the present thesis it was decided the collection of dosimetric data from the following CTA examinations: carotids, thoracic aorta and abdomen aorta. To make an examination the first step is the patient registration. The system prompts the user to select the appropriate protocol based on patient age, weight and exam type. Dose is displayed on the console prior to scanning for operator confirmation and validation. The automatic reduction of the dose to the patient is accomplished using acquisition data with ma modulation based on the scanogram. Active Collimation limits helical over-ranging, reducing dose delivered to the patient on all helical scans. The study involves data on CTDIvol and DLP values for two different systems; however, it is limited in its applicability as a DRL as it represents practice from only a single institution. The mean, median and range included in the results are an important component in local dose audit. However, in this study the 75th percentile of distribution of CTDIvol and DLP values will be estimated, as it will be this data that will be used to compare with a national 75th percentile level (DRL) to ensure they lie below it on most occasions, as well as with the existing bibliography. In this study, the patient data was obtained from the report of the systems and acquired totally over a 9-month period in patients participated in this study, 90 for each CT scanner. For each type of examination, a random sample of 30 standard sized patients was taken. Patient data sex, age, weight, height and body mass index (BMI), CTDIvol and DLP were collected. The BMI was calculated by dividing patient s weight by patient s height squared (in kg/m 2 ). [33]

40 5.1. Systems description Toshiba (Aquilion Prime) The CT scanner featuring a 72 kw generator, 7.5 MHU tube and minimum rotation time of 0.35 s. It is capable of imaging 80 slices per rotation. A summary of the main features of the scanner are presented below. Toshiba features Application Aquilion TM PRIME is a multislice helical CT system with an 80-row detector capable of generating 160 slices per rotation using the conexact TM reconstruction algorithm. Features Aquilion PRIME incorporates a variety of functions based on technologies that were developed for Aquilion ONE with the aim of significantly reducing the patient exposure dose, including Active Collimator, AIDR 3D (Adaptive Iterative Dose Reduction 3D), SURE Exposure TM 3D and Boost3D TM. During helical scanning, Active Collimator blocks X-rays not required for image reconstruction by limiting the extent of the X-ray beam at the start and end of the helical scan range. AIDR 3D uses an iterative algorithm to reduce image noise while maintaining details and structural edges. AIDR 3D can be applied to all acquisition modes for routine clinical use and is able to remove up to 50% of image noise, resulting in dose reduction of up to 75%. It delivers an integrated solution to facilitate diagnostic decision-making at the lowest possible radiation dose without compromising image quality. During helical scanning, SURE Exposure 3D continuously adjusts the exposure in the X, Y and Z direction based on the patient's body shape, reducing the patient dose to the lowest possible level. Boost3D allows the X-ray dose to be minimized for regions with high X-ray absorption, such as the shoulders and pelvis, while permitting acquisition of highly accurate images. Performance specifications 1) Scan parameters Scan types - Scanoscopy - Conventional scan: S & S: Mode with priority on time control between one scan and the next S & V: Mode with priority on image display after the scan - Volume scan: Mode for scanning the volume - Dynamic volume scan: Mode for continuously or intermittently scanning a volume - Helical scan: Mode for continuously scanning while the patient couch moves Rotation time [34]

41 - Conventional scan, Volume scan Half scan 0.23 s (conventional scan only) Full scan 0.35, 0.375, 0.4, 0.45, 0.5, 0.6, 0.75, 1.0, 1.5, 2.0, 3.0 s Slice thickness: - Conventional scans (S & S, S & V): 0.5, 1, 2, 3, 4, 5, 8 and 10 mm - Volume scan, Helical scan: 0.5 and 1 mm - Dynamic volume scan: 0.5, 1, 8 and 10 mm These slice thicknesses are implemented by stacking the data acquired in one of the following acquisition modes. Acquisition - Conventional scan (S & S, S & V) 4-row scan 0.5, 1, 2, 3, 4, 5, 8 and 10 mm 1-row scan 1, 2, 4, 6 and 8 mm - Volume scan 80-row scan 0.5 mm 40-row scan 1 mm - Dynamic volume scan 80-row scan 0.5 mm 40-row scan 1 mm 4-row scan 8 and 10 mm - Helical scan 80-row scan 0.5 mm 64-row scan 0.5 mm 40-row scan 0.5 and 1 mm 32-row scan 0.5 and 1 mm 20-row scan 0.5 and 1 mm 16-row scan 0.5 and 1 mm 4-row scan 0.5 mm Gantry tilt angle: ±30 o, remote control from the console is possible. Tube position for scanoscopy: 0 o, 90 o, 180 o and 270 o (preset), a desired angle can be specified (in 5 o increments). Gantry aperture: 780 mm in diameter Patient positioning projector: Laser, external and internal. 2) Dynamic volume scan Scan time: 0.35, 0.375, 0.4, 0.45, 0.5, 0.6, 0.75, 1, 1.5 s/360 Scan start time delay: Min. 0.5 s, Setting is possible in increments of 0.1 s. Image reconstruction-image interval: Reconstruction is posssible in increments of 0.05 s. 3) Helical scan Rotation time: 0.35, 0.375, 0.4, 0.45, 0.5, 0.6, 0.75, 1, 1.5 s/360 Continuous scan time: Max. 100 s Scan start time delay: Min. 1 s, setting is possible in increments of 0.1 s. Active Collimator: To reduce the exposure dose, the collimator operate asymmetrically at the start/end of scanning (except in the case of 4-row scanning). Couch-top speed: 0.8 mm/s to 160 mm/s [35]

42 CT pitch factor and Helical pitch - For all rotation speeds Scan rows CT pitch Helical pitch to to to to to to to to to to to to to to to to to to to to to to to to to to to to 6.0 Helical pitch = Couch-top movement (mm/rot.)/nominal scanning slice thickness (mm) CT pitch factor = Helical pitch/number of slices per rotation Note: Helical pitch setting is possible in increments of 0.1. (For 4-row scanning, setting is possible in increments of 0.5.) SURE Exposure 3D: Function for continuously varying the X-ray tube current to ensure the minimal X-ray dose during helical scanning Image reconstruction time: Up to 60 images/s with AIDR 3D (0.02 s/image) (depending on the scan and reconstruction conditions) Real-time helical reconstruction time: 12 images/s (0.083 s/image) (1 slice, matrix) Reconstruction position setting: Can be set in increments of a minimum of 0.1 mm by entering the couch-top position or using the scanogram. Reconstruction interval setting: Can be set in increments of a minimum of 0.1 mm. Reconstruction method MUSCOT reconstruction (4-row scanning) TCOT+ reconstruction (80-, 64-, 40-, 32-, 20-, and 16-row scanning; fast mode) V-TCOT reconstruction (80- and 40-row scanning; high image quality mode) 4) X-ray generation Max power: 72 kw X-ray beam shape Fan beam Channel-direction angle (fan angle): 49.2 X-ray exposure: Continuous [36]

43 X-ray tube voltage: 80, 100, 120 and 135 kv X-ray tube current: 10 ma to 600 ma (adjustable in 5-mA increments from 10 to 50 ma, and in 10-mA increments over 50 ma) X-ray tube heat capacity: 7.5 MHU X-ray tube cooling rate: Max khu/min (16.5kW) Actual 1008 khu/min (12.0 kw) Focal spot size-iec 60336: 2005, nominal: 0.9 mm 0.8 mm (small), 1.6 mm 1.4 mm (large) 5) X-ray detection Detection system: Solid-state detectors Main detector: elements Data acquisition: 896 channels 80 rows Reference detector: 1 set View rate: Max views/s 6) Data storage Magnetic disk - Raw data: 4000 rotations or more (for volume scan with 80 rows at 0.35 s) - Image data: images or more (when converting to pixel image) DVD-RAM: 4.7 Gbytes - DICOM images: 8000 DVD-R: 4.7 Gbytes - DICOM images: ) Image display Display monitor: 48.1 cm (19-inch) color LCD Monitor matrix: Image matrix: (max.) CT number: Display range: to Window width/level: Continuously variable Preset windows: 3/image Window types: Linear, non-linear (6 user-programmable), and double windows Image retrieval - Method: On-screen menus and keyboard - Mode: Image, series, and patient Multi-frame display: Reduction/cut-off display, ROI processing Inset scanogram display Information display: User selectable Cine display: Variable speed Scanogram/CT image switching: Show/hide scano line, zoom Slice-feed playback (CineView): High-speed image feeding using the mouse or keyboard [37]

44 GE (LightSpeed 16) The CT scanner featuring a 53.2 kw generator, 6.3 MHU tube and minimum rotation time of 0.5 s. It is capable of imaging 16 slices per rotation. A summary of the main features of the scanner are presented below. GE features Gantry # of detectors individual elements composed by : 8 rows of 1.25mm thickness and 16 rows of 0.625mm thickness, each containing 888 active patient elements ; 24 reference elements Type of detectors Ceramic Scintillator provides 98% x-ray absorption efficiency, 70% geometric efficiency Scan mode Axial Helical Gantry aperture 70cm Gantry angulation tilt +/ FOV Slice thickness Focus to Isocenter Distance Focus to Detector Distance 50cm 4 typical modes of data output: a) 8 x 1.25mm (uses center 16 rows) b) 8 x 2.5mm (uses all 24 rows) c) 16 x 1.25mm (uses all 24 rows) d) 16 x 0.625mm (uses center 16 rows) 54cm 95cm Table Vertical movement Longitudinal movement Maximum allowed weight από 51.6cm έως 99.1cm από 0cm έως 170cm a) 180Kg with +/-0.25mm positional accuracy b) 205Kg with +/-1.0 mm positional accuracy Generator - Tube Generator High-frequency on-board generator Continuous operation during scan 53.2 kw outpower Applied voltage kvp: 80, 100, 120, 140 Current ma: 10 to 440mA, 5mA increments Maximum ma for each kvp selection kvp max ma Rotation time in 0.5, 0.6, 0.7,0.8,0.9, 1.0, 2.0, 3.0, 4.0 s [38]

45 Focal spot size Filter Small Focal Spot 0.7mm(w) x 0.6mm(L) nominal value (IEC 336/93) 0.9mm(w) x 0.7mm(L) traditional methodology Large Focal Spot 0.9mm(w) x 0.9mm(L) nominal value (IEC 336/93) 1.2mm(w) x 1.2mm(L) traditional methodology tube insert: 1.52 mm Be tube housing: 0.32 mm Al. at 70kV Hardware Host computer Image processor Reconstruction time Reconstruction filter a) Dual SMP 2.66MHz Intel Xeon processors with 512KB L2 cache b) Intel Hyper threading technology c) 2GB DDR266 Dual channel Memory with a throughput of 4.2GB/sec a) Nvidia Quadro4 980 XGL AGP 8X graphics with 128MB Memory b) Graphics Processor Unit (GPU) clock 300MHz c) Graphics Memory clock 325MHz 0.17 sec image to image reconstruction in normal 16 slice reconstruction mode Soft tissue, Standard, Detail, Bone, Bone Plus, Lung and Edge Reconstruction Matrix 512 x 512 Display Matrix 512 x 512, 768 x 768, 1024 x 1024 HU range έως 3071 Storage media Total of 254GB system: a) Main system (host) disk drive 36GB b) 2 system disk drive (image disk) 73GB c) scan data disk drive 36GB d) standard MOD drive 2.3GB e) DVD Ram 4.7GB per side Software Programs a) Volume Viewer 3D, Volume Rendering b) MIP (Maximim Intensity Projection) c) Navigator d) Advanced Vessel Analysis e) Image Fusion f) SmartScore g) Cardiac Snapshot. h) SmartStep (CT Fluoro) [39]

46 5.2. Protocols In tables 4 and 5, the CT parameters of the CTA scan protocols for each anatomical region from both systems are presented. Table 4. Toshiba CTA parameters of scan protocols from specific anatomical regions Carotid Thoracic Abdomen Scan mode Helical Helical Helical kvp Rotation time Collimation 80 x x x 0.5 Pitch DFOV Alg AIDR 3D AIDR 3D AIDR 3D Filter FC 43 FC 08 FC 08 Phantom Body Body Body Table 5. GE CTA parameters of scan protocols from specific anatomical regions Carotid Thoracic Abdomen Scan mode Helical Helical Helical kvp Rotation time Collimation 16 x x x 1.25 Pitch DFOV Alg FBP FBP FBP Filter Std Std Std Phantom Head Body Body 5.3. Dose verification reports Prior to the collection of the dose indexes, a comparison of the CTDIw with measurements realized by the staff of the Medical Physics Department was accomplished. The deviations between the measured and the displayed values are within the limits proposed by the GAEC and are presented below. [40]

47 Toshiba Parameter Parameter`s explanation Control Instruments Control Elements Results Acceptable limits CTDI W Typical clinical prescription protocol (Toshiba display) The CTDI 100 value is obtained by placing the head (body) dosimetry phantom on the couch top to measure the dose Dosimeter: Barracuda Dosimetry phantoms: PMMA dosimetry phantoms with diameters of 160 mm and 320 mm Measure the dose at the dose measurement positions (3 o'clock, 6 o'clock, 9 o'clock, 12 o'clock, and center positions) with the center of the dosimeter (probe) located on the projector line. The dose values are established by averaging 10 timed-offset exposures using a dosimeter (probe) in integrate mode. The 10 timed-offset exposures are spaced at intervals of one tenth the rotation time. See Table 6 below ± 20 % of the measured value The CTDI 100 (peripheral) value is the average of the dose measured at four positions (3 o'clock, 6 o'clock, 9 o'clock, and 12 o'clock) and the CTDI 100 (center) value is the dose at the center position. Scan conditions a) Mode: Head Tube voltage: 120 kvp Tube current: 200 ma, Scan time: 1 s W-Volume: OFF FOV = 240 mm, Focus: S Slice thickness x rows: 1 mm x 40 Scan Range = 40 mm b) Mode: Body Tube voltage: 120 kvp Tube current: 300 ma, Scan time: 1 s W-Volume: OFF FOV = 400 mm, Focus: S Slice thickness x rows: 1 mm x 40 Scan Range = 40 mm [41]

48 Table 6. CTDI measurements from the Toshiba scanner Head mode Body mode Console (mgy) Measured (mgy) dev (%) Console (mgy) Measured (mgy) dev (%) CTDI w CTDI 100 (center) x x x x CTDI 100 (peripheral average) x x x x CTDI 100 (peripheral 12 o clock) x x x x CTDI 100 (peripheral 3 o clock) x x x x CTDI 100 (peripheral 6 o clock) x x x x CTDI 100 (peripheral 9 o clock) x x x x [42]

49 GE Parameter CTDI W Typical clinical prescription protocol (GE display) Parameter`s explanation The CTDI 100 value is obtained by placing the head (body) dosimetry phantom on the couch top to measure the dose Control Instruments Dosimeter: Barracuda Dosimetry phantoms : PMMA dosimetry phantoms with diameters of 160 mm and 320 mm Control Elements Results Acceptable limits Measurement of CTDI w for all clinical protocols static axial: 1s Scan range: I7.5-S7.5 Detector rows: 8 Axial thickness: 5 mm Images / rotation: 4 Detector config.: 8x2.5 Beam collim.: 20 mm SFOV: small (head) large (body DFOV: 25 cm (head) 50 cm (body) See Table 7 below ± 20 % of the measured value Table 7. CTDI measurements from the GE scanner phantom kvp ma CTDI c (mgy) [M] CTDI c (mgy) [N] CTDI p (mgy) [M] CTDI p (mgy) [N] CTDI w (mgy) [M] CTDI w (mgy) [N] CTDIw Dev % [M-N] Body Body Body Body Body Body Body Body Head [43]

50 6. RESULTS & DISCUSSION 6.1. Patient data collection Carotids - Toshiba The data acquired by the system regarding the carotid examination are given in Table 8. Besides the sex, age, weight, height and BMI of the patients, data related with the CTDIvol, SL and DLP of the CTA examination as well as the total CTDIvol and DLP are given. In addition, the ED from each CTA has been calculated using a normalized coefficient Carotids DLP (msv mgy -1 cm -1 ). This value was derived having in mind the following considerations. According to Table 3, for the head and neck, the normalized coefficient is different ( and respectively). Therefore, since the carotid examination encompass mainly the head and neck, the Carotids Head Neck normalized coefficient / 2. DLP DLP DLP The mean value of CTDIvol was 4.69 mgy, the median value was 4.30 mgy, the minimum value was 3.20 mgy, the maximum value was 7.30 mgy and the 75 th percentile was 5.45 mgy. The mean value of SL was cm, the median value was cm, the minimum value was cm, the maximum value was cm and the 75 th percentile was cm. These large values for the SL can be explained taking into account that the start level of the examination is located at the level of the aortic arch. The mean value of DLP was mgy cm, the median value was mgy cm, the minimum value was mgy cm, the maximum value was mgy cm and the 75 th percentile was mgy cm. The mean value of ED was 0.72 msv, the median value was 0.66 msv, the minimum value was 0.48 msv, the maximum value was 1.12 msv and the 75 th percentile was 0.81 msv. In addition, the mean value of the total CTDIvol was mgy, the median value was mgy, the minimum value was mgy, the maximum value was mgy and the 75 th percentile was mgy. Finally, the mean value of the total DLP was mgy cm, the median value was mgy cm, the minimum value was mgy cm, the maximum value was mgy cm and the 75 th percentile was mgy cm. At this point it should be mentioned that although the mean value of 4.69 mgy for the CTA CTDIvol corresponds to about 10% of the total CTDIvol, the mean value of [44]

51 mgy cm for the CTA DLP corresponds to about 95% of the total DLP. This means that the evaluation of the ED only from the CTA part of the comprehensive prescription protocol provides (with acceptable accuracy) the determination of ED. In Table 9, it is presented in a summary form, the values of the mean, median, along with the value of 75 th percentile of CTDIvol, SL, DLP and ED for the CTA examination. At Figure 9 an indicative dose report from carotid examination is presented. For display purposes, at Figure 10 the distribution of CTDIvol, SL, DLP and ED are given for the carotid CTA examination. In order to obtain a better sense about the variation of the distributions of the previously mentioned values, in Table 10 the mean, median and the range values of age, weight, height and BMI were estimated. Figure 11 shows the results of the correlation of BMI with the CTDIvol values in the specific exam. From this graph become obvious that as the BMI increases the CTDIvol increases also. This is normal as the carotid CTA examination (at our hospital) starts from the aortic arch. This means that the system uses the current modulation especially from the level of the aortic arch up to the level of the cervical vertebrae. In other case, if the carotid examination encompasses only the head and neck area, the CTDIvol would be stable as the alteration due to the size of these anatomical regions at first approximation is negligible. [45]

52 Table 8. Carotids data # Sex Age Weight (kg) Height (m) BMI (kg m -2 ) CTDI vol (mgy) [Body mode] SL (cm) CTA DLP (mgy cm) ED (msv) CTDI vol (mgy) Total DLP (mgy cm) 1. F F M M M M M M M M F M F M M F M M M M M F M M M M M F M M [46]

53 Table 9a. Carotids (CTA) derived data CTDIvol (mgy) SL (cm) DLP (mgy cm) ED (msv) Mean Median Range th percentile Table 9b. Carotids (CTA - Total) derived data CTDIvol (mgy) x DLP (mgy cm) x Mean x x Median x x Range x x 75 th percentile x x Figure 9. Indicative dose report from carotid examination (Toshiba). [47]

54 Figure 10. The distribution of CTDI vol, Scan Length, DLP and ED from carotid examinations. [48]

55 CTDI vol (mgy) Table 10. Patients characteristics (Carotids) Age Weight (kg) Height (m) BMI (kg m -2 ) Mean Median Range y = x x x R² = BMI (kg/m 2 ) Figure 11. The CTDI vol vs. BMI from carotid examinations. [49]

56 Chest - Toshiba The data acquired by the system regarding the chest examination are given in Table 11. Besides the sex, age, weight, height and BMI of the patients, data related with the CTDIvol, SL and DLP of the CTA examination as well as the total CTDIvol and DLP are given. In addition, the ED from each CTA has been calculated using a normalized coefficient Table 3. Chest DLP (msv mgy -1 cm -1 ). This value was taken from The mean value of CTDIvol was 4.96 mgy, the median value was 5.05 mgy, the minimum value was 2.80 mgy, the maximum value was 6.80 mgy and the 75 th percentile was 5.88 mgy. The mean value of SL was cm, the median value was cm, the minimum value was cm, the maximum value was cm and the 75 th percentile was cm. The mean value of DLP was mgy cm, the median value was mgy cm, the minimum value was mgy cm, the maximum value was mgy cm and the 75 th percentile was mgy cm. The mean value of ED was 3.29 msv, the median value was 3.38 msv, the minimum value was 1.82 msv, the maximum value was 4.72 msv and the 75 th percentile was 3.87 msv. In addition, the mean value of the total CTDIvol was mgy, the median value was mgy, the minimum value was mgy, the maximum value was mgy and the 75 th percentile was mgy. Finally, the mean value of the total DLP was mgy cm, the median value was mgy cm, the minimum value was mgy cm, the maximum value was mgy cm and the 75 th percentile was mgy cm. At this point it should be mentioned that although the mean value of 4.96 mgy for the CTA CTDIvol corresponds to about 9% of the total CTDIvol, the mean value of mgy cm for the CTA DLP corresponds to about 95% of the total DLP. This means that the evaluation of the ED only from the CTA part of the comprehensive prescription protocol provides (with acceptable accuracy) the determination of ED. In Table 12, it is presented in a summary form, the values of the mean, median, along with the value of 75 th percentile of CTDIvol, SL, DLP and ED for the CTA examination. At Figure 12 an indicative dose report from chest examination is [50]

57 presented. For display purposes, at Figure 13 the distribution of CTDIvol, SL, DLP and ED are given for the chest CTA examination. In order to obtain a better sense about the variation of the distributions of the previously mentioned values, in Table 13 the mean, median and the range values of age, weight, height and BMI were estimated. Figure 14 shows the results of the correlation of BMI with the CTDIvol values in the specific exam. From this graph become obvious that as the BMI increases the CTDIvol increases also. [51]

58 Table 11. Chest data # Sex Age Weight (kg) Height (m) BMI (kg m -2 ) CTDI vol (mgy) [Body mode] SL (cm) CTA DLP (mgy cm) ED (msv) CTDI vol (mgy) Total DLP (mgy cm) 1. M M F M M M M M M F M M M M M M M M M F F M M M M M M M M F [52]

59 Table 12a. Chest (CTA) derived data CTDIvol (mgy) SL (cm) DLP (mgy cm) ED (msv) Mean Median Range th percentile Table 12b. Chest (CTA - Total) derived data CTDIvol (mgy) x DLP (mgy cm) x Mean x x Median x x Range x x 75 th percentile x x Figure 12. Indicative dose report from chest examination (Toshiba). [53]

60 Figure 13. The distribution of CTDI vol, Scan Length, DLP and ED from chest examinations. [54]

61 CTDI vol (mgy) Table 13. Patients characteristics (Chest) Age Weight (kg) Height (m) BMI (kg m -2 ) Mean Median Range y = 0.002x x x R² = BMI (kg/m 2 ) Figure 14. The CTDI vol vs. BMI from chest examinations. [55]

62 Abdomen - Toshiba The data acquired by the system regarding the abdomen examination are given in Table 14. Besides the sex, age, weight, height and BMI of the patients, data related with the CTDIvol, SL and DLP of the CTA examination as well as the total CTDIvol and DLP are given. In addition, the ED from each CTA has been calculated using a normalized coefficient Table 3. Chest DLP (msv mgy -1 cm -1 ). This value was taken from The mean value of CTDIvol was 6.11 mgy, the median value was 5.65 mgy, the minimum value was 3.00 mgy, the maximum value was mgy and the 75 th percentile was 7.05 mgy. The mean value of SL was cm, the median value was cm, the minimum value was cm, the maximum value was cm and the 75 th percentile was cm. The mean value of DLP was mgy cm, the median value was mgy cm, the minimum value was mgy cm, the maximum value was mgy cm and the 75 th percentile was mgy cm. The mean value of ED was 4.33 msv, the median value was 3.82 msv, the minimum value was 2.08 msv, the maximum value was 8.34 msv and the 75 th percentile was 5.14 msv. In addition, the mean value of the total CTDIvol was mgy, the median value was mgy, the minimum value was mgy, the maximum value was mgy and the 75 th percentile was mgy. Finally, the mean value of the total DLP was mgy cm, the median value was mgy cm, the minimum value was mgy cm, the maximum value was mgy cm and the 75 th percentile was mgy cm. At this point it should be mentioned that although the mean value of 6.11 mgy for the CTA CTDIvol corresponds to about 10% of the total CTDIvol, the mean value of mgy cm for the CTA DLP corresponds to about 96% of the total DLP. This means that the evaluation of the ED only from the CTA part of the comprehensive prescription protocol provides (with acceptable accuracy) the determination of ED. In Table 15, it is presented in a summary form, the values of the mean, median, along with the value of 75 th percentile of CTDIvol, SL, DLP and ED for the CTA examination. At Figure 15 an indicative dose report from abdomen examination is [56]

63 presented. For display purposes, at Figure 16 the distribution of CTDIvol, SL, DLP and ED are given for the abdomen CTA examination. In order to obtain a better sense about the variation of the distributions of the previously mentioned values, in Table 16 the mean, median and the range values of age, weight, height and BMI were estimated. Figure 17 shows the results of the correlation of BMI with the CTDIvol values in the specific exam. From this graph become obvious that as the BMI increases the CTDIvol increases also. [57]

64 Table 14. Abdomen data # Sex Age Weight (kg) Height (m) BMI (kg m -2 ) CTDI vol (mgy) [Body mode] SL (cm) CTA DLP (mgy cm) ED (msv) CTDI vol (mgy) Total DLP (mgy cm) 1. F M F M M M M M M M M M M F M M M F M M M M M F M M F M F M [58]

65 Table 15a. Abdomen (CTA) derived data CTDIvol (mgy) SL (cm) DLP (mgy cm) ED (msv) Mean Median Range th percentile Table 15b. Abdomen (CTA - Toshiba) derived data CTDIvol (mgy) x DLP (mgy cm) x Mean x x Median x x Range x x 75 th percentile x x Figure 15. Indicative dose report from abdomen examination (Toshiba). [59]

66 Figure 16. The distribution of CTDI vol, Scan Length, DLP and ED from abdomen examinations. [60]

67 CTDI vol (mgy) Table 16. Patients characteristics (Abdomen) Age Weight (kg) Height (m) BMI (kg m -2 ) Mean Median Range y = x x x R² = BMI (kg/m 2 ) Figure 17. The CTDI vol vs. BMI from abdomen examinations. [61]

68 Carotids - GE The data acquired by the system regarding the carotid examination are given in Table 17. Data related with the CTDIvol, SL and DLP of the CTA examination as well as the total CTDIvol and DLP are given. In addition, the ED from each CTA has been calculated using a normalized coefficient Carotids DLP (msv mgy -1 cm -1 ). This value was derived having in mind the following considerations. According to Table 3, for the head and neck, the normalized coefficient is different ( and respectively). Therefore, since the carotid examination encompass mainly the head Carotids Head Neck and neck, the normalized coefficient / 2. DLP The mean value of CTDIvol was mgy, the median value was mgy, the minimum value was mgy, the maximum value was mgy and the 75 th percentile was mgy. The mean value of SL was cm, the median value was cm, the minimum value was cm, the maximum value was cm and the 75 th percentile was cm. These large values for the SL can be explained taking into account that the start level of the examination is located at the level of the aortic arch. The mean value of DLP was mgy cm, the median value was mgy cm, the minimum value was mgy cm, the maximum value was DLP mgy cm and the 75 th percentile was mgy cm. The mean value of ED was 5.03 msv, the median value was 4.72 msv, the minimum value was 3.88 msv, the maximum value was 6.93 msv and the 75 th percentile was 5.37 msv. In addition, the mean value of the total CTDIvol was mgy, the median value was mgy, the minimum value was mgy, the maximum value was mgy and the 75 th percentile was mgy. Finally, the mean value of the total DLP was mgy cm, the median value was mgy cm, the minimum value was mgy cm, the maximum value was mgy cm and the 75 th percentile was mgy cm. At this point it should be mentioned that although the mean value of mgy for the CTA CTDIvol corresponds to about 44% of the total CTDIvol, the mean value of mgy cm for the CTA DLP corresponds to about 96% of the total DLP. This means that the evaluation of the ED only from the CTA part of the comprehensive prescription protocol provides (with acceptable accuracy) the determination of ED. DLP [62]

69 In Table 18, it is presented in a summary form, the values of the mean, median, along with the value of 75 th percentile of CTDIvol, SL, DLP and ED for the CTA examination. At Figure 18 an indicative dose report from carotid examination is presented. For display purposes, at Figure 19 the distribution of CTDIvol, SL, DLP and ED are given for the carotid CTA examination. [63]

70 Table 17. Carotids data # Sex Age Weight (kg) Height (m) BMI (kg m -2 ) CTDI vol (mgy) [Head mode] SL (cm) CTA DLP (mgy cm) ED (msv) CTDI vol (mgy) Total DLP (mgy cm) 1. Μ x x x x Μ x x x x Μ x x x x F x x x x M x x x x M x x x x M x x x x M x x x x F x x x x M x x x x M x x x x M x x x x M x x x x M x x x x M x x x x M x x x x M x x x x M x x x x M x x x x M x x x x F x x x x M x x x x M x x x x M x x x x F x x x x M x x x x M x x x x F x x x x M x x x x F x x x x [64]

71 Table 18. Carotids (CTA) derived data CTDIvol (mgy) SL (cm) DLP (mgy cm) ED (msv) Mean Median Range th percentile Table 18. Carotids (CTA - Total) derived data CTDIvol (mgy) x DLP (mgy cm) x Mean x x Median x x Range x x 75 th percentile x x Figure 18. Indicative dose report from carotid examination (GE). [65]

72 Figure 19. The distribution of CTDI vol, Scan Length, DLP and ED from carotid examinations. [66]

73 Chest - GE The data acquired by the system regarding the chest examination are given in Table 19. Data related with the CTDIvol, SL and DLP of the CTA examination as well as the total CTDIvol and DLP are given. In addition, the ED from each CTA has been calculated using a normalized coefficient was taken from Table 3. Chest DLP (msv mgy -1 cm -1 ). This value The mean value of CTDIvol was mgy, the median value was mgy, the minimum value was 9.88 mgy, the maximum value was mgy and the 75 th percentile was mgy. The mean value of SL was cm, the median value was cm, the minimum value was cm, the maximum value was cm and the 75 th percentile was cm. The mean value of DLP was mgy cm, the median value was mgy cm, the minimum value was mgy cm, the maximum value was mgy cm and the 75 th percentile was mgy cm. The mean value of ED was 7.24 msv, the median value was 7.30 msv, the minimum value was 4.75 msv, the maximum value was 9.64 msv and the 75 th percentile was 7.75 msv. In addition, the mean value of the total CTDIvol was mgy, the median value was mgy, the minimum value was mgy, the maximum value was mgy and the 75 th percentile was mgy. Finally, the mean value of the total DLP was mgy cm, the median value was mgy cm, the minimum value was mgy cm, the maximum value was mgy cm and the 75 th percentile was mgy cm. At this point it should be mentioned that although the mean value of mgy for the CTA CTDIvol corresponds to about 35% of the total CTDIvol, the mean value of mgy cm for the CTA DLP corresponds to about 94% of the total DLP. This means that the evaluation of the ED only from the CTA part of the comprehensive prescription protocol provides (with acceptable accuracy) the determination of ED. In Table 20, it is presented in a summary form, the values of the mean, median, along with the value of 75 th percentile of CTDIvol, SL, DLP and ED for the CTA examination. At Figure 20 an indicative dose report from chest examination is [67]

74 presented. For display purposes, at Figure 21 the distribution of CTDIvol, SL, DLP and ED are given for the chest CTA examination. [68]

75 Table 19. Chest data # Sex Age Weight (kg) Height (m) BMI (kg m -2 ) CTDI vol (mgy) [Body mode] SL (cm) CTA DLP (mgy cm) ED (msv) CTDI vol (mgy) Total DLP (mgy cm) 1. M x x x x F x x x x M x x x x M x x x x M x x x x M x x x x M x x x x M x x x x F x x x x M x x x x M x x x x M x x x x F x x x x F x x x x M x x x x M x x x x M x x x x M x x x x M x x x x M x x x x F x x x x M x x x x M x x x x M x x x x M x x x x [69]

76 Table 20. Chest (CTA) derived data CTDIvol (mgy) SL (cm) DLP (mgy cm) ED (msv) Mean Median Range th percentile Table 20. Chest (CTA - Total) derived data CTDIvol (mgy) x DLP (mgy cm) x Mean x x Median x x Range x x 75 th percentile x x Figure 20. Indicative dose report from chest examination (GE). [70]

77 Figure 21. The distribution of CTDI vol, Scan Length, DLP and ED from chest examinations. [71]

78 Abdomen - GE The data acquired by the system regarding the abdomen examination are given in Table 21. Data related with the CTDIvol, SL and DLP of the CTA examination as well as the total CTDIvol and DLP are given. In addition, the ED from each CTA has been calculated using a normalized coefficient was taken from Table 3. Chest DLP (msv mgy -1 cm -1 ). This value The mean value of CTDIvol was mgy, the median value was mgy, the minimum value was mgy, the maximum value was mgy and the 75 th percentile was mgy. The mean value of SL was cm, the median value was cm, the minimum value was cm, the maximum value was cm and the 75 th percentile was cm. The mean value of DLP was mgy cm, the median value was mgy cm, the minimum value was mgy cm, the maximum value was mgy cm and the 75 th percentile was mgy cm. The mean value of ED was msv, the median value was msv, the minimum value was 8.25 msv, the maximum value was msv and the 75 th percentile was msv. In addition, the mean value of the total CTDIvol was mgy, the median value was mgy, the minimum value was mgy, the maximum value was mgy and the 75 th percentile was mgy. Finally, the mean value of the total DLP was mgy cm, the median value was mgy cm, the minimum value was mgy cm, the maximum value was mgy cm and the 75 th percentile was mgy cm. At this point it should be mentioned that although the mean value of mgy for the CTA CTDIvol corresponds to about 42% of the total CTDIvol, the mean value of mgy cm for the CTA DLP corresponds to about 97% of the total DLP. This means that the evaluation of the ED only from the CTA part of the comprehensive prescription protocol provides (with acceptable accuracy) the determination of ED. In Table 22, it is presented in a summary form, the values of the mean, median, along with the value of 75 th percentile of CTDIvol, SL, DLP and ED for the CTA examination. At Figure 22 an indicative dose report from abdomen examination is [72]

79 presented. For display purposes, at Figure 23 the distribution of CTDIvol, SL, DLP and ED are given for the abdomen CTA examination. [73]

80 Table 21. Abdomen data # Sex Age Weight (kg) Height (m) BMI (kg m -2 ) CTDI vol (mgy) [Body mode] SL (cm) CTA DLP (mgy cm) ED (msv) CTDI vol (mgy) Total DLP (mgy cm) 1. M x x x x M x x x x M x x x x M x x x x M x x x x F x x x x M x x x x F x x x x F x x x x M x x x x M x x x x M x x x x M x x x x M x x x x M x x x x M x x x x M x x x x M x x x x M x x x x M x x x x M x x x x M x x x x F x x x x M x x x x M x x x x M x x x x M x x x x M x x x x M x x x x M x x x x [74]

81 Table 22a. Abdomen (CTA) derived data CTDIvol (mgy) SL (cm) DLP (mgy cm) ED (msv) Mean Median Range th percentile Table 22b. Abdomen (CTA - Total) derived data CTDIvol (mgy) x DLP (mgy cm) x Mean x x Median x x Range x x 75 th percentile x Figure 22. Indicative dose report from abdomen examination (GE). [75]

82 Figure 23. The distribution of CTDI vol, Scan Length, DLP and ED from abdomen examinations. [76]