Home

Calibration & Clinical Measurements in Computed Tomography

1. Measured HVL Ratio with the nominal HVL 2 gt gt N on iN gt S ite ol ol ol T Al filtration T2 Ratio of signal with without Al filtration gt i oO On T 6 5 6 6 6 7 6 8 6 9 7 7 1 Thickness of Al filtration mm Figure 5 The ratio of the signals InvL r Io and Invt r Io with the correspondent thicknesses T and Tp of aluminium filtration o The interpolated value of what the ratio would be with the nominal HVL Q the measured HVL where the ratio is 0 5 O and the upper and lower limits for what is considered to be acceptable in dotted lines These are the values when RQT8 was acquired 10 2 1 2 Calibration of Detectors used in CT Two detectors were calibrated one 100 mm pencil ionization chamber Radcal Model no 10x5 3CT serial no 8319 and one semiconductor CT Dose Profiler CTDP from RTI Electronics serial no DP2 12120084 Calibrations were made in the newly acquired RQT qualities and in RQR qualities already established on the SSDL see Tab 2 All radiation qualities were based on IEC 61267 2005 10 To be able to calibrate the detectors one must accurately know what the air Kerma rate for reference is in the correct radiation qualities The reference air Kerma rate was determined with help from a reference pencil ionization chamber DCT 10 IBA Dosimetry GmbH serial no 253 The reference pencil ionization chamber was
2. 0 20 40 60 Effective diameter cm Figure 10 Conversion factors f for different effective diameters in cm The square characters are conversion factors valid for CTDI from the 32 cm body phantom The circular characters are conversion factors valid for CTDI from the 16 cm head phantom The characters are tabulated data reported in AAPM 204 table 1D amp 2D page 10 11 The lines are best fit exponential trendlines reported in AAPM 204 page 20 Conversion factors were not only produced for effective diameters They were also produced on using the Lateral and or the Anterior Posterior dimensions This project however only focus on the conversion factors valid for effective diameters 22 From knowing two parameters a size specific dose estimate SSDE could be calculated according to the following relationship er f dia CTDI CTDIL from Head phantom SSDE m Gy dose 16 orp dia CTD CTDL from Body phantom where firstly one needed to know the CTDI for one of the two conventional CTDI phan toms GYD for the Head phantom or CTD for the Body phantom Secondly one needed to obtain a conversion factor f shown in Fig 10 The conversion factor fore dia Was valid from CTDE and for a certain effective diameter of the patient In the same way 3 f dia WAS valid from CTDI and for a certain effective diameter of the patient The SSDE that was calculated according to Eq 16 was embedded with a d
3. 10 Body phantom GE Brightspeed Head phantom GE Brightspeed S i i 8 lt a y oy Py X N j hb Or oy uy uy CTDIj00 100 mAs Ol T T T T T T T p 0 625 p 0 875 p 1 35 p 1 675 p 0 625 p 0 875 p 1 35 p 1 675 Free in air GE Discovery o 3S CA 2 8 Ke so wn lt 15H e 10b J z GJ 5 7 a T T T p 0 531 p 0 969 p 1 375 Figure 18 CTDI100 100 mAs measured with the CTDP during helical scanning with different values of pitch p in the different phantoms and free in air The measurements in the phantoms were made on a GE Brightspeed Elite scanner in Kristiansand Norway and free in air on a GE Discovery 750HD scanner in Gothenburg Sweden The values of pitch differed between those scanners The measurements were made with 120kV and Large SFOV In the phantoms the collimation was 10mm and free in air it was 20mm 32 Tube voltage kV Collimation mm 60 b 60 F o exile J lt Somna o E Jol saol gpl heak S Sx S SF ISN SH inm coos So WAF SLY SYS Son 3 NYX 3 NN NYY Saga tom s 3 yee gis 3 AN FNN ses Q 20
4. s ae JA 20 7 FIS z 0F OF T T T T T T li T T 80kV 100kV 120kV 140kV 1 25mm 5mm 10mm 15mm 20mm SFOV y p s S te 2 07 Ss FF RE wae le Ps Noy Nos 4 ka gN NIN So 20F 5 ry E E Q 10 F H O 0 T T T Large Small Head E Converted from 32 10 cm phantom lI Converted from 16 10 cm phantom IMeasured in 10cm phantom Figure 19 CTDI 100mAs size corrected from the 32cm body phantom and the 16cm head phantom respectively to an effective diameter of 10 cm and the measured CTDI in the 10 cm pediatric phantom The data is shown when either tube voltage collimation or SFOV was the parameter of change and was measured with the Radcal pencil ionization chamber during axial scanning The measurements were made on a GE Brightspeed Elite scanner in Kristiansand Norway Large SFOV 10mm collimation and 120kV were the reference settings when another parameter was tested 33 4 Discussion 4 1 Calibration of Detectors used in CT Three detectors were used for calibration measurements one of which was the reference pencil chamber IBA DCT 10 The two detectors that were calibrated and the reference pencil chamber all handled the signal of response in a different way The reference pencil chamber was originally calibrated during partial irradiation a similar procedure that was established during this project and in a manner as described by IAEA 457 4 However the Radcal pencil chamber was originally calibrated dur
5. Di z _ Dalz mA t mA per mAs D1 z Da z tr 17 The CTDP measures an exposure rate hence the tube rotation time ty is cancelled out when CTDlIoo is normalized to the number of mAs The pencil chamber measures an exposure It should be noted that the relationship seen in Eq 17 is just valid when the exposure rate is independent of the X ray tubes position during the rotation This is why the CTDP not should be measured in the peripheral holes in the phantoms This approach that needs to be adopted when measuring with the CTDP also explains why the summation of dose values along 100mm Eq 12 needs to be multiplied with the pitch Or simulated pitch that the Mover system effectively makes Eq 13 A clear derivation for this pitch dependence has been shown before 15 It should be noted that the energy correction factors RQT 8 10 in Tab 5 that were used for the Radcal pencil chamber not perfectly simulate the radiation qualities arising either in the phantoms or when the tube voltage was lower than 100kV Also should be noted that when the tube voltage was 140kV during clinical measurements the energy correction factor valid for RQT 10 was chosen for the Radcal pencil chamber and a relatively higher difference compared to the CTDP was seen If the energy correction factors used for the Radcal pencil chamber were properly chosen is not further investigated in this project Although this is one factor that would b
6. 0 0700 0 0027 N 1 0 27 Quadratic sum 0 82 Combined uncertainty u y 0 91 2 Factors influencing use of the reference standard 2 1 Current ITA 6 66E 11 3 02E 04 N 1 2 2 Leakage ITA 6 66E 11 _7 18E 16 1 08E 05 N 1 000 2 3 Recombination k 1 0 0035 0 0020 R 1 2 4 Polarity effect k n 1 0 005 0 0029 R 1 2 5 Temperature K T 300 0 1000 0 0002 R 1 2 6 Pressure kPa p 101 3 0 0100 0 0001 R 1 2 7 Humidity k 1 0 001 0 0006 R 1 006 2 8 Chamber orientation 1 0 0012 R 1 2 9 Reproduction of source to chamber distance d cm 100 0 2 0 0008 T 1 0 08 2 10 Reproduction of irradiated length 5 cm 5 1 0 0 0000 R 1 000 Quadratic sum 0 15 Combined uncertainty u y 3 Factors influencing use of the chamber for calibration 3 1 Product readout P 1 19 3 52E 05 2 95E 05 N 1 ood 3 2 Constancy of Calibration factor on two different days Nx 1 19 0 0052942 0 0031 R 1 0 31 3 3 Leakage 0 0 0 N 0 0 00 3 4 Recombination k 1 0 0035 0 0020 R 0 0 00 3 5 Polarity effect ko 1 0 005 0 0029 R 0 000 3 6 Temperature K T 300 0 1000 0 0002 R 0 0 00 3 7 __ Pressure kPa p 101 3 0 0100 0 0001 R 0 0 00 3 8 Humidity k 1 0 001 0 0006 R 0 000 3 9 Chamber orientation 0 0012 R 1 3 10 Reproduction of source to chamber distance d cm 100 0 2 0 0008 T 1 3 11 Reproduction of irradiated length 5 cm 5 102 0 1 0 0113 R 1 Quadratic sum 40 Combined uncertainty u y 1 18 4 Total uncertainty Quadratic sum 2 31 Combined uncertainty u y Expande
7. 100 mAs vs different nominal collimations measured free in air with the different measuring techniques The measurements were made on a GE Discovery 750HD scanner in Gothen burg Sweden with Large SFOV and 120 kV During helical scanning only 20 and 40 mm collimation were available and the pitch were 0 531 and 0516 respectively for the collimations In Fig 18 it can be seen that different values of pitch not have any great effect on the CTDlIioo Different pitch measured in the phantoms were made on a GE Brightspeed Elite in Kristiansand Norway Different pitch measured free in air were made on a GE Discovery 750HD in Gothenburg Sweden Also note in Fig 18 that the collimations are not the same in the phantoms it was 10 mm and free in air it was 20 mm 3 4 Evaluate the size dependency of CTDIvol for different phantom sizes In Fig 19 it can be seen that the size corrected CTDI from the Head and Body phantom was quite near the measured values in the 10cm Pediatric phantom Especially for lower kV and larger collimations A protruding value is seen from the body phantom in the lowest collimation 1 25mm When SFOV is Small or Head the measured CTDI in the 10cm Pediatric phantom is a bit higher than the size corrected values Before any size corrections were made the measured CTDI for the Head and Body phantom were 84 and 42 respectively of the size corrected value 31 CTDI 00 100 mAs 20 F uy uy l l i i i l of 15
8. Body CTD Pirie mover m BodyCTDPhretical O 10 0 ji Tube voltage kV Figure 13 CTDI00 100 mAs vs different tube voltages measured in the different phantoms and different measurement techniques The measurements were made on a GE Brightspeed Elite scanner in Kristiansand Norway with Large SFOV and 10mm collimation and with a pitch of 0 625 during the helical measurements In the pediatric phantom measurements were only made with the Radcal pencil ionization chamber e Pediatric l Chamber axial 39 Head I Chamberaziat 2 e Head GCT D Pisialtmover HeadCTDPreticai S sol ie Body I Chamber arial a Body CT D Pirial mover BodyCTDPretical a E 10F O gl l 0 5 10 15 20 Collimation mm Figure 14 CTDIj00 100 mAs vs different nominal collimations measured in the different phantoms with the different measurement techniques The measurements were made on a GE Brightspeed Elite scanner in Kristiansand Norway with Large SFOV and 120kV In the Pediatric phantom measurements were only made with the Radcal pencil ionization chamber During helical scanning only 10 and 20mm collimation were available and the pitch was 0 625 The different curvature for the Body phantom is due to the larger focal spot tube current 400 mA In the other two phantoms only the small focal was used 29 40 L I Chamberazial OLD Pirial mover n e CT DPhelic
9. CT Dose Profiler The same company have also manufactured a Mover system that can be used with the CTDP Also a part of this project is to evaluate the Mover system 1 4 Chain of Calibrations Metrology is the science of measurement The aim in Metrology is to be able to measure phys ical units as reliable and accurate as possible An important part in this science is calibrations A calibration is performed in a way where a measuring instrument is read in conjunction with a known normal or reference This reference needs to be traceable to another higher level of reference In figure 3 a very simplified schematic is shown on how the traceability looks for radiation metrology 4 This schematic is in this project also called the chain of calibrations For every step down this hierarchy the uncertainties of measurement grow BIPM A Y PSDL A Y SSDL A Y User Figure 3 A very simplified schematic of the chain of calibrations for radiation metrology Bureau International des Poids et Mesures BIPM in Paris has the highest authority in ra diation metrology 4 Primary Standards Dosimetry Laboratories PSDL sets the standard to a degree of accuracy as high as possible The Secondary Standards Dosimetry Laboratories SSDL often sets the national standard for physical units when the country not have a PSDL The SSDL aims to set the standard to a degree of accuracy
10. When measuring during axial scanning with the Mover free in air it also consistently measured a CTDIi99 5 lower compared when measuring during helical scanning Fig 15 17 This possibly states that the Mover speed is 5 too high Note that this difference in CTDIj00 c was not seen when measuring in the phantoms Possibly because when measuring in the phantoms the whole exposure profile is not within 100mm In Fig 16 for 20mm collimation a protruding value was seen during helical measurement with the CTDP It gave approximately the same value as the Radcal pencil chamber which was not expected because of the deviation seen for the other measurements The reason for this is not exactly known Hence this result should be looked with some oversight 4 4 Evaluate the size dependency of CTDIvol for different phantom sizes Fig 19 shows that the measurements made in the 10cm phantom compared relatively well with the size corrected measurements made in the 16cm and 32cm phantoms Especially with lower tube voltage and larger collimation The measured CTDI was originally 80 and 40 of the size corrected value for the head phantom and body phantom respectively This says how much CTDI can underestimate the exposure for small patients such as children A protruding value is seen from the measurement made in the 32cm phantom during 1 25cm collimation A likely explanation for this can be seen in Fig 14 together with Fig 16 that the CTDlio0
11. as high as possible The SSDL often bridges the gap between the PSDL and the eventual User for example at the hospital Within calibration procedures it is important to be able to control all the parameters that will have an influence on the measured signal Both for the known reference and for the instrument that will be calibrated Within radiation metrology an important parameter is what radiation quality one measures in 4 Therefore radiation qualities for specific applications have been defined Some specific radiation qualities for applications in CT have been defined by the International Electrotechnical Comission IEC 10 This project was performed in two parts in the chain of calibration Fig 3 Namely in the SSDL and in being the User of the instruments In this project a calibration procedure for detectors used in CT was established For this procedure some specific radiation qualities needed to be acquired The calibration procedure in this study was performed at the Norwegian Radiation Protec tion Authority NRPA in the Secondary Standards Dosimetry Laboratory SSDL which is Norway s national reference for the units Gray Gy Becquerel Bq and Sievert Sv 1 5 Goals for this study e Establish a standardized calibration procedure for different detectors used in CT e Evaluate the newly released Mover system e Compare CTDIio9 measured clinically with a 100mm pencil ionization chamber and with the CT Dose Profiler Eval
12. c for the 32cm phantom do not follow the same curvature as the 16cm or 10cm 38 phantom This is due to the larger focal spot the CT system automatically chose when the effect was exceeding 24kW which was the case actually for all the measurements made in the 32cm phantom This highlights that for small collimations it is important to know what focal spot is chosen during the scanning if conversion factors like in this project will eventually be used in the clinic Actually the larger focal spot was also chosen when the tube voltage was 140 kV in the 16cm phantom Highlighting again how this project could have been performed better But with this in mind the size specific dose estimates was quite reliably matching the measured CTDI in the 10cm phantom with the big exception when the collimation was very narrow with a large focal spot The Small SFOV and Head SFOV was using the same bow tie filter Hence the measure ments gave nearly the exact same CTDI However looking into when these SFOV were used during measurements the size converted CTDI from the phantoms were consistently underestimating the measured CTDI in the 10 cm phantom This can perhaps be explained by with using this bow tie filter the exposure rate becomes lower the more peripheral from the isocenter one measures Remember CTDI is determined by 3 of the peripheral exposure Eq 2 But if this is correct needs to be investigated further According to AAPM 204 6 u
13. during this project calibrated at the na tional metrology institute of The Netherlands VSL The calibration procedure was performed by first measuring with the reference pencil chamber in all radiation qualities Tab 2 Then turning off the radiation and switch to one of the detectors that would be calibrated Which was repeated again for the remaining detector The calibration procedure was performed on two different days to see the stability of the detectors 20 Filtration RQT RQR Focal spot distance between Collimation Wa 5cm Focal spot Point of test Je vA dr 100 cm we Point of test VA P A _ Distance 2cm ae me Irradiated width w Ts 5 10 cm Figure 6 Explanatory drawing of the arrangement for calibration The anode angle of the X ray tube was 20 The exit of the collimation was 98cm from the focal spot and the point of test was 100cm from the focal spot This gave an irradiation width of 5 10 cm This was the arrangement for all detectors the reference pencil chamber DCT 10 the Radcal pencil chamber and the CTDP To reduce the length of the terms for the reference pencil ionization chamber and for the Radcal pencil ionization chamber They will now be called reference pencil chamber and Radcal pencil chamber 11 Figure 6 and 7 describes how the calibration measurements for all detectors were made The measurements for calibration were made with a square aperture 5 x 5cm which was
14. half value layer kV mm Cul mm Al mm Al mm Al RQT8 100 0 20 6 90 6 57 7 10 RQT9 120 0 25 8 40 8 13 8 64 RQT 10 150 0 30 10 10 9 64 10 64 The measurement arrangement was set up with a Capintec 30cc PM30 Ionization Chamber serial no 1232 With help from lasers the center of the Capintec chamber could be placed 100cm from the focal spot in the X ray tube Comet MXR 160 0 4 3 0 The X ray tube provides a continuos output of the radiation has a target angle of 20 and an inherent filtration of 0 8mm Be A plastic wheel with holes of 5cm in diameter around the wheel was used to put the filtration material for the radiation quality Goodfellow Cambridge Ltd The copper and aluminium filtration material was given to have an elemental purity of at least 99 9 The plastic wheel with its filtration holes was placed 18cm from the tube focus Five cm from the filtration hole an in house constructed mover was placed together with the half value layer test device The half value layer test device consisted of two different aluminium filtration packages one package with thickness T just below the nominal HVL and one package with thickness T just above the nominal HVL for the quality in mind see Tab 1 for the values of T and T for the different radiation qualities and Fig 4 for the measurement arrangement Figure 4 An explanatory picture of the arrangement for acquiring and measuring the HVL for the RQT qualities a Rotational pla
15. no 2 1 3 New dosimetric techniques for CT To be able to accommodate the problems with the CTDlIgo pencil ionization chambers of even longer lengths could be manufactured However even longer lengths would make them The article 9 refers to dose hence the choice of terminology But the dosimetric unit used in this thesis is air Kerma if not otherwise stated more expensive and more fragile New ideas of measuring the CTDI in other ways are instead being introduced 8 These ideas are based on the utilization of a small detector that can be approximated as a point dose detector To be able to measure the air Kerma profile during axial scanning the point like detector can accurately be moved through the beam free in air or in a phantom with a mover system During helical scanning the detector can be placed stationary on the table allowing it to be irradiated free in air or in a phantom along a length due to the table movement A clear advantage of this system is that a picture of the air Kerma profile can be achieved letting one know how much of the scatter tails is being missed RTI Electronics AB have manufactured an angular independent solid state detector called CT Dose Profiler CTDP with a measuring time resolution of 2000 dose values per second and with an effective length of 250 um that can be approximated as a point dose detector A part of this project is to compare the conventional 100mm pencil ionization chamber with the
16. pencil chamber it was assumed that RQT8 simulated most properly when the voltage was 80 kV or 100 kV RQT 9 was chosen when the voltage was 120 kV and RQT 10 for 140 kV 19 From all the measurements Tab 3 4 comparisons were made between the three different detector systems From the measurements in the CTDI phantoms comparison were made on CTDlIi00 measured centrally CTDIj00 lt Since the CTDP not was used for measurements in the peripheral holes in the phantoms Different tube rotation times ty and tube currents mA were chosen in the measurements and hence the CTDlioo was normalized to the tube current in mA multiplied with the tube rotation time t and divided with a factor of 100 i e normalized to the number of 100 mAs The Radcal pencil chamber with its electrometer gave a value Mradcat that needed to be corrected in the following way to get the CTDlioo and normalized to the number of 100 mAs MRadeat100 kp kg 100 nOTDIOO0paacai NT DN mGy 100 mAs 11 where Mradcai Was the read air Kerma value in mGy from the electrometer The term 100 corrected for the pencil chambers length in mm and NT was the nominal collimation in mm kp corrected for the temperature and pressure and kg for the energy correction factors used tr was the tube rotation time in seconds and 100 mA was the correction for the number of 100 mA When measuring during helical scanning with the CTDP the CTDlioo normalized to the number of 100mAs w
17. placed so its ending seen from the beams eye view was 98cm from the tube focus see Fig 6 The aperture was first centralized but then raised 1cm towards the cathode direction resulting in the field coming out from the aperture more homogeneous The central point of the chamber was then placed so it was in the center of the aperture and 100cm from the tube focus i e 2cm from the apertures ending Firstly the reference pencil chamber DCT 10 was irradiated using the different radiation qualities see Tab 2 and the current was measured in the same way as with the HVL measurements Measurements with the reference pencil chamber was repeated two times giving a total irradiation time of 100seconds All calibration measurements for all detectors were made with a tube current of 5mA Table 2 The radiation qualities based on IEC 61267 2005 used for calibration with their respective X ray tube voltages and their at SSDL measured half value layer in mm Al Also noted is the total filtration for each radiation quality Note that RQR qualities do not have any copper filtration Radiation X ray X ray HVL Total filtration Quality tube voltage tube current kV mA mm Al mm Al mm Cu RQT8 100 5 6 84 3 46 0 16 RQT9 120 5 8 48 3 49 0 22 RQT 10 150 5 10 36 3 37 0 30 RQR 9 120 5 4 98 4 04 0 The measured signal Mpey from the reference pencil chamber DCT 10 was a current with an order of magnitude around 1071 A To co
18. the material used is not considered to be water equivalent New ideas says that the phantoms need to be at least 45cm long 8 1 2 3 The scanning length When a patient is being scanned over a certain length the cumulative dose along this length can be explained as a convolution of the single slice dose profile along the defined length 8 9 If the scanning length is at least as long as one single slice physical dose profile an equilibrium value of the accumulated dose value in the centre of the scanning length will be achieved De 0 The actual integrated dose along the scanning length then can be explained as the multiplication of the equilibrium dose with the scanning length L De 0 L However the problem when measuring with the 100 mm pencil ionization chamber is that one will not measure the whole single slice physical dose profile which leads to an underestimation of both the equilibrium dose and the integrated dose along the scanning length Sin SNA 16x05 64x0 5 320 x 0 5 Figure 2 Explanatory picture of the evolution of increasing beam widths in CT From single slice widths of 4 5 mm to multi slice widths of 8 32 mm and increasing even more to 160 mm widths for Cone Beam CT scanners The picture is reproduced with permission by the IAEA IAEA publication International Atomic Energy Agency Status of Computed Tomography Dosimetry for Wide Cone Beam Scanners IAEA Human Health Reports 5 IAEA Vienna 2011 page
19. 9 2 55 6 48 5 19 2 48 4 19 1 41 7 48 4 19 1 48 5 19 2 27 8 48 4 19 2 48 5 19 2 18 5 48 5 19 2 47 8 18 9 47 8 19 0 14 9 47 8 19 0 3 3 Testings of the Detectors on Clinical CT scanners The FWHM was also measured using a clinical CT scanner The results are shown in Tab 8 It can be seen that there is a difference between the measured FWHM and the nominal collimation It is also seen that the FWHM is always smaller when measuring with the Mover during axial scanning compared to measuring during helical scanning The measured FWHM seems to be unaffected for different values of pitch For comparison the FWHM according to the reference manual is shown 14 Measuring during helical scanning seems to match closer according to the reference manual than during axial scanning with the Mover In average the measured FWHM with the Mover was 94 compared to the measured FWHM during a helical scan 25 Table 8 The nominal collimations on a GE Discovery 750HD scanner in Gothenburg Sweden and the measured FWHM Also noted is the FWHM according to the reference manual The measurements were made during axial scanning with the CTDP and Mover with a Mover speed of 83 3mm s Measurements were also made with the CTDP during helical scanning For helical scanning only 20 and 40 mm collimations were available and for 20mm collimation different values of pitch were tested Measured during 120kV and Large SFOV Nominal Measured FWHM Measured FWHM FW
20. Calibration amp Clinical Measurements in Computed Tomography An Evaluation of Different Dosimetric Methods Eric Gronlund M Sc Thesis Oslo Spring 2013 Supervisors Anne Thilander Klang Hilde Olerud Hans Bjerke University of Gothenburg Department of Radiation Physics Norwegian Radiation Protection Authority Department of Monitoring and Research 3University of Oslo The Institute of Physics E mail gusgroer student gu se Phone 0046 709 837 548 amp 0047 983 121 49 The Sahlgrenska Academy UNIVERSITY OF GOTHENBURG Abstract This thesis looked into the established CT dosimetry highlighting some problems with it In order to resolve some concerns with it new dosimetric methods are seemingly begin ning to be established One of these methods based on using point like detectors instead of the conventionally used pencil ionization chamber A calibration procedure for detectors in CT was established The procedure was more valid for the pencil ionization chamber than for a point like detector This thesis also compared how well a point like detector could measure the CTDIj09 compared to the pencil ionization chamber Measuring in the conventionally used CT phantoms the correspondence was good But when measuring free in air the correspon dence was not completely satisfying A dosimetric method to be able to give a size specific dose estimate based on the CTDI 9 was also tested This method correctly estimated the me
21. Detectors used in CT 02 2 0 20 000 4 24 3 2 Evaluating the Mover system 2 0 200000 a 24 3 3 Testings of the Detectors on Clinical CT scanners 25 3 4 Evaluate the size dependency of CTDIvol for different phantom sizes 31 Discussion 34 4 1 Calibration of Detectors used in CT aoaaa aaa a 34 4 1 1 Establishing a calibration procedure for detectors used in CT 34 4 2 Evaluating the Mover system aoaaa a 35 4 2 1 Evaluate how the plastic tube affects the measured signal with the CTDP 35 4 2 2 Measuring the FWHM 022000000045 35 4 3 Testings of the Detectors on Clinical CT scanners ooo a 36 4 4 Evaluate the size dependency of CTDIvol for different phantom sizes 38 Conclusion 40 Acknowledgement 41 References Appendix A 42 1 Introduction Computed Tomography CT is an X ray imaging technique used for medical diagnostics A CT scanner utilizes a rotating X ray source together with detectors on the opposite side of the source From the collected information images can be reconstructed with computer aided calculations resulting in tomographic images of the scanned object CT provides more diagnostic information compared to conventional planar X ray imaging since there will not be any overlapping tissue in the images However CT often gives a significantly higher radi ation dose compared with conventional planar X ray imaging 1 Also the frequency of CT ex
22. HM according to Collimation CTDParial mover CTDPhelical reference manual mm mm mm mm 1 25 2 7 2 8 5 7 0 7 4 10 12 0 12 4 20 21 0 22 4 22 3 22 3 21 9 K p 0 531 p 0 969 1 375 40 40 0 42 6 42 2 p 0 516 After the measuring had been done it was noted that the CT system GE Brightspeed Elite chose a larger focal spot when the effect P mA kV was exceeding 24kW The effect from this can clearly be seen in Fig 16 when measuring with the Radcal pencil chamber The different curvature for the Body phantom in Fig 14 is also due to this larger focal spot CTDI 00 100 mAs measured with the different measurement techniques in the phantoms for different tube voltages is seen in Fig 13 and for different nominal collimations in Fig 14 A relatively good correspondence is seen but with a slightly larger difference in Fig 13 when the tube voltage was 140 kV In average the CTDP combined both during helical scanning and axial scanning with the Mover measured 95 of what the Radcal pencil chamber measured The estimated standard deviation on this average value of 95 is 46 CTDIj00 100 mAs measured with the different measurement techniques free in air for dif ferent tube voltage is seen in Fig 15 and for different collimations in Fig 16 and 17 In Fig 15 a relatively larger difference is seen when the tube voltage was 140kV In Fig 16 and 17 for the smallest collimation the CTDP measured extra small value
23. al 3 30 S oS S 20 a ja i 0 ji ji i i ji 80 90 100 110 120 130 140 Tube voltage kV Figure 15 CTDlio0 100 mAs vs different kV measured free in air with the different measurement techniques The measurements were made on a GE Brightspeed Elite scanner in Kristiansand Norway with Large SFOV and 10mm collimation and with a pitch of 0 625 during the helical measurements e I Chamber arial 50 F l Large Focal spot CT D Parial mover CTD Phetical lt a 407 S a Small Focal spot S 30 z N H 20 E P a eRe 10 u 0 5 10 15 20 Collimation mm Figure 16 CTDIio0 100 mAs vs different collimations measured free in air with the different mea surement techniques For the smallest collimation 1 25 mm measurements with the Radcal chamber were made either with a small focal spot tube current 100mA or a large focal spot tube cur rent 300mA All other measurements referred to in this figure were made with a small focal spot The measurements were made on a GE Brightspeed Elite scanner in Kristiansand Norway with Large SFOV and 120kV During helical scanning only 10 and 20 mm collimation were available and the pitch was 0 625 30 e Chamberariai 40 GT D PE gaa mover n e CTDPhetical lt z j E 30 F 4 a ja O o 20 fF ji 4 E l l 0 10 20 30 40 Collimation mm Figure 17 CTDI00
24. aminations is increasing In 2010 it was stated that the frequency of CT examinations had doubled from 2002 to 2008 and that 80 of the population dose from medical imaging came from CT in the country of Norway 2 Because CT is an X ray based imaging technique that increasingly gives a significant contribution to the population dose it is of major importance to have Quality Assurance QA procedures with high validity that is up to date with the growing development of new techniques in CT To be able to make valid measurements for QA procedures the instruments that are used must be calibrated in a standardized and controlled manner Also of great importance is the validity of the instruments used in the QA procedure A part of this thesis is to address these issues This project is about to bring in standardized calibration procedures for detectors that can be used for dosimetric measurements in QA pro cedures for CT It is also about to compare and evaluate instruments and methods that can be used in the QA process 1 1 CT Dosimetry In CT the most common parameter for estimating the radiation dose is the CT Dose Index CTDI CTDI is the integral of air Kerma along the rotational symmetry axis for the X ray tube here denoted z divided with the number of simultaneously acquired slices N of nominal thickness T L 2 CTDI Nelo K z dz mGy 1 where the L in Eq 1 defines the length over which the integral is made ideally the lengt
25. as calculated in the following way n 100 aCTDhoverpp reticat Pd D i t mA mGy 100 mAs 12 where the pitch p was multiplied with the summation of measured dose from to n where n was the number of measured samplings within 100mm This was how the software Ocean automatically calculated the CTDIo0 12 The normalization to the number of 100 mAs was manually made with the ratio Pe When measuring with the CTDP and the Mover during axial scanning the CTDI09 was automatically calculated using Eq 13 The normalization was manually made with the ratio 100 tr mA Umitr 100 nCTDlio0crDP azial mover NT a t mA mGy 100 mAs 13 ve tr Essentially Eq 13 is the same as Eq 12 since pitch is defined as F see Eq 3 where vz is the table speed But now with the difference that no table was moved instead the detector was moved with the mover speed v in mm s One can notice from Eq 13 that the CTDlIio0 normalized to the number of 100 mAs is independent of the tube rotation time tr 20 Figure 9 Explanatory picture of how the measurements were performed with the Mover system This picture demonstrates when the CTDP was used with the Mover during axial scanning in a 32 cm body phantom The picture is from when the measurements was made using a GE Brighspeed Elite 16 Slice CT in Kristiansand Norway 21 2 4 Evaluate the size dependency of CTDI for different phantom sizes AAPM has rec
26. asured CTDI in a small CT phantom This seems to be a promising method to more correctly know the dose to a patient from a CT scan ABBREVIATIONS AND TERMS CT Axial Scanning Helical Scanning Pitch p SFOV CTDI HVL Homogeneity Coefficient CTDP CTDI AAPM SSDE IAEA IEC RQR qualities RQT qualities CTDI phantoms QA PSDL SSDL Metrology Aperture Coverage Factor K factor FWHM Computed Tomography Scanning without table translation during the tube rotation Scanning with table translation during the tube rotation The table translation for one tube rotation during a helical scan divided with the nominal collimation Scan Field of View CT Dose Index Half Value Layer Ratio of ACE CT Dose Profiler CT Dose Index American Association of Physicists in Medicine Size Specific Dose Estimate According to AAPM 204 International Atomic Energy Agency International Electrotechnical Comission Radiation qualities in radiation beams emerging from the X ray source assembly According to IEC 61267 2005 Radiation qualities based on Copper added filter According to IEC 61267 2005 Two conventionally used PMMA phantoms with a diameter of 16cm or 32cm for measuring the CTDI Quality Assurance Primary Standards Dosimetry Laboratories Secondary Standards Dosimetry Laboratories The science of Measurement Term synonymously used for a collimator in radiation Metrology Term used to get an expa
27. by IAEA 457 4 The established calibration procedure in this project is highly valid for the current dosimetric paradigm in CT where the pencil ionization chamber is indispensable But it is also a quite circumstantial calibration procedure meaning that it is easy to make errors However the established procedure also works for calibrating point like detectors But it should be noted that it introduces greater uncertainties for these kind of detectors due to the irradiated length that must be used to get the reference air Kerma To calculate the irradiated length as is done in this project is not the most credible way to handle the length of irradiation Film dosimetry is a possible way to handle this instead however it is not straightforward and in the end not as accurate as one could wish for Remember that it is irradiation lengths of tenths of a millimeter that needs to determined As is described in the introduction the current dosimetric paradigm in CT is loosing its validity A new dosimetric paradigm is starting to be established 8 13 This paradigm based on using point like detectors instead Since the whole dosimetry for CT is seemingly starting to be restructured in a way where it is valid for any beam width Perhaps also the calibration procedures which has a highly significant part in the dosimetry needs to be restructured This new calibration procedure simply based on knowing the air Kerma to a point with for example using a Farme
28. d uncertainty k 2 3 08 Sida 1
29. e 1 2 2 The width of the scanning beam and the phantoms Today often in the clinic the 100 mm pencil ionization chamber is used for dosimetric measure ments of the radiation output from a CT i e measurements of the CTDI 99 Measuring over this length is not sufficient to get the total accumulated air Kerma for in the clinic often used collimations such as 20 40mm Since the air Kerma profile width due to scattering effects extends beyond 100mm the measurement will not collect the whole air Kerma profile Monte Carlo simulations have shown that CTDIj99 compared to CTDI misses 60 in Body CTDI phantoms and 80 in Head CTDI phantoms for 10mm collimations in the central hole This percentage or efficiency was actually approximately the same for 40mm collimations For even larger collimations this efficiency was gradually decreasing until it reached 80mm collimation where it was starting to rapidly decrease even more for larger collimations 7 Clearly this is an issue that needs to be recognized when optimizing the dosimetry in CT Especially in that regard that Cone Beam CT scanners which becomes more common has beam widths often exceeding these widths see Fig 2 Looking at the CTDI phantoms in this aspect they are determined to be at least 14cm long and made from PMMA that has a density of 1 19 cm It can be said that 14cm is not suf ficient to simulate how the radiation scattering tails will spread in a real patient especially as
30. e 20 milliseconds Hence the displayed measuring time was used as the measuring time t The measured air Kerma rate from the reference pencil chamber Eq 5 was used to calculate a calibration factor for the other two detectors For the Radcal pencil chamber for each of the radiation qualities Tab 2 calibration factors were calculated in the following way K Nggo Ref Q mGy mGy 8 KQ raacai For the CT Dose Profiler calibration factors for each of the radiation qualities Q Tab 2 were calculated in the following way _ KrefQ N Hes G in Gaye mGy nC 9 14 Where Co was the mean collected charge in the unit of nC in the specific radiation qualities Q during the mean time t displayed from the three measurements 2 1 4 Normalization When calibrating detectors used for applications in CT the calibration coefficient is generally normalized to radiation quality RQT 9 4 Hence a correction factor kg is supplied for the other radiation qualities In this report normalization was also made to radiation quality RQT9 Radiation quality correction factors dependent for the radiation qualities for the calibrated detectors was calculated in the following way N kg 2 10 Nk RQT9 2 1 5 Uncertatinties during calibration The uncertainties associated with the calibrations was evaluated according to Guide to the Expression of Uncertainty in Measurement 11 see Appendix A For the measurements be tween t
31. e interesting to investigate further 36 Comparing the CTDP with the Radcal pencil chamber when measuring CTDIi00 inside the phantoms the consistency was relatively good Fig 13 14 In average the CTDP mea sured 95 of what the Radcal measured Any differences between measuring helically or axially with the Mover was not seen This should be recognized as a relatively good consis tency especially in that regard that so many input parameters were changed for the different measurement techniques Tab 3 4 This is also one factor that could have improved this project in keeping the input parameters more consistent between the measurements For example after the measurements were made it was noticed that the system changed to a bigger focal spot when the effect was exceeding 24kW This change of size in focal spot gave the greatest affect on small collimations and can be seen in Fig 14 where the difference in curvature from the body phantom is explained by this and in Fig 16 where a great difference was seen when measuring free in air The good correspondence that was seen when measuring in the phantoms also indirectly points to that the integrated energy corrections for the CTDP are valid It is important to keep in mind that attenuation from the patient table can affect the result the table effect Looking at Fig 12 b it can be seen how the table attenuates the irradiation during helical scanning This is not a major problem
32. e turning on and off for the irradiation was made with the lead shutter seen in Fig 7 The signal from the Radcal pencil chamber measured in the specific radiation qualities Q was corrected in the following way to get the air Kerma rate MRadcal i kp 10 1 w t KQ raacai mGy s 7 where Mradcai was the read value from the electrometer in mGy The term 10 was the ef fective length of the chamber in cm kp was the correction for temperature and pressure w 5 10 was the irradiated length in cm according to Eq 6 and t was the irradiation time in seconds received from the monitor chamber Measurements with the Radcal pencil chamber were performed 3times during 60 seconds of irradiation Averaging was made from the three measurements The CT Dose profiler CTDP can not measure a signal integrated over a length due to its small size of 250 um Instead it is assumed that the CTDP measures the air Kerma in a point The calibration of the CTDP was made with the same physical arrangement as with the pencil chambers see Fig 6 7 The CTDP with its corresponding measuring sys tem the Barracuda multimeter serial no BC1 08100077 and the software Ocean Version 2013 03 18 86 from RTI Electronics measured the collected charge in a timed mode for 20 seconds repeated three times Meaning that the system itself stopped measuring after 20 sec onds It also displayed for how long time it had been collecting signals which could deviat
33. each measuring sequence the relevant input parameters are shown When one of those input parameters was changed for a specific measurement it is noted with an underbrace Scanning Helical Axial Axial mode Scanning Scanning Scanning Measurement CT Dose Profiler CT Dose Profiler 100mm Radcal technique with Mover Pencil Chamber Pediatric phantom 10cm Free in air Head phantom 16cm Body phantom 32 cm 80mA tr 1s 80mA tr 4s Vm 83 3mm s 100mA tr 1s kV 80 100 120 140 Collimation mm 1 25 5 10 20 SFOV Large Small Head 100mA tr 1s 80 100 120 140 kV 80 100 120 140 80 100 120 140 SH 200mA Collimation mm 10 20 1 25 5 10 20 1 25 5 10 20 SE 300 amp 100mA SFOV Large Large Large Small Head Pitch 0 625 100mA tr 1s 200mA tr 3s Vm 83 3mm s 200mA tr 1s 80 100 120 140 SS 200mA kV 80 100 120 140 80 100 120 140 Collimation mm 10 20 1 25 5 10 20 1 25 5 10 _20 YY 150mA SFOV Large Large Large Small Head Pitch 0 625 0 875 1 35 1 675 100mA tr 1s 400m lt A tr 3s Vm 83 3mm s 400mA tr 1s kV 80 100 120 140 80 100 120 140 80 100 120 140 x p a a 200mA 380mA 380mA Collimation mm 10 20 1 25 5 10 20 1 25 5 10 20 SFOV Large Large Large Small Head Pitch 0 625 0 875 1 35 1 675 To com
34. ections supplied from its calibration certificate from VSL The CTDP shows the highest energy dependence and the Radcal pencil chamber has a slightly more uniform energy dependence than the reference pencil chamber Also noted is the estimated measurement uncertainties multiplied with a factor of two Cov erage Factor 2 resulting in a confidence interval of approximately 95 11 Table 5 The calibration coefficients Ng for the Radcal and CTDP detectors recieved with help from the reference detector DCT 10 and the expanded measurement uncertainties U with a coverage factor 2 Also noted is the radiation quality correction factors kg normalized to RQT 9 The radiation qualities in this table is based on IEC 61267 2005 DCT 10 The reference Radcal Calibrated CTDP Calibrated NgL 26 13 mcy cem ncy NK 1 016 tmcy mey NK 0 2881 mey ncy U 1 7 U 2 2 U 3 1 kg kg kg From VSL measured measured RQT 8 0 958 0 968 0 882 RQT9 1 1 1 RQT 10 1 083 1 075 1 244 RQR9Y 1 026 1 053 1 020 3 2 Evaluating the Mover system The plastic tube from the Mover system was studied if it had an effect on the signals from the CTDP The effect was significant and the results are shown in Tab 6 The greatest effect was seen in radiation quality RQR9 which has the lowest value of HVL This effect seems to be dependent on the HVL Tab 2 Table 6 The mean measured collected charge with the CTDP during three exposures of 20 seconds each Fr
35. ently released a report 6 where four different groups independent of each other have investigated how CTDI can be size converted to other sizes than the conventional CTDI phantoms A so called size specific dose estimate SSDE Their investigations were made by theoretical Monte Carlo simulations and by practical measurements in different phantoms With the results they independently achieved and in good agreement between each other conversion factors for CTDI dependent on effective diameter was produced The term effective diameter is mathematically explained with the following relationship 4A Effective diameter 4 9a cm 14 T and is introduced because patients are not completely round but can be assumed to be of elliptical shape Where the lateral width LAT and anterior posterior thickness AP in cm are the two parameters that can calculate the effective diameter according to the following relationship Effective diameter v AP LAT cm 15 The conversion factors f in AAPM 204 6 for CTDI dependent on effective diameter can be seen in Fig 10 The conversion factors are valid from knowing CTDI for the 32cm Body phantom or the 16cm Head phantom 4 F T T 0 03671937 Body Phantom _ e 9 x E o Head Phantom s 3 Fn A O amp g 2 F Be gp tH oO z Tf oO OL 1 874799 e7 0 03871313 x i
36. entre This was also substantiated from to the values in the reference manual The nominal col limation actually corresponds to an image parameter and not the physical beam width 8 The measured FWHM from the Mover was slightly below what the reference manual said The measured FWHM from helical scanning was slightly above but more close to what the reference manual said This could point to that the scanner collimation was affected whether it was in axial or helical scanning mode The values from the reference manual was measured during helical scanning However it is interesting to point out that when measuring with the Mover it always measured 95 of what was expected Both in the SSDL Tab 7 and compared with the measured FWHM during helical scanning Tab 8 This probably means that the Mover speeds are slightly higher than what is stated This also points to that the collimation on the scanner was unaffected whether it was in helical or axial scanning mode 35 To be able to measure FWHM on a clinical scanner to an accuracy that deserves justification instead of using the nominal collimation the accuracy of the measuring system should be known Measuring the FWHM with the CTDP during helical scanning seems to be accurate enough However when measuring during axial scanning with the Mover the accuracy is de pendent on how true the Mover speed is In conclusion it can be said that when using the Mover system the Mover itself could need a spec
37. fferent detectors were positioned in the isocentre completely free in air The plastic tube was on the CTDP for all measurements free in air When measuring with the Radcal pencil chamber measurements were performed in all holes in the phantoms According to Eq 3 and 4 the CTDI and CTDI could then be calcu lated The temperature and pressure was noted and the measurements with the Radcal pencil chamber were corrected for that This correction was not necessary for the CTDP since it is a solid state detector When measuring with the CTDP and its corresponding measuring system the Barracuda multimeter and the software Ocean measurements were only performed in the central hole in the phantoms This is explained by that with a pencil chamber one measures directly the integrated exposure along its effective length The CTDP is instead measuring the exposure rate from points along a length due to it being moved In a peripheral hole the measured exposure rate with the CTDP will vary dependent on where the X ray tube is relative to the CTDP However along a length in the central hole the exposure rate will be approximately the same during the whole tube rotation Hence the CTDP can validly be used for measurements in the central hole 12 To get the CTDI from measuring with the CTDP one multiplies the CTDlhoo with a k factor that is defined as k OTs 12 where CTDL is originally measured with a pencil chamber The scanner parameters that were
38. h of integration should be equal or longer than the actual physical air Kerma profile width CTDI can be interpreted as the whole air Kerma profile being deposited in the slice that is defined by the nominal beam collimation The following terminology on Dosimetric parameters in CT will be based on the International Standard IEC 60601 2 44 3 CTDI can both be measured free in air and ina CTDI phantom Regardless of in what one measures it is still air Kerma that is being measured 4 The CTDI phantoms most often used so far are cylindrical with a minimum length of 14cm and made from polymethylmethacrylate PMMA with a diameter of 32cm corresponding for a human body and 16cm corresponding for a human head 3 They have a central hole and peripheral holes every 90 1 cm below the surface where a detector can be put in the holes to measure the air Kerma see Fig 1 Most often the detector used is a pencil ionization chamber with an effective measuring length of 100mm This means that equation 1 needs to be integrated from L 2 50mm to L 2 50mm and the correct denotation is then CTDIi09 3 The reason for the peripheral holes and the central hole is because then a weighted CTDI can be calculated Figure 1 Picture of the two conventionally used CTDI phantoms The smaller Head phantom with a diameter of 16cm and the bigger Body phantom with a diameter of 32 cm from the measurements 3 1 3 where CTDlioo e is the CTDlio
39. he reference pencil chamber DCT 10 and the Radcal pencil chamber the irradiated length w was the same Hence no extra uncertainties in this regard was taken into account between those two detectors For the CTDP however the effective irradiated length was not the same since the diode in the CTDP was assumed to be a point The irradiated length w for the reference pencil chamber was needed according to Eq 5 to know the air Kerma to point For the irradiated length w according to Eq 6 a uniform distribution with an esti mated uncertainty of 0 1 cm was taken into account Which in turn gave an extra uncertainty for the CTDP but not for the Radcal pencil chamber in this regard 2 2 Evaluating the Mover system 2 2 1 Evaluating the accuracy of measuring the FWHM of beam profiles with the Mover system In order to measure the CTDI g9 on a CT scanner the dose profile needs to be integrated over 100 mm The CTDP can not directly measure over this integration of length since it assumed to be a point like detector Instead it is dependent to accurately be moved through this length of irradiation RTI Electronics have recently released a Mover system that can be used together with the CTDP This Mover system gives a solution to measure the CTDI when no table translation is possible The Mover used in this project was a prototype with serial number 001 In Fig 8 the principle of how the CTDP can be moved with the Mover is shown A steel wire is t
40. hreaded to the CTDP and an electrical motor moves the wire When measuring free in air a PMMA plastic tube is used to give support for the CTDP Not only measuring the CTDIjo09 or CTDI along any length of choice is possible with this system It is also possible to measure the actual full width at half maximum FWHM FWHM is often not the same as the nominal collimation determined from the manufacturer of the CT scanner 3Radiation quality correction factor Beam quality correction factor Energy correction factor are synony mously used terms Different sources use different terms The term Energy dependence is often colloquially used and also refers to the same meaning 15 Button to release the wire Plastic tube for free in air measurements Protection CT Dose tube Profiler probe Knob to manually move the wire Indicators of direction push and pull Quick mount with a ball head LED indicators and USB connector Base plate Figure 8 Explanatory picture of how the RTI Mover looks like and the principle of moving it with a steel wire The plastic tube is necessary to give support for the CTDP when measuring free in air The picture is reproduced with permission by RTI Electronics RTI Electronics Mover User s Manual English Version 1 0C page 11 The accuracy of measuring the FWHM with this system was performed at the SSDL The Mover together with the CTDP were set up in the same way a
41. ic tube and see the results Comparing the results in Fig 16 to Fig 17 a different curvature for different collimations is seen The reason for this is explained by that it was two different scanners The results between Fig 16 and Fig 17 should for that reason not directly be compared with each other Another thing to point out is that when the collimation was very narrow 1 25 mm the CTDP measured even lower values than the Radcal pencil chamber 65 But this effect has been recognized before 12 and is originated from that the diode in the CTDP needs a scattering contribution from wider fields When the fields are more narrow than 4 cm this effect is automatically corrected for However there is no valid correction for field sizes as narrow as 1 25 mm it just goes down to 3 mm This scattering contribution is not a problem when measuring in the CTDI phantoms since the scattering then is so considerable A small difference was also seen when measuring free in air Fig 15 17 between mea suring during helical scanning or with the Mover during axial scanning Measuring with the Mover during axial scanning the measurements were a bit lower As is described earlier from the FWHM measurements it is possible that the actual Mover speed was a bit higher than stated It is very interesting to point out that the measurement of the FWHM with the Mover was 5 lower than the true beam widths both clinically and in the SSDL Tab 7 8
42. ific speed calibration or a correction factor in the measuring software Ocean Also the many Mover speeds that can be chosen is not necessary two or three accurate Mover speeds should be enough 4 3 Testings of the Detectors on Clinical CT scanners CTDP needs another theoretical approach than measuring with the pencil chamber Looking at Fig 11 12 what is displayed is how the exposure rate varies for different times From knowing the speed then the positions can be determined It is easy to deceive oneself that it is an exposure profile displaying the exposure for different positions If one would for example measure the exposure profile with TLD detectors then this approach would be true But when measuring with the CTDP this approach needs to be changed For example holding the tube voltage and collimation constant but doubling the tube current and traveling speed one will measure the same total exposure As can be seen in Eq 13 it was shown that CTDIi00 100 mAs was independent of the tube rotation time tr This is not entirely intuitive but can be explained by the following thought experiment Assume that the dose to a point somewhere along the rotational symmetry axis z is measured with two different point like detectors The first detector measures the accu mulated dose from a tube rotation D z The second measures the dose rate D z during the tube rotation The following relationship can then be seen between the two detectors
43. in the conventional CTDI phantoms the results showed a good correspon dence Fourthly this project evaluated how size specific dose estimates could be used from CTDI o to more accurately know what the exposure will be for patients of different sizes Using these size specific dose estimates from AAPM 204 seems to be a promising way for the doctors and radiographers to more accurately know the exposure for any patient Especially in that regard that CTDI tends to be underestimated for small children 40 Acknowledgement Thanks to my supervisors e Hans Bjerke for his expertise in dosimetry calibration procedures and having interesting ideas e Hilde Olerud for her expertise in CT dosimetry positive spirit hospitality and for being the founder of this project e Anne Thilander Klang for sharing her clinical expertise and both very deep and broad knowledge e Also many thanks to RTI Electronics AB and Lars Herrnsdorf for their support e Lastly many thanks to Erlend Andersen at S rlandet Hospital in Kristiansand for shar ing a CT to this project 41 References 1 2 3 _ 4 5 6 7 8 9 10 11 Computed Tomography An Increasing Source of Radiation Exposure David J Brenner Eric J Hall New England Journal of Medicine 2007 357 2277 2284 2007 11 29 Radiologiske unders kelser i Norge per 2008 Trender i undersgkelsesf
44. ing fully irradiation then assuming it was a point detector that together with its corresponding electrometer was set to give a value in mGy This was why the Radcal pencil chamber during partial irradiation needed a correction for the irradiated length and also got the calibration factor in the unit of mGy mGy Tab 5 Lastly the CTDP do not share two of the physical attributes that the ion chambers have The radiation interacts in a solid material instead of a gas and the signal is not created along a length To be able to calibrate the CTDP from a reference pencil ionization chamber some extra considerations about the irradiated length needed to be made This gave an extra uncertainty to the calibrated CTDP In Tab 5 it shows that the CTDP has the biggest energy dependence It also shows that the energy dependence for the reference pencil chamber is a bit higher than the calibrated Radcal pencil chambers energy dependence It should also be noted that from a calibrating point of view although all parameters are kept under control to that extent that is possible for a SSDL Using a reference chamber that has the relative high energy dependence as in this project is not ideal 4 1 1 Establishing a calibration procedure for detectors used in CT In this project a calibration procedure was established It was based on two references firstly and foremostly in a manner on how the reference pencil chamber was originally calibrated But also as described
45. ion chambers and worked for a point like detector However it seems that the dosime try in CT will enter a new paradigm that will be consistent and robust for any type of CT scanner and beam width A dosimetry based on using point like detectors instead The estab lished calibration procedure turned out to highlight one very important aspect namely the length of irradiation This parameter is not as easily handled as one could think Perhaps calibration procedures in CT should be based on another more simple method A method of just knowing the air Kerma to a point Secondly the recently released Mover system from RTI Electronics AB was evaluated The Mover is developed to be used with a solid state point like detector called CT Dose Profiler CTDP When measuring the FWHM with the Mover system it seemed to measure a bit too small lengths Possibly it travels to fast However it seems to be a promising technique for dosimetric measurements on CT scanners when no table translation is possible Thirdly this project was used for a clinical comparison between the conventional pencil ion ization chamber and the CTDP The CTDP was used for measurements both during helical scanning and during axial scanning with the Mover The pencil ionization chamber was used during axial scanning Measuring free in air did not show the correspondence one could wish for Quite possibly the plastic tube to give support for the CTDP was a contributing factor When measuring
46. ion were made free in air but also made with the plastic tube on the CTDP It was studied how the the collected charge was affected due to the plastic tube 16 2 3 Testings of the Detectors on Clinical CT scanners To see how the calibrated detectors were performing for dosimetric measurements clinically they were tested on a clinical CT at S rlandet Hospital in Kristiansand Norway See Fig 9 to get an insight on how the measurements were performed The CT was a SPECT CT from GE Healthcare GE Discovery NM CT 670 but only the CT system was used GE Brightspeed Elite 16 Slice CT Essentially the two calibrated detectors were used the Radcal pencil cham ber and the CTDP But three different measurements techniques were used Firstly helical scanning and measuring with the CTDP Secondly axial scanning and measuring with the CTDP that was moved through the irradiation with the Mover Thirdly axial scanning and measuring with the Radcal pencil chamber Measurements with the detectors were performed free in air in the conventional Body and Head phantoms see Fig 1 and in a specific Pediatric phantom with a diameter of 10cm In the pediatric phantom measurements were only performed with the Radcal pencil chamber since the holes were a bit to small for the CTDP When measuring in the CTDI phantoms the phantoms were placed on the patient table Positioning was made with help from lasers integrated in the CT When measuring free in air the di
47. is is called Type B uncertainties according to GUM Often Type B uncertainties are estimated to have a uniform distribution of uncertainty often called a rectangular distribution Also a triangular distribution of uncertainty can be assumed All uncertainties associated with the calibration procedure was calculated according to the following equation VERAN 18 where the square root of the quadratic sum for all influencing factors u was multiplied with a coverage factor also called K factor in GUM 11 The K factor in this project was K 2 which results in a confidence interval of approximately 95 In the following page it can be seen how the uncertainties associated with the calibration of the CTDP was evaluated In several cases a scientific judgement was made for Type B uncertainties A rectangular or a triangular distribution was assumed Denoted R or T in the following page N is the denotation for a normal distribution Tabell1 Source of uncertainty Quanti stimate Deviation frel st uncert Prob d Sens Coeff X xX u u x c u x 1 Factors due to radiation field set up and calibration 1 1 X ray output stability Krab 1 0 0015 0 0015 N 1 1 2 Differences in energy spectra of radiation beams Kopec 1 0 0009 0 0005 R 1 1 3 Field inhomogeneity Kinnom 1 0 0000 0 0000 N 1 0o00 1 4 Uncertainty of the calibration coefficient reported by PSDL Ne 26 13 0 2221 0 0085 N 1 1 5 Constancy of the calibration coefficient 26
48. just the attenuation from the plastic tube As is shown in Tab 5 the energy dependency for the CTDP is relatively high Possibly the plastic tube differs the radiation quality that the diode in the CTDP is measuring and hence the difference maybe is explained When measuring with the CTDP and the Mover free in air with the plastic tube on perhaps it could need radiation quality factors specific for when the plastic tube is on or finding another solution on how to move the CTDP without a plastic tube 4 2 2 Measuring the FWHM When measuring the FWHM with the Mover in the SSDL the values seemed to be consistent for all speeds except the lowest one 14 9 mm s In this thesis the assumption is made that the true width of the collimated beam is explained by Eq 6 According to this assumption the true widths of the beam should be 51 0 and 20 4mm and the measured FWHM with the Mover system was approximately 48 5 and 19 2mm respectively Tab 7 The system showed a good precision between measurements but the accuracy of it needs to be investigated further A possible source of error is the accuracy of the Mover speed since it affects the time it is in the beam and further on the calculation of the FWHM When measuring the FWHM on a clinical scanner Tab 8 there was a difference between the nominal collimation and the measured FWHM This was also expected since the nominal collimation displayed from the scanner not equals the actual FWHM in the rotation c
49. mm just below the specified nominal HVL into the beam The current was then measured giving a measured value Ipvi T The mover moved attenuating aluminium with a thickness Tz mm just above the specified nominal HVL and the current was measured in the same way giving the measured value Ipvi T Lastly the mover moved away the attenuating aluminium out of the beam and the current was measured in the same way as without any attenuating material to see if the X ray system had drifted in any way giving the measured value I 2 The average value of the two measured values Io and I 2 was calculated giving Io Every measured value was temperature and pressure corrected and corrected for the leakage current The ratio of Iyvz r Io and IuvL T Io was calculated and these values were plotted on the Y axis with the corresponding thicknesses of the attenuating aluminum T and T on the X axis The nominal HVL was linearly interpolated in the plot between the values T and T2 to see what the ratio would be with the nominal HVL see Fig 5 If the ratio of the interpolated value for the nominal HVL was between 0 485 0 515 the Radiation Quality was considered to be obtained 10 If the ratio was not between these values the filtration on the plastic wheel was adjusted If the ratio was too high added filtration was needed and if the ratio was too low filtration needed to be removed S on ch on Al filtration T1 0 51 H
50. n the picture c The filtration wheel for the different radiation qualities and a lead shutter behind the filtration hole where the radiation comes out The lead shutter makes it possible to stop the irradiation without turning of the X ray tube 13 2 1 3 Calibration of the detectors After measuring with the reference pencil chamber the Radcal pencil chamber and its corre sponding electrometer Model No 9010 Serial no 90 1562 were calibrated The electrometer measured a signal MRaacal in the unit of mGy One could not measure any collected charge or current but only what the electrometer displayed with this system which was air Kerma The Radcal detector system was originally calibrated during fully irradiation assuming it was a point detector So the value of air Kerma that the electrometer displayed was valid if it was fully irradiated If the Radcal pencil chamber as in this case was partly irradiated it needed a correction of the irradiated length Also it needed to be a correctly determination of the irradiated time to measure the air Kerma rate This was determined using the Capintec 30cc PM30 chamber used for the HVL measurements as a monitor chamber behind the Radcal pencil chamber see Fig 7 The monitor chamber was connected to a Keithley 35040 elec trometer which can give an accurate value of the irradiation time since it turns on the timing when it detects radiation and turns of the timing when no radiation is detected Th
51. nd also during the whole tube rotation However this table effect is something to be aware of when measuring with the CTDP It should also be said that even though the Radcal pencil chamber was reliable for measuring the CTDlIoo it will always have that great disadvantage that it can not measure for longer lengths than 100mm The CTDP do not have this disadvantage and better yet one can see how much of the dose profile that is captured Also the CTDP is not limited to mea sure the CTDIj99 measuring the CTDI along any length of choice is possible with this system When measuring free in air Fig 15 17 the consistency was not so good anymore The assumption was made that the measurements with the Radcal pencil chamber was closest to the truth The difference seen with the CTDP can possibly partially be explained by that the plastic tube was always on the CTDP both during helical scanning and during axial scanning 37 with the Mover The reason for this was that this effect had not been recognized when the measurements were performed However looking at Tab 6 the maximum decrease in signal due to the plastic tube was 10 in RQR9 In average the CTDP measured 80 compared to the Radcal pencil chamber So logically it is not just the effect from the plastic tube that is seen What else that is contributing to this effect needs to be investigated further It would be very interesting to measure the CTDlIo9 free in air without the plast
52. nded uncertainty within a desired level of confidence Full Width at Half Maxiumum Contents 1 5 Introduction 1 1i CT Dosimetry 2 2 2 6 0 Sb oatn Oh eet pA ERE Se PED ee ee ROS EA 1 1 2 Issues about CTDIjg9 2 a a a 3 1217 Patient S17 3 5 eed ei eee we Rew eee 3 oe ee ees 3 1 2 2 The width of the scanning beam and the phantoms 3 1 2 3 The scanning length 02 0 0 00000000004 4 1 3 New dosimetric techniques for CT 0 0 000020 08 4 1 4 Chain of Calibrations s d a e vee eos RE Ee Ee 5 1 5 Goals for this Stud yos a x itd a ie a ode 2 A DA 7 Materials and Methods 8 2 1 Establishing a calibration procedure for detectors used in CT 8 2 1 1 Acquiring of the Radiation Qualities 2 8 2 1 2 Calibration of Detectors used in CT 2048 11 2 1 3 Calibration of the detectors 0 0 0 000000 202008 14 221A Normalization oss wien hie at te oh ee ae ae Bo 15 2 1 5 Uncertatinties during calibration ooo a 15 2 2 Evaluating the Mover system oaoa a 15 2 2 1 Evaluating the accuracy of measuring the FWHM of beam profiles with the Mover system ss 2b ein Bao ey e ee EOE a ee eo 15 2 2 2 Evaluate how the plastic tube affects the measured signal with the CTDP 16 2 3 Testings of the Detectors on Clinical CT scanners 17 2 4 Evaluate the size dependency of CTDI for different phantom sizes 22 Results 24 3 1 Calibration of
53. nt was 200 mA The difference in width between a and b is due to different preset of measuring times because of the different speeds the CTDP travelled through the beam Note that the width in the length scale is approximately the same but not in the time scale between a amp b The Heel effect is evident from the profiles 27 50mm 450mm a 50 mm 50 mm b Figure 12 Dose profiles from the CTDP in the measuring software Ocean when measuring in the Body phantom on a GE Brightspeed Elite scanner in Kristiansand Norway with 120kV and Large SFOV a 10mm collimation measured during axial scanning with the mover with a speed of 83 3 mm s the tube current was 400 mA b 10mm collimation measured during helical scanning with a pitch of 0 625 the tube current was 100mA The effect from table attenuation is evident from the dose profile in b But not in a since the tube rotation were slowed down to 3seconds and the Mover system with the CTDP in this case had time to capture the dose profile while the X ray tube where above the patient table 28 40 80 90 li li li l 100 110 120 130 140 e Pediatric I Chamberf arial s HeadI Chamberaziai 2 39 e Head CT DParial mover 2 cere Head CT DPhetical S BodyI Chamberaziat sl e
54. nvert the measured current to a reference air Kerma rate in the specific beam qualities Q the following relationship was used Mpref NKL kp kQ W mGy s 5 where Nr 26 13mGy cm nC was the calibration coefficient and kg corrected for the radiation quality Nx and kg was supplied from the calibration certificate from VSL kp was the correction for temperature and pressure w was the irradiated length of the chamber and was calculated from the following relationship Kref Q _ Wa dr w cm 6 where W 5cm was the width of the square aperture d 100 cm was the distance between the focal spot and the point of test and da 98cm was the distance between the focal spot and the exit of the square aperture see Fig 6 The irradiated length w in cm of the pencil chamber was calculated according to the geometrical relationship of similarity and also as described by IAEA 457 on how pencil ionization chambers should be calibrated 4 12 Figure 7 An explanatory picture of the arrangement for how calibrations of the detectors were performed a The reference pencil chamber IBA DCT 10 100cm from the tube focus put on a plastic holder that is fastened on a stand used when calibrating DAP meters It is positioned behind the 5 x 5cm square aperture which exit is 98cm the tube focus b The monitor chamber that can be used for determination of the irradiated time can be positioned here it is not completely seen i
55. o from the central hole and CTDIjo00 is the average CTDlioo from the peripheral holes The factors 2 and 3 represents the relative air Kerma contribution assuming a linear decrease of the air Kerma in the CTDI phantoms from the periphery to the centre 5 CTDI can be interpreted as the average air Kerma in the irradi ated cross section Further on to take account for any gaps between successive scanning or take account for successive scanning without table translation CTDI o is introduced 3 2 CTDI 3 CTDlio0 e Z CTDlioop mGy 2 x4 CTDL axial scanning NT CTDI CTDIyo1 Rd CTDL pitch helical scanning mGy 3 Ntr CTDI without table translation where N is the number of simultaneously acquired slices of nominal thickness T For axial sequential scanning Ad is the table translation between the consecutive scans For helical scanning Ad is the table translation for one tube rotation and the ratio od is called the pitch 3 A special case is when there is no table translation then the CTDI is defined as the number of tube rotations nir multiplied with the CTDI When a patient is scanned in a CT the scanning goes over the length that will be used for image reconstruction The scanned length is a bit longer than the reconstructed length To get a rendering of the whole radiation exposure across the total scanned length the term DLP Dose Length Product is introduced 3 CDT I 01 Ad Ns axial
56. om measuring with the CTDP completely free in air or with the plastic tube on Also shown is the ratio between the measurements in percent Charge nC Charge nC Ratio free in air with plastic tube Pisstictube RQT8 19 344 18 135 93 75 RQT 9 23 888 22 822 95 54 RQT 10 31 171 30 136 96 68 RQR 9 51 658 46 645 90 30 24 How accurate the FWHM could be measured was investigated for different Mover speeds and the results are shown in Tab 7 The precision between the measurements seems to be good but a deviation is found for the lowest mover speed 14 9mm s The accuracy of the measurements are not perfect when the collimations used were known very precisely The assumed true beam widths w from the collimations were calculated according to Eq 6 and was 51 0 and 20 4mm respectively In average when w 51 0 the measured FWHM was 95 of that value In average when w 20 4 the measured FWHM was 94 of that value Table 7 The measured FWHM with the CTDP for different mover speeds during 50 respective 20mm collimation in the SSDL The assumed true value of beam width w was 51 0 and 20 4mm respectively for the collimations This table shows the two measured values of FWHM except for the lowest mover speed 14 9mm s where three measurements were made Mover speed Measured FWHM Measured FWHM at 50mm collimation at 20mm collimation w 51 0 mm w 20 4 mm mm s mm mm 48 5 19 2 83 3 48 4 19 2 48 6 1
57. on of measurement data Guide to the Expression of Uncertainty in Measurement JCGM 100 2008 42 12 13 14 15 16 Manual for the CT Dose Profiler by RTI Electronics Inc CT Dose Profiler User s Manual English Version 5 1A september 2012 Review of Methods to Control Patient Doses and Image Qualty in Various CT Techniques Lars Herrnsdorf Proceedings of the International Conference Medical Physics 2012 8 10 November 2012 Kaunas Lithuania Discovery M CT750 HD Technical Reference Manual English 5432432 1EN Revision 1 2012 Evaluation of two thin CT dose profile detectors and a new way to perform QA in a CTDI head phantom Cederquist Bj rn Master Degree Thesis University of Gothenburg 2008 X ray mass attenuation coefficients in PMMA http physics nist gov PhysRefData XrayMassCoef ComTab pmma html 43 Appendix A The uncertainties associated with the calibration procedure was evaluated according to Guide to the Expression of Uncertainty in Measurement GUM 11 It was evaluated in three steps Firstly factors influencing due to radiation field set up and calibration Secondly factors influencing the use of the reference chamber Thirdly factors influencing the use of the detector for calibration Some uncertainties are known so called Type A uncertainties according to GUM Some uncer tainties can not exactly be known and a scientific judgement needs to be made th
58. osimetric conversion factor d factor of 1 073 mGy dose MGy air Kerma This gave the SSDE in dose to tissue and not air Kerma Measurements of the CTDlioo in both the central and peripheral holes were done with the Radcal pencil chamber in the 32cm body phantom 16cm head phantom and in the 10cm pediatric phantom This were done with the GE Brigthspeed Elite scanner in Kristiansand Norway From these mesurements the CTDI according to Eq 2 and CTDI 7 according to Eq 3 could be calculated for each phantom Testing the accuracy of using the size conversion factors f in Fig 10 were performed in the following way The measured CTDI from the 32 cm body phantom and from the 16 cm head phantom respectively was used in Eq 16 to calculate an SSDE for an effective diameter of 10cm The conversion factors f that were used was based from the equations seen in Fig 10 where x 10 in these cases The conversion factors were divided with the d factor to consistently get the unit in air Kerma The size converted CTDI from the head and body phantom to an effective diameter of 10cm was compared with the CTDI measured in the 10cm pediatric phantom 23 3 Results 3 1 Calibration of Detectors used in CT Detectors were calibrated in standardized radiation qualities from the reference pencil cham ber IBA DCT 10 Their measured calibration coefficients and energy corrections is shown in Tab 5 The reference pencil chamber had energy corr
59. plement some of the measurements testings between the CTDP and the Radcal pencil chamber was made on another scanner at Sahlgrenska University Hospital in Gothen burg Sweden The scanner used was from GE Healthcare GE Discovery CT 750HD The measurements were only performed free in air for different collimations Different values of pitch during helical scanning with the CTDP was also tested The FWHM that the CTDP can measure was also noted to see how the FWHM would be affected for different settings see Tab 4 The plastic tube was always on the CTDP for these complementary measurements Table 4 The different scanning modes and detector systems that were used for measuring the CTDIioo free in air during different parameter settings on a GE Discovery CT 750HD in Gothenburg Sweden The emphasized parameters correspond for the reference settings that was used when another parameter was tested Over each measuring sequence the relevant input parameters is shown When one of those input parameters was changed for a specific measurement it is noted with an underbrace During helical scanning at 20mm collimation different values of pitch were also tested For the CTDP both during helical scanning and axial scanning FWHM was noted Scanning Helical Axial Axial mode Scanning Scanning Scanning Measurement CT Dose Profiler CT Dose Profiler 100mm Radcal technique with Mover Pencil Chamber 100mA tr 1s 100mA tr 2s Vm 83 3mm s 100mA
60. r chamber The faithful Farmer chamber can in a PSDL be calibrated for air Kerma to knowing its accuracy within 4 o A much higher degree of certainty than for example the reference pencil chamber used in this project Tab 5 Calibration of pencil ionization chambers would still be valid with this procedure if they were 34 fully irradiated Then assuming they also are point like detectors Some of the contributing factors of uncertainty would then be how homogenous the radiation field is for the pencil chamber Also the stem effect and differences in ion collecting potential near the ends of the central electrode would contribute It should be investigated with how large uncertainties these factors would contribute Lastly one question that should be asked but not answered here is the question if not the whole dosimetry in CT should be restructured to knowing the absorbed dose to water instead 4 2 Evaluating the Mover system 4 2 1 Evaluate how the plastic tube affects the measured signal with the CTDP In Tab 6 it shows that using the plastic tube when measuring free in air affects the measured signal for the CTDP A rough estimate on radiation quality RQT8 with 100kV assuming a mean energy of 2 100 60keV and a wall thickness of 1 3 mm PMMA with attenuation coefficients received from NIST 16 the estimated attenuation due to the PMMA will be 3 This meaning that the measured differences shown in Tab 6 can not be fully explained with
61. rekvens og str ledoser til befolkningen Alm n A Friberg EG Widmark A Olerud HM StralevernRapport 2010 12 www nrpa no Publikasjoner StralevernRapport Medical Electrical Equipment Part 2 44 Particular requirements for basic safety and essential performance of X ray equipment for computed tomography IEC 60601 2 44 Edition 3 0 2009 02 IAEA Technical Reports Series No 457 Dosimetry in Diagnostic Radiology An International Code of Practice International Atomic Energy Agency Vienna 2007 Computed Tomography Dose Assessment A Practical Approach Leitz W Axelsson B Szendr6 G Radiation Protection Dosimetry Vol 57 Nos 1 4 pp 377 380 1995 Nuclear Technology Publishing Size Specific Dose Estimates SSDE in Pediatric and Adult Body CT Examinations Report of AAPM task group 204 2011 The trouble with CTDI100 Boone JM Med Phys 34 1364 1372 2007 Comprehensive Methodology for the Evaluation of Radiation Dose in X ray Computed Tomography A New Measurement Paradigm Based on a Unified Theory for Axial Helical Fan Beam and Cone Beam Scanning With or Without Longitudinal Translation of the Patient Table Report of AAPM task group 111 The future of CT Dosimetry 2010 A new look at CT dose measurement Beyond CTDI Dixon RL Med Phys 30 6 June 2003 Medical diagnostic X ray equipment Radiation conditions for use in the determination of characteristics IEC 61267 Second Edition 2005 11 Evaluati
62. s compared to the Radcal pencil chamber During 1 25 mm collimation small focal spot in Fig 16 and in Fig 17 the CTDP measured 63 and 69 respectively of what the Radcal chamber measured An un expected value for the CTDP is also seen in Fig 16 when measuring during helical scanning at 20mm collimation it measured approximately the same as the Radcal chamber Disre garding the more protruding values just noticed the CTDP measured during axial scanning with the Mover in average 78 of what the pencil chamber measured Disregarding the same protruding values again the CTDP measured during helical scanning in average 82 of what the pencil chamber measured The CTDP during axial scanning free in air with the Mover measured very consistently 5 less than during helical scanning free in air This dependence was not seen when measuring in the phantoms In the following two pages dose profiles from the measuring software Ocean are shown 26 b Figure 11 Dose profiles from the CTDP in the measuring software Ocean when measuring free in air on a GE Discovery 750HD scanner in Gothenburg Sweden with 120kV and Large SFOV a 40mm collimation measured during axial scanning with the Mover with a speed of 83 3 mm s the tube current was 100mA b 40 mm collimation measured during helical scanning with pitch 0 516 the tube curre
63. s when the calibration procedures were made i e with the 5 x 5cm lead aperture which was exactly 98 cm from the tube focus and the centre of the CTDP placed exactly 100 cm from the tube focus see Fig 6 Inside the 5 x 5cm opening a lead insert of 2cm width could be placed to further collimate the beam In the measuring software Ocean one can choose with what speed the Mover moves the CTDP The highest speed being 83 3mm s and the lowest 14 9mm s and several speeds to choose in between Measuring was repeated two times in both 50mm collimation and 20mm collimation with mover speeds 83 3 55 6 41 7 27 8 and 18 5 mm s But repeated three times with the lowest mover speed 14 9 mm s The CTDP total traveling with the Mover was set to be 100mm in all measurements letting the travel start and stop well outside the opening of the collimations The measured FWHM was automatically calculated in Ocean The measured FWHM was compared to the beam widths w The true beam widths was assumed to be described by Eq 6 Giving w 51 0 and w 20 4mm respectively for the collimations used 2 2 2 Evaluate how the plastic tube affects the measured signal with the CTDP The Mover system has a PMMA plastic tube for the CTDP to give support for it when measuring free in air see Fig 8 The wall thickness of the plastic tube was measured to be 1 3 mm How this plastic tube affects the signal that the CTDP measures was investigated All the measurements during calibrat
64. scanning DLP CDT Iyo1 L helical scanning mGy 7 cm 4 CTDIy 01 NT without table translation where for axial scanning Ad is the table translation between the consecutive scans and ns is the number of scans in the series For helical scanning L is the total table translation during the series The special case when no table translation is performed the length is defined as the number of simultaneously acquired slices N of nominal thickness T 1 2 Issues about CTDloo 1 2 1 Patient size CTDI is defined for the two CTDI phantoms of diameter 16cm or 32cm that are at least 14cm long 3 Using for example a standard scanning protocol used for scanning chests on adults the displayed CTDI and DLP on the scanner console will be based from the 32cm Body phantom This is independently on the actual size of the patient being scanned since it is only defined for the phantom However the smaller the patient keeping the same scanning parameters the greater the actual radiation exposure of the patient will be This is something to put extra consideration on especially when children are being scanned To be able to correct the CTDI for patients of different sizes the American Association of Physicists in Medicine AAPM has published a report to give more correctly size specific dose estimates SSDE for different patient sizes based on the CTDI 6 A part of this thesis is to evaluate how accurate these size specific dose estimates ar
65. sing this method to correct for size is related with an approxi mate uncertainty of 20 But the measurements in this project point to an accuracy greater than that With that said the correction factors are tended to be used from the CTDI displayed from the scanner console In this project although not given in the section of re sults the CTDI displayed from the scanner console was always higher than measured and in average 8 higher Also this investigation was made during ideal conditions i e the pediatric phantom used do not correctly simulate a small real patient With this in mind using the size specific dose estimates from the scanner console to estimate the real patients dose an uncertainty of 20 as stated is probably reasonable It should be noted that these size specific dose estimates are directed to be used from CTDI 9 They do not solve the propriety in using this parameter However CTDI is so far conven tionally used So with respect for justification and optimization when scanning patients and especially children size specific dose estimates should be introduced clinically so that the ra diographer and doctor knows the patients estimated dose to a higher accuracy than what is based on a standardized phantom 39 5 Conclusion This project can be structured in four parts Firstly it focused on calibration procedures for detectors in CT The calibration procedure that was established was highly valid for pencil ionizat
66. stic wheel with holes for the filtration material needed for different radiation quali ties b Capintec 30cc ionization chamber located 100 cm from the tube focus c The half value layer test device with the two aluminium packages one with thickness just below and one with thickness just above the nominal HVL 5 In house constructed mover system for the half value layer test device e The X ray tube The X ray voltage generator Pantax HF 320 160 was set to 100 kV and the tube current to 29mA to acquire the first radiation quality RQT 8 For RQT 9 the X ray voltage generator was set to 120 kV and the tube current to 25mA For RQT 10 the X ray voltage generator was set to 150kV and the tube current to 20mA The three qualities were established in the same following way The leakage current was measured before any radiation was turned on The sampling of current from the electrometer was automatically made in a computer program The sampling was made with 1 second intervals getting the instantaneous current in Ampere 50 times The average current of the 50 samplings was then calculated giving the measured current The estimated standard deviation was also calculated for the 50 samplings The radiation beam was turned on and a first measurement of the current without any attenuating aluminum was made giving the measured value Ip With the radiation beam still on the in house constructed mover moved attenuating aluminium with a thickness T
67. tested was tube voltage in kV nominal collimation and SFOV When measurements were performed during helical scanning with the CTDP also dif ferent values of pitch were tested Some of the collimations available for axial scanning were not available for helical scanning only 10 mm and 20 mm were available for helical scanning All scanning was made with the same scanning protocol Also noted for each measurement was the displayed nominal CTDI so it could be compared with the measured value Differ ent SFOV was only tested with the pencil chamber Different values of pitch during helical scanning with the CTDP was not tested free in air When measuring free in air with the Radcal pencil chamber two different values of tube current was used 100mA and 300mA See Tab 3 for the measurements made and also the relevant input parameters Such as tube current in mA the tube rotation time tr in seconds and mover speed Vm in mm s 17 Table 3 The different scanning modes measurement techniques and phantoms that were used for measuring the CTDlioo during different parameter settings on a GE Brightspeed Elite in Kristiansand Norway Mea surements in the pediatric phantom was only performed with the Radcal pencil chamber The Radcal pencil chamber was used for measuring in all holes in the phantoms The CTDP was only used in the central holes The emphasized parameters correspond to the reference settings used when another parameter was tested Over
68. tr 1s 120kV Large SFOV 120kV Large SFOV 120kV Large SFOV Collimation mm Free in air 20 40 1 25 5 10 20 40 1 25 2 5 5 10 20 40 pene pitch 0 516 t p 2s Different Pitch tested at Collimation 20 mm 0 531 0 969 1 375 The CTDP has integrated energy correction factors that are automatically taken into account when input parameters such as kV filtration and if measuring is performed in one of the CTDI phantoms or free in air are specified from the user in the software Ocean The energy correction factors are normalized to radiation quality RQR9 IEC 61267 1994 This quality was not used during the calibration procedure Since this is the reference quality for the CTDP the measured calibration factor and energy corrections from the calibration procedure was not taken into account for the clinical measurements Instead the supplied calibration factor from purchase was used which was N 0 2655mGy nC in RQR9 IEC 61267 1994 and calibrated at SWEDAC in 2012 12 10 During all measurements with the CTDP a total filtration of 7mm Al was specified in Ocean This was the default setting in Ocean and was not changed For the Radcal pencil chamber the performed calibration procedure could be utilized for the clinical measurements The measured calibration factor from the calibration procedure was used Some assumptions were made about what energy correction factor should be chosen For all measurements with the Radcal
69. uate how accurately CTDI can be size corrected 2 Materials and Methods 2 1 Establishing a calibration procedure for detectors used in CT 2 1 1 Acquiring of the Radiation Qualities In order to establish a standardized calibration procedure for detectors used in CT some radiation qualities specific for CT called RQT qualities needed to be acquired The RQT qualities are based on RQR qualities which application is in general radiography but with added copper filtration 4 10 RQR qualities only have aluminium filtration The radiation qualities that needed to be acquired for this study are defined by the Interna tional Electrotechnical Commission IEC and the terms definitions and method for acquiring these radiation qualities are described in the same report 10 The radiation qualities to be acquired are referred to as RQT 8 IEC 61267 2005 RQT 9 IEC 61267 2005 and RQT 10 IEC 61267 2005 but will be called RQT X further on in this report where X represents 8 9 or 10 see Tab 1 It should be noted that the RQT qualities are not defined with a nominal homogeneity coefficient hence no measurements of that parameter was made Table 1 The RQT qualities with the X ray tube voltage nominal added copper filtration in mm Cu the nominal first half value layer HVL in mm Al and the two different thicknesses of aluminium used when measuring the HVL Radiation X ray Nominal Nominal first T To Quality tube voltage added filter
70. when measuring during heli cal scanning for larger collimations Since the dose profile is evened out due to the relative small positional change for one tube rotation The lower the pitch will be in this regard the more it it will be evened out For smaller collimations however it can affect the results more The patient table can reduce the measured maximum exposure rate in the centre of the beam relatively more for smaller collimations With the Mover measurements were dependent on a single tube rotation hence the tube rotation was slowed down for these measurements As just described for helical scanning analogously the maximum exposure rate can be reduced if the tube is under the table at same the time when the CTDP is in the isocenter of the X ray tubes rotational symmetry axis i e in the centre of the beam This leading to a diminished value of the total exposure However with that said for example looking in Fig 18 different values of pitch did not have any great affect on the measured CTDlioo c If the collimation would be very narrow and measuring with the highest pitch this table effect diminishing the CTDI100 would probably have occurred between different scanning Positioning the phantom free in air would solve this problem and has been done before 15 The table effect is not relevant in the same way for the Radcal pencil chamber since it always measures along a length independently of where the tube is a

Calibration & Clinical Measurements in Computed Tomography

Contents

Download Pdf Manuals

Related Search

Related Contents

Calibration &amp; Clinical Measurements in Computed Tomography

Contents

Download Pdf Manuals

Related Search

Related Contents

Calibration & Clinical Measurements in Computed Tomography