BJR
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
British Journal of Radiology (2005) 78, 623-630
© 2005 British Institute of Radiology
doi: 10.1259/bjr/46029447
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hill, B
Right arrow Articles by Baldock, C
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hill, B
Right arrow Articles by Baldock, C

Full Paper

The dose response of normoxic polymer gel dosimeters measured using X-ray CT

B Hill, MSc 1,3 A Venning, MSc 1,2,3 and C Baldock, PhD 2,3

1 Medical Physics Section, Biomedical Engineering Services, The Canberra Hospital, POB 11 Woden, ACT 2606, 2 School of Physical and Chemical Sciences, Queensland University of Technology, Brisbane and 3 Institute of Medical Physics, School of Physics, University of Sydney, Sydney, Australia


   Abstract
Top
 Abstract
Introduction
 Method
 Results and discussion
 Conclusion
 References
 
X-ray CT was used to determine the dose response of normoxic polymer gel dosimeters. Normoxic polymer gel dosimeters were manufactured and irradiated up to 150 Gy. Up to 50 CT images were acquired on a Toshiba Aquilion Multislice CT scanner using protocols for 80 kV and 135 kV to determine dose response. HU-dose sensitivity, the linear regression of data for the HU versus dose for the linear part of the curve up to 60 Gy was 0.38±0.07 HU Gy–1 for 135 kV and 0.37±0.01 HU Gy–1 for 80 kV. Dose resolution was found to be < 1.3 Gy for an absorbed dose range up to 70 Gy for 135 kV, similar to that measured previously for polyacrylamide gel (PAG). Although the HU-dose sensitivity was lower than that previously measured for PAG gel dosimeters it had a greater range of absorbed dose indicating that normoxic polymer gel dosimeters have potential in CT gel dosimetry.


   Introduction
Top
 Abstract
 Introduction
Method
 Results and discussion
 Conclusion
 References
 
In 1993 a polymer gel dosimeter was developed that maintained spatial information following irradiation and could be visualized with the use of MRI [1]. The polymer gel dosimeter was based upon polymerization of acrylamide and N'N'–methylene-bis-acrylamide monomers infused in an aqueous agarose gel matrix [1]. The polymer gel was further developed using a gelatin matrix and given the acronym BANG [2]. This polyacrylamide gel (PAG) dosimeter is susceptible to atmospheric oxygen, due to the nature of the free radical chemistry [3] and is therefore normally manufactured under hypoxic atmospheric conditions.

In 2001 a polymer gel dosimeter was developed that could be produced in a normal oxygen environment or normoxic atmospheric conditions [4]. The primary reason these polymer gels are normoxic is because the polymer gel has a reduced sensitivity to oxygen as the anti-oxidant, ascorbic acid, binds dissolved oxygen within the polymer gel system into metallo-organic complexes in the gel prior to irradiation [5]. This and other normoxic gel dosimeters have been evaluated for dose response and spatial stability using MRI [4–6]. Normoxic polymer gel dosimeters have also been shown to be of use in the evaluation of clinical dose distributions [7].

Alternative methods of evaluation of polymer gel dosimeters to MRI have been investigated. These include optical CT [2, 8–10], vibrational spectroscopy [11, 12] and ultrasound [13–17].

X-ray CT has also been used to measure dose response and dose distributions of irradiated polymer gel dosimeters [18–20]. The resulting images, which are reconstructed from linear attenuation data, have CT numbers or Hounsfield units (HU) which are directly linked to physical density changes of the polymer gel dosimeters [21].

A limitation of using CT as an evaluation tool for polymer gel dosimeters is an increased uncertainty due to a low signal-to-noise ratio (SNR). SNR consists of both stochastic and structured noise in the image. Stochastic noise is a function of the parameters chosen in the image acquisition protocol. Image averaging and subtraction techniques reduce the stochastic noise component of the image [22]. This was demonstrated by averaging CT images of an irradiated polymer gel dosimeter and subsequently subtracting a pre-irradiated image to remove stochastic noise that obscured dose information [18]. The same method was used in a clinical study of polymer gel dosimeters [19]. Another approach was to subtract averaged images of water from averaged images of an irradiated polymer gel dosimeter to remove beam hardening and other artefacts [20]. Structured noise is related to inherent limitations of the CT scanner such as the image reconstruction method [23–27].

When using MRI to evaluate polymer gel dosimeters it is important to commission the scanner to ensure optimal performance of the imaging system and minimization of uncertainties in the measurement process [28]. To date no study has explored the commissioning requirements of CT scanners with respect to their use in the evaluation of polymer gel dosimeters. By investigating spatial information, dose and SNR, the optimal CT protocol can be determined for acquisition of images that are used for the evaluation of polymer gel dosimeters.

The aim of this work was to commission a CT scanner for polymer gel dosimetry and subsequently to investigate the dose response of normoxic polymer gel dosimeters using the CT scanner.


   Method
Top
 Abstract
 Introduction
 Method
Results and discussion
 Conclusion
 References
 
A Toshiba Aquilion Multislice CT scanner (Toshiba Corporation, Japan) was used for all CT measurements. It is a third generation CT with helical capability that can acquire four adjacent images in one rotation. The significant difference between this multislice CT compared with a single slice is that the detector is divided into 34 row segments along the z-axis (+z-axis direction is the direction that the couch moves into the CT scanner gantry). The four centre row segments are 0.5 mm thick with the 15 row segments adjacent to either side 1 mm thick allowing image acquisition of up to 32 mm (of the 34 row segments) in the z-axis with one rotation as opposed to 10 mm for a single slice system. This study used the scanner in axial mode only in which the 32 row segments are configured in 7 different ways. Signals from each of the 32 row segments are sent to four different digital acquisition systems where all four slices are reconstructed having either 0.5 mm, 1.0 mm, 2.0 mm, 4.0 mm, 5.0 mm, 6.0 mm, or 8.0 mm slice widths. The total slice widths that a user can define in a single rotation are subsequently 2.0 mm, 4.0 mm, 8.0 mm, 16.0 mm, 20.0 mm, 24.0 mm, or 32.0 mm. Total slice widths will be discussed throughout this paper.

The tube current available on the system is between 10 mA and 500 mA in steps of 10 mA. The tube voltages available are 80 kV, 100 kV, 120 kV and 135 kV. The rotation times are 0.5 s, 1.0 s, 1.5 s, 2.0 s and 3.0 s. The field of view (FOV) is 180 mm, 240 mm, 320 mm, 400 mm, 500 mm or as relative pixel size is 0.35 mm, 0.47 mm, 0.63 mm, 0.78 mm, 0.98 mm for a 512 x 512 pixel matrix.

Signal to noise ratio (SNR)
The SNR was investigated by comparing various image acquisition protocols to determine the highest SNR. Images of a uniform water phantom were acquired for protocols that varied in total slice width, 2 mm, 8 mm, 16 mm and 32 mm, tube current from 350 mA to 400 mA, tube voltage from 80 kV to 135 kV and the relative pixel size from 0.35 mm to 0.63 mm. The SNR was evaluated by averaging images from a single slice up to 100 slices from 100 images acquired for each protocol. The SNR was measured by taking a region of interest (ROI) of a constant number of pixel values in the averaged images and dividing by the standard deviation of pixel values for this ROI.

CT quality assurance
Accuracy of imaged slice width, low contrast detectability, limiting spatial resolution, image uniformity and HU accuracy were determined from images acquired using a Catphan® 500 CT phantom, in accordance with the Catphan® Manual (The Phantom Laboratory, Cambridge NY). CT dose index (CTDI) was calculated from data measured using dedicated Perspex head and body CTDI phantoms and a 100 mm pencil ionization chamber (RadCal Corporation, CA). The method used for measurement and calculations of CTDI was according to published guidelines [29] and compared with published data [30].

The imaging protocols for all of the Catphan® and CTDI measurements were for the highest SNR protocol determined on the CT scanner (400 mA, 135 kV, 1 s rotation and 0.35 mm x 0.35 mm pixel size).

Imaged slice width
Imaged slice width was measured using a Catphan® module CTP 401. The full-width half-maximum (FWHM) was determined according to the Catphan® procedure manual for slice widths 2 mm, 8 mm, 16 mm and 32 mm.

Low contrast detectability and limiting spatial resolution
Low contrast detectability was measured using a Catphan® module CTP 515. Limiting spatial resolution was measured using a Catphan® module CTP 528. The modulation transfer function (MTF) was calculated from line pairs acquired from an image of module CTP 528.

HU accuracy
HU accuracy was measured using a Catphan® module CTP 401. The mean HU was measured from ROIs within the four inserts (low density polyethylene, air, Teflon and acrylic) within the module and compared with the known HU of the materials.

CTDI
The CTDI in Perspex phantoms was measured using the 100 mm ionization chamber, inserted in the centre and periphery where the scan plane at the isocentre bisected the ionization chamber with all other cavities being filled with Perspex rods. Dose measurements at the centre were used to calculate the central CTDI. A mean of the peripheral dose measurements was used to calculate the peripheral CTDI.

Polymer gel dosimeters
Preparation of polymer gel dosimeters
A MAGIC normoxic polymer gel dosimeter was prepared in 500 ml batches using materials and methods described elsewhere [6]. The composition consisted of 82.9% triple distilled, de-ionized water by weight, 8% gelatin, 9% methyl acrylic acid, 1.0 mM ascorbic acid and 0.01 mM copper sulphate and 10 mM hydroquinone. All gels were poured into plastic containers (Packard, Meriden) of 60 mm height and 27 mm diameter and filled to the top to minimize oxygen entering the gel from the air space above the gel [6]. The vials were sealed with screw top lids.

Irradiation of polymer gel dosimeters
Prepared containers were irradiated up to 150 Gy using a 60Co Gammacell, previously calibrated [31].

Evaluation of polymer gel dosimeters
The previous phantom [21] was re-designed to allow larger polymer gel dosimeter phantoms to be used while reducing beam hardening artefacts in the CT images. The prepared plastic containers were placed at regular intervals in a plastic holder concentrically around the inner circumference of the cylindrical water tank at a distance of 50 mm within the inside edge (Figure 1Go). The plastic containers were centred in the phantom in the centre of the CT scanner gantry.



View larger version (101K):
[in this window]
[in a new window]
  		Figure 1. Water phantom and holder used to position the polymer gel dosimeters in the CT scanner. The polymer gel dosimeters are in place within the Perspex holder.

 
Two protocols were used to acquire images on the CT scanner for the evaluation of the polymer gel dosimeters: (i) 135 kV, 400 mA, 1 s rotation, 32 mm slice width and (ii) 80 kV, 400 mA, 1 s rotation, 32 mm slice width.

A single slice in the same position was averaged from 50 acquired images of the polymer gel dosimeters to obtain an averaged polymer gel dosimeter image. The polymer gel dosimeters were then removed, ensuring no movement of the external water phantom, and a further single slice in the same position was averaged from 50 acquired images of the water phantom to get an averaged water image. Dose contributions from the CT during evaluation [32] were not considered in this study.

The images were transferred to a PC and the images processed with modified Matlab® (MathworksTM) software written for gel dosimetry [33]. From the averaged polymer gel dosimeter image (Figure 2aGo) and the subtraction of the averaged water image (Figure 2bGo) a final polymer gel dosimeter image (Figure 2cGo) was obtained. ROIs were centred on the containers in the image, keeping the number of pixels constant for each study; the average HU was calculated and exported to Excel® (MicrosoftTM) and Origin® (MicrocalTM).



View larger version (49K):
[in this window]
[in a new window]
  		Figure 2. (a) The resulting image following image averaging of the polymer gel dosimeters. (b) The resulting image following image averaging of water. (c) The resulting image following image subtraction of the averaged water image from the averaged polymer gel dosimeter image.

 
The HU as a function of absorbed dose was fitted up to 150 Gy. The fit was used only as a guide to the eye to represent the shape of the data and not reflect an explanation of physical processes in the gel for the measured data. HU-dose sensitivity and standard error were calculated from the gradient of linear regression data from the linear part of the curve up to 60 Gy.

Uncertainty
The uncertainty for the data points was calculated using a previously described methodology [34]. Dose resolution, defined as the minimum difference that can be determined between two absorbed dose values for a 95% level of confidence, was also calculated [20, 21, 35].


   Results and discussion
Top
 Abstract
 Introduction
 Method
 Results and discussion
Conclusion
 References
 
Signal to noise ratio (SNR)
Figure 3aGo shows the SNR as a function of averaged images for up to 100 images with different slice widths. The SNR for the 2 mm slice width is greater than that for the 8 mm slice width due to focal spot size and reconstruction parameters. The SNR increase is greatest for the 32 mm slice width. Larger slice width yields better contrast or higher SNR, but the spatial resolution in the slice thickness dimension is reduced. The curve shows that SNR continues to increase beyond 100 averaged images. Previous papers found similar relationships for SNR [18, 21]. As the stochastic noise is highest for a smaller slice width, by using a larger slice width the stochastic noise in the measurement will be decreased. The limitation with using a larger slice width is that the image data are averaged over that width in the z-axis. This will limit the spatial resolution if a three-dimensional (3D) reconstruction of dose distribution were required. For example, for an image volume of a phantom of 200 mm diameter and 100 mm length using a 32 mm slice width, approximately three slices would be imaged. A 2 mm slice width over the same volume would increase the uncertainty in the x–y plane but give 50 slices and more information in the z-axis. A compromise must be considered when attempting to obtain high spatial information in the z-axis of the image and high spatial resolution when using CT as an evaluation tool and will depend on the clinical application requirements. To establish which evaluation tool is suited for clinical cases is subject to further work and is not addressed in this communication.



View larger version (33K):
[in this window]
[in a new window]
  		Figure 3. The signal to noise ratios (SNR) as a function of averaged images up to 100 with: (a) different slice widths; (b) different tube current; (c) different relative pixel sizes; (d) different tube voltage; and (e) for different slice width and varying relative pixel size.

 
Figure 3bGo shows the SNR as a function of averaged images up to 100 for images with different tube current. For the same pixel size and largest slice width of 32 mm the highest tube current provides a slight increase in SNR and thus a reduction in uncertainty. The tube current is directly related to X-ray photons used to produce the image and thus SNR [36]. It was predicted that for a 50 mA increase in tube current the SNR would increase by 7% and for 50 averaged images the SNR increased by 24%. A large change in SNR and thus uncertainty with a variation from 350 mA to 400 mA was demonstrated. Choosing a lower tube current than 350 mA will therefore reduce the SNR relative to that demonstrated with 400 mA. The limitation of using a high tube current, particularly when acquiring multiple images is that the X-ray tube heat capacity may be reached faster than with lower tube currents. This may be overcome by delaying the start of each image acquisition by several seconds.

Figure 3cGo shows the SNR as a function of averaged images up to 100 for images with different relative pixel size. For up to 20 averaged images the smallest relative pixel size had a lower SNR, however beyond this number of averaged images the SNR becomes lower for the larger relative pixel size of 0.63 mm. The spatial resolution of the CT scanner is 0.35 mm pixel size generated from a 512 x 512 matrix. Therefore the minimum object size that can be viewed on the monitor is 0.35 mm. However the field of view can be varied from 0.35 mm to 0.98 mm in pixel size. A change in field of view from 0.35 mm to 0.63 mm relative pixel size is simply a change in magnification of the image, by increasing the relative pixel size. With a larger relative pixel size the image has fewer pixels per area considering that the image matrix does not change from 512 x 512. Therefore the signal component of the SNR has fewer samples or pixels relative to the noise in the image. The result was, for a constant resolution, a reduction in contrast and thus SNR with larger relative pixel size as shown in Figure 3cGo.

Figure 3dGo shows the SNR as a function of averaged images for up to 100 images with different tube voltage. The lowest and highest tube voltage available on the CT scanner were chosen for comparison. 135 kV provides an increase of 55% in SNR and thus a reduction in uncertainty compared with 80 kV. The fitted curve clearly demonstrates that the higher tube voltage provides a greater SNR. Both tube voltages were used for the polymer gel evaluation for comparison.

Figure 3eGo shows the SNR as a function of averaged images for up to 100 with different slice widths and varying relative pixel size. When the 32 mm slice width was compared with the 8 mm slice width for a 0.35 mm pixel size a 53% increase in SNR or reduction in uncertainty was measured for 50 averaged images. This was a direct result of increasing the slice width or voxel volume as demonstrated in Figure 3aGo. When compared with a change of relative pixel size from 0.63 mm to 0.35 mm for the 8 mm slice width there was little benefit. The 32 mm slice width curve shows the largest increase in SNR. The increase in spatial information with the smallest pixel size and smaller slice width does not, in this study, outweigh the greater SNR shown by the largest slice width. The optimum SNR was produced with the largest slice width, highest tube voltage and current and for the smallest relative pixel size.

CT quality assurance
The results for Catphan® and CTDI measurements on the CT scanner are shown in Table 1Go. The Catphan® measurements were comparable with previous measurements on a GE multislice CT scanner [37]. CTDI measurements corresponded to results calculated using ImPACT software for the same CT scanner [38].


View this table:
[in this window]
[in a new window]
  		Table 1. Catphan and CT dose index (CTDI) results for Toshiba Aquilion multislice CT scanner with image acquisition parameters 135 kV, 400 mA, 1 s rotation and 0.35 mm x 0.35 mm pixel size

 
Polymer gel dosimeters
The optimal SNR scanning protocol at 135 kV was used to evaluate the normoxic polymer gel dosimeters. A comparison was made with a protocol at 80 kV. The HU and associated uncertainty are shown in Figure 4Go as a function of absorbed dose up to 150 Gy for 50 averaged slices for the different tube potentials.



View larger version (20K):
[in this window]
[in a new window]
  		Figure 4. The HU and associated uncertainty as a function of absorbed dose up to 150 Gy for a single slice in the same position averaged from 50 acquired images of the polymer gel dosimeters for different tube potentials.

 
At 80 kV the HU for the normoxic polymer gel dosimeters was lower than that for 135 kV indicating that the HU changes with tube potential. This was to be expected, as linear attenuation of the normoxic polymer gel dosimeter is energy dependent.

Linear regression of data for the HU-dose sensitivity curve up to 60 Gy was 0.38±0.07 HU Gy–1 for 135 kV and 0.37±0.01 HU Gy–1 for 80 kV. The HU-dose sensitivity was lower than previously published results of PAG polymer gel dosimeters as seen in Table 2Go. The absorbed dose range was greater than previous published PAG polymer gel dosimeters evaluated with CT scanners. An induction effect was noted within the first 5 Gy of both the sets of data. It was possible that any oxygen in the space between the gel and the screw top lids of the containers may have caused this as previously demonstrated with normoxic polymer gel dosimeters [6]. A further induction effect may have caused points 50 Gy and 60 Gy to have a low dose response. Above 100 Gy the dose response for both tube voltages starts to saturate as a result of monomers in the normoxic polymer gel dosimeter being consumed [6]. The CT absorbed dose response for 135 kV gives a lower uncertainty than that found for the 80 kV data.


View this table:
[in this window]
[in a new window]
  		Table 2. Polymer gel dosimeter compositions and measurement results evaluated with CT scanners. r2 and p-value taken from regression program in Microsoft® Excel. Dose resolution is defined in the text

 
Dose resolution is the minimal separation at which two dose points can be distinguished for a given standard deviation. This approximates to a value that would be twice the standard deviation as seen in Table 2Go for a dose resolution with 95% level of confidence as the standard deviation corresponds to a 52% level of confidence. The standard deviation alone does not provide an indication of optimization or comparison of polymer gel dosimeter evaluations. The measurement of uncertainty expressed as a standard deviation only provides information about that specific dose value. Dose resolution however, provides information about all of the measured data and allows subsequent measurements to be optimized according to the dose resolution required [35].

The dose resolution for an absorbed dose range is shown in Figure 5Go for the normoxic polymer gel dosimeter dose responses at 135 kV. Dose resolution for 80 kV was not presented because the high uncertainty associated with the data made it unrealistic to calculate a sensible value. Dose resolution for normoxic polymer gel dosimeter images at 135 kV is <1.3 Gy for an absorbed dose range up to 70 Gy. The data show that the lowest dose resolution occurs below 5 Gy. The dose resolution was similar to values found by Trapp et al [20] for PAG polymer gel dosimeters up to 10 Gy. The value quoted as dose resolution by Hilts et al [18] was the standard deviation of a single dose value with a 52% level of confidence and thus by this definition was not a measure of dose resolution.



View larger version (12K):
[in this window]
[in a new window]
  		Figure 5. Dose resolution for an absorbed dose range for normoxic polymer gel dosimeter for 135 kV.

 

   Conclusion
Top
 Abstract
 Introduction
 Method
 Results and discussion
 Conclusion
References
 
A multislice CT scanner was commissioned for polymer gel dosimeter evaluation. The protocol for optimal image acquisition was used to investigate the dose response of the normoxic polymer gel dosimeters. The dose responses shown demonstrate that X-ray CT was a suitable method of evaluation for normoxic polymer gel dosimeters. The normoxic polymer gel dosimeter evaluated on the multislice CT scanner provides a similar dose resolution to previously published data with the benefit of a greater absorbed dose range and was easier to manufacture.

Received for publication January 5, 2004. Revision received January 11, 2005. Accepted for publication January 25, 2005.


   References
Top
 Abstract
 Introduction
 Method
 Results and discussion
 Conclusion
 References
 

  1. Maryanski MJ, Gore JC, Kennan RP, Schulz RJ. NMR relaxation enhancement in gels polymerized and cross-linked by ionizing radiation: a new approach to 3D dosimetry by MRI. Magn Reson Imaging 1993;11:253–8.[Medline]
  2. Maryanski MJ, Schulz RJ, Ibbott GS, Gatenby JC, Xie J, Horton D, et al. Magnetic resonance imaging of radiation dose distributions using a polymer-gel dosimeter. Phys Med Biol 1994;39:1437–55.[CrossRef][Medline]
  3. Baldock C, Burford RP, Billingham N, Wagner GS, Patval S, Badawi RD, et al. Experimental procedure for the manufacture and calibration of polyacrylamide gel (PAG) for magnetic resonance imaging (MRI) radiation dosimetry. Phys Med Biol 1998;43:695–702.[Medline]
  4. Fong PM, Keil DC, Does MD, Gore JC. Polymer gels for magnetic resonance imaging of radiation dose distributions at normal room atmosphere. Phys Med Biol 2001;46:3105–13.[Medline]
  5. De Deene Y, Hurley C, Venning A, Vergote M, Mather M, Healy BJ, et al. A basic study of some normoxic polymer gel dosimeters. Phys Med Biol 2002;47:3441–63.[Medline]
  6. De Deene Y, Venning A, Hurley C, Healy BJ, Baldock C. Dose-response and spatial stability of various polymer gel dosimeters. Phys Med Biol 2002;47:2459–70.[Medline]
  7. Gustavsson H, Karlsson A, Back SAJ, Olsson LE. MAGIC-type polymer gel for three-dimensional dosimetry: intensity modulated radiation therapy verification. Med Phys 2003;30:1264–71.[Medline]
  8. Gore JC, Ranade M, Maryanski MJ, Schulz RJ. Radiation dose distributions in three dimensions from tomographic optical density scanning of polymer gels: I. Development of an optical scanner. Phys Med Biol 1996;41:2695–704.[CrossRef][Medline]
  9. Oldham M, Siewersden HH, Shetty A, Jaffray DA. High resolution gel-dosimetry by optical-CT and MR scanning. Med Phys 2001;23:699–705.[CrossRef]
  10. Maryanski MJ, Zastavker YZ, Gore JC. Radiation dose distributions in three dimensions from tomographic optical density scanning of polymer gels: II. Optical properties of the BANG polymer gel. Phys Med Biol 1996;41:2705–17.[Medline]
  11. Baldock C, Rintoul L, Lepage M, Menzies N, Back SAJ. Radiation dosimetry of polymer gel dosimeters using Raman microscopy. Australas Phys Eng Sci Med 2000;23:159.
  12. Rintoul L, Lepage M, Baldock C. Radiation dose distribution in polymer gels by Raman spectroscopy. Applied Spectroscopy 2003;57:51–7.[Medline]
  13. Mather LM, Whittaker AK, Baldock C. Ultrasound evaluation of polymer gel dosimeters. Phys Med Biol 2002;47:1449–58.[Medline]
  14. Mather LM, Charles PH, Baldock C. Measurement of ultrasonic attenuation coefficient in polymer gel dosimeters. Phys Med Biol 2003;48:269–75.[CrossRef]
  15. Mather LM, Collings AF, Bajenov N, Whittaker AK, Baldock C. Ultrasonic absorption in polymer gel dosimeters. Ultrasonics 2003;41:551–9.[Medline]
  16. Mather LM, De Deene Y, Whittaker AK, Simon GP, Rutgers R, Baldock C. Investigation of ultrasonic properties of PAG and MAGIC polymer gel dosimeters. Phys Med Biol 2002;47:4379–409.
  17. Mather LM, Baldock C. Ultrasound tomography imaging of radiation dose distributions in polymer gel dosimeters: preliminary study. Med Phys 2003;30:2140–8.[Medline]
  18. Hilts M, Audet C, Duzenli C, Jirasek A. Polymer gel dosimetry using x-ray computer tomography: a feasability study. Phys Med Biol 2000;45:2559–71.[Medline]
  19. Audet A, Hilts M, Jirasek A, Duzenli C. CT gel dosimetry technique: Comparison of a planned and measured 3D stereotactic dose volume. App Clin Med Phys 2002;3:110–8.
  20. Trapp J, Back SAJ, Lepage M, Michael G, Baldock C. An experimental study of the dose response of polymer gel dosimeters imaged with x-ray computer tomography. Phys Med Biol 2001;46:2939–51.[Medline]
  21. Trapp J, Back SAJ, Lepage M, Michael G, Menzies N, Bennett D, et al. X-ray computer tomography of gel dosimeters: Characterisation of gels. Australas Phys Eng Sci Med 2000;23:169.
  22. Hilts M, Duzenli C. Image filtering for improved dose resolution in CT polymer gel dosimetry. Med Phys 2004;31:39–49.[Medline]
  23. Cohen G, Di Bianca FA. The use of contrast-detail-dose evaluation of image quality in a computer tomographic scanner. J Comput Assist Tomogr 1979;3:189–95.[Medline]
  24. Fuchs T, Krause J, Schaller S, Flohr T, Kalender WA. Spiral interpolation algorithms for multislice spiral CT-part II: measurement and evaluation of slice sensitivity profiles and noise at a clinical multislice system. IEEE Trans Med Imaging 2000;19:835–47.[CrossRef][Medline]
  25. Kachelrie M, Watzke O, Kalender WA. Generalized multi-dimensional adaptive filtering for conventional and single-slice, multi-slice, and cone-beam CT. Med Phys 2001;28:475–90.[CrossRef][Medline]
  26. Katsuyuki T, Aradate H. Algorithm for image reconstruction in multi-slice helical CT. Med Phys 1998;25:550–61.[CrossRef][Medline]
  27. Schaller S, Flohr T, Krause J, Fuchs T, Kalender WA. Spiral interpolation algorithm for multislice spiral CT-part I: theory. IEEE Trans Med Imaging 2000;19:822–34.[CrossRef][Medline]
  28. De Deene Y. Fundamentals of MRI measurements for gel dosimtery. In: Baldock C, De Deene Y, editors. DOSEGEL 2001. Brisbane: Queensland University of Technology; 2001:49–65.
  29. EUR 16262 EN Brussels. European Commission. Quality Criteria for Computed Tomography; 1999.
  30. ImPACT. CT scanner acceptance testing. Leaflet No. 1 2001;v1.02.
  31. Baldock C, Fitchew R, Murry PJ, Murry M, Back P, Kron T. Calibration of Gammacell irradiator for irradiation of radiotherapy polymer dosimetry gel. Med Phys 1999;26:1130.
  32. Hill B, Venning A, Baldock C. Use of normoxic polymer gel dosimeters for measuring diagnostic doses on CT scanners. J Phys Conference Series 3. 2004:224–7.
  33. Murry P, Baldock C. Research software for MRI radiotherapy gel dosimetry. Aust Phys Eng Sci Med 2000;23:44–51.
  34. Trapp J, Michael G, Baldock C. Uncertainty calculations determined from calibration data for CT imaging of polymer gel dosimeters. In: Baldock C, De Deene Y, editors. DOSEGEL 2001. Brisbane, Australia: Queensland University of Technology; 2001:201–3.
  35. Baldock C, Lepage M, Bäck S, Murry P, Jayasekera P, Porter D, et al. Dose resolution in radiotherapy polymer gel dosimetry: effect of echo spacing in MRI pulse sequence. Phys Med Biol 2001;46:449–60.[CrossRef][Medline]
  36. Busberg JT, Siebert JA, Leidholdt EM, Boone JM. The essential physics of medical imaging (2nd edn). Philadelphia, PA: Lippincott, Williams and Wilkins, 2002.
  37. McCollough CH, Zink FE. Performance evaluation of a multi-slice CT system. Med Phys 1999;26:2223–30.[CrossRef][Medline]
  38. ImPACT. ImPACT Dose Calculation Software. 2003.




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hill, B
Right arrow Articles by Baldock, C
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hill, B
Right arrow Articles by Baldock, C

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
BJR DMFR IMAGING  ALL BIR JOURNALS