Embed Size (px)
Citation preview
Queensland University of Technology
School of Physical and Chemical Sciences
The Development of Normoxic Polymer Gel
Dosimetry using High Resolution MRI
Christopher Hurley M.App.Sci.(Med Phys), M.Ed.Admin., Grad.Dip.Ed., B.Eng.(Elec)
A thesis submitted at the Queensland University of Technology, in the School of
Physical and Chemical Sciences, in fulfillment of the requirements of the Doctor
of Philosophy.
2006
ii
Keywords
Polymer gel dosimetry, radiotherapy, brachytherapy, radiation dosimetry, PAG,
MAGIC, MAGAT, PAGAT, normoxic polymer gel dosimeters, high-resolution
MRI.
iii
Abstract
Dosimetry is a vital component of treatment planning in radiation therapy.
Methods of radiation dosimetry currently include the use of: ionization chambers,
thermoluminescent dosimeters (TLDs), solid-state detectors and radiographic
film. However, these methods are inherently either 1D or 2D and their use
involves the perturbation of the radiation beam. Although the dose distribution
within tissues following radiation therapy treatments can be modeled using
computerized treatment planning systems, a need exists for a dosimeter that can
accurately measure dose distributions directly and produce 3D dose maps. Some
radiation therapy and brachytherapy treatments require mapping the dose
distributions in high-resolution (typically < 1 mm). A dosimetry technique that is
capable of producing high resolution 3D dose maps of the absorbed dose
distribution within tissues is required.
Gel dosimetry is inherently a 3D integrating dosimeter that offers high spatial
resolution, precision and accuracy. Polymer gel dosimetry is founded on the basis
that monomers dissolved in the gel matrix polymerize due to the presence of free
radicals produced by the radiolysis of water molecules. The amount of
polymerization that occurs within a polymer gel dosimeter can be correlated to the
absorbed dose. The gel matrix maintains the spatial integrity of the polymers and
hence a dose distribution can be determined by imaging the irradiated polymer gel
dosimeter using an imaging modality such as MRI, x-ray computed tomography
(CT), ultrasound, optical CT or vibrational spectroscopy. Polymer gel dosimeters,
however, suffer from oxygen contamination. Oxygen inhibits the polymerization
reaction and hence polymer gel dosimeters must be manufactured, irradiated and
scanned in hypoxic environments.
iv
Normoxic polymer gel dosimeters incorporate an anti-oxidant into the formulation
that binds the oxygen present in the gel and allows the dosimeter to be made under
normal atmospheric conditions. The first part of this study was to provide a
comprehensive investigation into various formulations of polymer and normoxic
polymer gel dosimeters. Several parameters were used to characterize and assess
the performance of each formulation of polymer gel dosimeter including: spatial
resolution and stability, temporal stability of the R2-dose response, optimal R2-
dose response for changes in concentration of constituents and the effects of
oxygen infiltration. This work enabled optimal formulations to be determined that
would provide greater dose sensitivity. Further work was done to investigate the
chemical kinetics that take place within normoxic polymer gel dosimeters from
manufacture to post-irradiation. This study explored the functions that each of
the constituent chemicals plays in a polymer gel dosimeter. Although normoxic
polymer gel dosimeters exhibit very similar characteristics to polyacrylamide
polymer gel dosimeters, one important difference between them was found to be a
decrease in R2-dose sensitivity over time in the normoxic polymer gel dosimeter
compared to an increase in the polyacrylamide polymer gel dosimeters.
From an investigation into the function of anti-oxidants in normoxic polymer gel
dosimeters, alternatives were proposed. Several alternative anti-oxidants were
explored in this study that found that whilst some were reasonably effective,
tetrakis (hydroxymethyl) phosphonium chloride (THPC) had the highest reaction
rate. THPC was found not only to be an aggressive scavenger of oxygen, but also
to increase the dose sensitivity of the gel. Hence, a formulation of normoxic
polymer gel dosimeter was proposed, called MAGAT, that comprised:
methacrylic acid, gelatin, hydroquinone and THPC. This formulation was
examined in a similar fashion to the studies of the other formulations of polymer
and normoxic polymer gel dosiemeters. The gel was found to exhibit spatial and
temporal stability and an optimal formulation was proposed based on the R2-dose
response.
Applications such as IVBT require high-resolution dosimetry. Combined with
high-resolution MRI, polymer gel dosimetry has potential as a high-resolution 3D
integrated dosimeter. Thus, the second component of this study was to
v
commission a micro-imaging MR spectrometer for use with normoxic polymer
gel dosimeters and investigate artifacts related to imaging in high-resolutions.
Using high-resolution MRI requires high gradient strengths that, combined with
the Brownian motion of water molecules, was found to produce an attenuation of
the MR signal and hence lead to a variation in the measured R2. The variation in
measured R2 was found to be dependent on both the timing and amplitude of
pulses in the pulse sequence used during scanning. Software was designed and
coded that could accurately determine the amount of variation in measured R2
based on the pulse sequence applied to a phantom. Using this software, it is
possible to correct for differences between scans using different imaging
parameters or pulse sequences.
A normoxic polymer gel dosimeter was irradiated using typical brachytherapy
delivery and the resulting dose distributions compared with dose points predicted
by the computerized treatment planning system.The R2-dose response was
determined and used to convert the R2 maps of the phantoms to dose maps. The
phantoms and calibration vials were imaged with an in-plane resolution of 0.1055
mm/pixel and a slice thickness of 2 mm. With such a relatively large slice
thickness compared to the in-plane resolution, partial volume effects were
significant, especially in the region immediately adjacent the source where high
dose gradients typically exist. Estimates of the partial volume effects at various
distances within the phantom were determined using a mathematical model based
on dose points from the treatment planning system. The normalized and adjusted
dose profiles showed very good agreement with the dose points predicted by the
treatment planning system.
vi
Table of Contents
ABSTRACT................................................................................................................................................. III TABLE OF CONTENTS................................................................................................................................. VI LIST OF PUBLICATIONS:............................................................................................................................. IX LIST OF ABBREVIATIONS: ........................................................................................................................... X STATEMENT OF ORIGINAL AUTHORSHIP:................................................................................................... XI ACKNOWLEDGEMENTS ............................................................................................................................. XII CHAPTER 1 INTRODUCTION .........................................................................................................................1
1.1 Description of Research Problem Investigated..................................................................1 1.2 Overall Objective of the Study ...........................................................................................3 1.3 The Specific Aims of the Study ...........................................................................................3 1.4 Account of Scientific Progress Linking the Scientific Papers ............................................4 References................................................................................................................................6
CHAPTER 2 LITERATURE REVIEW................................................................................................................9
2.1 Radiotherapy and Brachytherapy ......................................................................................9 2.2 Radiation Dosimetry ........................................................................................................12
2.2.1 Clinical Dosimetry Requirements ...........................................................................................12 2.2.2 Dosimetry Detectors................................................................................................................13
2.2.2.1 Ionization Chambers .......................................................................................................13 2.2.2.2 Solid-State Detectors.......................................................................................................14 2.2.2.3 Thermoluminescent Dosimeters ......................................................................................14 2.2.2.4 Radiographic Film ..........................................................................................................15 2.2.2.5 Chemical Dosimeters ......................................................................................................15
2.3 Gel Dosimeters ................................................................................................................16 2.3.1 Ferrous Sulfate (Fricke) Gels ..................................................................................................17 2.3.2 Polymer Gel Dosimeters .........................................................................................................19 2.3.3 Normoxic Polymer Gel Dosimeters.........................................................................................23
2.4 Characteristics of Polymer Gel Dosimeters.....................................................................24 2.4.1 Effects of Oxygen....................................................................................................................24 2.4.2 Effect of Light .........................................................................................................................24 2.4.3 Temperature ............................................................................................................................25 2.4.4 Concentration of monomers ....................................................................................................26 2.4.5 Ageing of the gel .....................................................................................................................26
2.5 Evaluation of Polymer Gel Dosimeters............................................................................27 2.5.1 Magnetic Resonance Imaging .................................................................................................27 2.5.2 X-Ray Computed Tomography ...............................................................................................29 2.5.3 Optical Computed Tomography ..............................................................................................31 2.5.4 Ultrasound ...............................................................................................................................32 2.5.5 Vibrational Spectroscopy ........................................................................................................33
2.6 High-Resolution MRI in Polymer Gel Dosimetry ............................................................34 2.7 Brachytherapy Applications of Gel Dosimetry ................................................................37 2.8 Sources of Uncertainty in Polymer Gel Dosimeters ........................................................40 2.9 Conclusion .......................................................................................................................42 References..............................................................................................................................44
CHAPTER 3 DOSE-RESPONSE STABILITY AND INTEGRITY OF THE DOSE DISTRIBUTION OF VARIOUS POLYMER GEL DOSIMETERS ......................................................................................................63
Abstract..................................................................................................................................64 3.1 Introduction .....................................................................................................................64 3.2 Materials and methods.....................................................................................................65
vii
3.2.1 Gel fabrication......................................................................................................... 65 3.2.2 Irradiation................................................................................................................ 66 3.2.3 Scanning................................................................................................................... 66
3.3 Results.............................................................................................................................. 67 3.3.1 R2-Dose stability...................................................................................................... 67 3.3.2 Integrity of dose distribution.................................................................................... 68
3.4 Discussion........................................................................................................................ 71 3.4.1 R2-Dose stability...................................................................................................... 71 3.4.2 Integrity of the dose distribution.............................................................................. 73
3.5 Conclusions ..................................................................................................................... 74 References.............................................................................................................................. 75
CHAPTER 4 A BASIC STUDY OF SOME NORMOXIC POLYMER GEL DOSIMETERS .......................................77
Abstract.................................................................................................................................. 78 4.1 Introduction ..................................................................................................................... 79 4.2 Materials and Methods .................................................................................................... 79
4.2.1 MAGIC gel components ........................................................................................... 80 4.2.2 Dose distribution of half-blocked field..................................................................... 81 4.2.3 Anti-oxidants ............................................................................................................ 82 4.2.4 Potentiometric oxygen measurements...................................................................... 82
4.3 Results.............................................................................................................................. 82 4.3.1 MAGIC gel components ........................................................................................... 82 4.3.2 Dose distribution of half-blocked field..................................................................... 86 4.3.3 Anti-oxidants ............................................................................................................ 87 4.3.4 Potentiometric oxygen measurements...................................................................... 89
4.4 Discussion........................................................................................................................ 90 4.4.1 Acrylic polymer gels ................................................................................................ 92 4.4.2 Normoxic polymer gels ............................................................................................ 93
4.4 Conclusions ..................................................................................................................... 98 References.............................................................................................................................. 99
CHAPTER 5 THE EFFECTS OF MOLECULE SELF-DIFFUSION OF WATER ON QUANTITATIVE MRI MEASUREMENTS IN HIGH-RESOLUTION POLYMER GEL DOSIMETRY ......................................................101
Abstract................................................................................................................................ 102 5.1 Introduction ................................................................................................................... 103 5.2 Theory............................................................................................................................ 104 5.3 Methods ......................................................................................................................... 106
5.3.1 Sample preparation................................................................................................ 106 5.3.2 R2 measurements ................................................................................................... 107 5.3.3 Self-diffusion coefficient measurements................................................................. 107 5.3.4 Computer simulations of the diffusion effect on R2 ............................................... 107 5.3.5 Computer simulations of the diffusion effect on phase encoding ........................... 108
5.4 Results............................................................................................................................ 108 5.4.1 R2 measurements ................................................................................................... 108 5.4.2 Computer simulations of the diffusion effect on R2 ............................................... 111 5.4.3 Computer simulations of the diffusion effect on phase encoding ........................... 112
5.5 Discussion...................................................................................................................... 115 5.6 Conclusions ................................................................................................................... 116 References............................................................................................................................ 116
CHAPTER 6 A STUDY OF A NORMOXIC POLYMER GEL DOSIMETER COMPRISING METHACRYLIC ACID, GELATIN, AND TETRAKIS (HYDROXYMETHYL) PHOSPHONIUM CHLORIDE (MAGAT)...................119
Abstract................................................................................................................................ 120 6.1 Introduction ................................................................................................................... 120 6.2 Materials and Methods .................................................................................................. 122
6.2.1 Formulation ........................................................................................................... 122 6.2.1.1 Investigation of Concentration of THPC and HQ ........................................... 122 6.2.1.2 Investigation of Concentration of Gelatin and MAA....................................... 122
6.2.2 R2-Dose Response ................................................................................................. 123 6.2.3 Spatial Stability...................................................................................................... 123
viii
6.2.4 Scanning and Processing .......................................................................................123 6.3 Results and Discussion ..................................................................................................123
6.3.1 Formulation ...........................................................................................................123 6.3.2 Concentrations of THPC and HQ ..........................................................................123 6.3.3 Concentrations of Gelatin and MAA......................................................................124 6.3.4 R2-Dose Response..................................................................................................125 6.3.5 Spatial Stability ......................................................................................................129
6.4 Conclusions....................................................................................................................132 References............................................................................................................................132
CHAPTER 7 HIGH-RESOLUTION GEL DOSIMETRY OF A HDR BRACHYTHERAPY SOURCE USING NORMOXIC POLYMER GELS ....................................................................................................................135
Abstract................................................................................................................................136 7.1 Introduction ...................................................................................................................136 7.2 Materials and Methods ..................................................................................................138 7.3 Results and Discussion ..................................................................................................140
7.3.1 Calibration .............................................................................................................140 7.3.2 Dose maps ..............................................................................................................142 7.3.3 Dose profiles ..........................................................................................................143 7.3.4 Agreement between dose maps and treatment plans ..............................................143
7.4 Conclusion .....................................................................................................................144 References............................................................................................................................145
CHAPTER 8 GENERAL DISCUSSION ..........................................................................................................147
8.1 The Principal Significance of the Findings....................................................................149 8.1.1 Analyzing and Optimizing Polymer Gel Dosimeter Formulations........................................149 8.1.2 Chemical Properties of Normoxic Polymer Gel Dosimeters .................................................150 8.1.3 A Normoxic Polymer Gel Dosimeter Using THPC...............................................................152 8.1.4 Evaluating Polymer Gel Dosimeters using High-Resolution MRI ........................................153 8.1.5 Application of Normoxic Polymer Gel Dosimeters using High-Resolution MRI to Brachytherapy Treatment Plans .....................................................................................................154
8.2 Conclusions and Future Work .......................................................................................155 References............................................................................................................................159
APPENDIX A ............................................................................................................................................161 LISTING OF THE CODE FOR DETERMINING VARIATIONS IN R2 DUE TO THE APPLICATION OF PULSE SEQUENCES DURING HIGH-RESOLUTION MRI
ix
List of Publications:
1. De Deene, Y., Venning, A., Hurley, C., Healy, B. J., Baldock, C., Dose-
response stability and integrity of the dose distribution of various polymer
gel dosimeters, Phys Med Biol, 2002. 47(14) 2459-2470.
2. De Deene, Y., Hurley, C., Venning, A., Vergote, K., Mather, M., Healy,
B.J., Baldock, C., A basic study of some normoxic polymer gel
dosimeters. Phys Med Biol, 2002. 47, 3441-63.
3. Hurley, C., De Deene, Y., Meder, R., Pope, J.M., Baldock, C., The effects
of molecular self-diffusion of water on quantitative MRI measurements in
high-resolution polymer gel dosimetry, Phys Med Biol, 2003. 48: 3043–
3058.
4. Hurley, C., Venning, A. and Baldock, C., A Study of a Normoxic Polymer
Gel Dosimeter comprising Methacrylic Acid, Gelatin and Tetrakis
(Hydroxymethyl) Phosphonium Chloride (MAGAT), App Radiat and Iso,
2005. 63: 443 - 456.
5. Hurley, C., McLucas, C., Pedrazzini, G., and Baldock, C., High-
Resolution Gel Dosimetry of a HDR Brachytherapy Source Using
Normoxic Polymer Gels: Preliminary Study, Nucl Instr Meth Phys Res A,
2006. 565: 793 – 803 (In Press).
x
List of Abbreviations: AAPM American Association of Physicists in Medicine ABAGIC Acrylamide, methylene-Bis-acrylamide, Ascorbic acid, Gelatin,
Hydroquinone, and copper(II) sulphate gel CT Computed Tomography HDR High Dose Rate HEA 2-Hydroxyethyl Acrylate gel ICRU International Commission on Radiation Units IMRT Intensity Modulated Radiotherapy IVBT Intravascular Brachytherapy LDR Low Dose Rate MAGAS Methacrylic Acid, Gelatin, AScorbic acid MAGAT Methacrylic Acid, Gelatin, Ascorbic Acid and THPC MAGIC Methacrylic Acid, Gelatin, Initiated by Copper Sulphate (the first
normoxic polymer gel proposed by Fong et al (2001). MRI Magnetic Resonance Imaging PAG PolyAcrylamide Gelatin gel PAGAS PolyAcrylamide, Gelatin, AScorbic acid PDR Pulsed Dose Rate R1 MRI Longitudinal Relaxation Rate (measured in s-1) R2 MRI Transverse Relaxation Rate (measured in s-1) T1 MRI Longitudinal Relaxation Time (measured in s) T2 MRI Transverse Relaxation Time (measured in s) TE MRI Echo Time ΔTE MRI Inter-Echo Time THPC Tetrakis (Hydroxymethyl) Phosphonium Chloride TLD Thermoluminescent Dosimeter TR MRI Relaxation Time
xi
Statement of Original Authorship:
The work contained in this thesis has not been previously submitted to meet
requirements for an award at this or any other higher education institution. To the
best of my knowledge and belief, the thesis contains no material previously
published or written by another person except where due reference is made.
Christopher A. Hurley
17th August, 2006
xii
Acknowledgements I would like to sincerely thank the following persons for their invaluable
contributions to the project:
Clive Baldock, my supervisor, for inspiring me to excellence, providing
invaluable advice and for the countless hours of support in helping me to see the
bigger picture in research. I am deeply and sincerely grateful for your endless
enthusiasm and perseverance as you challenged me to achieve ever greater heights
in this project!
Yves De Deene for the incredible knowledge and experiences that you shared
with me, helping me to understand and appreciate the complex world of MRI.
Your ability to open my mind to understanding the world of MRI and Physics is
extraordinary.
Jim Pope for your exceptional ability to understand exactly what was going on
when I was exploring the field of high-resolution MRI. Your advice was always
exactly what I needed.
Thanks to Cameron McLucas and Greg Pedrazzini for your patient work in the
irradiation of the phantoms and your expert technical advice in brachytherapy and
dosimetry. Thanks to Southern X-Ray Clinic of the Wesley Hospital for the use
of MRI scanner, brachytherapy afterloader and linear accelerators and the Wesley
Research Institute. Your commitment to research in medicine is highly
commendable.
Brian Thomas and Elizabeth Stein for your assistance and guidance throughout
the course. Your willingness to sit and talk at any time, encouraged me to see the
course through to the end.
1
Chapter 1
Introduction
1.1 Description of Research Problem Investigated
The use of radiation to supply a lethal dose to tissue affected by disease has been a
practice used for many years. A common part of most major hospitals in the
world, the radiation therapy (or radiation oncology) department has formed an
integral component of medicine’s fight against cancer. The aim of radiation
therapy has been to deliver a dose of ionizing radiation to a tumour or lesion,
whilst minimizing the dose that may be delivered to the surrounding healthy
tissue. Radiation therapy has advanced significantly over the past decades and
now offers a wide range of techniques that can conform a radiation beam to
maximize dose to a targeted tissue whilst minimizing dose to surrounding tissues.
Conformal radiotherapy and intensity-modulated radiotherapy (IMRT) have made
considerable advances in shaping the doses in three dimensions (3D) delivered to
tissues. In a similar fashion, brachytherapy, which involves a radioactive source
being placed directly inside target tissues within a patient’s body, has enabled the
localization of ionizing radiation to a small volume of tissue, again effectively
minimizing the dose being delivered to healthy tissues that may surround the
tumour or lesion. With these advances in dose delivery, the target volumes can
now incorporate complex geometries with high dose gradients.
Although the amount of radiation delivered to the patient’s body can been easily
determined through well-known and derived equations, it is more difficult to
accurate determine the distribution of absorbed dose within the patient’s body.
Absorbed dose distributions are typically determined using computerized
2
treatment planning software that is based on radiation models and simulations of
the absorbed dose. Direct measurements of the absorbed dose have traditionally
been determined using ionization chambers, thermoluminescent dosimeters
(TLDs), solid-state detectors and radiographic film. However, ionization
chambers, TLDs and solid-state detectors usually exhibit poor spatial resolution
(due to the size of the measuring device), radiographic film is inherently 2D and
all these detectors perturb the radiation beam. Gel dosimetry provides a method
by which the spatial, 3D distribution of the absorbed dose can be determined
[1,2].
Currently there are several different formulations for gel dosimeters under
investigation. Each comprises an aqueous gel matrix to provide spatial stability,
cross-linker monomers that polymerize when irradiated and other constituents that
function to maintain the chemical stability or improve the performance of the gel
as a dosimeter [3-6]. Polyacrylamide polymer gel dosimeters suffer from the
effects of oxygen infiltration which prevents the polymerization process that can
be correlated to absorbed dose. Normoxic polymer gel dosimeters, that use an
anti-oxidant to bind free oxygen, have recently been proposed but require further
investigation and development before use in clinical practice.
Following irradiation, phantoms of polymer gel dosimeter are evaluated using
imaging modalities such as MRI [2], x-ray computed tomography (CT) [7],
optical CT [8] and ultrasound [9]. To date, MRI has been the most frequently
used scanning technique for gel dosimetry. Using high-resolution MRI to
evaluate polymer gel dosimeters has the potential to provide dose distributions
with high spatial resolutions in the order of sub-millimeters (~ 100 microns).
However, the evaluation of polymer gel dosimeters using high-resolution MRI is
only in its infancy and there is still much work to be done. To date, high-
resolution MRI has not been used to evaluate normoxic polymer gel dosimeters.
3
1.2 Overall Objective of the Study
The objective of this study was to further develop normoxic polymer gel
dosimetry using high-resolution magnetic resonance imaging and assess its
feasibility in the verification of high-resolution treatment plans, such as those used
in intravascular brachytherapy.
1.3 The Specific Aims of the Study
The specific aims of this study included:
• Exploration of different formulations of polymer gel dosimeters and
normoxic polymer gel dosimeters in order to obtain normoxic polymer gel
dosimeters with optimal characteristics for use in gel dosimetry.
• Measurement of the effects of physical and chemical properties of various
formulations of normoxic polymer gel dosimeters on dose maps obtained
using gel dosimetry with high-resolution MRI to assess their performance
for use in radiotherapy dosimetry.
• Investigation of high-resolution magnetic resonance imaging, to achieve
in-plane spatial resolutions ~ 100 microns, for its potential use in
evaluating polymer gel dosimeters accurately and efficiently.
• Examination of the effects of molecular diffusion on images produced
using high-resolution MRI in order to eliminate the effects causing errors
in the calculated dose maps of polymer gel dosimeters.
• Application and assessment of normoxic polymer gel dosimeters
irradiated using typical brachytherapy deliveries and evaluated using high-
resolution MRI.
4
1.4 Account of Scientific Progress Linking the Scientific
Papers
Normoxic polymer gel dosimeters present a significant advance in gel dosimetry
and show good potential to the verification of radiation therapy and brachytherapy
dose distributions in 3D. However, their development is still in its infancy and
normoxic polymer gel dosimeters have yet to be incorporated into clinical
practice. Using high-resolution MRI provides the potential for extending the
applications of gel dosimetry to include radiation therapy and brachytherapy
treatments that require verification with resolutions at the sub-millimeter level
(typically ~ 100 microns). This study has been broken into two aspects: the
analysis and development of normoxic polymer gel dosimeters and secondly, the
investigation of the use of high-resolution MRI to evaluate normoxic polymer gel
dosimeters.
Significant changes of the polymer structure are known to occur in polymer gel
dosimeters following irradiation [10]. These changes ultimately affect the
chemical and physical properties of a gel and hence its suitability for use in gel
dosimetry. Chapter 3 investigates some different formulations of polymer gel
dosimeter, including polyacrylamide gel (PAG), polymer gel dosimeters made
with 2-hydroxyethyl acrylate (HEA), and normoxic polymer gel dosimeters
including the MAGIC gel (methacrylic and ascorbic acid in gelatin initiated by
copper) [11]. The effects of varying the concentrations of the constituent
components of the gel on the R2-dose response was explored to produce an
optimal formulation. The temporal stability of the R2-dose response was also
examined by relating changes in the R2-intercept and the R2-dose sensitivity
(slope) to reactions within the gel. The spatial stability was investigated using
dose profiles through a phantom exposed to a half-blocked field.
Chapter 4 further investigates normoxic polymer gel dosimeters through a
chemical analysis of the MAGIC gel. The role of the different chemicals and
reactions kinetics are explored as they affect the R2-dose response and spatial and
5
temporal stability of the MAGIC gel. A comprehensive investigation of the
chemical reactions that take place within a normoxic polymer gel dosimeter is
presented in order to explain the role that each constituent component plays in the
overall gel. An understanding of these reactions assists in the development of
more optimal formulations. In addition, alternative anti-oxidants to ascorbic acid
are proposed and investigated as to their effectiveness for oxygen scavenging in
normoxic polymer gel dosimeters.
High-resolution MRI requires high gradient strengths that can significantly vary
the R2 values obtained due to molecular diffusion of water molecules within the
gel itself. The degree of variation is affected by the imaging parameters chosen
by an operator. Chapter 5 investigates the extent to which the imaging parameters
alter the R2 values and techniques that can be incorporated to provide an accurate
dose map using high-resolution MRI in polymer gel dosimetry. Software is
developed that can predict the variation in R2 due to the application of MRI pulse
sequences that are used.
Chapter 6 investigates a normoxic polymer gel dosimeter comprising tetrakis
(hydroxymethyl) phosphonium chloride (THPC) as an alternative anti-oxidant to
the ascorbic acid that is used in MAGIC polymer gel dosimeters. This gel,
composed of methacrylic acid, gelatin, and THPC, was called MAGAT. This
chapter evaluates the R2-dose response, the temporal stability of the R2-dose
response, the spatial stability and provides an optimal formulation for the
MAGAT polymer gel dosimeter using high-resolution MRI.
Finally, chapter 7 examines the application of a normoxic polymer gel dosimeter
using high-resolution MRI to typical brachytherapy deliveries. Using a line and a
point irradiation pattern as a plan, dose distribution maps were produced and
compared with dose points predicted by the treatment planning system. Adjusting
for partial volume effects, the dose profiles show good agreement to dose points
predicted by the computerized treatment planning system.
6
References
[1] Maryanski, M.J., Gore, J.C., Kennan, R.P. and Schulz, R.J., 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(2)
253-8.
[2] Maryanski, M.J., Gore, J.C. and Schulz, R.J., US Patent: Three-dimensional
detection, dosimetry and imaging of an energy field by formation of a
polymer in a gel. 1994, Patent Number 5,321,357. United States.
[3] Baldock, C., Burford, R.P., Billingham, N., Cohen, D. and Keevil, S.F.,
Polymer gel composition in MRI dosimetry. Med Phys, 1996. 23 1070.
[4] De Deene, Y., Hanselaer, P., De Wagter, C., Achten, E. and De Neve, W.,
An investigation of the chemical stability of a monomer/polymer gel
dosimeter. Phys Med Biol, 2000. 45(4) 859-78.
[5] Lepage, M., Whittaker, A.K., Rintoul, L., Back, S.A. and Baldock, C.,
Modelling of post-irradiation events in polymer gel dosimeters. Phys Med
Biol, 2001. 46(11) 2827-39.
[6] Lepage, M., Whittaker, A.K., Rintoul, L., Back, S.A. and Baldock, C., The
relationship between radiation-induced chemical processes and transverse
relaxation times in polymer gel dosimeters. Phys Med Biol, 2001. 46(4)
1061-74.
[7] Trapp, J.V., Back, S.A., Lepage, M., Michael, G. and Baldock, C., An
experimental study of the dose response of polymer gel dosimeters imaged
with x-ray computed tomography. Phys Med Biol, 2001. 46(11) 2939-51.
[8] Gore, J.C., Ranade, M., Maryanski, M.J. and Schulz, R.J., 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(12) 2695-704.
[9] Mather, M.L., Whittaker, A.K. and Baldock, C., Ultrasound evaluation of
polymer gel dosimeters. Phys Med Biol, 2002. 47(9) 1449-58.
7
[10] Lepage, M., Whittaker, A.K., Rintoul, L. and Baldock, C., 13C-NMR, 1H-
NMR, and FT-Raman study of radiation-induced modifications in radiation
dosimetry polymer gels. J App Poly Sci, 2001. 79 1572-1581.
[11] Fong, P.M., Keil, D.C., Does, M.D., and Gore, J.C., Polymer gels for
magnetic resonance imaging of radiation dose distributions at normal room
atmosphere. Phys Med Biol, 2001. 46(12) 3105-13.
8
9
Chapter 2
Literature Review
2.1 Radiotherapy and Brachytherapy Since the development of radiation therapy and related fields, patients have been
exposed to radiation for the treatment of malignant disease. The approach taken
to achieve this has involved the exposure of the affected tissue area to a radiation
beam of sufficient energy to cause significant damage to the cells in the affected
area without affecting the surrounding normal tissue to any great extent [1].
Exposing tissues affected by cancer to sufficient levels of radiation causes
irreversible cell damage therefore preventing the cancerous cells from further
growth and metastasis [2].
There are currently several methods by which radiation can be delivered to a
targeted area. The most widely used method is external beam irradiation. In this
approach, a linear accelerator (linac) generates megavoltage energy photon or
electron beams that are focused and aimed onto a patient’s body. Using
collimation and rotation, linacs are able to confine the radiation exposure to a
particular region of the patient’s body. Using x-ray computed tomography (CT) it
is possible to obtain scans of a patient enabling greater anatomical delineation.
This information can then be used to ensure that high doses of radiation are
delivered to target volume whilst sparing the surrounding healthy tissues [3].
Magnetic resonance imaging (MRI) and positron emission tomography (PET) can
enhance the process of tumour identification giving a more precise geometric
definition of the tumour. A more precise tumour definition may lead to improved
irradiation of the true extent of the tumour. This process is known as conformal
radiation therapy (CRT). Multi-leaf collimators (MLCs) can now be used to
10
control the shape of the radiation beam used to treat a patient [4]. Current MLCs
typically have between 40 to 120 leaves of varying widths (0.5 to 1 cm) across the
leaf range. These leaves, made of tungsten, can be moved in front of the beam at
specific lengths to define the radiation beam. In this way, treatment exposures can
be made to conform more precisely to target volumes using sophisticated
computer software that manipulates radiotherapy equipment in real time. One key
objective of the clinical implementation of conformal radiotherapy is assuring that
the complex manipulations of the radiotherapy equipment required for the therapy
are actually performed, and that the dose distributions calculated by treatment
planning systems and delivered by treatments are correct [5]. Correctly
measuring the delivered dose is central to the process of assuring the accuracy of
treatment plans and treatment deliveries.
Intensity modulated radiation therapy (IMRT) provides the ability to vary the
radiation fluence within each radiation beam during treatment. IMRT can further
enhance the ability of linac to control the radiation distribution within a targeted
volume. The dose distribution in this case is non-uniform across several radiation
beams, which summate to produce an optimal dose distribution [6]. Combined
with conformal radiation therapy using multi-leaf collimators, treatment plans can
now incorporate complex geometries with non-uniform dose distributions to be
delivered to a patient.
With the increase in treatment complexity, the need for verification of computer
generated treatment plans is most significant. The dose distribution required by a
treatment plan is calculated by complex computer algorithms that model the
radiation system. Monte Carlo models have the potential to predict the dose in
complicated geometries using a variety of different types of radiation (electrons,
photons, scatter from collimators, scatter within the patient body, etc.) as applied
to the internal geometry of the region of the patient body [7]. Despite the
advances in Monte Carlo modeling of radiotherapy deliveries, there is still a need
to verify the absorbed dose at points within the distribution using direct
measurements.
11
Brachytherapy involves the use of radioactive sources that are either placed
directly on the patient’s skin or, more commonly, inserted into the patient and
positioned close to the affected tissues [1]. In this way, high radiation doses can
be delivered directly to the tumour site thereby concentrating dose in regions
requiring treatment and effectively minimizing the irradiation of surrounding
healthy tissues. Brachytherapy treatments can be classified as either high dose
rate (HDR) or low dose rate (LDR) depending on the radioactive source used.
HDR sources must be inserted into the target tissue for a short period of time,
whereas LDR sources must be inserted for substantially longer periods of time to
deliver sufficient dose to the targeted tissues. New brachytherapy source designs
are often commissioned using a Monte Carlo-based analysis of the dose
distribution surrounding the source [8-15].
More recently, a combination of the HDR and LDR brachytherapy has been used
called pulsed dose rate brachytherapy (PDR) [14,16,17]. PDR involves the use of
stronger radiation sources than those used in LDR brachytherapy given in a series
of short exposures of 10 to 30 minutes every hour to approximate the same overall
dose as with LDR brachytherapy. Typically sources include gamma and beta
emitters such as 192Ir, 32P, 90Sr/Y and 125I. Each source has its own advantages
and is generally chosen to match a specific treatment. These sources may be used
either as wire stents, seeds inserted through catheters or liquids that can be
injected into balloons during angioplasties.
Intravascular Brachytherapy (IVBT) emerged from the need to resolve the
problem of restenosis, a re-closing of arteries following angioplasty [18-20]. In
1995, the American Association of Physicists in Medicine (AAPM) established a
task group committee to investigate and report on the dosimetry of interstitial
brachytherapy sources. Their findings were presented in a report, called TG-43,
that examined the role of photon-emitting sources used for interstitial
brachytherapy [21]. This work was followed-up by a second report in 1999 by
TG-60 that investigated the physics in intravascular brachytherapy [22]. This
report examined the relative advantages of different sources available for use in
brachytherapy. It found that beta emitters, such as 32P and 90Sr/Y have advantages
in terms of high activity and dose rate, good radiation safety and a long half-life.
12
Gamma emitters, such as 192Ir, have advantages in terms of radial dose uniformity,
high dose rate, and reasonably long half-life. One of the key concerns to be raised
in this report was dose inhomogeneity due to non-centering of the source. This
concern is reflected in the need to assure the positional accuracy of the radiation
delivery system that is particularly significant in brachytherapy [23]. Although
Monte Carlo calculations have been effective in the determination of dose
distributions surrounding brachytherapy sources, these must be verified for further
optimization of procedures to be possible.
2.2 Radiation Dosimetry
Dosimetry has been at the core of the development of radiation therapy and there
currently exists many different methods of measuring the absorbed dose delivered
to tissue (and other mediums) [24,25]. Today, complex computer algorithms are
used to determine the dose distributions required in a particular treatment plan.
These algorithms aim to assure that the calculated dose distributions accurately
reflect the dosimetry requirements of a treatment plan. In addition, with computer
controlled radiation delivery, software must assure that the complex
manipulations of radiotherapy equipment are actually performed correctly [5,26].
The determination of absorbed dose in 3D is fundamental to clinical
environments, but few methods exist by which 3D measurements can be made
easily and accurately. Computer algorithms model the effective dose absorbed
into specific volumes of tissue and other structures in the patient’s body [1,22].
However useful the computer calculations might be in the prediction of absorbed
dose, the problem of easily and accurately measuring the actual absorbed dose
distribution in 3D still remains, even if only to provide a verification of the
computed calculations [27].
2.2.1 Clinical Dosimetry Requirements
Advances in conformal radiation therapy treatments have enabled target volume
to be defined with complex geometries. Similarly, radiation delivery can consist
13
of varied and non-uniform irradiation fields. These complexities make radiation
dosimetry a difficult process that must meet stringent criteria in order to be
effective in producing accurate and detailed maps of the dose distribution. The
following parameters are typically significant in the design of dosimetry systems:
measurement sensitivity, accuracy, precision, spatial resolution, energy and dose-
rate insensitivity, tissue equivalence, non-directionality, ease of use and able to
produce three-dimensional, integrated dose maps [7].
The International Commission on Radiation Units (ICRU) recommends that the
overall accuracy in delivered dose be within 5 % of the true dose [28,29]. To be
effective in a variety of radiation therapy applications, dosimeters need to measure
dose accurately and to have high spatial resolution. This is particularly true for
brachytherapy, where high resolution is essential since steep dose gradients exist
close to the source. These high dose gradients can occur within very small
regions of interest, typically < 2 mm [30]. Detectors should also exhibit tissue
equivalence in order to not perturb the radiation field and hence have a significant
effect on the measured dose. Non-tissue equivalent dosimeters require the
application of correction factors in order to determine absorbed dose. These may
introduce uncertainties. Dosimeters should also be able to integrate dose for a
number of sequential fields to accommodate the time varying doses delivered to a
patient.
2.2.2 Dosimetry Detectors
There are currently many different dosimetry detectors that are used to determine
radiation dose distribution delivered in radiotherapy treatments each with its own
advantages and disadvantages based on criteria listed above.
2.2.2.1 Ionization Chambers
Currently the most commonly used dosimeter for external beam measurements is
the ionization chamber [31,32]. Ionization chambers provide one dimensional
point dose measurements in radiation therapy applications. Ionization chambers
14
consist of a cavity containing gas, usually air, in which charge is liberated by the
ionization of a gas within the chamber by radiation. This charge is collected by
electrodes, typically composed of aluminum or carbon based material. The charge
can then be correlated to the delivered dose through calibration factors traceable
to a standards laboratory. Although these detectors have high accuracy and
practicality, they do introduce a perturbation of the photon beam which must be
corrected for. The active volumes of ionization chambers are typically between
0.1 and 0.6 cm3 resulting in poor spatial resolution making them unsuitable for
dosimetry in the near-zone of brachytherapy sources. Measuring the dose
distribution in 3D using ionization chambers is a laborious process. Dose
distributions resulting highly conformal external beam delivery typically require
spatial resolution beyond the capabilities of ionization detectors.
2.2.2.2 Solid-State Detectors
Solid-state detectors, such as semiconductor diodes, can be manufactured with an
active volume around 0.1 mm3, and hence are more suited to applications such as
brachytherapy. Solid-state detectors function by measuring the amount of charge
liberated by the passage of ionizing radiation in solid semiconductors [33-36].
Arrays of solid-state detectors can be spatially arranged or translated around static
fields to achieve 2D and 3D maps of dose distributions. These detectors have the
ability to measure dose distributions with higher resolutions than ionization
chambers, in real-time and are easy to use, however, they are not tissue equivalent
and require independent calibration. Solid-state detectors also suffer from an
energy dependence and their response drifts over time due to radiation damage.
2.2.2.3 Thermoluminescent Dosimeters
Thermoluminescence refers to the emission of light from an irradiated crystalline
material following heating. The amount of light emitted from a crystal can be
correlated to absorbed dose. Thermoluminescent dosimeters (TLDs) consist of
small crystals available in a range of sizes, some as small as 1 mm2, that act as
point detectors. When exposed to radiation, these crystals store a small amount of
the energy in the crystal lattice. Upon heating, these crystals release this stored
energy as light, which can be detected using a photomultiplier tube. They can be
used to provide higher resolutions using a spatially arranged array of closely
15
packed TLDs. The main advantages of TLDs include: their wide useful dose
range, small physical size, reuseability and economy for most radiation types [37-
40].
2.2.2.4 Radiographic Film
Radiographic films are used to capture a 2D image of a dose distribution.
Conventional films are based on silver-halide (typically silver bromide) and have
a strong energy dependence at photon energies in the 10 to 200 keV range [41], an
effect derived from the high atomic number of the silver in the film and
absorption due to the photoelectric effect that is significant in this energy range.
In addition, silver-halide films are non-tissue equivalent. They function by
converting silver ions to silver upon irradiation. The bromine is removed during
developing leaving opaque clusters of silver on the film in irradiated regions.
Radiochromic films are relatively tissue equivalent and do not require chemical
developing, however, they exhibit significant temperature dependence and their
sensitivity varies photon energy. They develop a specific colouring in irradiated
regions as a result of a dye-forming or polymerization process in which energy is
transferred from an energetic photon or particle to the receptive part of a leuko-
dye or colourless photomonomer molecule [41]. Gaf-chromic films, a type of
radiochromic film, were developed with a more uniform response with photon
energy. Gaf-chromic films are popular in clinical radiation dosimetry as they also
exhibit high sensitivity and high spatial resolution [42,43].
Film dosimeters are inherently 2D and can be used to obtain dose information by
relating the absorbed dose to the optical density of the film. Although sheets of
film can be stacked between tissue equivalent material to provide a 3D dosimetry
system, the process is cumbersome and time consuming. The use of radiographic
films in dosimetry is complicated by their non-tissue equivalence, uncertainties in
film processing and their inherent 2D nature [44].
2.2.2.5 Chemical Dosimeters
Chemical dosimeters function on the phenomenon that chemical changes occur in
the dosimeters when exposed to ionizing radiation. The absorbed dose delivered
16
to the dosimeter can be correlated to the extent of radiation-induced chemical
change within the dosimeter itself. These chemical changes can be measured as a
change in the spin-lattice or spin-spin relalaxation rates, the change in
concentration of ions present in solution or the optical turbidity within the sample.
One of the more widely known chemical dosimeters is the Fricke dosimeter,
proposed by Fricke and Morse in 1927 [45]. The Fricke dosimeter functions by
the conversion of ferrous ions (Fe2+) to ferric ions (Fe3+) in solution when
irradiated.
A second type of chemical dosimeter uses the radiolysis of water within a gel
matrix to initiate polymerization reactions. The degree of polymerization can be
correlated to the absorbed dose [46]. Chemical dosimeters are capable of high
precision dose measurements and can provide spatial information when dissolved
into an aqueous gel matrix.
2.3 Gel Dosimeters
As outlined above, there is a need for a dosimeter that does not perturb the
radiation beam, is inherently three-dimensional and has the ability to integrate
radiation doses over time. The use of radiation sensitive gels to fulfill these
requirements has great potential. Radiation sensitive gels were first considered
for use in radiation dosimetry in the 1950s by Day et al. who were investigating
radiation induced colour changes in dyes [47,48]. With the addition of gelling
agents, a chemical dosimeter could be made to be spatially stable and hence to
provide spatial dose information. With the development of methods to image the
chemical changes within gel dosimeters, this became the basis of modern gel
dosimetry.
Gel dosimetry has advanced significantly over the past two decades. Now gel
dosimeters have the ability to measure three-dimensional dose distributions with
high resolution (less than 1 mm in-plane resolution) [49] and the dose sensitivity
of the gel is independent of the energy of the irradiating beam and the dose rate
17
used to irradiate the gel [50]. Most importantly, the gel is the dosimeter and thus
does not perturb the radiation beam like conventional dosimetry techniques [51].
In addition, gel dosimeters are also tissue equivalent. The gel dosimeter can be
used to simulate the tissue of the human body undergoing radiation therapy, by
pouring into anthropomorphic phantoms [52].
There are two main varieties of gel dosimeter: ferrous sulfate gel dosimeters and
polymer gel dosimeters. More recently, normoxic polymer gel dosimeters have
become popular due to their ability to be manufactured under normal atmospheric
conditions.
2.3.1 Ferrous Sulfate (Fricke) Gels
In 1984, Gore et al. proposed the use of nuclear magnetic resonance imaging of
ferrous sulfate gel dosimeters, also called a Fricke gel, which exhibit a change in
their paramagnetic species as a result of exposure to ionizing radiation [53]. Gore
et al. also proposed the potential of this chemical response as a dosimeter capable
of producing 3D dose distributions of the nature required for use in radiation
therapy. The use of a Fricke gel has been developed into a method founded on the
principle that ferrous ions (Fe2+) are oxidized to ferric ions (Fe3+) when subjected
to free radicals produced by exposing water to ionizing radiation. To achieve this,
a gel consisting of an aerated dilute solution of ammonium ferrous sulfate is
suspended in an aqueous gel, such as agarose or gelatin. The gel matrix provides
the support structure by which the dosimeter can maintain a spatial arrangement,
and hence provide spatial information about the dose distribution within the
irradiated gel dosimeter. This was significant for gel dosimetry as Gore et al.
were able to show that the radiation-induced change could be detected using
nuclear magnetic resonance (NMR) [53].
The change from ferrous to ferric ions that occurs in regions exposed to the
radiation, provides changes in the gel dosimeter’s nuclear magnetic resonance
(NMR) spin-lattice relaxation rate, R1, and spin-spin relaxation rate, R2, as both
18
ferrous and ferric ions are paramagnetic species capable of reducing the proton
relaxation times of water. In particular, the ferric ions exhibit a stronger
paramagnetic enhancement of the water-proton NMR relaxation rates. Gore et al.
were also able to show that by spatially distributing the ferrous ions, the spatial
distribution of dose could be imaged using MRI [53]. This was a significant
advance in dosimetry as this was the first evidence for a truly 3D dosimeter for
clinical radiation therapy. Since then, many studies have investigated combining
Fricke chemical dosimeters with gelling agents and imaging using MRI [54-56].
Optical tomography, an alternative to MRI, has also been investigated by many
authors as an imaging modality for Fricke gel dosimetry [57-60].
A great deal of research has been performed to investigate various aspects of
Fricke gel dosimetry, including the effects of: ferrous ion concentration, radiation
dose rate, beam energy, oxygenation, agarous concentration and acid content [61-
64]. Additionally, the use of alternative gelling agents, such as gelatin [65-67]
and polyvinyl alcohol (PVA) [68], have been explored.
There are two major drawbacks with the use of Fricke gels for 3D dosimetry. The
first limitation of ferrous gels is the continual diffusion of the ferric ions through
the gel post-irradiation (see references cited in Baldock et al. 2001 [69]). This
diffusion leads to a blurring of dose distributions over time, and hence a
degradation of spatial integrity of the dosimetry. However, with the use of
chelating agents, this diffusion can be reduced to better maintain the spatial
information over extended time periods [69]. The addition of saccarides to
ferrous-agarose-xylenol orange gels have also improved dose sensitivity [70].
The problems experienced with Fricke gel dosimeters, have prompted an
alternative gel formulation that could provide a more stable dosimeter that would
provide both a better spatial resolution as well as being relatively insusceptible to
problems caused by ageing [71]. A more detailed discussion of Fricke gel
dosimeters can be found in Back et al. 1999 [72] and Schreiner 2004 [73].
19
2.3.2 Polymer Gel Dosimeters
In 1993, Maryanski et al. introduced the use of radiation initiated polymerization
into gel dosimetry [74]. It was well-known that polymerization and cross-linking
could be initiated by irradiation and that the degree of polymerization could be
correlated to the amount of radiation delivered [75-78]. Maryanski et al. used this
knowledge to construct a gel dosimeter based on the polymerization [74]. This
polymer gel dosimeter used an agarose gel infused with acrylamide and N,N’-
methylene-bis-acrylamide (bis) co-monomers. It functioned on the premise that
ionizing radiation would initiate the polymerization of the co-monomers and
induce cross-linking by way of the bis forming a connection between two
acrylamide chains. Nitrous oxide was used to saturate the solution in order to
remove any oxygen. Likewise, the manufacture of these dosimeters had to occur
in a hypoxic environment as oxygen was shown to inhibit the polymerization
process [74]. The use of polymer gel dosimeters provided solutions to some of
the problems that had been encountered with Fricke gels and hence provided the
ability to conduct improved dosimetry in 3D. Polymer gel dosimeters do not
suffer the diffusion problems observed in Fricke gels and the range of doses that
the polymer gel dosimeters responded to can be engineered by varying the
chemical constituents that make up the gel [5]. Another desirable property of
polymer gel dosimeters is their optical characteristics, in particular the observable
transparency of regions of the gel that are unexposed to ionizing radiation and the
opaque regions where ionizing radiation has affected the gel provided immediate
visual clues as the distribution of absorbed dose. The opacity of regions within
the irradiated regions of gel are due to the formation of polymer aggregates
initiated by free radicals formed by the radiolysis of water molecules.
Polymer gel dosimeters were found to exhibit significant changes in both the
NMR spin-lattice relaxation rate (R1) and the spin-spin relaxation rate (R2).
However, unlike the Fricke gel, in polymer gel dosimeters, a change in R2 is
much more pronounced than the change in R1 [74,79-81]. Therefore, the spin-
spin relaxation rates (R2 = 1/T2) determined from a suitable magnetic resonance
20
image and are correlated to the absorbed dose. One of the main aspects to emerge
from the investigation of polymer gel dosimetry using MRI was a quantitative
method of measuring the performance of the polymer gel as a dosimeter. The
slope of the linear region of the R2-dose response over the range 0 to 10 Gy,
provides a measure called the R2-dose sensitivity of the particular gel dosimeter
(see figure 1). The R2-dose sensitivity is a useful quantity to compare different
gel formulations and MRI imaging techniques. The original formulation of the
polymer gel exhibited a dose sensitivity of 0.28 s-1 Gy-1 [74].
0 2 4 6 8 10
1
2
3
4
R2 = 0.96146 + 0.28574*Dose(r2 = 0.9953, p < 0.0001)
R2 (
s-1)
Absorbed Dose (Gy)
Figure 1. A typical R2-dose response from a polyacrylamide gel dosimeter
showing a linear fit with corresponding formula from which the dose sensitivity
can be determined.
Maryanski et al. then went on to further develop the formulation of polymer gel
dosimeters. A new polymer gel dosimeter, based on Bis, Acrylamide, Nitrogen
and Gelatin (called BANG) was produced [82-84]. The reason for changing the
gel matrix from agarose to gelatin lay in attempting to reduce the component of
the R2 that was due to the gel matrix itself. The gelatin produced an R2 an order
of magnitude lower than the agarose gel. This reduction in the magnitude of R2
significantly reduced the zero dose baseline in MRI measurement [82]. The R2
magnitude then became dominantly controlled by the polymerization that had
occurred in the gel due the absorbed dose itself. The R2-dose sensitivity of the
polymer gel dosimeter based on gelatin was found to be only 0.25 s-1 Gy-1 over
21
the range of 0 to 8 Gy for a magnetic field strength of 1.5 T, but it did demonstrate
good reproducibility.
Since this initial work various concentrations of polymer gel dosimeters have been
explored. The replacement of acrylamide with acrylic acid led to the development
of a polymer gel dosimeter partially suited to the verification of stereotactic
radiosugery and high dose rate (HDR) brachytherapy dose delivery [50]. The
study demonstrated that the dose response was independent of irradiation
conditions. The R2-dose sensitivity of this polymer gel dosimeter was found to be
0.335 s-1 Gy-1 for doses up to 12 Gy. The response of the gel was independent of
the energy, up to 15 MeV, and of the dose rate, over the range 0.003 to 0.067 Gy
s-1.
The optical characteristics of gelatin based polymer gel dosimeters based on
gelatin were further explored using optical tomographic densitometry by
Maryanski and Gore [85,86]. The method is founded on the scattering of light
that occurs due to the presence of micro-particles that are produced due to the
polymerization of co-monomers initiated by the irradiation. The measurement
procedure consisted of the use of a specially designed optical scanner that
comprised a He-Ne laser that imaged the gel in much the same fashion as first
generation x-ray computed tomography using filtered back-projection to
reconstruct images of the absorbed dose within the gel. This method of imaging
gels was further developed into a method of dosimetry for complex stereotactic
radiosurgery [87]. This paper detailed the benefits that polymer gel dosimetry had
over conventional dosimetry techniques that had been used previously in
radiosurgery, including: thermoluminescent dosimeters (TLDs) and small volume
ionization chambers. These devices suffer from the inherent problems of poor
spatial resolution due to the size of the device and the perturbation of the ionizing
radiation due to the physical presence of the device when measuring. Knisley et
al. showed the polymer gel dosimeter to be effective 3D dosimeters exhibiting
high resolution, precision and accuracy [87].
Pappas et al. investigated the use of N-vinylpyrrolidone combined with bis in the
manufacture of an alternative polymer gel dosimeter called VIPAR [88,89].
22
Nitrogen was replaced with argon in this study as argon is heavier than air and
thus decreased the likelihood of air diffusion through the seals on the vessels
containing the dosimeter. Formulations including either gelatin or agarose were
explored, however, the agarose produced an increase in the turbidity of the gel
and, thus, gelatin was prefered. The R2-dose sensitivity of this polymer gel
dosimeter was found to be ~ 0.1 s-1 Gy-1 which, although less than half that of
acrylamide gels, remained constant with time between irradiation and imaging and
showed good reproducibility. The R2-dose sensitivity of VIPAR exhibited a
quasi-linearity over the range 0 to 12 Gy and thus validated the use of polymer gel
dosimeters based on N-vinylpyrrolidone as a satisfactory formulation for
applications in gel dosimetry.
The replacement of acrylamide with sodium methacrylate was investigated by
Murphy et al. [90]. The sodium methacrylate formulation exhibited several
advantages over the acrylamide formulations including: reduced toxicity, a higher
R2-dose response and the inclusion of a distinct NMR signal due to the presence
of a methyl group in the monomer. The methyl group is later consumed in the
polymerization process. Proton spectroscopy had been previously used to study
polyacrylamide gel dosimeters and had shown that the overall loss of monomer
could be determined spectroscopically as a function of dose [91]. Murphy et al.
were able to demonstrate the same or enhanced possibility in polymer gel
dosimeters composed of sodium methacrylate [90]. It was also found that the pH
of the gel had a major effect on the overall dose sensitivity of the gel. If the pH
was left unchanged during the manufacture of the gel, the R2-dose sensitivity was
found to be only half that of polyacrylamide gel dosimeter, however, by adding
sodium hydroxide and raised the pH to 7.7, the R2-dose sensitivity was made
equivalent to that of the polyacrylamide gels [90].
In summary, polymer gel dosimeters possess many of the desired characteristics
that were required for dosimetry in radiation therapy. Polymer gel dosimeters
have the ability to integrate dose without perturbing the radiation beam, they are
tissue-equivalent, independent of radiation energy over a wide range of photon
energies and inherently 3D. However, clinical acceptance has been limited in part
23
because they were not easy to manufacture, or use, due to the requirement of a
strict hypoxic environment during their manufacture [92].
2.3.3 Normoxic Polymer Gel Dosimeters
The next significant step in development of gel dosimetry came with the advent of
polymer gel dosimeters that could be manufactured under normal atmospheric
(normoxic) conditions. The first normoxic polymer gel dosimeter was proposed
by Fong et al. in 2001 [93]. Called MAGIC, it comprised: methacrylic, ascorbic
acid, hydroquinone, gelatin and copper(II) sulphate. The main feature introduced
in normoxic polymer gel dosimeters was the addition of an anti-oxidant, in this
case ascorbic acid, into the gel formulation. As noted earlier, the polymerization
process within polymer gel dosimeters is inhibited by the presence of oxygen
which scavenges the free radicals produced by the radiolysis of water. It is
usually these free radicals that initiate the polymerization reaction. With the
inclusion of an anti-oxidant in the formulation of a polymer gel dosimeter, oxygen
present in the gel dosimeter can be bound into metallo-organic complexes. Once
the oxygen is bound, it is prevented from binding the free radicals and hence
inhibiting polymerization reaction essential for polymer gel dosimetry.
More recently, various studies have investigated the addition of other anti-
oxidants to a range of different polymer gel dosimeter formulations. Tetrakis
(hydroxymethyl) phosphonium chloride (THPC) has been added as an anti-
oxidant to various formulations including methacrylic acid gelatin gel dosimeters
with copper(II) sulphate and hydroquinone (MAGAT), polyacryamide gelatin gel
dosimeters (PAGAT) [94-96], and just methacrylic acid, gelatin gel dosimeters
(MAGAS) [97]. Venning et al. have performed an extensive analysis of
radiological properties the MAGIC, MAGAS and MAGAT gels using Monte
Carlo modelling [94]. They were able to ascertain that the gel exhibited many of
the characteristics necessary for use in radiotherapy gel dosimetry, including
tissue equivalence. There are still many variations of formulation for normoxic
24
polymer gel dosimeters yet to be fully investigated. The further characterization
of normoxic polymer gel dosimeters is a major component of this thesis.
2.4 Characteristics of Polymer Gel Dosimeters 2.4.1 Effects of Oxygen
The process of polymerization is initiated by free radicals formed from the
radiolysis of water in the gel composition. These free radicals combine with the
monomers making them reactive. Molecular oxygen, however, acts as a
scavenger of these free radicals and hence prevents them from initiating the
polymerization process [51,74,82,92]. Even trace amounts of oxygen in the gel
mixture can lead to the failure of the gel as an effective dosimeter. An important
component of the manufacture of polymer gel dosimeters is the removal of
oxygen from either a reaction flask or a glove box by the bubbling of an inert gas,
for example nitrogen or argon, through the water that is to be used in the
formulation before mixing the other ingredients [98,99]. It is, therefore, important
to ensure the type and quality of the seals used on the vessels do not allow the
diffusion of oxygen into the vessel. Maintaining a strict hypoxic environment has
been a significant drawback of polymer gel dosimeters in the past and made the
process of polymer gel dosimetry awkward to implement into clinical practice.
However, with the advent of normoxic polymer gel dosimeters, as described
above, the strict hypoxic environment is no longer required. Normoxic polymer
gel dosimeters can be manufactured under normal atmospheric conditions on the
bench top. However, the development of normoxic polymer gel dosimeters is in
their infancy and much work is still to be done to be able to fully understand and
integrate normoxic polymer gel dosimeters into clinical practice.
2.4.2 Effect of Light
The initiation of the polymerization process should be caused by the radiolysis of
water that leads to the production of free radicals, as discussed above. However, a
25
number of alternative initiators exist. Bright light, especially sunlight, can initiate
photopolymerisation of the gel before it is irradiated and consequently degrade the
sensitivity of the gel [51,100]. Polymer gel dosimeters should, therefore, be
manufactured, irradiated and stored away from strong light sources.
2.4.3 Temperature
There are several places where temperature plays a significant role in the
manufacture of the gel. The first step in the manufacturing procedure requires
high temperature to facilitate mixing of the gelatin and water. The gelatin must be
added whilst the water is at room temperature to avoid the gelatin forming lumps.
Once the gelatin has soaked into the water, the mixture is then heated to ~ 50 °C
to ensure that the gelatin has completely dissolved into the water [98]. The
temperature of the mixture must be kept below 55 °C when mixing the monomers
to avoid pre-polymerisation that may be caused due to the temperature of the
solution. Following the manufacture, the temperature of the gel should be kept
low to ensure the gel sets in the vessel it has been placed in. Salomons et al.
(2002) have showed that a temperature increase occurs within a polyacrylamide
gel dosimeter during and immediately after irradiation due to the exothermic
polymerization reactions [101]. This temperature change can affect
polymerization reactions within the gel dosimeter and hence may lead to
inaccurate calibration of gel dosimeter images. In order to minimize the effect of
this artifact on the dose maps produced by gel dosimeters, the size, shape and
temperature of the gel dosimeters must be controlled.
During magnetic resonance imaging, the temperature of the polymer gel has a
very significant effect of the overall R2-dose sensitivity of the polymer gel
dosimeter. Several authors have found an increase in the R2-dose sensitivity of
the gel as temperature decreases [102,103]. This effect is thought due to a change
in the proton exchange rates in the gel as the temperature is varied. As the
temperature is decreased, the motion of the water protons becomes slower. This
increases the exchange rate of energy between protons [51]. It has also been
found that a change of even 1 °C within a phantom can give rise to dose
26
uncertainties of approximately 50 cGy in dose maps derived from gel dosimeters
imaged using MRI [102]. The temperature of a gel undergoing the imaging
process must be kept constant to avoid changes in the relaxation rates over the
time of imaging. The gel should be kept in temperature controlled conditions,
such as those of the MRI room, for at least 12 hours before imaging to allow time
for the dosimeter to equilibrate to the scanning temperature.
2.4.4 Concentration of monomers
The R2-dose sensitivity of a gel may be increased by increasing the total
monomer content of the gel, typically symbolized using %T [102,104]. The
typical concentrations of monomers range from 3 % to 9 % of the total weight.
An increase in the monomer concentration is limited, though, by the low solubility
of the bis and crystallization of the gel that can occur during storage [74]. Also,
as some monomers used in polymer gel dosimetry may be strong acids, high
concentrations can adversely affect the gelatin within the gel dosimeter over time.
In order to produce the highest R2-dose sensitivity, the relative proportion of each
of the individual co-monomers, typically symbolized using %C, was found to be
50 % of the total co-monomer content [102,105]. This result was supported by
investigating the effect of chemical exchange on magnetization transfer in
polyacrylamide gels [106].
2.4.5 Ageing of the gel
Unlike the problems encountered by Fricke gels where there is a diffusion of the
ferric ions over time thus degrading the spatial information contained in the gel
dosimeter, polymer gel dosimeters are not as limited by time constraints [107].
One exception is a time evolution of the dose response that occurs following the
irradiation of the gel as polymerisation processes are occurring at the greatest rate
and may lead to errors in the use of separate calibration vials and phantom [51].
De Deene et al., in an investigation into the stability of the polymer gel dosimeter
structure, found that the initial 12 hours post-irradiation yield significant errors
due to the chemical instability of the polymer gel [108]. After around 12 hours,
27
most polymer gel dosimeters maintain a reasonable temporal stability over a
period of several days.
2.5 Evaluation of Polymer Gel Dosimeters
Following irradiation, polymer gel dosimeters must be imaged to produce maps of
the absorbed dose distribution. To evaluate polymer gel dosimeters, several
imaging modalities have been explored and include: MRI, x-ray CT, optical CT
and ultrasound. In order to be effective in evaluating polymer gel dosimeters, an
imaging modality must be capable of sufficiently detecting the radiation-induced
changes that occur within the polymer gel dosimeter. To date, MRI has been the
most commonly investigated imaging modality used to evaluate polymer gel
dosimeters.
2.5.1 Magnetic Resonance Imaging
To date, MRI has been the most common imaging modality used to evaluate gel
dosimeters. The polymer chains within a polymer gel dosimeter, initiated by
irradiation, affect the mobility of the water molecules. Water molecules attached
to polymer chains undergo very slow and restricted motions compared to bulk
water and water molecules in hydrated monomer. MRI detects this difference by
measuring the larger relaxation rates exhibited by water molecules attached to the
polymer chains. The extent of the polymer network within a polymer gel
dosimeter is related to the absorbed dose. As the polymer network is spatially
arranged within an aqueous gel matrix, the dose distribution within the gel matrix
can be imaged using MRI. In order to relate the measured relaxation rates to
absorbed dose, a R2-dose calibration curve is used. This method provides an
absolute dosimetry. In relative dosimetry, a calibration curve is not required
provided the delivered dose is within the linear R2-dose response range of the
dosimeter.
28
The radiation-induced changes that occur within a polymer gel dosimeter have an
effect on the spin-lattice relaxation rate, R1, and the spin-spin relaxation rate, R2.
These changes can be correlated to the absorbed dose within a polymer gel
dosimeter and hence a calibration curve can be calculated. The change in R2 due
to irradiation of polymer gel dosimeters has been found to be more pronounced
than that of R1 and hence evaluations of polymer gel dosimeters are usually based
on R2 [109]. MRI has been widely demonstrated to be capable of imaging simple
and complex dose distributions in polymer gel dosimeters [49,51,74,82,103,110-
112].
Relaxation rates are obtained in MRI by applying a radio frequency (RF) pulse
exciting the magnetization of the spin system within water protons. As the
magnetization of the water protons returns to its equilibrium state, it can be
sampled. A pulse sequence is a set of RF pulses that are applied to a sample to
produce a nuclear magnetic resonance signal, called an echo. A multiple spin-
echo pulse sequence acquires a train of equally spaced spin-echoes. In this case,
the magnetization signal is refocused using 180° pulses. Each echo that results
occur at a specific echo time that can then be used to determine the spin-lattice or
spin-spin relaxation times, T1 and T2 respectively, within the sample. In 2D, a set
of base images can be attained through a specified cross-section of the sample
being imaged. T2 maps can then be calculated using an exponential fit of the time
course decay of each pixel value within consecutive base images.
Kaurin et al. (1999) demonstrated the effectiveness of a polymer gel dosimeter for
verifying conformal radiotherapy treatments in a four-field oblique and lateral
wedged field technique [113]. Likewise, polymer gel dosimeters have been used
to verify treatments involving IMRT and treatments using dynamic multi-leaf
collimators. De Deene et al. (2000) demonstrated the use of a polymer gel
dosimeter for the verification of dose distributions produced by conformal
radiation of a mediastinal tumour located near the oesophagus [111]. Oldham et
al. (1998) showed good results were obtained using polymer gel dosimeters for
the verification of a nine field tomotherapy irradiation [49]. The use of MRI to
evaluate dose distributions within polymer gel dosimeters has show good spatial
29
agreement with doses specified by treatment planning calculations. MRI is
known for its high contrast and ability to generate images in 3D.
MRI is limited, however, by practical and technical complications; in particular,
MRI equipment is expensive and often access is limited. Technical limitations
relate to the accuracy of dose distributions determined via MRI being susceptible
to temperature variations during scanning, as discussion earlier. Further
complications exist due to artifacts inherent to MRI. De Deene et al. have
demonstrated the spatial accuracy of dose distributions can be affected by
geometrical distortions caused by eddy currents, magnetic field inhomogeneities
and magnetic field gradient inhomogeneities [114,115]. The dose response was
also found to be dependent on the scanning orientation and the size of the field of
view [116]. Typically, MRI pulse sequences used in conventional diagnostic
imaging are not optimal for imaging polymer gel dosimeters. Sub-optimal pulse
sequences have been reported to be another source of uncertainty in dose
distributions produced by MRI equipment [117,118]. These studies have
demonstrated that it is possible, however, to optimize pulse sequences for
producing dose maps from imaging gel dosimeters [118,119]. These
complications make the routine use of MRI and polymer gel dosimetry in clinics
difficult.
2.5.2 X-Ray Computed Tomography
The potential of x-ray computed tomography (CT) to evaluate polymer gel
dosimeters has also been investigated [119-122]. The radiation-induced changes
within a polymer gel dosimeter lead to density changes which, in turn, affect the
x-ray linear attenuation coefficients [120,121]. Investigations have explored the
correlation between Hounsfield Units (HU) from x-ray equipment and absorbed
dose. These investigations have also explored the relationship between the x-ray
linear attenuation coefficient and absorbed dose given changes in the composition
of polymer gel dosimeters.
30
Audet et al. (2002) have more recently investigated the use of x-ray CT to
evaluate complex dose distributions in a polymer gel dosimeter. In their study, a
dose delivery based on a stereotactic treatment plan was used to expose a polymer
gel dosimeter. The dosimeter was later imaged using a clinical x-ray CT scanner
[122]. The results showed that a polymer gel dosimeter could be used to delineate
irradiated regions within a volume of unirradiated gel. There was good agreement
found between the measured dose distribution and the treatment plan, with spatial
agreement for the 50 % and 80 % isodose curves to within 3 mm. One key
advantage of using x-ray CT to evaluate polymer gel dosimeters is its relative
temperature insensitivity. Unlike the highly-temperature sensitivity of MRI, a
temperature change of 4 °C to 23 °C within a polymer gel dosimeter evaluated
using x-ray CT results in a change in dose sensitivity of less than 10 % [119].
Other advantages are economy and accessibility, especially compared to MRI and
its high spatial accuracy capable of sub-millimeter spatial resolutions.
More recently, there have been studies evaluating normoxic polymer gel
dosimeters using x-ray CT [96,123-125]. Brindha et al. (2004) have investigated
the linear attenuation coefficients for normoxic polymer gel dosimeters. In
particular, they have investigated normoxic polymer gel dosimeters made from
polyacrylamide or methacrylic acid and tetrakis (hydroxymethyl) phosphonium
chloride as an anti-oxidant, respectively PAGAT and MAGAT gels [96]. They
found that the CT-dose response was linear up to 15 Gy for the PAGAT gel and
up to 10 Gy for the MAGAT gel and concluded the two normoxic polymer gel
dosimeters were suitable for use in radiotherapy dosimetry. Hill et al. have
investigated the dose response of a MAGIC gel dosimeter using x-ray CT using
doses up to 150 Gy [123,124]. They then applied the methodology to the noval
measurement of computer tomography dose indices on diagnostic x-ray CT
scanners [125]. They concluded that normoxic polymer gel dosimetry was a valid
technique for use in x-ray CT quality assurance and hence showed promise to be
incorporated into clinical practice.
There are, however, a couple of limitations to the effective use of x-ray CT for
evaluating polymer gel dosimeters. X-ray CT suffers from a lower dose
sensitivity and a lower contrast-to-noise ratio (CNR) compared to the signal-to-
31
noise ratio (SNR) of MRI [119]. The CNR of an image can be improved by
averaging multiple scans, however, this leads to a dramatic increase in the
scanning time and tube loading. This in turn, increases the cost associated with
imaging gel dosimeters. Also, the linear effective dose range of a polymer gel
dosimeter evaluated using x-ray CT is relatively small compared to MRI, being
linear only up to around 10 Gy.
2.5.3 Optical Computed Tomography
As the optical properties of a polymer gel dosimeter change when irradiated, it has
been possible to detect absorbed dose within the dosimeter using optical methods
and correlating the optical density to absorbed dose. In addition by using a
computed tomography (CT) system, it is possible to reconstruct an image of the
dose distribution within a polymer gel dosimeter [85,86,126,127]. An optical CT
system functions by passing light through the polymer gel dosimeter via differing
angles encircling the dosimeter. The change in opacity within a polymer gel
dosimeter is a result of the radiation induced polymerization. As the polymer
aggregates form in the dosimeter, they become scatter centres causing the optical
attenuation to increase with dose [86]. The dose distribution within a polymer gel
dosimeter can be determined by reconstructing the tomographic images of optical
density and correlating to absorbed dose.
The advantages of an optical CT system to evaluate polymer gel dosimeters
include their low noise and hence a superior SNR compared to MRI images,
simple low cost equipment and capability to produce 3D dose maps with
sufficient spatial resolution, accuracy and precision. Oldham et al. (2001)
demonstrated the use of optical CT to evaluate a polymer gel dosimeter for the
verification of complex radiotherapy treatment plans such as those used in
radiosurgery and IMRT [126]. They found that 3D dose maps could be produced
in under an hour with sub-millimeter spatial resolution, 3 % accuracy and less
than 1 % precision. Xu et al. (2002) have also shown the capability of optical CT
32
to image steep dose gradients, such as those encountered in brachytherapy, within
a polymer gel dosimeter [128].
There are, however, significant technical limitations to the use of optical CT for
polymer gel dosimetry. Loss of signal can occur at container boundaries due to
reflection or refraction of laser light used for scanning. Likewise, with increasing
depth of the phantom, the attenuation of light can dramatically reduce the
available information. This places a significant restriction on the size of phantoms
that can be evaluated using optical CT. Additionally, with increasing dose, the
optical density of the gel becomes so significant that scanning is ineffective,
essentially narrowing the potential dynamic range of absorbed dose that can be
imaged to between 0 and 10 Gy. Also, the total optical density across a phantom
cannot exceed a certain maximum as determined by the dynamic range of the light
detection system and its corresponding SNR [128]. When imaging of polymer gel
dosimeters is restricted to only low doses (< 10 Gy ), the contrast-to-noise ratio is
also reduced, which can diminish the dose accuracy.
2.5.4 Ultrasound
It is well-known that ultrasound can be used to determine the characteristics of
materials [129]. The ultrasonic parameters of most interest include: ultrasonic
attenuation, ultrasonic speed of propagation and ultrasonic impedance [129].
Ultrasound has been well studied in its capability to characterize polymer
structures [129-132] and the changes in polymer structures due to irradiation [133-
135]. With changes in polymer structure due to irradiation being quantifiable via
a number of different ultrasound parameters, it has been possible to correlate
parameters to dose and hence build an effective dosimeter.
Maryanski et al. (1999) were the first to propose the use of ultrasound to evaluate
polymer gel dosimeters [136]. Extensive research into the use of ultrasound to
evaluate polymer gel dosimeters was conducted by Mather et al. [137] who found
that several ultrasound characteristics including: speed of propagation, attenuation
33
and transmitted signal intensity all display a strong variation with absorbed dose
that continues beyond doses of 15 Gy. Ultrasound was applied successfully to
both polyacrylamide polymer gel dosimeters (PAG) and normoxic polymer gel
dosimeters. However, correlated to ultrasonic speed, the dose response in PAG
and normoxic polymer gel dosimeters was found to be fundamentally different
due to differences in the dependence of the gel’s elastic properties on dose at
ultrasonic frequencies. It is from this investigation that Mather et al. were able to
construct a prototype ultrasound system capable of imaging absorbed dose
distributions in polymer gel dosimeters.
Currently, the use of ultrasound to evaluate polymer gel dosimeters is in its
infancy and further work is required to validate reproducibility and to determine
the sources of inconsistencies found in the absolute values of ultrasonic speed and
attenuation in polymer gel dosimeters [137].
2.5.5 Vibrational Spectroscopy
Fourier Transform (FT) Raman spectroscopy can be a useful tool for investigating
radiation induced changes in polymer gel dosimeters [138-142]. Typical
monomers, such as acrylamide and bis-acrylamide, have been identified through
vibrational bands. Changes to these vibrational bands can be correlated to
absorbed dose. Rintoul et al. used Raman microscopy to investigate dose
distributions with spatial resolutions approaching 1 μm [142]. Gustavsson et al.
have developed and optimized a polymer gel dosimeter based on 2-hydroxyethyl
acrylate (HEA) using FT-Raman spectroscopy [143]. From the FT-Raman
spectroscopy results, Gustavsson et al. were able to examine the chemical
structure and properties of the HEA gel and study how these effects contribute to
the relaxation process that is imaged using MRI.
The use of vibration spectroscopy to evaluate 3D dose in polymer gel dosimeters
is also in its early stages. There is still significant work that must be done for
vibrational spectroscopy to become an effective evaluation tool for dosimetry.
34
Although it has the potential for providing dose distributions with very high
spatial resolutions, it may be limited by the achievable penetration depth of light
into the dosimeter.
2.6 High-Resolution MRI in Polymer Gel Dosimetry MRI has been the gold standard of the many techniques that have been used to
evaluate polymer gel dosimeters post-irradiation. Some applications, such as
brachytherapy, require high resolution dose maps that are beyond the capabilities
of many clinical MRI scanners. To achieve high resolution in MRI, gradient
strengths must be increased, which, in turn, have a significant effect on the
relaxation rates of samples being imaging leading to unwanted artifacts. The
gradient strengths applied to a sample are dictated by the specific pulse sequence
used.
In MRI there are many different approaches in scanning a gel dosimeter, including
measuring the spin-lattice relaxation time, T1, the spin-spin relaxation time, T2,
and the apparent diffusivity, Dapp. T2 is predominantly measured in polymer gel
dosimetry. In measuring T2, however, the pulse sequence, and in particular the
echo spacing can be varied to optimize the imaging for the particular gel. Over
the lifetime of the application of polymer and Fricke gels to radiation therapy
dosimetry, many different pulse sequences have been investigated. Gore et al.,
the first to propose the use of MRI to image gel dosimeters, used an inversion
recovery pulse sequence to measure T1 and a Carr Purcell Meiboom Gill (CPMG)
pulse sequence to measure T2 [53]. Figure 2 shows a typical CMPG sequence
that images samples in three dimensions. As changes in T2 dominate those in T1
for polymer gel dosimeters, CPMG was selected by Maryanski et al. to image a
polymer gel dosimeter [74]. With a long recovery time (6000 ms) to allow the
longitudinal signal to completely recover, Maryanski et al. varied the echo time in
order to assess the most optimal echo spacing for the particular gel being scanned.
35
Figure 2. A typical CPMG pulse sequence where GRO, GSS and GPE are the
pulse amplitudesin each of the three dimensios, NT is the pulse duration, TE the
inter-echo time and n the number of echoes acquired.
Using a polyacrylamide polymer gel, Maryanski et al. used a series of Hahn spin
echo pulse sequences [82]. The Hahn spin echo pulse sequence is a single echo
sequence that is usually executed many times using different τ to obtain data to
determine the T2 of a sample. A multi-echo Carr-Purcell pulse sequence, where a
[τ-180°-τ-acquire] sequence is performed repeatedly following a 90° pulse to
obtain several images, was discussed as a seemingly more appropriate option,
however, it had previously been established that this pulse sequence gave
significant errors in T2 caused by the imperfect refocusing of 180° pulses. A
Hahn spin echo pulse sequence avoids the RF inhomogeneities caused by standing
wave effects and attenuation that leads to errors in T2 measurements [50,144].
However, the diffusion of spins can occur between the application of the pulses
and the echo formation thus leading to a loss in the signal, especially for longer
echo times [145]. The optimization of MRI for the purpose of gel dosimetry has
become one of the focuses of research into the improvement and implementation
of MRI gel dosimetry in the clinical setting [103,117,145].
A circularly polarized head coil is often used as both transmitter and receiver to
provide a more adequate signal-to-noise ratio (SNR) [146]. Ideally the main body
36
coil would be used for excitation (as the magnetic field is more uniform) and the
head coil used for detection. However, many clinical scanners will not allow this
approach [146]. The use of a circularly polarized head coil and a standard
multiple spin echo sequence, based on the CPMG pulse sequence, seems to be
favoured by many polymer dosimetry groups [49,88,90,147,148]. A multiple
spin-echo with phase alternating phase shift (PHAPS) pulse sequence has been
used by De Deene et al. [103,108,109,114,115]. This pulse sequence involves the
sampling of k-space (the raw data matrix) twice, once with a Carr-Purcell (CP)
encoding and once with a Carr-Purcell-Meiboom-Gill (CPMG) encoding [111].
The PHAPS scheme compensates for ghosting and mirror artifacts due to
stimulated echoes [103]. The application of a gradient train preceding each pulse
cycle of the sequence will reduce deviations in the measured T2 and consequently
the calculated absorbed dose [114]. The effect of this gradient train is to bring the
eddy current field offset into a steady state. Variations in the eddy current offset
are the cause of many errors in the measurement of T2 and artifacts in the
corresponding images.
It is well-known that the Brownian motion of water molecules in the presence of a
strong magnetic field gradient will have an attenuating effect on an MR signal
[149,150]. When applying a multiple spin-echo sequence to a sample, the echoes
are attenuated differently and hence a change in the measured R2 of the sample
results. The diffusion weighting factor, commonly referred to as the b-factor, is
used to characterize the effects of self-diffusion of water molecules on the MR
signal [151-153]. Polymer gel dosimetry usually involves very small echo times
of the order 10 to 30 ms [117,118] and, to achieve high-resolutions, very high
gradients strengths. These conditions can introduce significant artifacts into the
use of high-resolution MRI. For high-resolution MRI to be effective in polymer
gel dosimetry, an investigation is required into the effects of molecular self-
diffusion within polymer gel dosimeters due to the application of high gradient
strengths and the use of small echo times in the pulse sequence. Once validated,
high-resolution evaluation of polymer gel dosimeters will provide a valuable
dosimetry tool capable of assessing applications such as intravascular
brachytherapy that typically require resolutions less than 100 μm.
37
Ertl et al. (2000) used high-resolution MRI to evaluate a polyacrylamide polymer
gel dosimeter used in stereotactic radiation therapy [154]. In this study, Ertl et al.
irradiated a head phantom filled with a polyacrylamide polymer gel dosimeter
using a Leksell Gamma Knife. The phantom was then scanned using a 3 T whole-
body MR-tomograph to achieve an in-plane resolution of 234 μm and slice
thickness of 1 mm. Their results showed good agreement between film dosimetry
and calculated data. Berg et al. (2001) attempted high-resolution imaging on a 3
T whole-body scanner with a methacrylic polymer gel dosimeter [155]. In this
study, Berg et al. used T1, T2 and diffusivity to measure the dose response and
found that T2 produced the highest sensitivity. Berg et al. achieved an in-plane
resolution of less than 117 μm with a slice thickness of 0.8 to 1 mm [155]. Their
results from gel measurements were in good agreement with those predicted by
the treatment planning system at high dose levels.
2.7 Brachytherapy Applications of Gel Dosimetry The use of gel dosimeters has been applied to many areas of radiation therapy
[112,154,156-158]. However, its application to brachytherapy has been of
particular interest due to the difficulties faced by other dosimetric methods when
attempting to produce maps of the absorbed dose distribution surrounding a
radioactive source. There are two predominant issues that make dosimetry
difficult in brachytherapy: the high spatial resolution required due to the typical
dimensions of target volumes and the high dose gradients that exist in close
proximity to the source. These issues are more significant in intravascular
brachytherapy where radiation doses are delivered to artery walls only millimeters
away. In conventional brachytherapy tumours of the order of centimeters are
treated, while in intravascular brachytherapy arteries with diameters on the order
of millimeters are treated [159].
The use of gel dosimetry in brachytherapy has been investigated by a number of
authors [147,160-171]. In 1994, Schreiner et al. investigated the use of Fricke-
gelatin dosimeters using NMR to evaluate HDR brachytherapy dose distributions
[160]. They used a microSelectron HDR remote afterloader with a 192Ir source to
expose Fricke gel dosimeters to a variety of irradiation patterns. Results from the
38
Fricke-gelatin dosimeters showed good agreement with dose points predicted by
the treatment planning system.
The use of ferrous sulphate gel dosimeters in intracavitary brachytherapy was
investigated by Knutsen, et. al. in 1997 [163]. This investigation involved the use
of an 192Ir source in a remote afterloading device, however, the investigation was
limited by the diffusion of the ferric ions in the gel following irradiation. It was
shown, however, that the isodose lines produced by both computerized treatment
planning software and the measured experimental results were within 2 mm of
each other. The results of the two methods, computerized treatment planning and
Fricke gel dosimetry, were superimposed on each other to provide a means of
comparison.
Hafeli et al. (2000) measured the dose distribution of a 188W / 188Re beta line
source in an endovascular brachytherapy application using a number of different
methods, one of which was a polymer gel dosimeter [162]. Although there was
approximate agreement between the methods used (TLD, Gafchromic film,
polymer gel dosimeter and computer simulation), the polymer gel dosimeter did
not show the same trends as the other methods. However, it was noted that the
polymer gel dosimeter provided 3D data with good resolution.
Brachytherapy may be classed as either low dose rate (LDR) using sources that
deliver around 30 to 90 cGy.h-1 or high dose rate (HDR) where the sources deliver
around 60 to 300 cGy.h-1 [24,159]. The use of a LDR 137Cs source was explored
by Farajollahi et al. (1999) using a polymer gel dosimeter [161]. It was concluded
that polymer gel dosimeters were a suitable dosimeter for verifying LDR
brachytherapy and that good spatial resolution could be achieved. In this study,
the experiment simulated treatment of the intrauterine tube and ovoids, but the
potential of the gel to study more complex geometries was mentioned. The
susceptibility of the gel to oxygen contamination was seen as a main drawback for
the use of polymer gel dosimetry in clinical practice.
High dose rate brachytherapy was explored by McJury et al. (1999) using an 192Ir
source [147]. Again, the results indicated that gel dosimetry is appropriate for
39
measuring the dose distribution surrounding a brachytherapy source. Cross-
section planes were used to examine the distributions around the radioactive
source and good agreement existed between the gel dosimetric measurements and
computer treatment plans. One of the main concerns with regards to the
experiment performed was the possible heating of the gel over time during
imaging. The RF fields used inside MRI equipment deposit energy in the object
being imaged. Changes in the temperature of an object during the process of
imaging lead to changes in the measured T2 and hence lead to inaccuracies in the
measured dose. Papagiannis et al. (2001) investigated the dose distributions close
to an 192Ir source using a N-vinylpyrrolidone-based polymer gel dosimeter and
found good agreement with Monte Carlo dose calculations down to within 3 mm
of the source [166]. In their study, Papagiannis et al. found that the volume
averaging effects close to the source were significant and suggested that a
reduction in slice thickness should improve on the accuracy of the measured dose
distributions.
De Deene et al. (2001) conducted a comprehensive study of the accuracy of
polymer gel dosimetry on a phantom irradiated using brachytherapy [168]. In this
study, De Deene et al. examined the problems associated with oxygen permeation
into polymer gel dosimeters and the subsequent effects on the measured dose. It
was found that oxygen causes a dose threshold in the dose-R2 curve that was
linearly correlated with oxygen concentration. Oxygen was found to permeate
through the catheter and hence De Deene et al. concluded that the inhibitory
effects of oxygen could only be prevented by keeping the phantom in a nitrogen
environment during manufacture and irradiation [168]. A second effect was
found to be the diffusion of monomers post-irradiation in regions where high dose
gradients exist. This diffusion leads to dose overshoots in regions of high dose or
high dose gradients [169]. Susceptibility effects were found to deform the dose
map at locations very close to the catheter. The extent of the deformation was
related to catheter size, the position of the catheter with respects to the direction of
the main field of the MRI scanner and the receiver bandwidth of the imaging
sequence [168]. In using polymer gel dosimetry, it was also found that
discrepancies between where the centre of the source was assumed to be and
where it actually was could lead to significant errors in the measured dose
40
distribution. In addition, partial volume effects were found to be significant in
regions immediately adjacent the central catheter. The combination of these two
effects led to extreme variations in the measured dose distribution.
Amin et al. (2003) compared the results of polyacrylamide polymer gel
dosimeters with radiochromic film to investigate an intravascular brachytherapy 90Sr/90Y source [170]. They found good agreement between the gel dosimeter and
film with a resolution of 0.4 mm/pixel. However, when the resolution was
increased to 0.2 mm/pixel, Amin et al. found significant variations in the
measured dose of the gel dosimeter. It is possible that partial volume effects over
steep dose gradients may have led to these variations. They also found that
oxygen permeation into the gel caused considerable problems and hence made the
technique of gel dosimetry using polyacrylamide polymer gel dosimeters
unsuitable for implementation into clinical practice. It was concluded that
although polymer gel dosimetry did produce good dose distributions,
radiochromic film was preferred in clinical practice due to ease of use.
2.8 Sources of Uncertainty in Polymer Gel Dosimeters
There are potentially many sources of error in each of the steps involved in gel
dosimetry: the manufacture of the gel, the irradiation, the storage of the gel, the
scanning, the post-processing of the images and the calibration of the images.
Each stage in this process introduces stochastic and deterministic uncertainties
that can affect the results. The application of gel dosimetry to a clinical setting
requires the uncertainty in the overall procedure be well known and accounted for
[172,173].
Oldham et al. (1998) discussed a method of improving the calibration accuracy in
gel dosimetry by using depth dose data to obtain a number of average T2 values
for each dose level [174]. They showed that when a large number of calibration
points are obtained, the uncertainty in the slope of the linear R2 response, α, can
be lowered by a factor of about 4 and the uncertainty in the y-intercept, R0, can be
41
lowered by a factor of 10. Baldock et al. (1999) discussed the implications of this
measure of the uncertainty [172]. They pointed out that the r2 fitting parameter is
not a good method of indicating the relative quality of the calibrations. It was
shown that the most significant reductions in the overall uncertainty would be
achieved by reducing the noise in the R2 map.
An investigation into the noise in MRI polymer gel dosimetry was also conducted
by Low et al. (2000) [175]. They concluded that there was little difference in dose
maps produced from small and large vessels of gel imaged using MRI, however, it
was noted that this was inconsistent with some previous research [176]. Volume
averaging reduced dose errors at higher doses, but the systematic errors due to the
MRI scanning artifacts limited the reduction of the overall error.
The concept of dose resolution, pDΔ , was introduced by Baldock et al. (2001)
[117]. It is defined as the minimal separation between two absorbed doses so that
they may be distinguished with a given level of confidence, p. The selection of an
appropriate echo spacing for a given range of doses was determined by the echo
spacing that maintained the lowest dose resolution. This method of evaluating the
quality of gel dosimeters represented an improvement of the conventional dose
sensitivity, as it incorporates the uncertainty of the imaging and the calibration
curve into the measure. Figure 3 shows the 95 % dose resolution for a
polyacrylamide gelatin gel dosimeter.
0 1 2 3 4 5 6 7 8 9 10
0.2
0.4
0.6
0.8 20 ms 25 ms 30 ms
DΔ95
% (G
y)
Absorbed Dose (Gy)
42
Figure 3. The 95% Dose Resolution of a polyacrylamide gelatin gel measured
using a CPMG pulse sequence with inter-echo times of 20, 25 and 30
milliseconds.
Lepage et al. (2001) have investigated the variations in T2 obtained using MRI
due to non-uniformity in the main magnetic field [146]. These effects have also
been studied in ferrous sulphate gel dosimeters and have been found to give
significant uncertainties in T2 [177]. The sources of non-uniformity of the
measured T2 can occur due to imperfect main, or B0, magnetic fields, gradient
eddy currents, the inhomogeneities in the RF signal and the receiver filters.
Lepage et al. proposed that a correction matrix be constructed that would measure
the amount of non-uniformity in a particular MRI scanner for a given echo
spacing by making multiple scans of a number of uniform phantoms with varying
T2 values [146]. Using a linear interpolation, a large matrix could be constructed
of the measured differences between different regions within the field of view and
future images corrected using the matrix. It was found that within a region of
around 78 cm2 in the center of the coronal plane, there was good consistency in
the measured T2 indicating good uniformity in this the central region of this MRI
scanner.
Watanabe et al. (2005) investigated the use of linear and non-linear equations in
the process of calibrating dose distributions from polymer gel dosimeters and
radiographic film [178]. They found that linear equations could be used without
knowing the actual calibration equation derived from the R2-dose response curve.
It was found that using a transformation method, relative dosimetry contained less
uncertainty than the dose distribution determined by gel dosimetry or film,
provided the dosimetry method was at least quasi-linear or could undergo a
variable transformation using logarithmic functions to exhibit a quasi-linearity.
2.9 Conclusion
Normoxic polymer gel dosimetry shows great promise as an effective dosimeter
specifically in its potential to verify irradiation treatment plans. This study
43
provides further exploration of the physical and chemical properties of various
formulations of normoxic polymer gel dosimeters and seeks to apply the
technology to applications requiring high resolution, such as intravascular
brachytherapy.
44
References
[1] Khan, F.M., The physics of radiation therapy. 2nd ed. 1994, Baltimore:
Williams & Wilkins, pp 542-560.
[2] Johns, H.E. and Cunningham, J.R., The Physics of Radiology. 1983,
Springfield: Charles C Thomas.
[3] Webb, S., The Physics of Conformal Radiotherapy. 1997, Bristol: Institute
of Physics Publishing.
[4] Powlis, W.D., Smith, A.R., and Cheng, E., Initiation of multileaf
collimator conformal radiation therapy. Int J Radiat Oncol Biol Phys,
1991. 25: p. 171-179.
[5] Schreiner, L.J., Gel Dosimetry: motivation and historical foundations, in
DosGel'99 Proceedings of the 1st International Workshop on Radiation
Therapy Gel Dosimetry. 1999, Canadian Organization of Medical
Physicists: Lexington, Kentucky.
[6] Nutting, C., Dearnaley, D.F., and Webb, S., Intensity modulated radiation
therapy: a clinical review. Br J Radiol, 2000. 73: p. 459-469.
[7] Kron, T., Radiation therapy requirements: what do we expect from gel
dosimetry, in Proceedings of DosGel 2001: The second international
conference on radiotherapy gel dosimetry, Baldock, C. and De Deene, Y.,
Editors. 2001, Queensland University of Technology: Brisbane. p. 2-9.
[8] Karaiskos, P., Angelopoulos, A., Pantelis, E., Papagiannis, P., Sakelliou,
L., Kouwenhoven, E., and Baltas, D., Monte Carlo dosimetry of a new
192Ir pulsed dose rate brachytherapy source. Med Phys, 2003. 30: p. 9-16.
[9] Papagiannis, P., Angelopoulos, A., Pantelis, E., Sakelliou, L., Karaiskos,
P., and Shimizu, Y., Monte Carlo dosimetry of 60Co HDR brachytherapy
sources. Med Phys, 2003. 30: p. 712-721.
[10] Perez-Calatayud, J., Ballester, F., Serrano-Andres, M.A., Lluch, J.L.,
Puchades, V., Limami, Y., and Casal, E., Dosimetric characteristics of the
45
CDC-type miniature cylindrical 137Cs brachytherapy sources. Med Phys,
2002. 29: p. 538-543.
[11] Ballester, F., Lluch, J.L., Serrano-Andres, M.A., Casal, E., Puchades, V.,
and Limami, Y., Monte Carlo calculations of dose rate distributions
around the Walstam CDC.K-type 137Cs sources. Phys Med Biol, 2001.
46: p. 2029-2040.
[12] Williamson, J.F., Monte Carlo modeling of the transverse-axis dose of the
Model 200 103Pd interstitial brachytherapy source. Med Phys, 2000. 27:
p. 643-654.
[13] Williamson, J.F., Monte Carlo-based dose-rate tables for the Amersham
CDCS.J and 3M model 6500 137Cs tubes. Int J Radiat Oncol Biol Phys,
1998. 41: p. 959-970.
[14] Williamson, J.F. and Li, Z., Monte Carlo aided dosimetry of the
microselectron pulsed and high dose-rate 192Ir sources. Med Phys, 1995.
22(6): p. 809-19.
[15] Williamson, J.F., Comparison of measured and calculated dose rates in
water near I-125 and Ir-192 seeds. Med Phys, 1991. 18: p. 776-786.
[16] Valicenti, R.K., Kirov, A.S., Meigooni, A.S., Mishra, V., Das, R.K., and
Williamson, J.F., Experimental validation of Monte Carlo dose
calculations about a high-intensity Ir-192 source for pulsed dose-rate
brachytherapy. Med Phys, 1995. 22(6): p. 821-9.
[17] Skowronek, J. and Piotrowski, T., Pulsed dose rate brachytherapy: a
method description and review of clinical application. Przegl Lek, 2002.
59(1): p. 31-36.
[18] Balter, S., Chan, R.C., and Shope, T.B., Introduction and Overview:
Intravascular Brachytherapy - Fluoroscopically Guided Interventions, in
Intravascular Brachytherapy, Fluoroscopically Guided Interventions,
Balter, S., Chan, R.C., and Shope, T.B., Editors. 2002, American
Association of Physicists in Medicine: Montreal. p. 1-16.
[19] Teirstein, P.S., Massullo, V., Jani, S., Popma, J.J., Mintz, G.S., Russo,
R.J., Schatz, R.A., Guarneri, E.M., Steuterman, S., Morris, N.B., Leon,
M.B., and Tripuraneni, P., Catheter-based radiotherapy to inhibit
restenosis after coronary stenting. N Engl J Med, 1997. 336: p. 1697-1703.
46
[20] Waksman, R. and Schwartz, R.S., Vascular Brachytherapy for Prevention
of Restenosis: A Brief History and Overview. J Invasive Cardiol, 1999.
11(1): p. 33-35.
[21] Nath, R., Anderson, L.L., Luxton, G., Weaver, K.A., Williamson, J.F., and
Meigooni, A.S., Dosimetry of interstitial brachytherapy sources:
recommendations of the AAPM Radiation Therapy Committee Task
Group No. 43. American Association of Physicists in Medicine. Med
Phys, 1995. 22(2): p. 209-34.
[22] Nath, R., Amols, H., Coffey, C., Duggan, D., Jani, S., Li, Z., Schell, M.,
Soares, C., Whiting, J., Cole, P.E., Crocker, I., and Schwartz, R.,
Intravascular brachytherapy physics: report of the AAPM Radiation
Therapy Committee Task Group no. 60. American Association of
Physicists in Medicine. Med Phys, 1999. 26(2): p. 119-52.
[23] Chakri, A. and Thomadsen, B., A quality management program in
intravascular brachytherapy. Med Phys, 2002. 29(12): p. 2850-2860.
[24] Thwaites, D.I. and Williams, J.R., Radiotherapy physics in practice. 2nd
ed. 2000, Oxford: Oxford University Press,. xxii, 332 , 2 of plates.
[25] Attix, F.H., Roesch, W.C., and Tochilin, E., Radiation Dosimetry. 2nd ed.
1966, New York: Academic Press.
[26] Leavitt, D.D. and Starkschall, G. Proceedings of the XIIth International
Conference on the Use of Computers in Radiation Therapy. 1998.
Madison WI: Medical Physics Publ.
[27] Back, S.A., Magnusson, P., Olsson, L.E., Montelius, A., Fransson, A., and
Mattsson, S., Verification of single beam treatment planning using a
ferrous dosimeter gel and MRI (FeMRI). Acta Oncol, 1998. 37(6): p. 561-
6.
[28] (ICRU), I.C.o.R.U.a.M., Determination of Absorbed Dose in a Patient
Irradiated by Beams of X or Gamma Rays in Radiotherapy Procedures.
Technical Report 24. 1979: ICRU Publishing.
[29] Brahme, A., Dosimetric precision requirements in radiation therapy. Acta
Radiol Oncol, 1984. 23: p. 379-391.
[30] Van Dyk, J., The Modern Technology of Radiation Oncology. 1999,
Wisconsin: Medical Physics Publishing.
47
[31] (IAEA), I.A.E.A., Absorbed dose determination in photon and electron
beams, in An international code of practice. 1987, IAEA Technical
Reports Series No. 277: Vienna, Austria.
[32] Svensson, H. and Brahme, A., Recent advances in electron and photon
dosimetry, in Radiation Dosimetry: Physical and Biological Aspects.
1986, Plenum Press: New York.
[33] Rikner, G. and Grusell, E., General specifications for silicon
semiconductors for use in radiation dosimetry. Phys Med Biol, 1987. 32:
p. 870-873.
[34] Rikner, G. and Grusell, E., Patient dose measurements in photon fields by
means of silicon semiconductor detectors. Med Phys, 1987. 17: p. 870-
873.
[35] Nilsson, B., Ruden, B., and Sorcini, B., Characteristics of silicon diodes as
patient dosimeters in external radiation therapy. Radiother Oncol, 1988.
11: p. 279-288.
[36] Li, C., Lamel, L.S., and Tom, D., A patient dose verification program
using diode detectors. Med Dosim, 1995. 20: p. 209-214
[37] Horowitz, Y.S., ed. Thermoluminescence and Thermoluminescent
Dosimetry. Vol I - III. 1984, CRC Press: Florida.
[38] Kron, T., Thermoluminescence dosimetry and its applications in medicine
- Part 1: Physics, materials and equipment. Australas Phys Eng Sci Med,
1994. 17: p. 175-199.
[39] Kron, T., Thermoluminescence dosimetry and its applications in medicine
- Part 2: History and applications. Australas Phys Eng Sci Med, 1995. 18:
p. 1-25.
[40] Kron, T., Clinical thermoluminescence dosimetry: how do expectations
and results compare? Radiother Oncol, 1993. 26: p. 151-161.
[41] Niroomand-Rad, A., Blackwell C.R., Coursey, B.M., Gall, K.P., Galvin,
J.M, McLaughlin, W.L., Meigooni, A.S., Nath, R., Rodgers, J.E., Soares,
C.G.,Radiochromic Film Dosimetry: Recommendations of the AAPM
Radiation Therapy Committee Task Group No. 55. American Association
of Physicists in Medicine. Med Phys, 1998. 25(11): pp. 2093-2115.
[42] McLaughlin, W., Soares, C., Sayeg, J., McCullough, E., Kline, R., Wu, A.,
and Maitz, A., The use of a radiochromic detector for the determination of
48
stereotactic radiosurgery dose characteristics. Med Phys, 1994. 21(3): p.
379-388.
[43] Dini, S.A., Koona, R.A., Ashburn, J.R., and Meigooni, A.S., Dosimetric
evaluation of GAFCHROMIC XR tyhpe T and XR type R films. J. Appl
Clin Med Phys, 2005. 6(1): p. 114-134.
[44] Back, S.A.J., Implementation of MRI Gel Dosimetry in Radiation
Therapy, PhD Thesis, Department of Radiation Physics. 1998, Lund
University: Malmo, Sweden.
[45] Fricke, H. and Morse, S., The chemical action of roentgen rays on dilute
sulfate solutions as a measure of radiation dose. Am J Roentgenol Radium
Therapy Nucl Med, 1927. 18: p. 430-432.
[46] Chapiro, A., Radiation chemistry of polymeric system. High polymers ; v.
15. 1962, New York: Interscience,. xvii, 712.
[47] Day, M.J. and Stein, G., Chemical effects of ionising radiation in some
gels. Nature, 1950. 166: p. 146-147.
[48] Day, M.J., Andrews, H.L., Murphy, R.E., and Lebrun, E.J., Gel dosimeter
for depth dose measurements. Review of Scientific Instruments, 1957. 28:
p. 329-332.
[49] Oldham, M., I. Baustert, C. Lord, T.A. Smith, M. McJury, A.P.
Warrington, M.O. Leach, and S. Webb, An investigation into the
dosimetry of a nine-field tomotherapy irradiation using BANG-gel
dosimetry. Phys Med Biol, 1998. 43(5): p. 1113-32.
[50] Maryanski, M.J., G.S. Ibbott, P. Eastman, R.J. Schulz, and J.C. Gore,
Radiation therapy dosimetry using magnetic resonance imaging of
polymer gels. Med Phys, 1996. 23(5): p. 699-705.
[51] McJury, M., M. Oldham, V.P. Cosgrove, P.S. Murphy, S. Doran, M.O.
Leach, and S. Webb, Radiation dosimetry using polymer gels: methods
and applications. Br J Radiol, 2000. 73(873): p. 919-29.
[52] Keall, P. and C. Baldock, A theoretical study of the radiological properties
and water equivalence of Fricke and polymer gels used for radiation
dosimetry. Australas Phys Eng Sci Med, 1999. 22(3): p. 85-91.
[53] Gore, J.C., Kang, Y.S., and Schulz, R.J., Measurement of radiation dose
distributions by nuclear magnetic resonance (NMR) imaging. Phys Med
Biol, 1984. 29(10): p. 1189-97.
49
[54] Back, S.A., J. Medin, P. Magnusson, P. Olsson, E. Grusell, and L.E.
Olsson, Ferrous sulphate gel dosimetry and MRI for proton beam dose
measurements. Phys Med Biol, 1999a. 44(8): p. 1983-96.
[55] Reichl, B., D. Matthaei, J. Richter, and A. Haase, An automated fast MR-
imaging method for localized measurements of dose distributions using
NMR-Fricke gel dosimetry. Evaluation of influences on the measurement
accuracy. Strahlenther Onkol, 1996. 172(6): p. 312-9.
[56] Chu, W.C., W.Y. Guo, M.C. Wu, W.Y. Chung, and D.H. Pan, The
radiation induced magnetic resonance image intensity change provides a
more efficient three-dimensional dose measurement in MRI- Fricke-
agarose gel dosimetry. Med Phys, 1998. 25(12): p. 2326-32.
[57] Healy, B., Brindha, S., Zahmatkesh, M., Baldock, C., Characterisation of
the ferrous-xylenol orange-gelatin (FXG) gel dosimeter, in Third
International Conference on Radiotherapy Gel Dosimetry, De Deene, Y.
and Baldock, C., Editors. 2004, J Phys Conf Ser 3: Ghent. p. 142-145.
[58] Tarte, B.J., Jardine, P.A., and van Doorn, T., Laser scanned agarose gel
sections for radiation field mapping. Int J Radiat Oncol Biol Phys, 1996.
36: p. 175-179.
[59] Appleby, A. and Leghrouz, A., Imaging of radiation dose by visible colour
development in ferrous agarose xylenol orange gels. Med Phys, 1991. 18:
p. 309-312.
[60] Kelly, R.U., Jordan, K.J., and Battista, I., Optical CT reconstruction of 3D
dose distributions using the ferrous benzoic-xylenol (FBX) gel dosimeter.
Med Phys, 1998. 25: p. 1741-1750.
[61] Olsson, L.E., Petersson, S., Ahlgren, L., and Mattsson, S., Ferrous
sulphate gels for determination of absorbed dose distributions using MRI
techniques: Basic studies. Phys Med Biol, 1989. 34: p. 43-52.
[62] Olsson, L.E., Fransson, A., Ericsson, A., and Mattsson, S., MR imaging of
absorbed dose distributions from radiotherapy using ferrous sulphate gels.
Phys Med Biol, 1990. 35: p. 1623-1631.
[63] Olsson, L.E., Radiation Dosimetry using Magnetic Resonance Imaging, in
Department of Radiation Physics. 1990, Lund University: Malmo,
Sweden.
50
[64] Schulz, R.J., deGuzman, A.F., Nguyen, D.B., and Gore, J.C., Dose-
response curves for Fricke-infused agarose gels as obtained by nuclear
magnetic resonance. Phys Med Biol, 1990. 35(12): p. 1611-22.
[65] Hazle, J.D., Hefner, L., Nyerick, C.E., et al., Dose-response characteristics
of a ferrous-sulphate-doped gelatin system for determining radiation
absorbed dose distributions by magnetic resonance imaging (FeMRI).
Phys Med Biol, 1991. 36: p. 1117-1125.
[66] Schreiner, L.J., Parker, W., Evans, M.D.C., et al., Brachytherapy dose
verification using MRI and Fricke gelatin: a quality assurance technique.
Proc. Of the Conf. Brachytherapy: Today and the Future, 1994, Scotsdale
AZ, 127-133.
[67] Duzenli, C., Sloboda, R., Robinson, D., Material properties of gelatin
affecting spin-spin relaxation rate in the ferrous sulphate gel NMR
dosimeter. Phys Med Biol, 1994. 39: p. 1577-1592.
[68] Chu, K.C., Jordan, K.J., Battista, J.J., Van Dyk, J., and Rutt, B.K.,
Polyvinyl alcohol-Fricke hydrogel and cryogel: two new gel dosimetry
systems with low Fe3+ diffusion. Phys Med Biol, 2000. 45(4): p. 955-69.
[69] Baldock, C., Harris, P.J., Piercy, A.R., and Healy, B., Experimental
determination of the diffusion coefficient in two- dimensions in ferrous
sulphate gels using the finite element method. Australas Phys Eng Sci
Med, 2001. 24(1): p. 19-30.
[70] Healy, B.J., Zahmatkesh, M.H., Nitschke, K.N., and Baldock, C., Effect of
saccharide additives on response of ferrous-agarose-xylenol orange
radiotherapy gel dosimeters. Med Phys, 2003. 30(9): p. 2282-2291.
[71] Schreiner, L.J., Gel Dosimetry: motivation and historical foundations, in
DosGel'99 Proceedings of the 1st International Workshop on Radiation
Therapy Gel Dosimetry, Schreiner, L.J., Editor. 1999, Canadian
Organisation of Medical Physicists: Lexington, Kentucky.
[72] Back, S.A.J. and L.E. Olsson. Ferrous gels (FeGel) and MRI for absorbed
dose measurements: 15 years of development. in DosGel'99 Proceeings of
the 1st International Workshop on Radiation Therapy Gel Dosimetry.
1999. Lexington, Kentucky: Canadian Organisation of Medical Physicists.
51
[73] Schreiner, L.J., Review of Fricke gel dosimeters, in Third International
Conference on Radiotherapy Gel Dosimetry, De Deene, Y. and Baldock,
C., Editors. 2004, J Phys Conf Ser 3: Ghent. p. 9-21.
[74] Maryanski, M.J., J.C. Gore, R.P. Kennan, and R.J. Schulz, 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(2): p. 253-8.
[75] Dole, M., The Radiation Chemistry of Macromolecules Vol 1 and 2. 1973,
New York: Academic Press.
[76] Platzer, N., Irradiation of Polymers. 1967, Washington, DC: American
Chemical Society.
[77] Wilson, J.E., Radiation Chemistry of Monomers, Polymers and Plastics.
1974, New York: Marcel Dekker Inc.
[78] Charlesby, A., Radiation Effects in Materials. 1960, Oxford: Pergamon
Press.
[79] Audet, C. The NMR relaxometry of radiotherapy gel dosimeters. in
DosGel'99 Proceedings of the 1st International Workshop on Radiation
Therapy Gel Dosimetry. 1999. Lexington, Kentucky: Canadian
Organization of Medical Physicists.
[80] Lepage, M., A.K. Whittaker, L. Rintoul, S.A. Back, and C. Baldock, The
relationship between radiation-induced chemical processes and transverse
relaxation times in polymer gel dosimeters. Phys Med Biol, 2001. 46(4): p.
1061-74.
[81] Gochberg, D.F., P.M. Fong, and J.C. Gore, Studies of magnetization
transfer and relaxation in irradiated polymer gels--interpretation of MRI-
based dosimetry. Phys Med Biol, 2001. 46(3): p. 799-811.
[82] Maryanski, M.J., R.J. Schulz, G.S. Ibbott, J.X. Gatenby, D. Horton, and
J.C. Gore, Magnetic resonance imaging of radiation dose distributions
using a polymer-gel dosimeter. Phys Med Biol, 1994. 39: p. 1437-1455.
[83] Maryanski, M.J., J.C. Gore, and R.J. Schulz, US Patent: Three-
Dimensional Detection, Dosimetry and Imaging of an Energy Field by
Formation of a Polymer in a Gel, in US Patented Number 5,321,357. 1994:
United States.
52
[84] Maryanski, M.J., Ibbott, G.S., Schulz, R.J., Xie, J., and Gore, J.C.,
Magnetic resonance imaging of radiation dose distributions in tissue-
equivalent polymer-gel dosimeters, in Proc Soc Magn Reson Med. 1994.
p. 204.
[85] Maryanski, M.J., Y.Z. Zastavker, and J.C. Gore, 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(12): p. 2705-17.
[86] Gore, J.C., M. Ranade, M.J. Maryanski, and R.J. Schulz, 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(12): p. 2695-704.
[87] Knisely, J.P.S., L. Liu, M.J. Maryanski, M. Ranade, R.J. Schulz, and J.C.
Gore, Three-dimensional dosimetry for complex stereotactic radiosurgery
using a tomographic optical density scanner and BANG polymer gels.
Radiosurgery, 1997. 2: p. 251-260.
[88] Pappas, E., T. Maris, A. Angelopoulos, M. Paparigopoulou, L. Sakelliou,
P. Sandilos, S. Voyiatzi, and L. Vlachos, A new polymer gel for magnetic
resonance imaging (MRI) radiation dosimetry. Phys Med Biol, 1999.
44(10): p. 2677-84.
[89] Pappas, E., I. Seimenis, A. Angelopoulos, P. Georgolopoulou, M.
Kamariotaki-Paparigopoulou, T. Maris, L. Sakelliou, P. Sandilos, and L.
Vlachos, Narrow stereotactic beam profile measurements using N-
vinylpyrrolidone based polymer gels and magnetic resonance imaging.
Phys Med Biol, 2001. 46(3): p. 783-97.
[90] Murphy, P.S., V.P. Cosgrove, M.O. Leach, and S. Webb, A modified
polymer gel for radiotherapy dosimetry: assessment by MRI and MRS.
Phys Med Biol, 2000a. 45(11): p. 3213-23.
[91] Murphy, P.S., V.P. Cosgrove, A.J. Schwarz, S. Webb, and M.O. Leach,
Proton spectroscopic imaging of polyacrylamide gel dosimeters for
absolute radiation dosimetry. Phys Med Biol, 2000. 45(4): p. 835-45.
[92] Hepworth, S.J., M.O. Leach, and S.J. Doran, Dynamics of polymerization
in polyacrylamide gel (PAG) dosimeters: (II) modeling oxygen diffusion.
Phys Med Biol, 1999. 44(8): p. 1875-84.
53
[93] Fong, P.M., Keil, D.C., Does, M.D., and Gore, J.C., Polymer gels for
magnetic resonance imaging of radiation dose distributions at normal
room atmosphere. Phys Med Biol, 2001. 46(12): p. 3105-13.
[94] Venning, A.J., Nitschke, K., Keall, P.J., and Baldock, C., Radiological
properties of normoxic polymer gel dosimeters. Med Phys, 2005. 32(4): p.
1047-1053.
[95] Venning, A.J., Hill, B., Brindha, S., Healy, B.J., and Baldock, C.,
Investigation of the PAGAT polymer gel dosimeter using magnetic
resonance imaging. Phys Med Biol, 2005. 50: p. 3875-3888.
[96] Brindha, S., Venning, A.J., Hill, B., and Baldock, C., Experimental study
of attenuation properties of normoxic polymer gel dosimeters. Phys Med
Biol, 2004. 49: p. N353-361.
[97] Venning, A.J., Mather, M.L., and Baldock, C., Investigation of vacuum
pumping on the dose response of the MAGAS normoxic polymer gel
dosimeter. Australas Phys Eng Sci Med, 2005. 28(2): p. 105-110.
[98] Baldock, C., R.P. Burford, N. Billingham, G.S. Wagner, S. Patval, R.D.
Badawi, and S.F. Keevil, Experimental procedure for the manufacture and
calibration of polyacrylamide gel (PAG) for magnetic resonance imaging
(MRI) radiation dosimetry. Phys Med Biol, 1998. 43(3): p. 695-702.
[99] Haraldsson, P., S.A. Back, P. Magnusson, and L.E. Olsson, Dose response
characteristics and basic dose distribution data for a polymerization-based
dosemeter gel evaluated using MR. Br J Radiol, 2000. 73(865): p. 58-65.
[100] Seymour, R.B., C.E. Carraher, and R.B. Seymour, Seymour/Carraher's
polymer chemistry : an introduction. 4th , rev. and expanded ed.
Undergraduate chemistry ; 13. 1996, New York: M. Dekker,. xxvii, 688.
[101] Salomons, G.J., Park, Y.S., McAuley, K.B., Schreiner, L.J., Temperature
increases associated with polymerization of irradiated PAG dosimeters.
Phys Med Biol, 1992. 47: p. 1435-1448.
[102] Maryanski, M.J., C. Audet, and J.C. Gore, Effects of crosslinking and
temperature on the dose response of a BANG polymer gel dosimeter. Phys
Med Biol, 1997. 42(2): p. 303-11.
[103] De Deene, Y., C. De Wagter, B. Van Duyse, S. Derycke, W. De Neve, and
E. Achten, Three-dimensional dosimetry using polymer gel and magnetic
54
resonance imaging applied to the verification of conformal radiation
therapy in head-and-neck cancer. Radiother Oncol, 1998. 48(3): p. 283-91.
[104] Baldock, C., R.P. Burford, N. Billingham, D. Cohen, and S.F. Keevil,
Polymer gel composition in MRI dosimetry. Med Phys, 1996. 23: p. 1070.
[105] Fuxman, A.M., McAuley, K.B., Schreiner, L.J., Modeling of free-radical
crosslinking copolymerization of acrylamide and N,N’-
methylenebis(acrylamide) for radiation dosimetry. Macromol Theory
Simul, 2003. 12: p. 647-662.
[106] Kennan, R.P., K.A. Richardson, J. Zhong, M.J. Maryanski, and J.C. Gore,
The effects of cross-link density and chemical exchange on magnetisation
transfer in polyacrylamide gels. J Magn Reson, 1996. 110(B): p. pp. 267 -
277.
[107] McJury, M., M. Oldham, M.O. Leach, and S. Webb, Dynamics of
polymerization in polyacrylamide gel (PAG) dosimeters: (I) ageing and
long-term stability. Phys Med Biol, 1999b. 44(8): p. 1863-73.
[108] De Deene, Y., P. Hanselaer, C. De Wagter, E. Achten, and W. De Neve,
An investigation of the chemical stability of a monomer/polymer gel
dosimeter. Phys Med Biol, 2000. 45(4): p. 859-78.
[109] De Deene, Y., Fundamentals of MRI measurements, in Proceedings of 2nd
International Conference on Radiotherapy Gel Dosimetry, Baldock, C. and
De Deene, Y., Editors. 2001, Queensland University of Technology:
Brisbane, Australia.
[110] Hasson, B.F., Chemical dosimetry in the near-zone of brachytherapy
sources. Med Phys, 1996. 23: p. 803.
[111] De Deene, Y., De Wagter, C., Van Duyse, B., Derycke, S., Mersseman,
B., De Gersem, W., Voet, T., Achten, E., and De Neve, W., Validation of
MR-based polymer gel dosimetry as a preclinical three dimensional
verification tool in conformal radiotherapy. Magn Reson Med, 2000.
43(1): p. 116-25.
[112] Ibbott, G.S., Maryanski, M.J., Eastman, P., Holcomb, S.D., Zhang, Y.,
Avison, R.G., Sanders, M., and Gore, J.C., Three-dimensional
visualization and measurement of conformal dose distributions using
magnetic resonance imaging of BANG polymer gel dosimeters. Int J
Radiat Oncol Biol Phys, 1997. 38(5): p. 1097-103.
55
[113] Kaurin, D.G., Maryanski, M.J., Duggan, D.M., Morton, K.C., and Coffey,
C.W., Use of polymer gel dosimetry, pelvic phantom and virtual
simulation to verify setup and calculated three-dimensional dose
distributions for a prostate treatment, in Proceedings of 1st International
Workshop on Radiation Therapy Gel Dosimetry, Schreiner, L.J., Editor.
1999, Canadian Organization of Medical Physicists: Lexington, Kentucky.
p. 184-189.
[114] De Deene, Y., De Wagter, C., De Neve, W., and Achten, E., Artefacts in
multi-echo T2 imaging for high-precision gel dosimetry: I. Analysis and
compensation of eddy currents. Phys Med Biol, 2000. 45(7): p. 1807-23.
[115] De Deene, Y., De Wagter, C., De Neve, W., and Achten, E., Artefacts in
multi-echo T2 imaging for high-precision gel dosimetry: II. Analysis of
B1-field inhomogeneity. Phys Med Biol, 2000. 45(7): p. 1825-39.
[116] De Deene, Y. and De Wagter, C., Artefacts in multi-echo T2 imaging for
high-precision gel dosimetry: Effects of temperature drift during scanning.
Phys Med Biol, 2001. 46: p. 2697-2711.
[117] Baldock, C., Lepage, M., Back, S.A.J., Murry, P.J., Jayasekera, P.M.,
Porter, D., and Kron, T., Dose resolution in radiotherapy polymer gel
dosimetry: effect of echo spacing in MRI pulse sequence. Physics in
Medicine and Biology, 2001. 46: p. 449-460.
[118] De Deene, Y. and Baldock, C., Optimization of multiple spin-echo
sequences for 3D polymer gel dosimetry. Phys Med Biol, 2002. 47: p.
3117-3141.
[119] Hilts, M., Audet, C., Duzenli, C., and Jirasek, A., Polymer gel dosimetry
using x-ray computed tomography: a feasibility study. Phys Med Biol,
2000. 45(9): p. 2559-71.
[120] Trapp, J.V., Back, S.A.J., Lepage, M., Michael, G., and Baldock, C.,
Investigation of the x-ray computed tomography absorbed dose response
characteristics of various polymer gel dosimeter compositions. Phys Med
Biol, 2001. 46: p. 2939-2951.
[121] Trapp, J.V., Michael, G., De Deene, Y., and Baldock, C., Attenuation of
diagnostic energy photons by polymer gel dosimeters. Phys Med Biol,
2002. 47: p. 4247-4258.
56
[122] Audet, C., Hilts, M., Jirasek, A., and Duzenli, C., CT gel dosimetry
technique: Comparison of a planned and measured 3D stereotactic dose
volume. J Appl Clin Med Phys, 2002. 3: p. 110-118.
[123] Hill, B., Venning, A.J., and Baldock, C., Use of normoxic polymer gel
dosimeters for measuring diagnostic doses on CT scanners, in Third
International Conference on Radiotherapy Gel Dosimetry, De Deene, Y.
and Baldock, C., Editors. 2004, J Phys Conf Ser: Ghent. p. 224-227.
[124] Hill, B., Venning, A., and Baldock, C., The dose response of normoxic
polymer gel dosimeters measured using X-ray CT. Br J Radiol, 2005. 78:
p. 623-630.
[125] Hill, B., Venning, A.J., and Baldock, C., A preliminary study of the novel
application of normoxic polymer gel dosimeters for the measurement of
CTDI on diagnostic x-ray CT scanners. Med Phys, 2005. 32(6): p. 1589-
1597.
[126] Oldham, M., Siewerdsen, J.H., Shetty, A., and Jaffray, D.A., High
resolution gel-dosimetry by optical-CT and MR scanning. Med Phys,
2001. 23: p. 699-705.
[127] Maryanski, M.J. and Ranade, M.K., Laser micro-beam scanning of
dosimetry gels. Proc SPIE Int Soc Optic Eng, 2001. 4320: p. 764-767.
[128] Xu, Y., Wuu, C., and Maryanski, M.J., An algorithm for optimized CT
scanning of gel dosimeters. Med Phys, 2002. 29: p. 1205.
[129] Krstelj, V., Ultrasonic testing of polymer materials. Insight, 1996. 38: p.
14-20.
[130] Luprano, V.A.M., Ramires, P.A., Montagna, G., and Milella, E., Non-
destructive characterization of hydrogels. J Mater Sci Mater Med, 1997. 8:
p. 175-178.
[131] Taylor, K.J.W. and Wells, P.N.T., Tissue characterization. Ultrasound in
Med and Biol, 1989. 15: p. 421-428.
[132] Maffezzoli, A., Luprano, V.A., Montagna, G., Esposito, F., and Nicolais,
L., Ultrasonic wave attenuation during water sorption in poly(2-
hydroxyethyl methacrylate) hydrogels. Poly Eng Sci, 1996. 36: p. 1832-
1838.
57
[133] Goldshteins, A.Y., Dolatsis, A., and Zavitska, R.F., Ultrasonic wave
propagation in a polymerizing medium. Mekhanika Polimerov, 1974. 2: p.
204-208.
[134] Zahran, R.R., Kandeil, A.Y., Higazy, A.A., and Kassem, M.E., Ultrasonic
and thermal properties of gamma irradiated low-density polyethylene. J
Appl Poly Sci, 1993. 49: p. 1291-1297.
[135] Zahran, R.R., Effects of gamma irradiation on the ultrasonic and structural
properties of polyoxymethylene. Materials Letters, 1998. 37: p. 83-89.
[136] Maryanski, M.J., Radiation-sensitive polymer gels: properties and
manufacturing, in Proceedings of the First International Workshop on
Radiation Therapy Gel Dosimetry. 1999, Canadian Organsiation of
Medical Physicists: Lexington, Kentucky.
[137] Mather, M.L., Whittaker, A.K., and Baldock, C., Ultrasound evaluation of
polymer gel dosimeters. Phys Med Biol, 2002. 47(9): p. 1449-58.
[138] Baldock, C., Rintoul, L., Keevil, S.F., Pope, J.M., and George, G.A.,
Fourier transform Raman spectroscopy of polyacrylamide gels (PAGs) for
radiation dosimetry. Phys Med Biol, 1998. 43(12): p. 3617-27.
[139] Lepage, M., Whittaker, A.K., Rintoul, L., Back, S.A., and Baldock, C.,
Modelling of post-irradiation events in polymer gel dosimeters. Phys Med
Biol, 2001. 46(11): p. 2827-39.
[140] Jirasek, A. and Duzenli, C., Relative effectiveness of polyacrylamide gel
dosimeters applied to proton beams: Fourier transform Raman
observations and track structure calculations. Med Phys, 2002. 29(4): p.
569-77.
[141] Jirasek, A.I., Duzenli, C., Audet, C., and Eldridge, J., Characterization of
monomer/crosslinker consumption and polymer formation observed in FT-
Raman spectra of irradiated polyacrylamide gels. Phys Med Biol, 2001.
46(1): p. 151-65.
[142] Rintoul, L., Lepage, M., and Baldock, C., Radiation dose distribution in
polymer gels by Raman spectroscopy. Appl Spectrosc, 2003. 57(1): p. 51-
7.
[143] Gustavsson, H., Back, S.A.J., Lepage, M., Rintoul, L., and Baldock, C.,
Development and optimization of a 2-hydroxyethylacrylate MRI polymer
gel dosimeter. Phys Med Biol, 2004. 49: p. 227-241.
58
[144] Duzenli, C. and Robinson, D., Correcting for rf inhomogeneities in
multiecho pulse sequence MRI dosimetry. Med Phys, 1995. 22(10): p.
1645-50.
[145] Baustert, I.C., Oldham, M., Smith, T.A., Hayes, C., Webb, S., and Leach,
M.O., Optimized MR imaging for polyacrylamide gel dosimetry. Phys
Med Biol, 2000. 45(4): p. 847-58.
[146] Lepage, M., Tofts, P.S., Back, S.A., Jayasekera, P.M., and Baldock, C.,
Simple methods for the correction of T2 maps of phantoms. Magn Reson
Med, 2001. 46(6): p. 1123-9.
[147] McJury, M., Tapper, P.D., Cosgrove, V.P., Murphy, P.S., Griffin, S.,
Leach, M.O., Webb, S., and Oldham, M., Experimental 3D dosimetry
around a high-dose-rate clinical 192Ir source using a polyacrylamide gel
(PAG) dosimeter. Phys Med Biol, 1999. 44(10): p. 2431-44.
[148] Lepage, M., Whittaker, A.K., Rintoul, L., and Baldock, C., 13C-NMR,
1H-NMR, and FT-Raman study of radiation-induced modifications in
radiation dosimetry polymer gels. Journal of Applied Polymer Science,
2001. 79: p. 1572-1581.
[149] Brandl, M. and Haase, A., Molecular diffusion in NMR microscopy. J
Magn Reson, 1994. 103(Series B): p. 162-167.
[150] Carr, H.Y. and Purcell, E.M., Effects of diffusion on free precession in
nuclear magnetic resonance experiments. Physical Review, 1954. 94: p.
630-638.
[151] Mattiello, J., Basser, P.J., LeBihan, D., Analytical expressions for the b
matrix in NMR diffusion imaging and spectroscopy. Journal of Magnetic
Resonance, 1994. 108(Series A): p. 131-141.
[152] Karger, J., Pfeifer, H., Heink, W., Principles and applications of self-
diffusion measurements by nuclear magnetic resonance. Advances in
Magnetic Resonance, 1988. 12: p. 1 - 89.
[153] Stejskal, E.O. and Tanner, J.E., Spin diffusion measurements: spin echoes
in the presence of a time-dependent field gradient. J Chem Phys, 1965. 42:
p. 288 - 292.
[154] Ertl, A., Berg, A., Zehetmayer, M., and Frigo, P., High-resolution dose
profile studies based on MR imaging with polymer BANG(TM) gels in
59
stereotactic radiation techniques. Magn Reson Imaging, 2000. 18(3): p.
343-9.
[155] Berg, A., Ertl, A., and Moser, E., High-resolution polymer gel dosimetry
by parameter selective MR- microimaging on a whole body scanner at 3T.
Med Phys, 2001. 28(5): p. 833-43.
[156] Hepworth, S.J., McJury, M., Oldham, M., Morton, E.J., and Doran, S.,
Dose mapping of inhomogeneities positioned in radiosensitive polymer
gels. Nuclear Instruments and Methods in Physics Research, 1999.
422(A): p. 756-760.
[157] Farajollahi, A.R., Bonnett, D.E., Tattam, D., and Green, S., The potential
use of polymer gel dosimetry in boron neutron capture therapy. Phys Med
Biol, 2000. 45(4): p. N9-14.
[158] Ramm, U., Weber, U., Bock, M., Kramer, M., Bankamp, A., Damrau, M.,
Thilmann, C., Bottcher, H.D., Schad, L.R., and Kraft, G., Three-
dimensional BANG gel dosimetry in conformal carbon ion radiotherapy.
Phys Med Biol, 2000. 45(9): p. N95-102.
[159] Soares, C., Dosimetric Issues in Vascular Brachytherapy (TG-43/60), in
Intrascular Brachytherapy - Fluoroscopically Guided Interventions, Balter,
S., Chan, R.C., and Shope, T.B., Editors. 2002, American Association of
Physicists in Medicine: Montreal. p. 321-372.
[160] Schreiner, L.J., Crooks, I., Evans, M.D.C., Keller, B.M., Parker, W.A.,
Imaging of HDR brachytherapy dose distributions using NMR Fricke-
gelatin dosimetry. Magn Reson Imaging, 1994. 12: p. 901-907.
[161] Farajollahi, A.R., Bonnett, D.E., Ratcliffe, A.J., Aukett, R.J., and Mills,
J.A., An investigation into the use of polymer gel dosimetry in low dose
rate brachytherapy. Br J Radiol, 1999. 72(863): p. 1085-92.
[162] Hafeli, U.O., Roberts, W.K., Meier, D.S., Ciezki, J.P., Pauer, G.J., Lee,
E.J., and Weinhous, M.S., Dosimetry of a W-188/Re-188 beta line source
for endovascular brachytherapy. Med Phys, 2000. 27(4): p. 668-75.
[163] Knutsen, B.H., Skretting, A., Hellebust, T.P., and Olsen, D.R.,
Determination of 3D dose distribution from intracavitary brachytherapy of
cervical cancer by MRI of irradiated ferrous sulphate gel. Radiother
Oncol, 1997. 43(2): p. 219-27.
60
[164] Ibbott, G.S. and Nath, R., Dose-rate constant for Imagyn 125I
brachytherapy source. Med Phys, 2001. 28(4): p. 705.
[165] Baras, P., Seimenis, I., Kipouros, P., Papagiannis, P., Angelopoulos, A.,
Sakelliou, L., Pappas, E., Baltas, D., Karaiskos, P., Sandilos, P., and
Vlachos, L., Polymer gel dosimetry using a three-dimensional MRI
acquisition technique. Med Phys, 2002. 29(11): p. 2506-16.
[166] Papagiannis, P., Pappas, E., Kipouros, P., Angelopoulos, A., Sakelliou, L.,
Baras, P., Karaiskos, P., Seimenis, I., Sandilos, P., and Baltas, D.,
Dosimetry close to an 192Ir HDR source using N-vinylpyrrolidone based
polymer gels and magnetic resonance imaging. Med Phys, 2001. 28(7): p.
1416-26.
[167] Wuu, C.S., Schiff, P., Maryanski, M.J., Liu, T., Borzillary, S., and
Weinberger, J., Dosimetry study of Re-188 liquid balloon for intravascular
brachytherapy using polymer gel dosimeters and laser-beam optical CT
scanner. Med Phys, 2003. 30(2): p. 132-7.
[168] De Deene, Y., Reynaert, N., and De Wagter, C., On the accuracy of
monomer/polymer gel dosimetry in the proximity of a high-dose-rate
192Ir source. Phys Med Biol, 2001. 46(11): p. 2801-25
[169] Fuxman, A.M., McAuley, K.B., Schreiner, L.J., Modelling of
polyacrylamide gel dosimeters with spatially non-uniform radiation dose
distribution. Chem Eng Sci, 2005. 60: p. 1277-1293.
[170] Amin, M.N., Horsfield, M.A., Bonnett, D.E., Dunn, M.J., Poulton, M., and
Harding, P.F., A comparison of polyacrylamide gels and radiochromic
film for source measurements in intravascular brachytherapy. Br J Radiol,
2003. 76(911): p. 824-31.
[171] Gore, E., Gillin, M.T., Albano, K., and Erickson, B., Comparison of high
dose-rate and low dose-rate dose distributions for vaginal cylinders. Int J
Radiat Oncol Biol Phys, 1995. 31(1): p. 165-70.
[172] Baldock, C., Murry, P., and Kron, T., Uncertainty analysis in polymer gel
dosimetry. Phys Med Biol, 1999. 44(11): p. N243-6.
[173] (IAEA), I.A.E.A., Absorbed dose determination in photon and electron
beams, in An international code of practice for dosimetry. 2000, Technical
Report Series 400 IAEA: Vienna.
61
[174] Oldham, M., McJury, M., Baustert, I.B., Webb, S., and Leach, M.O.,
Improving calibration accuracy in gel dosimetry. Phys Med Biol, 1998.
43(10): p. 2709-20.
[175] Low, D.A., Markman, J., Dempsey, J.F., Mutic, S., Oldham, M.,
Venkatesan, R., Haacke, E.M., and Purdy, J.A., Noise in polymer gel
measurements using MRI. Med Phys, 2000. 27(8): p. 1814-7.
[176] Low, D.A., Dempsey, J.F., Venkatesan, R., Mutic, S., Markman, J., Mark
Haacke, E., and Purdy, J.A., Evaluation of polymer gels and MRI as a 3-D
dosimeter for intensity- modulated radiation therapy. Med Phys, 1999.
26(8): p. 1542-51.
[177] Magnusson, P., Back, S.A., and Olsson, L.E., MRI image plane
nonuniformity in evaluation of ferrous sulphate dosimeter gel (FeGel) by
means of T1-relaxation time. Magn Reson Imaging, 1999. 17(9): p. 1357-
70.
[178] Watanabe, Y., Variable transformation of calibration equations for
radiation dosimetry. Phys Med Biol, 2005. 50(6): p. 1221-34.
62
63
Chapter 3
Dose-response Stability and Integrity of
the Dose Distribution of Various
Polymer Gel Dosimeters
Full-text available at:
http://dx.doi.org/10.1088/0031-9155/47/14/307
77
Chapter 4
A Basic Study of Some Normoxic
Polymer Gel Dosimeters
Full-text available at: http://dx.doi.org/10.1088/0031-9155/47/19/301
101
Chapter 5
The Effects of Molecule Self-Diffusion
of Water on Quantitative MRI
Measurements in High-Resolution
Polymer Gel Dosimetry
Full-text available at:
http://dx.doi.org/10.1088/0031-9155/48/18/306
119
Chapter 6
A Study of a Normoxic Polymer Gel Dosimeter
comprising methacrylic acid, gelatin, and
Tetrakis (Hydroxymethyl) Phosphonium
Chloride (MAGAT)
Full-text available at: http://dx.doi.org/10.1016/j.apradiso.2005.03.014
135
Chapter 7
High-Resolution Gel Dosimetry of a
HDR Brachytherapy Source Using
Normoxic Polymer Gels
Full-text available at: http://dx.doi.org/10.1016/j.nima.2006.05.167
147
Chapter 8
General Discussion
Radiation therapy has seen significant advances in the effective irradiation of
tumours using methods that can conform the irradiation pattern to the affected
target tissue more precisely than previously possible [1]. The approach is to
deliver the maximum dose to the target volume whilst sparing the healthy
surrounding tissues. Conformal radiotherapy, IMRT, dynamic multi-leaf
collimation and on-line portal imaging have all aided this development and hence
treatment delivery can now comprise very complex geometries. Likewise,
brachytherapy treatments can now be achieved with sub-millimeter scales using
high dose gradients. Currently dose distributions are calculated using treatment
planning systems that incorporate complex computer algorithms. With the
increase in complexity there needs to be an effective means by which to verify
that the planned treatment is delivered through direct measurements. Polymer gel
dosimetry has been shown to be an effective technique for accurate 3D dosimetry
with high spatial resolution. It is capable of integrating dose throughout a
dynamic delivery without perturbing the radiation beam [2,3]. One key clinical
limitation to polymer dosimeters is lack of ease of manufacture and use due to
strict hypoxic requirements that are necessary to prevent oxygen from infiltrating
the gel, hence inhibiting the polymerization process that is used to correlate to
absorbed dose to produce dose distributions [4].
The work of this thesis has involved the investigation of normoxic polymer gel
dosimeters with high-resolution MRI. Normoxic polymer gel dosimeters include
an anti-oxidant in the formulation to bind dissolved oxygen preventing it from
inhibiting the polymerization process [5]. The first aspect of this study involved a
comprehensive investigation of different formulations of polymer and normoxic
148
polymer gel dosimeters, in order to determine the optimal composition for use in
polymer gel dosimetry. The second aspect of this research was to investigate
high-resolution MRI for normoxic polymer gel dosimetry with high spatial
resolution. The key application of this work has been intravascular brachytherapy
(IVBT) that aims to deliver doses to the walls of arteries to prevent restenosis
following percutaneous transluminal coronary angioplasty (PTCA). Annually,
there are 450 000 coronary angioplasties performed on de novo and restonotic
lesions in the United States alone [6,7]. Approximately 70 to 80 % of cases
require stent implants to reduce restenosis and 15 to 35 % of these patients still
develop restenosis [6,8]. IVBT has been found to be effective in the treatment of
restenosis.
The following aspects were studied to enable a full assessment of normoxic
polymer gel dosimeters using high-resolution MRI as an effective medical
dosimeter:
• An exploration of different formulations of polymer gel dosimeters and
normoxic polymer gel dosimeters to obtain optimal characteristics for use
in gel dosimetry.
• The determination of the physical and chemical properties of various
formulations of normoxic polymer gel dosimeters on dose maps obtained
using gel dosimetry with MRI.
• An examination of the chemical mechanisms that occur when normoxic
polymer gel dosimeters are manufactured, irradiated and subsequently
evaluated.
• The investigation of high-resolution magnetic resonance imaging and its
potential for use in evaluating polymer gel dosimeters. This involved the
commissioning a micro-imaging magnetic resonance spectrometer for
high-resolution imaging of polymer gel dosimeters.
149
• An examination of the effects that high gradient strengths combined with
molecular diffusion, have on R2 and hence dose maps produced using
high-resolution MRI. It was found that with higher resolution MRI, there
were significant changes in the measured R2. Software was developed
that could use differences in the MRI parameters applied during imaging
to calculate quantitatively the effects of molecular self-diffucion of water
on the measured R2 within a polymer gel dosimeter. A listing of the
developed software is included in appendix A.
• The application and evaluation of normoxic polymer gel dosimeters
irradiated using high-resolution MRI on typical brachytherapy deliveries.
In particular a point source and line irradiation patterns were explored and
compared with predictions from computer treatment planning software.
8.1 The Principal Significance of the Findings
8.1.1 Analyzing and Optimizing Polymer Gel Dosimeter Formulations
Polymer gel dosimeters have the potential to provide accurate, integrated 3D maps
of dose distributions with high spatial resolution. However, several different
formulations of polymer gel dosimeter have been investigated and proposed over
the years. In order to be able to compare different polymer gel dosimeters,
specific criteria are used that relate to requirements for clinical dosimetry of
radiation therapy treatment planning. Before polymer and normoxic polymer gel
dosimetry can be implemented in clinical or routine practice, comprehensive
studies must be conducted to explore, in detail, the chemical and physical aspects
of gel dosimetry. This thesis is an investigation of the common formulations of
polymer gel dosimeters and the recently proposed normoxic polymer gel
dosimeters.
The investigation of each polymer gel dosimeter involves a temporal study of the
changes in both the slope and R2-intercept of the R2-dose response curve. The
150
changes in the R2-intercept at zero dose are related to the ageing of the gelatin
network. It was found that the ageing process of the gelatin network was related
to the total concentration of gelatin molecules. An increase in gelatin was also
found to result in a decrease in the characteristic time for the R2-dose sensitivity
(slope) to saturate. However, the total degree of post-irradiation polymerization
or restructuring was found to be unaffected in the long term by the concentration
of gelatin. The ratio of cross-linker monomer to linear monomer was found to
have a significant effect on the stability of the R2-dose sensitivity but was found
not to affect the R2 intercept. Additionally, whilst the R2-dose sensitivity
increased over time for PAG gel dosimeter, it was found to decrease for normoxic
polymer gel dosimeters. The temporal stability of polyacrylamide gel (PAG)
dosimeters was also found to differ from that of normoxic polymer gel
dosimeters. These differences indicate that normoxic polymer gel dosimeters
have a different polymer structure to PAG gel dosimeters.
The spatial stability of the gel was also explored by examining the measured dose
at the dose edge in a gel exposed to a half-beam blocked field. This enabled
monomer diffusion to be measured and hence used as an indicator of stability.
Both the PAG gel dosimeters and the normoxic polymer gel dosimeters were
found to be spatially stable for doses up to 12 Gy, whilst the HEA gel dosimeter
demonstrated an instability in the form of a dose overshoot that varied over time.
The spatial stability of the PAG and normoxic polymer gel dosimeters was found
to be stable with an apparent penumbra between 3.1 mm and 4.4 mm for gels of
different formulation. The normoxic polymer gel dosimeters exhibited a smaller
apparent penumbra (3.2 to 4.3 mm) than the PAG gel dosimeters (4.05 to 4.4
mm), however, the HEA gel exhibited the lowest apparent penumbra at around 3.1
mm. It was concluded that waiting periods of 10 h and 30 h be respected between
irradiating and scanning PAG and normoxic polymer gel dosimeters respectively.
8.1.2 Chemical Properties of Normoxic Polymer Gel Dosimeters
151
The first normoxic polymer gel dosimeter was proposed by Fong et al. in 2001
[5]. It was comprised of methacrylic acid, ascorbic acid, hydroquinone, copper
(II) sulphate and gelatin, and it was called MAGIC. Although specific
concentrations were provided by Fong et al., no explanation or justification was
given for the choice of concentrations. This study involved varying the
concentration of each constituent in order to map a dose response between 0 and 5
Gy. A region of optimal dose response was determined by examining a difference
R2 map between unirridiated and irradiated gels given varying concentration of
each constituent.
Advances in the understanding of the chemical behaviour and properties of
normoxic polymer gel dosimeters have been made through the identification of
the reactions that occur from manufacture to post-irradiation. It was found that
the reaction between two constituents, copper (II) sulphate and ascorbic acid,
leads to the formation of a ascorbate-copper complex. Some initial
polymerization takes place due to the creation of radicals in the ascorbate-copper-
oxygen complex. Ascorbic acid (an anti-oxidant) was found to scavenge and bind
the free oxygen within the gel dosimeter, whilst copper acts as a catalyst in the
oxygen scavenging. With high concentrations of copper (II) sulphate the
polymerization reaction is terminated by a redox reaction in which copper is
reduced. Hydroquinone, when used in low quantities, facilitates the radiation-
induced polymerization reaction, however, at high concentrations it acts as an
inhibitor.
In this thesis, five anti-oxidants were explored for potential use in normoxic
polymer gel dosimetry; only three were found to be effective. The oxygen
scavenging ability of an anti-oxidant could be measured independently and
provided basis for determining the characteristic time for a normoxic polymer gel
dosimeter to become radiation sensitive. Of the three viable anti-oxidants, tetrakis
(hydroxymethyl) phosphonium chloride (THPC) was found to have the highest
reaction rate. THPC was also found to increase the R2-dose sensitivity of the
normoxic polymer gel dosimeter.
152
8.1.3 A Normoxic Polymer Gel Dosimeter Using THPC
Tetrakis (hydroxymethyl) phosphonium chloride (THPC) was found to be a very
aggressive scavenger of dissolved oxygen. Hence, a normoxic polymer gel
dosimeter made using methacrylic acid, gelatin, hydroquinone and THPC (called
MAGAT) was proposed and investigated. In a similar study to section 8.1.2, the
constituents of the MAGAT normoxic polymer gel dosimeter were optimized by
varying their concentrations and exploring the effect on the dose response of the
gel. Firstly, the concentrations of THPC and hydroquinone were varied and
assessed for optimal R2-dose response. Secondly, concentrations of methacrylic
acid and gelatin were also varied to determine the optimal concentration that
produced the highest dose response. The R2-dose response was found to varying
slightly over a 7 day period. The R2-intercept steadily increased over this time,
whilst the R2-dose sensitivity (slope) steadily decreased. These results concurred
with those of the previous study that examined the R2-dose response of normoxic
polymer gel dosimeter made with ascorbic acid as the anti-oxidant.
In additional, the spatial stability of the MAGAT polymer gel dosimeter was
investigated by exposing a phantom to a half-beam blocked field and examining
the dose profile through the boundary between the irradiated and unirradiated
sections of the phantom. The spatial stability of the MAGAT polymer gel
dosimeter was found to exhibit an apparent penumbra of around 3.5 mm.
Previous studies of the normoxic polymer gel dosimeters, found that the
penumbra was between 3.2 mm and 3.7 mm for normoxic polymer gel dosimeter
with different formulations (see section 8.1.1). It was found that the addition of
small quantities of hydroquinone aided in maintaining the spatial stability.
It was concluded that the MAGAT normoxic polymer gel dosimeter exhibited a
R2-dose response that was stable over a 7 day period and spatially stable when
imaged 24 h after irradiation. This concurs with the previous study which
concluded that normoxic polymer gel dosimeters should be scanned about 30 h
after irradiation.
153
8.1.4 Evaluating Polymer Gel Dosimeters using High-Resolution MRI
In order to evaluate polymer gel dosimeters at high-resolution, a 4.7 T magnetic
resonance spectrometer was commissioned for use in scanning calibration vials
and phantoms designed for simulating typical brachytherapy treatments. This
commissioning involved investigating various pulse sequences and the pulse
parameters of various optimized pulse sequences to obtain suitable images. Using
MRI to evaluate polymer gel dosimeters at high-resolution introduces artifacts
into the images that are often negligible in MRI at typical clinical resolutions.
The predominant effect was found to be the self-diffusion through steep magnetic
field gradients of water molecules caused by Brownian motion. This self-
diffusion had a significant effect on measured R2 values at high-resolution. It was
found that changes in the imaging parameters, including echo time, which is often
used to optimize pulse sequences in polymer gel dosimetry [9,10], led to
significant changes in measured R2 values due to the self-diffusion.
Software was written to evaluate the extent of the variation in measured R2 values
for specific parameters of a MRI pulse sequence (see appendix A). The software
is capable of incorporating changes in timing parameters and gradient strengths
and subsequently predicting the variation in measured R2 as a result of the water
self-diffusion given the particular pulse sequence. Gel dosimetry typically
involves imaging a set of calibration vials and a phantom. This is usually done
separately, and thus, any differences in the imaging parameters between the
calibration vials and phantoms may introduce significant differences in the
measured R2 values when scanned at high-resolution. These variations in the
measured R2 values will give rise to significant errors in the measured absorbed
dose. Correction factors can be determined using the software and applied to R2
maps of calibration vials and phantoms thereby adjusting for errors that arise due
to self-diffusion at high-resolution. On a simpler level, keeping the imaging
parameters the same for both the calibration vials and the phantoms prevents any
differences in R2 occuring due to self-diffusion of water molecules in high-
resolution MR imaging.
154
8.1.5 Application of Normoxic Polymer Gel Dosimeters using High-
Resolution MRI to Brachytherapy Treatment Plans
An optimal formulation of a normoxic polymer gel dosimeter was manufactured
and irradiated using a brachytherapy HDR remote afterloader system. The dose
deliveries were set by typical brachytherapy plans that included a single dwell
position simulating a point source irradiation and a line irradiation pattern. A set
of calibration vials was also filled with normoxic polymer gel dosimeter from the
same batch. Post-irradiation, both the calibration vials and the phantoms were
evaluated using high-resolution MRI to produce R2 maps. These R2 maps were
then converted to relative dose maps and compared to the dose distribution
predicted by the computer treatment planning system.
The investigation found differences between the dose distributions measured in
the normoxic polymer gel dosimeter and the predicted dose distributions from the
computer treatment planning system. However, an exploration of various artifacts
revealed that partial volume averaging had a significant effect on the measured
dose given the relatively higher slice thickness compared to in-plane resolution
and the high dose gradients at radial distances close to the source (< 10 mm). An
approximation of the uncertainty due to partial volume effects could be
determined from dose points predicted by the computerized treatment planning
system. These approximations of the amount of variation due to partial volume
effects were used to calculate an adjustment that was applied to measured dose
profiles. Very good agreement was found between the adjusted, normalized dose
distributions measured using the gel dosimeter and dose points predicted by the
computerized treatment planning system. The results indicate that normoxic
polymer gel dosimeters evaluated with high-resolution MRI can be effective in
applications requiring high-resolution such as intravascular brachytherapy.
Likewise they could be the tool of choice for verifying treatment plans involving
complex geometries or new source designs.
155
8.2 Conclusions and Future Work
Polymer gel dosimeters have been developed to map dose distributions resulting
from irradiations such as radiation therapy and brachytherapy treatments.
Radiation therapy and brachytherapy applications that incorporate high dose
gradients, extremely asymmetric fields or complex geometries are difficult to
measure directly using conventional dosimetry techniques. However, polymer gel
dosimetry offers a technique capable of producing 3D integrated dose maps with
high spatial resolution. Whilst polymer gel dosimeters based on polyacrylamide
suffer from the effects of oxygen infiltration, normoxic polymer gel dosimeters
can easily be fabricated on the bench top under normal atmospheric conditions
and show great potential as effective dosimeters.
An integral part of the polymer gel dosimetry process is the evaluation using an
imaging modality such as MRI or optical CT. Applications such as intravascular
brachytherapy, however, require high-resolution dose information due to the small
dimensions of the target volume. Hence, the application of high-resolution MRI
techniques for evaluating normoxic polymer gel dosimeters has been investigated.
At high-resolution, MRI suffers from a number of artifacts, the most significant of
which is the self-diffusion of molecules during scanning.
The aims of this study have been achieved and the main conclusions are as
follows:
• Various formulations of polymer gel dosimeters and normoxic polymer
gel dosimeters have been optimized to achieve specific characteristics
desired for use in radiation therapy and brachytherapy dosimetry. These
included: R2-dose response, temporal stability of the R2 intercept and the
R2-dose sensitivity (slope) of the R2-dose response, spatial stability and
reduced oxygen effects. Optimal formulations included the MAGIC
normoxic polymer gel dosimeter (9 % methacrylic acid, 1 mM ascorbic
acid, 8 % gelatin, 0.01 mM copper(II) sulphate, 0.01 mM hydroquinone).
156
• The chemical mechanisms that take place in normoxic polymer gel
dosimeter formulation have been extensively explored and used to explain
the role that constituents of a specific formulation have. It has been
shown using chemical models that only low concentrations of copper are
effective in catalyzing the oxygen scavenging ability of ascorbic acid.
Hydroquinone in low concentrations was also found to facilitate the
radiation-induced polymerization reaction. This information assists in the
process of developing optimal formulations and explaining phenomena
associated with normoxic polymer gel dosimetry.
• The desired chemical and physical properties listed above were used to
propose a new formulation of normoxic polymer gel dosimeter. This
normoxic polymer gel dosimeter, called MAGAT, uses tetrakis
(hydroxymethyl) phosphonium chloride (THPC) as an anti-oxidant. The
optimal composition of the MAGAT gel was found to be 6 % methacrylic
acid, 6 % gelatin, 10 mM THPC and 0.05 mM hydroquinone based on R2
response. This dosimeter was found to exhibit promising temporal
stability of its R2-dose response and spatial stability.
• High-resolution magnetic resonance imaging has been implemented on a
micro-imaging spectrometer using a pulse sequence optimized for use
with polymer gel dosimetry. A multi-slice, multi-echo CPMG sequence
was found to be suitable for use with gel dosimetry. Small inter-echo
times are essential for imaging normoxic polymer gel dosimeters due to
relatively small characteristic T2 times. Likewise, to achieve high-
resolution imaging, a spectrometer must be capable of large magnetic
gradients.
• Artifacts were found to occur due to scanning at high-resolutions, the most
significant of which was found to be the effect on the measured R2 due to
self-diffusion of water molecules during scanning. Using knowledge of
the gradient strengths applied to a sample, the diffusion weighting could
157
be calculated and the R2 corrected for this effect. Software was
developed to analyze the timing and magnitude of pulses within a pulse
sequence in order to calculate correction factors that could be applied (a
listing is included in appendix A).
• A normoxic polymer gel dosimeter was fabricated and used to verify
typical brachytherapy treatments plans. In particular a line irradiation
pattern and irradiation due to a single dwell position effectively simulating
a point source. Dose distributions were produce using high-resolution
MRI (achieving an in-plane resolution of 0.1055 mm/pixel) and compared
to predictions of dose points from the computerized treatment planning
system.
• Partial volume effects were found to be significant at high resolution or
when large voxel sizes are used that have been exposed to large dose
gradients. By approximating the amount of variation within each voxel
due to partial volume effects, adjustments could be applied to dose
profiles that produced very good agreement between measured and
predicted doses from the computerized treatment planning system.
This thesis is the first study to comprehensively investigate normoxic polymer gel
dosimeters and to apply them to high-resolution brachytherapy applications. The
investigation of the theory behind the chemical kinetics and polymerization
processes that take place in polymer and normoxic polymer gel dosimetry support
the experimental findings. The implementation of normoxic polymer gel
dosimetry in the verification of high-resolution applications has been
demonstrated with good agreement between polymer gel dosimeter dose
measurements and dose calculations using a computerized treatment planning
system. It is therefore concluded that normoxic polymer gel dosimetry using
high-resolution MRI can be effectively applied to the verification of applications
requiring high-resolution, such as intravascular brachytherapy. Likewise,
normoxic polymer gel dosimetry could be used in applications where complex
geometries or severely asymmetric fields exist. Of particular significance is the
potential for normoxic polymer gel dosimetry using high-resolution MRI to be
158
applied to new source designs and more complex treatment plans given the
advances that have occurred in conformal radiation therapy over the past decade.
Future work could include an investigation into the reproducibility of the results.
Of particular interest would be an investigation of increasing resolution in the
MRI evaluation of normoxic polymer gel dosimeters. In this study, a resolution
of 0.1055 × 0.1055 × 2.0 mm3 was used. With smaller slice thickness (< 2 mm),
it would be possible to further reduce the partial volume effect and hence decrease
the uncertainty of the measured doses. Differences between MRI scans of the
calibration tubes and the phantoms may also have an effect on the measure dose
values and hence require further investigation.
Although the anti-oxidants used in this study to reduce the oxygen sensitivity of
polymer gel dosimeters exhibited good results, there are many anti-oxidants that
are yet to be explored for their effectiveness in normoxic polymer gel dosimetry.
The nature of the chemical processes that occur in normoxic polymer gel
dosimetry using these other anti-oxidants would need to be explored in order to
propose alternative formulations. It was noted in this study that the polymer
structure of normoxic polymer gel dosimeters differs from that of polyacrylamide
polymer gel dosimeters. The different structure still requires further exploration
in order to understand the differences in the temporal stability of normoxic
polymer gel dosimeters.
It has been concluded that the use of normoxic polymer gel dosimeters with high-
resolution MRI for the verification of radiation therapy and brachytherapy
treatment plans, is valid and shows great potential as an effective 3D integrating
dosimeter. The ease of manufacture of normoxic polymer gel dosimeters on the
bench top makes them a viable contender for implementation in routine clinical
treatment planning.
159
References
[1] Kron, T., Radiation therapy requirements: what do we expect from gel
dosimetry, in Proceedings of DosGel 2001: The second international
conference on radiotherapy gel dosimetry, Baldock, C. and De Deene, Y.,
Editors. 2001, Queensland University of Technology: Brisbane. p. 2-9.
[2] Maryanski, M.J., Ibbott, G.S., Schulz, R.J., Xie, J., and Gore, J.C.,
Magnetic resonance imaging of radiation dose distributions in tissue-
equivalent polymer-gel dosimeters, in Proc Soc Magn Reson Med. 1994.
p. 204.
[3] Schreiner, L.J., Gel Dosimetry: motivation and historical foundations, in
DosGel'99 Proceedings of the 1st International Workshop on Radiation
Therapy Gel Dosimetry. 1999, Canadian Organisation of Medical
Physicists: Lexington, Kentucky.
[4] Maryanski, M.J., Radiation-sensitive polymer gels: properties and
manufacturing, in Proceedings of the First International Workshop on
Radiation Therapy Gel Dosimetry. 1999, Canadian Organsiation of
Medical Physicists: Lexington, Kentucky.
[5] Fong, P.M., Keil, D.C., Does, M.D., and Gore, J.C., Polymer gels for
magnetic resonance imaging of radiation dose distributions at normal
room atmosphere. Phys Med Biol, 2001. 46(12): p. 3105-13.
[6] Baim, D.S., Cutlip, D.E., Midei, M., Linnemeier, T.J., Schreiber, T., Cox,
D., Kereiakes, D., Popma, J.J., Robertson, L., Prince, R., Lansky, A.J., Ho,
K.K., and Kuntz, R.E., Final results of a randomized trial comparing the
MULTI-LINK stent with the Palmaz-Schatz stent for narrowing in native
coronary arteries. Am J Cardiol, 2001. 87: p. 157-162.
[7] Waksman, R., Vascular Brachytherapy For Restenosis: Unresolved Issues,
in Intravascular Brachtherapy, Fluoroscopically Guided Interventions,
160
Balter, S., Chan, R.C., and Shope, T.B., Editors. 2002, American
Association of Physicists in Medicine: Montreal, Quebec.
[8] Baim, D.S., Cutlip, D.E., O'Shaughnessy, C.D., Hermiller, J.B., Kereiakes,
D., Giambartolomei, A., Katz, S., Lansky, A.J., Fitzpatrick, M., Popma,
J.J., Ho, K.K., Leon, M.B., and Kuntz, R.E., Final results of a randomized
trial comparing the NIR stent to the Palmaz-Schatz stent for narrowing in
native coronary arteries. Am J Cardiol, 2001. 87(152-156).
[9] Baldock, C., Lepage, M., Back, S.A.J., Murry, P.J., Jayasekera, P.M.,
Porter, D., and Kron, T., Dose resolution in radiotherapy polymer gel
dosimetry: effect of echo spacing in MRI pulse sequence. Phys Med Biol,
2001. 46: p. 449-460.
[10] De Deene, Y. and Baldock, C., Optimization of multiple spin-echo
sequences for 3D polymer gel dosimetry. Phys Med Biol, 2002. 47: p.
3117-3141.
161
Appendix A
Listing of the Code for Determining
Variations in R2 due to the Application
of Pulse Sequences during High-
Resolution MRI
The following Java code was written in order to quantitatively determine the
magnitudes of variation in R2 values when imaging a polymer gel dosimeter using
high-resolution MRI. The code uses information about the pulse sequence, in
particular the pulse timing and strengths in order to calculate the self-diffusion
weightings that would be applied to sample during high-resolution MRI imaging.
162
package diffcalc3; import javax.swing.UIManager; import java.awt.*; public class diffcalc3 { boolean packFrame = false; public diffcalc3() { Frame1a frame = new Frame1a("Diffusion Weighting Calculator"); if (packFrame) { frame.pack(); } else { frame.validate(); } Dimension screenSize =
Toolkit.getDefaultToolkit().getScreenSize(); Dimension frameSize = frame.getSize(); if (frameSize.height > screenSize.height) { frameSize.height = screenSize.height; } if (frameSize.width > screenSize.width) { frameSize.width = screenSize.width; } frame.setLocation((screenSize.width - frameSize.width) / 2,
(screenSize.height - frameSize.height) / 2); frame.setVisible(true); } public static void main(String[] args) { try { UIManager.setLookAndFeel(UIManager.getSystemLookAndFeelClassName()); } catch(Exception e) { e.printStackTrace(); } new diffcalc3(); } }
Class: diffcalc3 Author: Christopher Hurley Date: June, 2002
163
public class Echo { public int te = 0; public int echotime = 0; public double height = 0; public boolean enabled = false; public Echo() { } public Echo(int te, int echotime, double height) { this.te = te; this.echotime = echotime; this.height = height; enabled = true; } public void show() { enabled = true; } public void hide() { enabled = false; } } import java.io.*; import java.util.*; import java.awt.*; import java.awt.event.*; import javax.swing.*; public class Frame1a extends JFrame implements ActionListener{ // Constants.... public int nEchoes = 128; public Vector graphData = new Vector(); public int TE = 5050; // in useconds public double Gmax; // in T/m public double gamma; // in rad/s/T public double D; // in m^2/s public double TimeInc = 0.000001; // in useconds public int P0; public int P1; public int aqq;
Class: Frame1a Author: Christopher Hurley Date: June, 2002
Class: Echo Author: Christopher Hurley Date: June, 2002
164
public int d2; public int d3; public int d4; public int d5; public int d12; public int d14; public int d18; public double GROa; public double GROb; public double GSSa; public double GSSb; public double GSSc; public double GSSd; public double GSSe; public double GSSf; public double GPHa; public double GPHb; int c0, c1, c2, c3, c4, c4_5, c5, c6, c7, c8, c8_5, c9, c10, c11,
c12; int cTE; // Cycle TE calculated for repeat section int p0, p1, p2; // Pulse centres for P0 and P1 and Echo (time) double h0 = 3.0; // Arbitrary height for P0 double h1 = 6.0; // Arbitrary height for P1 int inclGrad = 5; // Default to include SS & RO gradients MyInt Loc = new MyInt(5); surface1 canvas = new surface1(); JPanel contentPane; BorderLayout borderLayout1 = new BorderLayout(); JPanel jPanel1 = new JPanel(new BorderLayout()); JTextArea output = new JTextArea(); JMenuBar jMenuBar1 = new JMenuBar(); JMenu jMenu1 = new JMenu(); JMenuItem jMenuItem1 = new JMenuItem(); JMenuItem jMenuItem2 = new JMenuItem(); JMenuItem jMenuItem3 = new JMenuItem(); JMenuItem jMenuItem4 = new JMenuItem(); JMenu jMenu2 = new JMenu(); JMenuItem jMenuItem5 = new JMenuItem(); JMenuItem jMenuItem6 = new JMenuItem(); JMenuItem jMenuItem7 = new JMenuItem(); JMenu jMenu3 = new JMenu(); JMenuItem jMenuItem8 = new JMenuItem(); JMenuItem jMenuItem9 = new JMenuItem(); JMenuItem jMenuItem10 = new JMenuItem(); JMenu jMenu4 = new JMenu(); JMenuItem jMenuItem11 = new JMenuItem(); JMenuItem jMenuItem12 = new JMenuItem(); public Frame1a(String title) { super(title); enableEvents(AWTEvent.WINDOW_EVENT_MASK); try {
165
jbInit(); } catch(Exception e) { e.printStackTrace(); } } private void jbInit() throws Exception { contentPane = (JPanel) this.getContentPane(); contentPane.setLayout(borderLayout1); this.setSize(new Dimension(700, 500)); jMenu1.setText("File"); jMenuItem1.setActionCommand("OpenFile"); jMenuItem1.setText("Open"); jMenuItem1.setAccelerator(javax.swing.KeyStroke.getKeyStroke(79,
0, false)); jMenuItem2.setActionCommand("SaveFile"); jMenuItem2.setText("Save"); jMenuItem2.setAccelerator(javax.swing.KeyStroke.getKeyStroke(83,
0, false)); jMenuItem3.setActionCommand("CloseFile"); jMenuItem3.setText("Close"); jMenuItem3.setAccelerator(javax.swing.KeyStroke.getKeyStroke(67,
0, false)); jMenuItem4.setText("Constants"); jMenu2.setText("Parameters"); jMenuItem5.setActionCommand("EditParam"); jMenuItem5.setText("Edit"); jMenuItem6.setActionCommand("LoadParam"); jMenuItem6.setText("Load"); jMenuItem7.setActionCommand("SaveParam"); jMenuItem7.setText("Save"); jMenu3.setText("Gradients"); jMenuItem8.setActionCommand("EditGrad"); jMenuItem8.setText("Edit"); jMenuItem9.setActionCommand("LoadGrad"); jMenuItem9.setText("Load"); jMenuItem10.setActionCommand("SaveGrad"); jMenuItem10.setText("Save"); jMenu4.setText("Calculation"); jMenuItem11.setActionCommand("GradCalculate"); jMenuItem11.setText("Calculate"); jMenuItem12.setActionCommand("GradInclude"); jMenuItem12.setText("Inclusions"); jMenuBar1.add(jMenu1); jMenuBar1.add(jMenu2); jMenuBar1.add(jMenu3); jMenuBar1.add(jMenu4); jMenuBar1.add(jMenuItem4); jMenu1.add(jMenuItem1); jMenu1.add(jMenuItem2); jMenu1.add(jMenuItem3);
166
jMenuItem1.addActionListener(this); jMenuItem2.addActionListener(this); jMenuItem3.addActionListener(this); jMenuItem4.addActionListener(this); jMenu2.add(jMenuItem5); jMenu2.add(jMenuItem6); jMenu2.add(jMenuItem7); jMenuItem5.addActionListener(this); jMenuItem6.addActionListener(this); jMenuItem7.addActionListener(this); jMenu3.add(jMenuItem8); jMenu3.add(jMenuItem9); jMenu3.add(jMenuItem10); jMenuItem8.addActionListener(this); jMenuItem9.addActionListener(this); jMenuItem10.addActionListener(this); jMenu4.add(jMenuItem11); jMenu4.add(jMenuItem12); jMenuItem11.addActionListener(this); jMenuItem12.addActionListener(this); jPanel1.add(output,BorderLayout.NORTH); jPanel1.add(canvas,BorderLayout.CENTER); output.setRows(5); contentPane.add(jPanel1, BorderLayout.CENTER); contentPane.add(jMenuBar1, BorderLayout.NORTH); readConstants(); displayGrad(); } public void displayGrad() { canvas.repaint();
Pulse pulP0 = new Pulse(TE,p0-20,p0+20,h0); canvas.pulseList.addElement(pulP0);
Pulse pulP1 = new Pulse(TE,p1-20,p1+20,h1); canvas.pulseList.addElement(pulP1); Pulse groA = new Pulse(TE,c2,c3,GROb); canvas.gradRO.addElement(groA); Pulse groB = new Pulse(TE,c8,c9,GROa); canvas.gradRO.addElement(groB); Pulse gssA = new Pulse(TE,c0,c1,GSSa); canvas.gradSS.addElement(gssA); Pulse gssB = new Pulse(TE,c2,c3,GSSb); canvas.gradSS.addElement(gssB); Pulse gssC = new Pulse(TE,c4,c5,GSSc); canvas.gradSS.addElement(gssC); Pulse gssD = new Pulse(TE,c5,c6,GSSd); canvas.gradSS.addElement(gssD); Pulse gssE = new Pulse(TE,c7,c8,GSSe); canvas.gradSS.addElement(gssE);
167
Pulse gssF = new Pulse(TE,c9,c10,GSSf); canvas.gradSS.addElement(gssF); canvas.firstEcho = new Echo(TE, p2, 6); canvas.setBounds(200,200,100,100); canvas.setForeground(Color.white); canvas.setBackground(Color.blue); } // END DisplayGrad public void CalculateDiffusion() { // Setup gradient lists... // i measured in microseconds. double integ1; double integ1prev = 0; double integ2; double integ2prev = 0; double sum = 0; int pn = 1; long i; // Front section... (with 90 pulse) for(i = c0; i < c4; i++) { // for(2) if((i>=c0) && (i<c1)) sum = getGrad(1); else if((i>=c2) && (i<c3)) sum = getGrad(3); else sum = getGrad(4); sum = sum*pn*Gmax/100; integ1 = integ1prev + (sum*TimeInc); integ1prev = integ1; integ2 = integ2prev + (integ1*integ1*TimeInc); integ2prev = integ2; } // END for(2) // Recirculating section... (no 90 pulse) // Start at c4 which is set to zero. for(i = 0; i < (cTE*nEchoes); i++) { // for(3) if((i%cTE) == (c5-c4)) pn *= -1; if(((i%cTE)>=(c4_5-c4)) && (((i%cTE) < (c5-c4)))) sum = getGrad(5); else if (((i%cTE)>=(c5-c4)) && (((i%cTE) < (c6-c4)))) sum = getGrad(6); else if (((i%cTE)>=(c7-c4)) && (((i%cTE) < (c8-c4)))) sum = getGrad(8); else if (((i%cTE)>=(c8-c4)) && (((i%cTE) < (c9-c4)))) sum = getGrad(9); else if (((i%cTE)>=(c9-c4)) && (((i%cTE) < (c10-c4)))) sum = getGrad(10); else sum = getGrad(11); sum = sum*pn*Gmax/100; integ1 = integ1prev + (sum*TimeInc); integ1prev = integ1; integ2 = integ2prev + (integ1*integ1*TimeInc); integ2prev = integ2;
168
if ((i%cTE) == (c8_5-c4)) { graphData.addElement(new Double(integ2*gamma*gamma*D)); } // END if } // END for(3) // Print results of bD to screen... output.append(" \n"); for(i=0; i<nEchoes; i++) { output.append(String.valueOf(((Double) graphData.elementAt((int)i)).doubleValue()) + " "); } // END for try { FileOutputStream stream = new FileOutputStream("calcs.dat"); PrintWriter pW = new PrintWriter(stream,true); for(i=0; i<nEchoes; i++) { pW.println(String.valueOf(((Double) graphData.elementAt((int)i)).doubleValue())); } stream.close(); } catch(IOException ie) { JOptionPane.showMessageDialog(this,"Error opening 'calcs.dat': "
+ ie.toString()); } graphData.removeAllElements(); } // END method: calculateDiffusion. public double getGrad(int loc) { switch(loc) { case 1: switch(inclGrad) { case 1: case 2: case 4: return 0; case 3: case 5: case 6: case 7: return GSSa; } case 2: return 0; case 3: switch(inclGrad) { case 1: case 4: return GROa; case 2: return 0; case 3: case 6: return GSSb; case 5: case 7: return GROa + GSSb; }
169
case 4: return 0; case 5: case 6: switch(inclGrad) { case 1: case 2: case 4: return 0; case 3: case 5: case 6: case 7: return GSSc; } case 7: return 0; case 8: switch(inclGrad) { case 1: return 0; case 2: case 4: return GPHa; case 3: case 5: return GSSe; case 6: case 7: return GSSe + GPHa; } case 9: switch(inclGrad) { case 1: case 4: case 5: case 7: return GROb; case 2: case 3: case 6: return 0; } case 10: switch(inclGrad) { case 1: return 0; case 2: case 4: return GPHb; case 3: case 5: return GSSf; case 6: case 7: return GPHb + GSSf; } case 11: return 0; } // end switch(loc) return 0;
170
} // END getGrad protected void processWindowEvent(WindowEvent e) { super.processWindowEvent(e); if (e.getID() == WindowEvent.WINDOW_CLOSING) { System.exit(0); } } public void actionPerformed(ActionEvent e) { String source = e.getActionCommand(); if(source == "OpenFile") { JFrame fileframe = new frame2(); fileframe.setSize(400,300); fileframe.setLocation(200,100); fileframe.setVisible(true); } else if (source == "SaveFile") { } else if (source == "CloseFile") { } else if (source == "GradCalculate") { readConstants(); CalculateDiffusion(); } else if (source == "GradInclude") { JDialog incGrad = new frame6(Loc); inclGrad = Loc.getValue(); incGrad.setSize(200,300); incGrad.setLocation(200,100); incGrad.setVisible(true); output.append(" Inc: " + Loc.getValue()); } else if (source == "EditParam") { JDialog paramFrame = new frame4(); paramFrame.setSize(200,300); paramFrame.setLocation(200,100); paramFrame.setVisible(true); readConstants(); displayGrad(); } else if (source == "EditGrad") { JDialog gradFrame = new frame5(); gradFrame.setSize(200,300); gradFrame.setLocation(200,100); gradFrame.setVisible(true); readConstants(); displayGrad(); } else if (source == "Constants") { JDialog constFrame = new frame3(); constFrame.setSize(200,150); constFrame.setLocation(200,100); constFrame.setVisible(true); readConstants(); displayGrad(); } } public void readConstants() { // Read in constants...
171
output.setText(""); try { File constsFile = new File("consts.dat"); BufferedReader reader = new BufferedReader(new
FileReader(constsFile)); String line = reader.readLine(); D = Double.parseDouble(line); line = reader.readLine(); Gmax = Double.parseDouble(line); line = reader.readLine(); gamma = Double.parseDouble(line); reader.close(); output.append(String.valueOf(D) + "m^2/s " +
String.valueOf(Gmax) + "T/cm " + String.valueOf(gamma));
} catch (IOException ie) { output.append(ie.toString()); } // Read in parameters... try { File paramsFile = new File("params.dat"); BufferedReader reader = new BufferedReader(new
FileReader(paramsFile)); String line = reader.readLine(); TE = Integer.parseInt(line); line = reader.readLine(); P0 = Integer.parseInt(line); line = reader.readLine(); P1 = Integer.parseInt(line); line = reader.readLine(); aqq = Integer.parseInt(line); line = reader.readLine(); d2 = Integer.parseInt(line); line = reader.readLine(); d3 = Integer.parseInt(line); line = reader.readLine(); d4 = Integer.parseInt(line); line = reader.readLine(); d5 = Integer.parseInt(line); line = reader.readLine(); d12 = Integer.parseInt(line); line = reader.readLine(); d14 = Integer.parseInt(line); line = reader.readLine(); d18 = Integer.parseInt(line); reader.close(); output.append("\n" + String.valueOf(TE) + " " + String.valueOf(P0) + " " + String.valueOf(P1) + " " + String.valueOf(aqq) + " " +
172
String.valueOf(d2) + " " + String.valueOf(d3) + " " + String.valueOf(d4) + " " + String.valueOf(d5) + " " + String.valueOf(d12) + " " + String.valueOf(d14) + " " + String.valueOf(d18)); } catch (IOException ie) { output.append(ie.toString()); } // Read in gradients... try { File gradisFile = new File("gradis.dat"); BufferedReader reader = new BufferedReader(new
FileReader(gradisFile)); String line = reader.readLine(); GROa = Double.parseDouble(line); line = reader.readLine(); GROb = Double.parseDouble(line); line = reader.readLine(); GSSa = Double.parseDouble(line); line = reader.readLine(); GSSb = Double.parseDouble(line); line = reader.readLine(); GSSc = Double.parseDouble(line); line = reader.readLine(); GSSd = Double.parseDouble(line); line = reader.readLine(); GSSe = Double.parseDouble(line); line = reader.readLine(); GSSf = Double.parseDouble(line); reader.close(); output.append("\n" + String.valueOf(GROa) + " " +
String.valueOf(GROb) + " " + String.valueOf(GSSa) + " " + String.valueOf(GSSb) + " " + String.valueOf(GSSc) + " " + String.valueOf(GSSd) + " " + String.valueOf(GSSe) + " " + String.valueOf(GSSf)); } catch (IOException ie) { output.append(ie.toString()); } // Recalculated delays... // Start at half way through 90 pulse... c0 = 0; c1 = P0/2; c2 = c1 + d4; c3 = c2 + d3; c4 = c3 + d4 + d5; // Repeat section for Number of Echoes c4_5 = c4 + 1; c5 = c4_5 + d4 + d18 + P1/2; // First half of 180 pulse c6 = c4_5 + d4 + d18 + P1 + d18; // Second half of 180 pulse c7 = c6 + d4 + 1 + d5; c8 = c7 + d2;
173
c8_5 = c8 + d14 + d14 + aqq/2; c9 = c8 + d14 + d14 + aqq; c10 = c9 + d12 + d12; c11 = c10 + d4 + d5; // END of repeat section. cTE = c11 - c4; output.append(" Calculated TE: " + String.valueOf(cTE) + " "); p0 = (c0 + c1)/2; p1 = (c4 + c6)/2; p2 = (p0 + p1 + p1); } // END ReadConstants } import javax.swing.*; import java.awt.*; import java.awt.event.*; public class frame2 extends JFrame { JFileChooser jFileChooser1 = new JFileChooser(); public frame2() { try { jbInit(); } catch(Exception e) { e.printStackTrace(); } } private void jbInit() throws Exception { this.setTitle("Open File"); jFileChooser1.addActionListener(new
java.awt.event.ActionListener() { public void actionPerformed(ActionEvent e) { jFileChooser1_actionPerformed(e); } }); this.getContentPane().add(jFileChooser1, BorderLayout.NORTH); } void jFileChooser1_actionPerformed(ActionEvent e) { String source = e.getActionCommand(); this.setTitle(source); if (source == "CancelSelection") { this.dispose(); } else if (source == "ApproveSelection") { }
Class: frame2 Author: Christopher Hurley Date: June, 2002
174
} } import javax.swing.*; import java.awt.*; import java.awt.event.*; import java.io.*; public class frame3 extends JDialog { JPanel contentPane; JPanel jPanel1 = new JPanel(); JPanel jPanel2 = new JPanel(); JButton jButton1 = new JButton(); JButton jButton2 = new JButton(); JLabel labelD = new JLabel("D"); JTextField fieldD = new JTextField(" "); JLabel unitsD = new JLabel("m^2/s"); JLabel labelGM = new JLabel("Gmax"); JTextField fieldGM = new JTextField(" "); JLabel unitsGM = new JLabel("T/m"); JLabel labelGA = new JLabel("Gamma"); JTextField fieldGA = new JTextField(" "); JLabel unitsGA = new JLabel("rad/s/T"); BorderLayout borderLayout1 = new BorderLayout(); FlowLayout flowLayout1 = new FlowLayout(); GridBagLayout gridBagLayout1 = new GridBagLayout(); public frame3() { try { jbInit(); } catch(Exception e) { e.printStackTrace(); } } private void jbInit() throws Exception { this.setTitle("Set Constants"); contentPane = (JPanel) this.getContentPane(); contentPane.setLayout(borderLayout1); jButton1.setText("Accept"); jButton1.addActionListener(new java.awt.event.ActionListener() {
Class: frame3 Author: Christopher Hurley Date: June, 2002
175
public void actionPerformed(ActionEvent e) { jButton1_actionPerformed(e); } }); jButton2.setText("Cancel"); jButton2.addActionListener(new java.awt.event.ActionListener() { public void actionPerformed(ActionEvent e) { jButton2_actionPerformed(e); } }); jPanel1.setLayout(flowLayout1); jPanel1.add(jButton1, null); jPanel1.add(jButton2, null); contentPane.add(jPanel2, BorderLayout.CENTER); contentPane.add(jPanel1, BorderLayout.SOUTH); jPanel2.setLayout(gridBagLayout1); GridBagConstraints c[] = new GridBagConstraints[9]; for(int i=0; i<9;i++) { c[i] = new GridBagConstraints(); c[i].gridwidth = 1; c[i].gridheight = 1; c[i].ipadx = 10; c[i].ipady = 1; c[i].gridx = i%3; c[i].gridy = i/3; } jPanel2.add(labelD,c[0]); jPanel2.add(fieldD,c[1]); jPanel2.add(unitsD,c[2]); jPanel2.add(labelGM,c[3]); jPanel2.add(fieldGM,c[4]); jPanel2.add(unitsGM,c[5]); jPanel2.add(labelGA,c[6]); jPanel2.add(fieldGA,c[7]); jPanel2.add(unitsGA,c[8]); } void jButton1_actionPerformed(ActionEvent e) { // Accept // save values... try { FileOutputStream stream = new FileOutputStream("consts.dat"); PrintWriter pW = new PrintWriter(stream,true); pW.println(fieldD.getText().trim()); pW.println(fieldGM.getText().trim()); pW.println(fieldGA.getText().trim()); stream.close(); } catch(IOException ie) { JOptionPane.showMessageDialog(this,"Error opening 'consts.dat':
" + ie.toString()); } // exit
176
this.dispose(); } void jButton2_actionPerformed(ActionEvent e) { // Cancel // exit without saving this.dispose(); } void jTextField1_actionPerformed(ActionEvent e) { } } import javax.swing.*; import java.awt.*; import java.awt.event.*; import java.io.*; public class frame4 extends JDialog { JPanel contentPane; JPanel mainPanel = new JPanel(new BorderLayout()); JPanel buttonPanel = new JPanel(new FlowLayout()); JPanel paramPanel = new JPanel(new GridLayout(11,2)); JPanel descPanel = new JPanel(new FlowLayout()); JButton jButton1 = new JButton(); JButton jButton2 = new JButton(); JLabel labelTE; JLabel labelP0; JLabel labelP1; JLabel labelaqq; JLabel labeld2; JLabel labeld3; JLabel labeld4; JLabel labeld5; JLabel labeld12; JLabel labeld14; JLabel labeld18; JTextField fieldTE; JTextField fieldP0; JTextField fieldP1; JTextField fieldaqq; JTextField fieldd2; JTextField fieldd3; JTextField fieldd4; JTextField fieldd5; JTextField fieldd12; JTextField fieldd14;
Class: frame4 Author: Christopher Hurley Date: June, 2002
177
JTextField fieldd18; public frame4() { try { jbInit(); } catch(Exception e) { e.printStackTrace(); } } private void jbInit() throws Exception { this.setTitle("Set Parameters"); contentPane = (JPanel) this.getContentPane(); mainPanel = (JPanel) this.getContentPane(); jButton1.setText("Accept"); jButton1.addActionListener(new java.awt.event.ActionListener() { public void actionPerformed(ActionEvent e) { jButton1_actionPerformed(e); } }); jButton2.setText("Cancel"); jButton2.addActionListener(new java.awt.event.ActionListener() { public void actionPerformed(ActionEvent e) { jButton2_actionPerformed(e); } }); labelTE = new JLabel("TE"); labelP0 = new JLabel("P0"); labelP1 = new JLabel("P1"); labelaqq = new JLabel("aqq"); labeld2 = new JLabel("d2"); labeld3 = new JLabel("d3"); labeld4 = new JLabel("d4"); labeld5 = new JLabel("d5"); labeld12 = new JLabel("d12"); labeld14 = new JLabel("d14"); labeld18 = new JLabel("d18"); fieldTE = new JTextField(); fieldP0 = new JTextField(); fieldP1 = new JTextField(); fieldaqq = new JTextField(); fieldd2 = new JTextField(); fieldd3 = new JTextField(); fieldd4 = new JTextField(); fieldd5 = new JTextField(); fieldd12 = new JTextField(); fieldd14 = new JTextField(); fieldd18 = new JTextField(); paramPanel.add(labelTE); paramPanel.add(fieldTE); paramPanel.add(labelP0); paramPanel.add(fieldP0); paramPanel.add(labelP1); paramPanel.add(fieldP1); paramPanel.add(labelaqq); paramPanel.add(fieldaqq);
178
paramPanel.add(labeld2); paramPanel.add(fieldd2); paramPanel.add(labeld3); paramPanel.add(fieldd3); paramPanel.add(labeld4); paramPanel.add(fieldd4); paramPanel.add(labeld5); paramPanel.add(fieldd5); paramPanel.add(labeld12); paramPanel.add(fieldd12); paramPanel.add(labeld14); paramPanel.add(fieldd14); paramPanel.add(labeld18); paramPanel.add(fieldd18); buttonPanel.add(jButton1); buttonPanel.add(jButton2); JLabel desc = new JLabel("Enter in the Delays in useconds ..."); descPanel.add(desc); mainPanel.add(buttonPanel,BorderLayout.SOUTH); mainPanel.add(paramPanel,BorderLayout.CENTER); mainPanel.add(descPanel,BorderLayout.NORTH); } void jButton1_actionPerformed(ActionEvent e) { // Accept // save values... try { FileOutputStream stream = new FileOutputStream("params.dat"); PrintWriter pW = new PrintWriter(stream,true); pW.println(fieldTE.getText().trim()); pW.println(fieldP0.getText().trim()); pW.println(fieldP1.getText().trim()); pW.println(fieldaqq.getText().trim()); pW.println(fieldd2.getText().trim()); pW.println(fieldd3.getText().trim()); pW.println(fieldd4.getText().trim()); pW.println(fieldd5.getText().trim()); pW.println(fieldd12.getText().trim()); pW.println(fieldd14.getText().trim()); pW.println(fieldd18.getText().trim()); stream.close(); } catch(IOException ie) { JOptionPane.showMessageDialog(this,"Error opening 'params.dat':
" + ie.toString()); } // exit this.dispose(); // exit this.dispose(); } void jButton2_actionPerformed(ActionEvent e) { // Cancel // exit without saving this.dispose(); } void jTextField1_actionPerformed(ActionEvent e) { }
179
} import javax.swing.*; import java.awt.*; import java.awt.event.*; import java.io.*; public class frame5 extends JDialog { JPanel contentPane; JPanel mainPanel = new JPanel(new BorderLayout()); JPanel buttonPanel = new JPanel(new FlowLayout()); JPanel paramPanel = new JPanel(new GridLayout(8,2)); JPanel descPanel = new JPanel(new FlowLayout()); JButton jButton1 = new JButton(); JButton jButton2 = new JButton(); JTextField fieldGROa; JTextField fieldGROb; JTextField fieldGSSa; JTextField fieldGSSb; JTextField fieldGSSc; JTextField fieldGSSd; JTextField fieldGSSe; JTextField fieldGSSf; public frame5() { try { jbInit(); } catch(Exception e) { e.printStackTrace(); } } private void jbInit() throws Exception { this.setTitle("Set Parameters"); contentPane = (JPanel) this.getContentPane(); mainPanel = (JPanel) this.getContentPane(); jButton1.setText("Accept"); jButton1.addActionListener(new java.awt.event.ActionListener() { public void actionPerformed(ActionEvent e) { jButton1_actionPerformed(e); } }); jButton2.setText("Cancel"); jButton2.addActionListener(new java.awt.event.ActionListener() { public void actionPerformed(ActionEvent e) {
Class: frame5 Author: Christopher Hurley Date: June, 2002
180
jButton2_actionPerformed(e); } }); JLabel labelGROa = new JLabel("GROa"); JLabel labelGROb = new JLabel("GROb"); JLabel labelGSSa = new JLabel("GSSa"); JLabel labelGSSb = new JLabel("GSSb"); JLabel labelGSSc = new JLabel("GSSc"); JLabel labelGSSd = new JLabel("GSSd"); JLabel labelGSSe = new JLabel("GSSe"); JLabel labelGSSf = new JLabel("GSSf"); fieldGROa = new JTextField(); fieldGROb = new JTextField(); fieldGSSa = new JTextField(); fieldGSSb = new JTextField(); fieldGSSc = new JTextField(); fieldGSSd = new JTextField(); fieldGSSe = new JTextField(); fieldGSSf = new JTextField(); paramPanel.add(labelGROa); paramPanel.add(fieldGROa); paramPanel.add(labelGROb); paramPanel.add(fieldGROb); paramPanel.add(labelGSSa); paramPanel.add(fieldGSSa); paramPanel.add(labelGSSb); paramPanel.add(fieldGSSb); paramPanel.add(labelGSSc); paramPanel.add(fieldGSSc); paramPanel.add(labelGSSd); paramPanel.add(fieldGSSd); paramPanel.add(labelGSSe); paramPanel.add(fieldGSSe); paramPanel.add(labelGSSf); paramPanel.add(fieldGSSf); buttonPanel.add(jButton1); buttonPanel.add(jButton2); JLabel desc = new JLabel("Enter in the Gradient Strength % ..."); descPanel.add(desc); mainPanel.add(buttonPanel,BorderLayout.SOUTH); mainPanel.add(paramPanel,BorderLayout.CENTER); mainPanel.add(descPanel,BorderLayout.NORTH); } void jButton1_actionPerformed(ActionEvent e) { // Accept // save values... try { FileOutputStream stream = new FileOutputStream("gradis.dat"); PrintWriter pW = new PrintWriter(stream,true); pW.println(fieldGROa.getText().trim()); pW.println(fieldGROb.getText().trim()); pW.println(fieldGSSa.getText().trim()); pW.println(fieldGSSb.getText().trim()); pW.println(fieldGSSc.getText().trim()); pW.println(fieldGSSd.getText().trim()); pW.println(fieldGSSe.getText().trim()); pW.println(fieldGSSf.getText().trim()); stream.close(); }
181
catch(IOException ie) { JOptionPane.showMessageDialog(this,"Error opening 'gradis.dat':
" + ie.toString()); } // exit this.dispose(); } void jButton2_actionPerformed(ActionEvent e) { // Cancel // exit without saving this.dispose(); } void jTextField1_actionPerformed(ActionEvent e) { } } import javax.swing.*; import java.awt.*; import java.awt.event.*; import java.io.*; public class frame6 extends JDialog { JPanel contentPane; JPanel mainPanel = new JPanel(new BorderLayout()); JPanel buttonPanel = new JPanel(new FlowLayout()); JPanel paramPanel = new JPanel(new GridLayout(3,1)); JPanel descPanel = new JPanel(new FlowLayout()); JButton jButton1 = new JButton(); JButton jButton2 = new JButton(); JCheckBox RO; JCheckBox PH; JCheckBox SS; JLabel labelRO; JLabel labelPH; JLabel labelSS; MyInt Loc; public frame6(MyInt Loc) { this.Loc = Loc; try { jbInit(); } catch(Exception e) { e.printStackTrace();
Class: frame6 Author: Christopher Hurley Date: June, 2002
182
} } // frame6 private void jbInit() throws Exception { this.setTitle("Inclusions"); contentPane = (JPanel) this.getContentPane(); mainPanel = (JPanel) this.getContentPane(); jButton1.setText("Accept"); jButton1.addActionListener(new java.awt.event.ActionListener() { public void actionPerformed(ActionEvent e) { jButton1_actionPerformed(e); } }); jButton2.setText("Cancel"); jButton2.addActionListener(new java.awt.event.ActionListener() { public void actionPerformed(ActionEvent e) { jButton2_actionPerformed(e); } }); JLabel labelGSSf = new JLabel("GSSf"); RO = new JCheckBox("Read Out Gradient"); PH = new JCheckBox("Phase Encode Gradient"); SS = new JCheckBox("Slice Select Gradient"); buttonPanel.add(jButton1); buttonPanel.add(jButton2); paramPanel.add(RO); paramPanel.add(PH); paramPanel.add(SS); JLabel desc = new JLabel("Tick the gradients to include..."); descPanel.add(desc); mainPanel.add(buttonPanel,BorderLayout.SOUTH); mainPanel.add(paramPanel,BorderLayout.CENTER); mainPanel.add(descPanel,BorderLayout.NORTH); } // END jbInit void jButton1_actionPerformed(ActionEvent e) { // Accept if (!RO.isSelected() && !PH.isSelected() && !SS.isSelected()) Loc.setValue(0); else if (RO.isSelected() && !PH.isSelected() && !SS.isSelected()) Loc.setValue(1); else if (!RO.isSelected() && PH.isSelected() && !SS.isSelected()) Loc.setValue(2); else if (!RO.isSelected() && !PH.isSelected() && SS.isSelected()) Loc.setValue(3); else if (RO.isSelected() && PH.isSelected() && !SS.isSelected())
183
Loc.setValue(4); else if (RO.isSelected() && !PH.isSelected() && SS.isSelected()) Loc.setValue(5); else if (!RO.isSelected() && PH.isSelected() && SS.isSelected()) Loc.setValue(6); else if (RO.isSelected() && PH.isSelected() && SS.isSelected()) Loc.setValue(7); else Loc.setValue(0); this.dispose(); } void jButton2_actionPerformed(ActionEvent e) { // Cancel // exit without saving this.dispose(); } void jTextField1_actionPerformed(ActionEvent e) { } } public class GradLine { int time; int data; int integ1; int square; public GradLine() { data = 0; time = 0; } public GradLine(int time, int data) { this.time = time; this.data = data; } } public class MyInt {
Class: GradLine Author: Christopher Hurley Date: June, 2002
Class: MyInt Author: Christopher Hurley Date: June, 2002
184
int value; public MyInt() { value = 0; } public MyInt(int i) { value = i; } public void setValue(int i) { value = i; } public int getValue() { return value; } } public class Pulse { public int start = 0; // in microseconds public int finish = 0; // in microseconds public double height = 0; // in percentage public int te = 0; // in microseconds public boolean enabled = false; public Pulse() { } public Pulse(int te, int start, int finish, double height) { this.te = te; this.start = start; this.finish = finish; this.height = height; enabled = true; } public void show() { enabled = true; } public void hide() { enabled = false; } } public class surface1 extends Canvas { Vector pulseList = new Vector();
Class: Pulse Author: Christopher Hurley Date: June, 2002
Class: surface1 Author: Christopher Hurley Date: June, 2002
185
Vector gradRO = new Vector(); Vector gradSS = new Vector(); Echo firstEcho; public int timeOffset = 50; public surface1() { } public void paint(Graphics g) { Dimension d = getSize(); int start = 0; int finish = 0; g.setColor(Color.yellow); int w = d.width; int h = d.height; g.drawString("PULSE",5,h/4); g.drawString("GRO",5,h/2); g.drawString("GSS",5,3*h/4); if(pulseList.isEmpty()) { g.drawLine(0, h/4, w, h/4); } else { for(int i=0; i<pulseList.size();i++) {
finish = drawPulse(g,(Pulse)pulseList.elementAt(i),h,w,start,h/4);
start = finish; } // Draw Echo... int eSF = drawEcho(g,w,h/4); g.drawLine(start,h/4,eSF-30,h/4); g.drawLine(eSF+30,h/4,w,h/4); } start = 0; if(gradRO.isEmpty()) { g.drawLine(0, h/2, w, h/2); } else { for(int i=0; i<gradRO.size();i++) { finish =
drawPulse(g,(Pulse)gradRO.elementAt(i),h,w,start,h/2); start = finish; } g.drawLine(start,h/2,w,h/2); } start = 0; if(gradSS.isEmpty()) { g.drawLine(0, 3*h/4, w, 3*h/4); } else { for(int i=0; i<gradSS.size();i++) { finish =
drawPulse(g,(Pulse)gradSS.elementAt(i),h,w,start,3*h/4);
start = finish; } g.drawLine(start,3*h/4,w,3*h/4);
186
} } public int drawPulse(Graphics g, Pulse pulse, int h, int w, int
lastF, int line) { int dh = (int)((h*pulse.height)/100); int te = pulse.te; int dstart = pulse.start; int dfinish = pulse.finish; int ds = w*dstart/((int) (1.5*te)) + timeOffset; int df = w*dfinish/((int)(1.5*te)) + timeOffset; g.drawLine(lastF,line,ds,line); g.drawLine(ds,line,ds,line - dh); g.drawLine(ds,line-dh,df,line-dh); g.drawLine(df,line,df,line-dh); return df; } public int drawEcho(Graphics g, int w, int line) { int x = (firstEcho.echotime*w/((int) (1.5 * firstEcho.te))) +
timeOffset - 60; int y = line - 60; g.drawArc(x-30,y+60,60,60,0,90); g.drawArc(x-30,y,60,60,0,-90); g.drawArc(x+30,y,60,60,-90,-90); g.drawArc(x+30,y+60,60,60,90,90); return firstEcho.echotime*w/((int)(1.5*firstEcho.te))+timeOffset-
30; } }
