skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: SU-E-J-147: Monte Carlo Study of the Precision and Accuracy of Proton CT Reconstructed Relative Stopping Power Maps

Abstract

Purpose: The quantification of the intrinsic performances of proton computed tomography (pCT) as a modality for treatment planning in proton therapy. The performance of an ideal pCT scanner is studied as a function of various parameters. Methods: Using GATE/Geant4, we simulated an ideal pCT scanner and scans of several cylindrical phantoms with various tissue equivalent inserts of different sizes. Insert materials were selected in order to be of clinical relevance. Tomographic images were reconstructed using a filtered backprojection algorithm taking into account the scattering of protons into the phantom. To quantify the performance of the ideal pCT scanner, we study the precision and the accuracy with respect to the theoretical relative stopping power ratios (RSP) values for different beam energies, imaging doses, insert sizes and detector positions. The planning range uncertainty resulting from the reconstructed RSP is also assessed by comparison with the range of the protons in the analytically simulated phantoms. Results: The results indicate that pCT can intrinsically achieve RSP resolution below 1%, for most examined tissues at beam energies below 300 MeV and for imaging doses around 1 mGy. RSP maps accuracy of less than 0.5 % is observed for most tissue types within the studied dosemore » range (0.2–1.5 mGy). Finally, the uncertainty in the proton range due to the accuracy of the reconstructed RSP map is well below 1%. Conclusion: This work explores the intrinsic performance of pCT as an imaging modality for proton treatment planning. The obtained results show that under ideal conditions, 3D RSP maps can be reconstructed with an accuracy better than 1%. Hence, pCT is a promising candidate for reducing the range uncertainties introduced by the use of X-ray CT alongside with a semiempirical calibration to RSP.Supported by the DFG Cluster of Excellence Munich-Centre for Advanced Photonics (MAP)« less

Authors:
; ;  [1];  [2]; ;  [3]; ;  [4]
  1. Ludwig Maximilians University Munich, Garching, DE (Germany)
  2. Universite de Strasbourg, Strasbourg (France)
  3. Universite Lyon 1, Institut de Physique Nucleaire de Lyon, Lyon (France)
  4. Universite Lyon 1, INSA Lyon, CREATIS, Lyon (France)
Publication Date:
OSTI Identifier:
22494158
Resource Type:
Journal Article
Resource Relation:
Journal Name: Medical Physics; Journal Volume: 42; Journal Issue: 6; Other Information: (c) 2015 American Association of Physicists in Medicine; Country of input: International Atomic Energy Agency (IAEA)
Country of Publication:
United States
Language:
English
Subject:
60 APPLIED LIFE SCIENCES; ACCURACY; ALGORITHMS; BIOMEDICAL RADIOGRAPHY; MONTE CARLO METHOD; PERFORMANCE; PHANTOMS; PLANNING; PROTON BEAMS; PROTON COMPUTED TOMOGRAPHY; RADIATION DOSES; RADIOTHERAPY; STOPPING POWER

Citation Formats

Dedes, G, Asano, Y, Parodi, K, Arbor, N, Dauvergne, D, Testa, E, Letang, J, and Rit, S. SU-E-J-147: Monte Carlo Study of the Precision and Accuracy of Proton CT Reconstructed Relative Stopping Power Maps. United States: N. p., 2015. Web. doi:10.1118/1.4924232.
Dedes, G, Asano, Y, Parodi, K, Arbor, N, Dauvergne, D, Testa, E, Letang, J, & Rit, S. SU-E-J-147: Monte Carlo Study of the Precision and Accuracy of Proton CT Reconstructed Relative Stopping Power Maps. United States. doi:10.1118/1.4924232.
Dedes, G, Asano, Y, Parodi, K, Arbor, N, Dauvergne, D, Testa, E, Letang, J, and Rit, S. Mon . "SU-E-J-147: Monte Carlo Study of the Precision and Accuracy of Proton CT Reconstructed Relative Stopping Power Maps". United States. doi:10.1118/1.4924232.
@article{osti_22494158,
title = {SU-E-J-147: Monte Carlo Study of the Precision and Accuracy of Proton CT Reconstructed Relative Stopping Power Maps},
author = {Dedes, G and Asano, Y and Parodi, K and Arbor, N and Dauvergne, D and Testa, E and Letang, J and Rit, S},
abstractNote = {Purpose: The quantification of the intrinsic performances of proton computed tomography (pCT) as a modality for treatment planning in proton therapy. The performance of an ideal pCT scanner is studied as a function of various parameters. Methods: Using GATE/Geant4, we simulated an ideal pCT scanner and scans of several cylindrical phantoms with various tissue equivalent inserts of different sizes. Insert materials were selected in order to be of clinical relevance. Tomographic images were reconstructed using a filtered backprojection algorithm taking into account the scattering of protons into the phantom. To quantify the performance of the ideal pCT scanner, we study the precision and the accuracy with respect to the theoretical relative stopping power ratios (RSP) values for different beam energies, imaging doses, insert sizes and detector positions. The planning range uncertainty resulting from the reconstructed RSP is also assessed by comparison with the range of the protons in the analytically simulated phantoms. Results: The results indicate that pCT can intrinsically achieve RSP resolution below 1%, for most examined tissues at beam energies below 300 MeV and for imaging doses around 1 mGy. RSP maps accuracy of less than 0.5 % is observed for most tissue types within the studied dose range (0.2–1.5 mGy). Finally, the uncertainty in the proton range due to the accuracy of the reconstructed RSP map is well below 1%. Conclusion: This work explores the intrinsic performance of pCT as an imaging modality for proton treatment planning. The obtained results show that under ideal conditions, 3D RSP maps can be reconstructed with an accuracy better than 1%. Hence, pCT is a promising candidate for reducing the range uncertainties introduced by the use of X-ray CT alongside with a semiempirical calibration to RSP.Supported by the DFG Cluster of Excellence Munich-Centre for Advanced Photonics (MAP)},
doi = {10.1118/1.4924232},
journal = {Medical Physics},
number = 6,
volume = 42,
place = {United States},
year = {Mon Jun 15 00:00:00 EDT 2015},
month = {Mon Jun 15 00:00:00 EDT 2015}
}
  • Purpose: In standard proton therapy clinical practice, proton stopping power uncertainties are in the order of 3.5%, which affects the ability of placing the proton Bragg peak at the edge of the tumor. The innovating idea of this project is to approach the uncertainty problem in RSP by using combined information from X-ray CT and proton radiography along a few beam angles. In addition, this project aims to quantify the systematic error introduced by the theoretical models (Janni, ICRU49, Bischel) for proton stopping power in media. Methods: A 3D phantom of 36 cm3 composed of 9 materials randomly placed ismore » created. Measured RSP values are obtained using a Gammex phantom with a proton beam. Theoretical RSP values are calculated with Beth-Block equation in combination with three databases (Janni, ICRU49 and Bischel). Clinical RSP errors are simulated by introducing a systematic (1.5%, 2.5%, 3.5%) and a random error (+/−0.5%) to the theoretical RSP. A ray-tracing algorithm uses each of these RSP tables to calculate energy loss for proton crossing the phantom through various directions. For each direction, gradient descent (GD) method is done on the clinical RSP table to minimize the residual energy difference between the simulation with clinical RSP and with theoretical RSP. The possibility of a systematic material dependent error is investigated by comparing measured RSP to theoretical RSP as calculated from the three models. Results: Using 10,000 iterations on GD algorithm, RSP differences between theoretical values and clinical RSP have converged (<1%) for each error introduced. Results produced with ICRU49 have the smallest average difference (0.021%) to the measured RSP. Janni (1.168%) and Bischel (−0.372%) database shows larger systematic errors. Conclusion: Based on these results, ray-tracing optimisation using information from proton radiography and X-ray CT demonstrates a potential to improve the proton range accuracy in a clinical environment.« less
  • Purpose: To determine the suitability of dual-energy CT (DECT) to calculate relative electron density (RED) of tissues for accurate proton therapy dose calculation. Methods: DECT images of RED tissue surrogates were acquired at 80 and 140 kVp. Samples (RED=0.19−2.41) were imaged in a water-equivalent phantom in a variety of configurations. REDs were calculated using the DECT numbers and inputs of the high and low energy spectral weightings. DECT-derived RED was compared between geometric configurations and for variations in the spectral inputs to assess the sensitivity of RED accuracy versus expected values. Results: RED accuracy was dependent on accurate spectral inputmore » influenced by phantom thickness and radius from the phantom center. Material samples located at the center of the phantom generally showed the best agreement to reference RED values, but only when attenuation of the surrounding phantom thickness was accounted for in the calculation spectra. Calculated RED changed by up to 10% for some materials when the sample was located at an 11 cm radius from the phantom center. Calculated REDs under the best conditions still differed from reference values by up to 5% in bone and 14% in lung. Conclusion: DECT has previously been used to differentiate tissue types based on RED and Z for binary tissue-type segmentation. To improve upon the current standard of empirical conversion of CT number to RED for treatment planning dose calculation, DECT methods must be able to calculate RED to better than 3% accuracy throughout the image. The DECT method is sensitive to the accuracy of spectral inputs used for calculation, as well as to spatial position in the anatomy. Effort to address adjustments to the spectral calculation inputs based on position and phantom attenuation will be required before DECT-determined RED can achieve a consistent level of accuracy for application in dose calculation.« less
  • Purpose: To develop a calibration curve that includes and minimizes the variations of Hounsfield Unit (HU) from a CT scanner to Relative Stopping Power (RSP) of tissues along the proton beam path. The variations are due to scanner and proton energy, technique, phantom size and placement, and tissue arrangement. Methods: A CIRS 062 M phantom with 10 plugs of known relative electron density (RED) was scanned through a 16 slice GE Discovery CT Simulator scanner. Three setup combinations of plug distributions and techniques clinically implemented for five treatment regions were scanned with energies of 100, 120, and 140 kV. Volumetricmore » HU values were measured for each plug and scan. The RSP values derived through the Bethe-Bloch formula are currently being verified with parallel-plate ionization chamber measurements in water using 80, 150, and 225 MeV proton beam. Typical treatment plans for treatment regions of brain, head-&-neck, chest, abdomen, and pelvis are being planned and dose delivered will be compared with film and Optically Stimulated Luminescence (OSL) measurements. Results: Percentage variations were determined for each variable. For tissues close to water, variations were <1% from any given parameter. Tissues far from water equivalence (lung and bone) showed the greatest sensitivity to change (7.4% maximum) with scanner energy and up to 5.3% with positioning of the phantom. No major variations were observed for proton energies within the treatment range. Conclusion: When deriving a calibration curve, attention should be placed to low and high HU values. A thorough verification process of calculated vs. water-phantom measured RSP values at different proton energies, followed by dose validation of planned vs. measured doses in phantom with film and OSL detectors are currently being undertaken.« less
  • Purpose: The purpose of this study is to evaluate any effects of converted CT density variation in treatment planning system (TPS) of spot scanning proton therapy with an IROC proton prostate phantom at our new ProteusOne Proton Therapy Center. Methods: A proton prostate phantom was requested from the Imaging and Radiation Oncology Core Houston (IROC), The University of Texas MD Anderson Cancer Center, Houston, TX, where GAF Chromic films and couples of thermo luminescent dosemeter (TLD) capsules in target and adjacent structures were embedded for imaging and dose monitoring. Various material such as PVC, PBT HI polystyrene as dosimetry insertsmore » and acrylic were within phantom. Relative stopping power (SP) were provided. However our treatment planning system (TPS) doesn’t require SP instead relative density was converted relative to water in TPS. Phantom was irradiated and the results were compared with IROC measurements. The range of relative density was converted from SP into relative density of water as a new assigned material and tested. Results: The summary of TLD measurements of the prostate and femoral heads were well within 2% of the TPS and met the criteria established by IROC. The film at coronal plane was found to be shift in superior-inferior direction due to locking position of cylinder insert was off and was corrected. The converted CT density worked precisely to correlated relative stopping power. Conclusion: The proton prostate phantom provided by IROC is a useful methodology to evaluate our new commissioned proton pencil beam and TPS within certain confidence in proton therapy. The relative stopping power was converted into relative physical density relatively to water and the results were satisfied.« less
  • Purpose: This research investigates application of secondary prompt gamma (PG) emission spectra, resulting from nuclear reactions induced by protons, to characterize tissue composition along the particle path. The objective of utilizing the intensity of discrete high-energy peaks of PG is to improve the accuracy of relative stopping power (RSP) values available for proton therapy treatment planning on a patient specific basis and to reduce uncertainty in dose depth calculations. Methods: In this research, MCNP6 was used to simulate PG emission spectra generated from proton induced nuclear reactions in medium of varying composition of carbon, oxygen, calcium and nitrogen, the predominantmore » elements found in human tissue. The relative peak intensities at discrete energies predicted by MCNP6 were compared to the corresponding atomic composition of the medium. Results: The results have shown a good general agreement with experimentally measured values reported by other investigators. Unexpected divergence from experimental spectra was noted in the peak intensities for some cases depending on the source of the cross-section data when using compiled proton table libraries vs. physics models built into MCNP6. While the use of proton cross-section libraries is generally recommended when available, these libraries lack data for several less abundant isotopes. This limits the range of their applicability and forces the simulations to rely on physics models for reactions with natural atomic compositions. Conclusion: Current end-of-range proton imaging provides an average RSP for the total estimated track length. The accurate identification of tissue composition along the incident particle path using PG detection and characterization allows for improved determination of the tissue RSP on the local level. While this would allow for more accurate depth calculations resulting in tighter treatment margins, precise understanding of proton beam behavior in tissue of various compositions is necessary requiring detailed simulations with a high degree of accuracy.« less