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

Title: SU-E-T-399: Evaluation of Selection Criteria for Computational Human Phantoms for Use in Out-Of-Field Organ Dosimetry for Radiotherapy Patients

Abstract

Purpose: To quantify the dosimetric uncertainty due to organ position errors when using height and weight as phantom selection criteria in the UF/NCI Hybrid Phantom Library for the purpose of out-of-field organ dose reconstruction. Methods: Four diagnostic patient CT images were used to create 7-field IMRT plans. For each patient, dose to the liver, right lung, and left lung were calculated using the XVMC Monte Carlo code. These doses were taken to be the ground truth. For each patient, the phantom with the most closely matching height and weight was selected from the body size dependent phantom library. The patient plans were then transferred to the computational phantoms and organ doses were recalculated. Each plan was also run on 4 additional phantoms with reference heights and or weights. Maximum and mean doses for the three organs were computed, and the DVHs were extracted and compared. One sample t-tests were performed to compare the accuracy of the height and weight matched phantoms against the additional phantoms in regards to both maximum and mean dose. Results: For one of the patients, the height and weight matched phantom yielded the most accurate results across all three organs for both maximum and mean doses.more » For two additional patients, the matched phantom yielded the best match for one organ only. In 13 of the 24 cases, the matched phantom yielded better results than the average of the other four phantoms, though the results were only statistically significant at the .05 level for three cases. Conclusion: Using height and weight matched phantoms does yield better results in regards to out-of-field dosimetry than using average phantoms. Height and weight appear to be moderately good selection criteria, though this selection criteria failed to yield any better results for one patient.« less

Authors:
;  [1];  [2]; ;  [3];  [4]
  1. East Carolina University, Greenville, NC (United States)
  2. University of Michigan, Ann Arbor, MI (United States)
  3. National Cancer Institute, Rockville, MD (United States)
  4. University of Pittsburgh Medical Center, Pittsburgh, PA (United States)
Publication Date:
OSTI Identifier:
22548446
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; 61 RADIATION PROTECTION AND DOSIMETRY; ACCURACY; COMPARATIVE EVALUATIONS; COMPUTERIZED TOMOGRAPHY; DOSIMETRY; ERRORS; GROUND TRUTH MEASUREMENTS; IMAGE PROCESSING; LIVER; LUNGS; MONTE CARLO METHOD; PATIENTS; PHANTOMS; RADIATION DOSES; RADIOTHERAPY

Citation Formats

Pelletier, C, Jung, J, Lee, C, Pyakuryal, A, Lee, C, and Kim, J. SU-E-T-399: Evaluation of Selection Criteria for Computational Human Phantoms for Use in Out-Of-Field Organ Dosimetry for Radiotherapy Patients. United States: N. p., 2015. Web. doi:10.1118/1.4924760.
Pelletier, C, Jung, J, Lee, C, Pyakuryal, A, Lee, C, & Kim, J. SU-E-T-399: Evaluation of Selection Criteria for Computational Human Phantoms for Use in Out-Of-Field Organ Dosimetry for Radiotherapy Patients. United States. doi:10.1118/1.4924760.
Pelletier, C, Jung, J, Lee, C, Pyakuryal, A, Lee, C, and Kim, J. 2015. "SU-E-T-399: Evaluation of Selection Criteria for Computational Human Phantoms for Use in Out-Of-Field Organ Dosimetry for Radiotherapy Patients". United States. doi:10.1118/1.4924760.
@article{osti_22548446,
title = {SU-E-T-399: Evaluation of Selection Criteria for Computational Human Phantoms for Use in Out-Of-Field Organ Dosimetry for Radiotherapy Patients},
author = {Pelletier, C and Jung, J and Lee, C and Pyakuryal, A and Lee, C and Kim, J},
abstractNote = {Purpose: To quantify the dosimetric uncertainty due to organ position errors when using height and weight as phantom selection criteria in the UF/NCI Hybrid Phantom Library for the purpose of out-of-field organ dose reconstruction. Methods: Four diagnostic patient CT images were used to create 7-field IMRT plans. For each patient, dose to the liver, right lung, and left lung were calculated using the XVMC Monte Carlo code. These doses were taken to be the ground truth. For each patient, the phantom with the most closely matching height and weight was selected from the body size dependent phantom library. The patient plans were then transferred to the computational phantoms and organ doses were recalculated. Each plan was also run on 4 additional phantoms with reference heights and or weights. Maximum and mean doses for the three organs were computed, and the DVHs were extracted and compared. One sample t-tests were performed to compare the accuracy of the height and weight matched phantoms against the additional phantoms in regards to both maximum and mean dose. Results: For one of the patients, the height and weight matched phantom yielded the most accurate results across all three organs for both maximum and mean doses. For two additional patients, the matched phantom yielded the best match for one organ only. In 13 of the 24 cases, the matched phantom yielded better results than the average of the other four phantoms, though the results were only statistically significant at the .05 level for three cases. Conclusion: Using height and weight matched phantoms does yield better results in regards to out-of-field dosimetry than using average phantoms. Height and weight appear to be moderately good selection criteria, though this selection criteria failed to yield any better results for one patient.},
doi = {10.1118/1.4924760},
journal = {Medical Physics},
number = 6,
volume = 42,
place = {United States},
year = 2015,
month = 6
}
  • Currently, two classes of the computational phantoms have been developed for dosimetry calculation: (1) stylized (or mathematical) and (2) voxel (or tomographic) phantoms describing human anatomy through mathematical surface equations and three-dimensional labeled voxel matrices, respectively. Mathematical surface equations in stylized phantoms provide flexibility in phantom design and alteration, but the resulting anatomical description is, in many cases, not very realistic. Voxel phantoms display far better anatomical realism, but they are limited in terms of their ability to alter organ shape, position, and depth, as well as body posture. A new class of computational phantoms - called hybrid phantoms -more » takes advantage of the best features of stylized and voxel phantoms - flexibility and anatomical realism, respectively. In the current study, hybrid computational phantoms representing reference 15-year male and female body anatomy and anthropometry are presented. For the male phantom, organ contours were extracted from the University of Florida (UF) 14-year series B male voxel phantom, while for the female phantom, original computed tomography (CT) data from two 14-year female patients were used. Polygon mesh models for the major organs and tissues were reconstructed for nonuniform rational B-spline (NURBS) surface modeling. The resulting NURBS/polygon mesh models representing body contour and internal anatomy were matched to anthropometric data and reference organ mass data provided by the Centers for Disease Control and Prevention (CDC) and the International Commission on Radiation Protection (ICRP), respectively. Finally, two hybrid 15-year male and female phantoms were completed where a total of eight anthropometric data categories were matched to standard values within 4% and organ masses matched to ICRP data within 1% with the exception of total skin. To highlight the flexibility of the hybrid phantoms, 10th and 90th weight percentile 15-year male and female phantoms were further developed from the 50th percentile phantoms through adjustments in the body contour to match the total body masses given in CDC pediatric growth curves. The resulting six NURBS phantoms, male and female phantoms representing their 10th, 50th, and 90th weight percentiles, were used to investigate the influence of body fat distributions on internal organ doses following CT imaging. The phantoms were exposed to multislice chest and abdomen helical CT scans, and in-field organ absorbed doses were calculated. The results demonstrated that the use of traditional stylized phantoms yielded organ dose estimates that deviate from those given by the UF reference hybrid phantoms by up to a factor of 2. The study also showed that use of reference, or 50th percentile, phantoms to assess organ doses in underweight 15-year-old children would not lead to significant organ dose errors (typically less than 10%). However, more significant errors were noted (up to {approx}30%) when reference phantoms are used to represent overweight children in CT imaging dosimetry. These errors are expected to only further increase as one considers CT organ doses in overweight and obese individuals of the adult patient population, thus emphasizing the advantages of patient-sculptable phantom technology.« less
  • Purpose: Recent advances in cancer treatments have greatly increased the likelihood of post-treatment patient survival. Secondary malignancies, however, have become a growing concern. Epidemiological studies determining secondary effects in radiotherapy patients require assessment of organ-specific dose both inside and outside the treatment field. An essential input for Monte Carlo modeling of particle transport is radiological images showing full patient anatomy. However, in retrospective studies it is typical to only have partial anatomy from CT scans used during treatment planning. In this study, we developed a multi-step method to extend such limited patient anatomy to full body anatomy for estimating dosemore » to normal tissues located outside the CT scan coverage. Methods: The first step identified a phantom from a library of body size-dependent computational human phantoms by matching the height and weight of patients. Second, a Python algorithm matched the patient CT coverage location in relation to the whole body phantom. Third, an algorithm cut the whole body phantom and scaled them to match the size of the patient. Then, merged the two anatomies into one whole body. We entitled this new approach, Anatomically Predictive Extension (APE). Results: The APE method was examined by comparing the original chest-abdomen-pelvis CT images of the five patients with the APE phantoms developed from only the chest part of the CAP images and whole body phantoms. We achieved average percent differences of tissue volumes of 25.7%, 34.2%, 16.5%, 26.8%, and 31.6% with an average of 27% across all patients. Conclusion: Our APE method extends the limited CT patient anatomy to whole body anatomy by using image processing and computational human phantoms. Our ongoing work includes evaluating the accuracy of these APE phantoms by comparing normal tissue doses in the APE phantoms and doses calculated for the original full CAP images under generic radiotherapy simulations. This research was supported by the NIH Intramural Research Program.« less
  • Following the previously developed Korean tomographic phantom, KORMAN, two additional whole-body tomographic phantoms of Korean adult males were developed from magnetic resonance (MR) and computed tomography (CT) images, respectively. Two healthy male volunteers, whose body dimensions were fairly representative of the average Korean adult male, were recruited and scanned for phantom development. Contiguous whole body MR images were obtained from one subject exclusive of the arms, while whole-body CT images were acquired from the second individual. A total of 29 organs and tissues and 19 skeletal sites were segmented via image manipulation techniques such as gray-level thresholding, region growing, andmore » manual drawing, in which each of segmented image slice was subsequently reviewed by an experienced radiologist for anatomical accuracy. The resulting phantoms, the MR-based KTMAN-1 (Korean Typical MAN-1) and the CT-based KTMAN-2 (Korean Typical MAN-2), consist of 300x150x344 voxels with a voxel resolution of 2x2x5 mm{sup 3} for both phantoms. Masses of segmented organs and tissues were calculated as the product of a nominal reference density, the prevoxel volume, and the cumulative number of voxels defining each organs or tissue. These organs masses were then compared with those of both the Asian and the ICRP reference adult male. Organ masses within both KTMAN-1 and KTMAN-2 showed differences within 40% of Asian and ICRP reference values, with the exception of the skin, gall bladder, and pancreas which displayed larger differences. The resulting three-dimensional binary file was ported to the Monte Carlo code MCNPX2.4 to calculate organ doses following external irradiation for illustrative purposes. Colon, lung, liver, and stomach absorbed doses, as well as the effective dose, for idealized photon irradiation geometries (anterior-posterior and right lateral) were determined, and then compared with data from two other tomographic phantoms (Asian and Caucasian), and stylized ORNL phantom. The armless KTMAN-1 can be applied to dosimetry for computed tomography or lateral x-ray examination, while the whole body KTMAN-2 can be used for radiation protection dosimetry.« less
  • Purpose: ImpactMC (CT Imaging, Erlangen, Germany) is a Monte Carlo (MC) software package that offers a GPU enabled, user definable and validated method for 3D dose distribution calculations for radiography and Computed Tomography (CT). ImpactMC, in and of itself, offers limited capabilities to perform batch simulations. The aim of this work was to develop a framework for the batch simulation of absorbed organ dose distributions from CT scans of computational voxel phantoms. Methods: The ICRP 110 adult Reference Male and Reference Female computational voxel phantoms were formatted into compatible input volumes for MC simulations. A Matlab (The MathWorks Inc., Natick,more » MA) script was written to loop through a user defined set of simulation parameters and 1) generate input files required for the simulation, 2) start the MC simulation, 3) segment the absorbed dose for organs in the simulated dose volume and 4) transfer the organ doses to a database. A demonstration of the framework is made where the glandular breast dose to the adult Reference Female phantom, for a typical Chest CT examination, is investigated. Results: A batch of 48 contiguous simulations was performed with variations in the total collimation and spiral pitch. The demonstration of the framework showed that the glandular dose to the right and left breast will vary depending on the start angle of rotation, total collimation and spiral pitch. Conclusion: The developed framework provides a robust and efficient approach to performing a large number of user defined MC simulations with computational voxel phantoms in CT (minimal user interaction). The resulting organ doses from each simulation can be accessed through a database which greatly increases the ease of analyzing the resulting organ doses. The framework developed in this work provides a valuable resource when investigating different dose optimization strategies in CT.« less
  • Estimating organ residence times is an essential part of patient-specific dosimetry for radioimmunotherapy (RIT). Quantitative imaging methods for RIT are often evaluated using a single physical or simulated phantom but are intended to be applied clinically where there is variability in patient anatomy, biodistribution, and biokinetics. To provide a more relevant evaluation, the authors have thus developed a population of phantoms with realistic variations in these factors and applied it to the evaluation of quantitative imaging methods both to find the best method and to demonstrate the effects of these variations. Using whole body scans and SPECT/CT images, organ shapesmore » and time-activity curves of 111In ibritumomab tiuxetan were measured in dosimetrically important organs in seven patients undergoing a high dose therapy regimen. Based on these measurements, we created a 3D NURBS-based cardiac-torso (NCAT)-based phantom population. SPECT and planar data at realistic count levels were then simulated using previously validated Monte Carlo simulation tools. The projections from the population were used to evaluate the accuracy and variation in accuracy of residence time estimation methods that used a time series of SPECT and planar scans. Quantitative SPECT (QSPECT) reconstruction methods were used that compensated for attenuation, scatter, and the collimator-detector response. Planar images were processed with a conventional (CPlanar) method that used geometric mean attenuation and triple-energy window scatter compensation and a quantitative planar (QPlanar) processing method that used model-based compensation for image degrading effects. Residence times were estimated from activity estimates made at each of five time points. The authors also evaluated hybrid methods that used CPlanar or QPlanar time-activity curves rescaled to the activity estimated from a single QSPECT image. The methods were evaluated in terms of mean relative error and standard deviation of the relative error in the residence time estimates taken over the phantom population. The mean errors in the residence time estimates over all the organs were <9.9% (pure QSPECT), <13.2% (pure QPLanar), <7.2% (hybrid QPlanar/QSPECT), <19.2% (hybrid CPlanar/QSPECT), and 7%-159% (pure CPlanar). The standard deviations of the errors for all the organs over all the phantoms were <9.9%, <11.9%, <10.8%, <22.0%, and <107.9% for the same methods, respectively. The processing methods differed both in terms of their average accuracy and the variation of the accuracy over the population of phantoms, thus demonstrating the importance of using a phantom population in evaluating quantitative imaging methods. Hybrid CPlanar/QSPECT provided improved accuracy compared to pure CPlanar and required the addition of only a single SPECT acquisition. The QPlanar or hybrid QPlanar/QSPECT methods had mean errors and standard deviations of errors that approached those of pure QSPECT while providing simplified image acquisition protocols, and thus may be more clinically practical.« less