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

Title: SU-F-T-450: The Investigation of Radiotherapy Quality Assurance and Automatic Treatment Planning Based On the Kernel Density Estimation Method

Abstract

Purpose: To develop a fast automatic algorithm based on the two dimensional kernel density estimation (2D KDE) to predict the dose-volume histogram (DVH) which can be employed for the investigation of radiotherapy quality assurance and automatic treatment planning. Methods: We propose a machine learning method that uses previous treatment plans to predict the DVH. The key to the approach is the framing of DVH in a probabilistic setting. The training consists of estimating, from the patients in the training set, the joint probability distribution of the dose and the predictive features. The joint distribution provides an estimation of the conditional probability of the dose given the values of the predictive features. For the new patient, the prediction consists of estimating the distribution of the predictive features and marginalizing the conditional probability from the training over this. Integrating the resulting probability distribution for the dose yields an estimation of the DVH. The 2D KDE is implemented to predict the joint probability distribution of the training set and the distribution of the predictive features for the new patient. Two variables, including the signed minimal distance from each OAR (organs at risk) voxel to the target boundary and its opening angle with respectmore » to the origin of voxel coordinate, are considered as the predictive features to represent the OAR-target spatial relationship. The feasibility of our method has been demonstrated with the rectum, breast and head-and-neck cancer cases by comparing the predicted DVHs with the planned ones. Results: The consistent result has been found between these two DVHs for each cancer and the average of relative point-wise differences is about 5% within the clinical acceptable extent. Conclusion: According to the result of this study, our method can be used to predict the clinical acceptable DVH and has ability to evaluate the quality and consistency of the treatment planning.« less

Authors:
; ; ;  [1]
  1. Fudan University Shanghai Cancer Center, Shanghai, Shanghai (China)
Publication Date:
OSTI Identifier:
22649041
Resource Type:
Journal Article
Resource Relation:
Journal Name: Medical Physics; Journal Volume: 43; Journal Issue: 6; Other Information: (c) 2016 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; KERNELS; MAMMARY GLANDS; PATIENTS; PLANNING; PRODUCTIVITY; QUALITY ASSURANCE; RADIATION DOSES; RADIOTHERAPY; TRAINING

Citation Formats

Fan, J, Fan, J, Hu, W, and Wang, J. SU-F-T-450: The Investigation of Radiotherapy Quality Assurance and Automatic Treatment Planning Based On the Kernel Density Estimation Method. United States: N. p., 2016. Web. doi:10.1118/1.4956635.
Fan, J, Fan, J, Hu, W, & Wang, J. SU-F-T-450: The Investigation of Radiotherapy Quality Assurance and Automatic Treatment Planning Based On the Kernel Density Estimation Method. United States. doi:10.1118/1.4956635.
Fan, J, Fan, J, Hu, W, and Wang, J. 2016. "SU-F-T-450: The Investigation of Radiotherapy Quality Assurance and Automatic Treatment Planning Based On the Kernel Density Estimation Method". United States. doi:10.1118/1.4956635.
@article{osti_22649041,
title = {SU-F-T-450: The Investigation of Radiotherapy Quality Assurance and Automatic Treatment Planning Based On the Kernel Density Estimation Method},
author = {Fan, J and Fan, J and Hu, W and Wang, J},
abstractNote = {Purpose: To develop a fast automatic algorithm based on the two dimensional kernel density estimation (2D KDE) to predict the dose-volume histogram (DVH) which can be employed for the investigation of radiotherapy quality assurance and automatic treatment planning. Methods: We propose a machine learning method that uses previous treatment plans to predict the DVH. The key to the approach is the framing of DVH in a probabilistic setting. The training consists of estimating, from the patients in the training set, the joint probability distribution of the dose and the predictive features. The joint distribution provides an estimation of the conditional probability of the dose given the values of the predictive features. For the new patient, the prediction consists of estimating the distribution of the predictive features and marginalizing the conditional probability from the training over this. Integrating the resulting probability distribution for the dose yields an estimation of the DVH. The 2D KDE is implemented to predict the joint probability distribution of the training set and the distribution of the predictive features for the new patient. Two variables, including the signed minimal distance from each OAR (organs at risk) voxel to the target boundary and its opening angle with respect to the origin of voxel coordinate, are considered as the predictive features to represent the OAR-target spatial relationship. The feasibility of our method has been demonstrated with the rectum, breast and head-and-neck cancer cases by comparing the predicted DVHs with the planned ones. Results: The consistent result has been found between these two DVHs for each cancer and the average of relative point-wise differences is about 5% within the clinical acceptable extent. Conclusion: According to the result of this study, our method can be used to predict the clinical acceptable DVH and has ability to evaluate the quality and consistency of the treatment planning.},
doi = {10.1118/1.4956635},
journal = {Medical Physics},
number = 6,
volume = 43,
place = {United States},
year = 2016,
month = 6
}
  • Purpose: To use 3D printers to design and construct complex geometrical phantoms for commissioning treatment planning systems, dose calculation algorithms, quality assurance (QA), dose delivery, and patient dose verifications. Methods: In radiotherapy, complex geometrical phantoms are often required for dose verification, dose delivery and calculation algorithm validation. Presently, fabrication of customized phantoms is limited due to time, expense and challenges in machining of complex shapes. In this work, we designed and utilized 3D printers to fabricate two phantoms for QA purposes. One phantom includes hills and valleys (HV) for verification of intensity modulated radiotherapy for photons, and protons (IMRT andmore » IMPT). The other phantom includes cylindrical cavities (CC) of various sizes for dose verification of inhomogeneities. We evaluated the HV phantoms for an IMPT beam, and the CC phantom to study various inhomogeneity configurations using photon, electron, and proton beams. Gafcromic ™ films were used to quantify the dose distributions delivered to the phantoms. Results: The HV phantom has dimensions of 12 cm × 12 cm and consists of one row and one column of five peaks with heights ranging from 2 to 5 cm. The CC phantom has a size 10 cm × 14 cm and includes 6 cylindrical cavities with length of 7.2 cm and diameters ranging from 0.6 to 1.2 cm. The IMPT evaluation using the HV phantom shows good agreement as compared to the dose distribution calculated with treatment planning system. The CC phantom also shows reasonable agreements for using different algorithms for each beam modalities. Conclusion: 3D printers with submillimiter resolutions are capable of printing complex phantoms for dose verification and QA in radiotherapy. As printing costs decrease and the technology becomes widely available, phantom design and construction will be readily available to any clinic for testing geometries that were not previously feasible.« less
  • The current standard for brachytherapy dose calculations is based on the AAPM TG-43 formalism. Simplifications used in the TG-43 formalism have been challenged by many publications over the past decade. With the continuous increase in computing power, approaches based on fundamental physics processes or physics models such as the linear-Boltzmann transport equation are now applicable in a clinical setting. Thus, model-based dose calculation algorithms (MBDCAs) have been introduced to address TG-43 limitations for brachytherapy. The MBDCA approach results in a paradigm shift, which will require a concerted effort to integrate them properly into the radiation therapy community. MBDCA will improvemore » treatment planning relative to the implementation of the traditional TG-43 formalism by accounting for individualized, patient-specific radiation scatter conditions, and the radiological effect of material heterogeneities differing from water. A snapshot of the current status of MBDCA and AAPM Task Group reports related to the subject of QA recommendations for brachytherapy treatment planning is presented. Some simplified Monte Carlo simulation results are also presented to delineate the effects MBDCA are called to account for and facilitate the discussion on suggestions for (i) new QA standards to augment current societal recommendations, (ii) consideration of dose specification such as dose to medium in medium, collisional kerma to medium in medium, or collisional kerma to water in medium, and (iii) infrastructure needed to uniformly introduce these new algorithms. Suggestions in this Vision 20/20 article may serve as a basis for developing future standards to be recommended by professional societies such as the AAPM, ESTRO, and ABS toward providing consistent clinical implementation throughout the brachytherapy community and rigorous quality management of MBDCA-based treatment planning systems.« less
  • Purpose: In the context of national calls for reorganizing cancer clinical trials, the National Cancer Institute sponsored a 2-day workshop to examine challenges and opportunities for optimizing radiotherapy quality assurance (QA) in clinical trial design. Methods and Materials: Participants reviewed the current processes of clinical trial QA and noted the QA challenges presented by advanced technologies. The lessons learned from the radiotherapy QA programs of recent trials were discussed in detail. Four potential opportunities for optimizing radiotherapy QA were explored, including the use of normal tissue toxicity and tumor control metrics, biomarkers of radiation toxicity, new radiotherapy modalities such asmore » proton beam therapy, and the international harmonization of clinical trial QA. Results: Four recommendations were made: (1) to develop a tiered (and more efficient) system for radiotherapy QA and tailor the intensity of QA to the clinical trial objectives (tiers include general credentialing, trial-specific credentialing, and individual case review); (2) to establish a case QA repository; (3) to develop an evidence base for clinical trial QA and introduce innovative prospective trial designs to evaluate radiotherapy QA in clinical trials; and (4) to explore the feasibility of consolidating clinical trial QA in the United States. Conclusion: Radiotherapy QA can affect clinical trial accrual, cost, outcomes, and generalizability. To achieve maximum benefit, QA programs must become more efficient and evidence-based.« less
  • Purpose: Dosimetry using film, CR, electronic portal imaging, or other 2D detectors requires calibration of the raw image data to obtain dose. Typically, a series of known doses are given to the detector, the raw signal for each dose is obtained, and a calibration curve is created. This calibration curve is then applied to the measured raw signals to convert them to dose. With the advent of IMRT, film dosimetry for quality assurance has become a routine and labor intensive part of the physicist's day. The process of calibrating the film or other 2D detector takes time and additional filmmore » or images for performing the calibration, and comes with its own source of errors. This article studies a new methodology for the relative dose calibration of 2D imaging detectors especially useful for IMRT QA, which relies on the treatment plan dose image to provide the dose information which is paired with the raw QA image data after registration of the two images (plan-based calibration). Methods: Validation of the accuracy and robustness of the method is performed on ten IMRT cases performed using EDR2 film with conventional and plan-based calibration. Also, for each of the ten cases, a 5 mm registration error was introduced and the Gamma analysis was reevaluated. In addition, synthetic image tests were performed to test the limits of the method. The Gamma analysis is used as a measure of dosimetric agreement between plan and film for the clinical cases and a dose difference metric for the synthetic cases. Results: The QA image calibrated by the plan-based method was found to more accurately match the treatment plan doses than the conventionally calibrated films and also to reveal dose errors more effectively when a registration error was introduced. When synthetic acquired images were systematically studied, localized and randomly placed dose errors were correctly identified without excessive falsely passing or falsely failing pixels, unless the errors were concentrated in a majority of pixels in a contiguous narrow dose band. Irregularities seen in the calibration curve expose these errors. Conclusions: The plan-based calibration method was found to be an accurate, efficient procedure, capable of detecting IMRT QA relative dosimetry errors as well as, or better than conventional calibration methods.« less
  • This report summarizes the consensus findings and recommendations emerging from 2007 Symposium, 'Quality Assurance of Radiation Therapy: Challenges of Advanced Technology.' The Symposium was held in Dallas February 20-22, 2007. The 3-day program, which was sponsored jointly by the American Society for Therapeutic Radiology and Oncology (ASTRO), American Association of Physicists in Medicine (AAPM), and National Cancer Institute (NCI), included >40 invited speakers from the radiation oncology and industrial engineering/human factor communities and attracted nearly 350 attendees, mostly medical physicists. A summary of the major findings follows. The current process of developing consensus recommendations for prescriptive quality assurance (QA) testsmore » remains valid for many of the devices and software systems used in modern radiotherapy (RT), although for some technologies, QA guidance is incomplete or out of date. The current approach to QA does not seem feasible for image-based planning, image-guided therapies, or computer-controlled therapy. In these areas, additional scientific investigation and innovative approaches are needed to manage risk and mitigate errors, including a better balance between mitigating the risk of catastrophic error and maintaining treatment quality, complimenting the current device-centered QA perspective by a more process-centered approach, and broadening community participation in QA guidance formulation and implementation. Industrial engineers and human factor experts can make significant contributions toward advancing a broader, more process-oriented, risk-based formulation of RT QA. Healthcare administrators need to appropriately increase personnel and ancillary equipment resources, as well as capital resources, when new advanced technology RT modalities are implemented. The pace of formalizing clinical physics training must rapidly increase to provide an adequately trained physics workforce for advanced technology RT. The specific recommendations of the Symposium included the following. First, the AAPM, in cooperation with other advisory bodies, should undertake a systematic program to update conventional QA guidance using available risk-assessment methods. Second, the AAPM advanced technology RT Task Groups should better balance clinical process vs. device operation aspects-encouraging greater levels of multidisciplinary participation such as industrial engineering consultants and use-risk assessment and process-flow techniques. Third, ASTRO should form a multidisciplinary subcommittee, consisting of physician, physicist, vendor, and industrial engineering representatives, to better address modern RT quality management and QA needs. Finally, government and private entities committed to improved healthcare quality and safety should support research directed toward addressing QA problems in image-guided therapies.« less