Varian Medical Systems
  • Palo Alto, United States
Recent publications
Background: Patients with high-risk prostate cancer (HRPC) have multiple accepted treatment options. Because there is no overall survival benefit of one option over another, appropriate treatment must consider patient life expectancy, quality of life, and cost. Methods: The authors compared quality-adjusted life years (QALYs) and cost effectiveness among treatment options for HRPC using a Markov model with three treatment arms: (1) external-beam radiotherapy (EBRT) delivered with 20 fractions, (2) EBRT with 23 fractions followed by low-dose-rate (LDR) brachytherapy boost, or (3) radical prostatectomy alone. An exploratory analysis considered a simultaneous integrated boost according to the FLAME trial (ClinicalTrials.gov identifier NCT01168479). Results: Treatment strategies were compared using the incremental cost-effectiveness ratio (ICER). EBRT with LDR brachytherapy boost was a cost-effective strategy (ICER, $20,929 per QALY gained). These results were most sensitive to variations in the biochemical failure rate. However, the results still demonstrated cost effectiveness for the brachytherapy boost paradigm, regardless of any tested parameter ranges. Probabilistic sensitivity analysis demonstrated that EBRT with LDR brachytherapy was favored in 52% of 100,000 Monte Carlo iterations. In an exploratory analysis, EBRT with a simultaneous integrated boost was also a cost-effective strategy, resulting in an ICER of $62,607 per QALY gained; however, it was not cost effective compared with EBRT plus LDR brachytherapy boost. Conclusions: EBRT with LDR brachytherapy boost may be a cost-effective treatment strategy compared with EBRT alone and radical prostatectomy for HRPC, demonstrating high-value care. The current analysis suggests that a reduction in biochemical failure alone can result in cost-effective care, despite no change in overall survival.
Aim This study aims to report preclinical validation, and the first clinical treatment of total bone marrow irradiation (TMI) and total bone marrow and lymph nodal irradiation (TMLI) using Volumetric modulated arc therapy in Halcyon-E ring gantry linear accelerator. Preclinical validation includes simulation, planning, patient-specific QA, and dry run. Material and method Four patients, two female and two male, with body weights of 116 kg, 52 kg, 64 kg, and 62 kg; with two with chronic myeloid leukemia, one each with acute lymphoblastic leukemia and acute myeloid leukemia (AML) were simulated and planned for TMI/TMLI. Patients were immobilized with a full-body vacuum bag. Head first supine (HFS) and Feet first supine (FFS) CT scans were acquired from head to knee and knee to toe. Planning target volume (PTV) was created with a uniform margin of 6 mm over the total bone marrow/bone marrow + lymph nodes. HFS and FFS PTVs were optimized independently using 6MV unflatten energy for 12 Gy in 6 fractions. Plans were merged to create the resultant dose distribution using a junction bias dose matching technique. The total number of isocenters was ≤ 10 per CT set, and two to four full arcs were used for each isocenter. A junction dose gradient technique was used for dose feathering between arcs between adjacent isocenters. Result Only one female patient diagnosed as AML received the TMLI treatment, while the other three patients dropped out due to clinical complications and comorbidities that developed in the time between simulation and treatment. The result presented has been averaged over all four patients. For PTV, 95% dose was normalised to 95% volume, PTV_V107% receiving 3.3 ± 3.1%. Total lung mean and V12Gy were 1048.6 ± 107.1 cGy and 19.5 ± 12.1%. Maximum lens doses were 489.5 ± 35.5 cGy (left: L) and 497 ± 69.2 cGy (right: R). The mean cardiac and bilateral kidney doses were 921.75 ± 89.2 cGy, 917.9 ± 63.2 cGy (L), and 805.9 ± 9.7 cGy (R). Average Monitor Unit was 7738.25 ± 1056.6. The median number of isocenters was 17(HFS+FFS), average MU/Dose (cGy) ratio per isocenter was 2.28 ± 0.3. Conclusion Halcyon-E ring gantry linear accelerator capable of planning and delivering TMI/TMLI.
Purpose: Current radiation therapy (RT) treatment planning relies mainly on pre-defined dose-based objectives and constraints to develop plans that aim to control disease while limiting damage to normal tissues during treatment. These objectives and constraints are generally population-based, in that they are developed from the aggregate response of a broad patient population to radiation. However, correlations of new biologic markers and patient-specific factors to treatment efficacy and toxicity provide the opportunity to further stratify patient populations and develop a more individualized approach to RT planning. We introduce a novel intensity-modulated radiation therapy (IMRT) optimization strategy that directly incorporates patient-specific dose response models into the planning process. In this strategy, we integrate the concept of utility-based planning where the optimization objective is to maximize the predicted value of overall treatment utility, defined by the probability of efficacy (e.g., local control) minus the weighted sum of toxicity probabilities. To demonstrate the feasibility of the approach, we apply the strategy to treatment planning for non-small cell lung cancer (NSCLC) patients. Methods and materials: We developed a prioritized approach to patient-specific IMRT planning. Using a commercial treatment planning system (TPS), we calculate dose based on an influence matrix of beamlet-dose contributions to regions-of-interest. Then, outside of the TPS, we hierarchically solve two optimization problems to generate optimal beamlet weights that can then be imported back to the TPS. The first optimization problem maximizes a patient's overall plan utility subject to typical clinical dose constraints. In this process, we facilitate direct optimization of efficacy and toxicity trade-off based on individualized dose-response models. After optimal utility is determined, we solve a secondary optimization problem that minimizes a conventional dose-based objective subject to the same clinical dose constraints as the first stage but with the addition of a constraint to maintain the optimal utility from the first optimization solution. We tested this method by retrospectively generating plans for five previously treated NSCLC patients and comparing the prioritized utility plans to conventional plans optimized with only dose metric objectives. To define a plan utility function for each patient, we utilized previously published correlations of dose to local control and grade 3-5 toxicities that include patient age, stage, microRNA levels, and cytokine levels, among other clinical factors. Results: The proposed optimization approach successfully generated RT plans for five NSCLC patients that improve overall plan utility based on personalized efficacy and toxicity models while accounting for clinical dose constraints. Prioritized utility plans demonstrated the largest average improvement in local control (16.6%) when compared to plans generated with conventional planning objectives. However, for some patients the utility-based plans resulted in similar local control estimates with decreased estimated toxicity. Conclusion: The proposed optimization approach, where the maximization of a patient's RT plan utility is prioritized over the minimization of standardized dose metrics, has the potential to improve treatment outcomes by directly accounting for variability within a patient population. The implementation of the utility-based objective function offers an intuitive, humanized approach to biological optimization in which planning trade-offs are explicitly optimized. This article is protected by copyright. All rights reserved.
Background and purpose We describe a multicenter cross validation of ultra-high dose rate (UHDR) (>= 40 Gy/s) irradiation in order to bring a dosimetric consensus in absorbed dose to water. UHDR refers to dose rates over 100-1000 times those of conventional clinical beams. UHDR irradiations have been a topic of intense investigation as they have been reported to induce the FLASH effect in which normal tissues exhibit reduced toxicity relative to conventional dose rates. The need to establish optimal beam parameters capable of achieving the in vivo FLASH effect has become paramount. It is therefore necessary to validate and replicate dosimetry across multiple sites conducting UHDR studies with distinct beam configurations and experimental set-ups. Materials and methods Using a custom cuboid phantom with a cylindrical cavity (5 mm diameter by 10.4 mm length) designed to contain three type of dosimeters (thermoluminescent dosimeters (TLDs), alanine pellets, and Gafchromic films), irradiations were conducted at expected doses of 7.5 to 16 Gy delivered at UHDR or conventional dose rates using various electron beams at the Radiation Oncology Departments of the CHUV in Lausanne, Switzerland and Stanford University, CA. Results Data obtained between replicate experiments for all dosimeters were in excellent agreement (+/- 3 %). In general, films and TLDs were in closer agreement with each other, while alanine provided the closest match between the expected and measured dose, with certain caveats related to absolute reference dose. Conclusion In conclusion, successful cross-validation of different electron beams operating under different energies and configurations lays the foundation for establishing dosimetric consensus for UHDR irradiation studies, and, if widely implemented, decrease uncertainty between different sites investigating the mechanistic basis of the FLASH effect.
Objectives: Treatment of advanced and recurrent endometrial cancer (EC) has traditionally been based on factors, such as histology and stage. Recent studies suggest that tumor molecular profiles may predict treatment outcomes and guide treatment choice. For example, pembrolizumab (PMB) use in patients with mismatch repair deficient (MMRd) tumors versus combined pembrolizumab and lenvatinib (PL) in patients with mismatch repair proficient tumors (MMRp). Furthermore, POLE mutated (POLEmut) tumors demonstrate a favorable prognosis even with minimal treatment. Prospective trials using molecular profiling for treatment assignment are ongoing, but tumor molecular testing (TMT) is costly and time-consuming. We, therefore, sought to evaluate the cost-effectiveness of TMT in stage III EC. Methods: A Markov decision model was created to compare two testing strategies in patients with stage III EC: TMT versus no testing (NT). TMT patients had sequential POLE next-generation sequencing (NGS), mismatch repair (MMR) immunohistochemistry (IHC), and p53 IHC. Adjuvant therapies included radiation (RT) alone for POLEmut and chemoradiation (CTRT) for NT, MMRd, p53 mutated (p53mut), and no specific molecular profile (NSMP) using PORTEC-3 data. All patients received six cycles of carboplatin and paclitaxel (CP) for first recurrence using GOG 209 outcomes. MMRd patients received PMB alone for the second recurrence, whereas all other patients received PL. National database and registry values, literature review, and expert input were used for cost and utility values. A healthcare payer perspective, willingness-to-pay (WTP) threshold of $100,000, and 5-year time horizon were used. Data analysis was done with TreeAge Pro Software 2021. Outcomes were reported in quality-adjusted life-years (QALYs) and incremental cost-effectiveness ratios (ICERs). Sensitivity analyses tested model robustness, and a structural sensitivity analysis was performed in which controls underwent MMR IHC alone. Results: Compared to NT, TMT was less costly and more effective. While the model was most sensitive to median overall survival with PL, probability of MMRd status, and cost of lenvatinib, TMT remained cost-saving over all parameters tested. On Monte Carlo analysis, TMT was favored in 79% and 80% of iterations with a WTP of $50,000 and $100,000, respectively. When controls had MMR IHC alone, TMT was no longer cost-effective with an incremental cost of $3100 and incremental effectiveness of 0.017 QALYs, or $184,942/QALY gained. Conclusions: For patients with stage III EC, this model suggests TMT is associated with cost savings compared to NT. However, when compared to MMR IHC alone, a common clinical approach, TMT was no longer a cost-effective strategy at a WTP threshold of $100,000. Further prospective studies are needed to determine the prognostic value of molecular profiles in endometrial cancer. It will be essential to continue assessing the cost-effectiveness of these molecular tests as new data becomes available.
Purpose To introduce and validate a newly developed deep-learning (DL) auto-segmentation algorithm for head and neck (HN) organs at risk (OARs) and to compare its performance with a published commercial algorithm. Methods A total of 864 HN cancer cases were available to train and evaluate a prototype algorithm. The algorithm is based on a fully convolutional network with combined U-Net and V-net. A Dice loss plus Cross-Entropy Loss function with Adam optimizer was used in training. For 75 validation cases, OAR sets were generated with three DL-based models (A: the prototype model trained with gold data, B: a commercial software trained with the same data, and C: the same software trained with data from another institution). The auto-segmented structures were evaluated with Dice similarity coefficient (DSC), Hausdorff distance (HD), voxel-penalty metric (VPM) and DSC of area under dose-volume histograms. A subjective qualitative evaluation was performed on 20 random cases. Results Overall trend was for the prototype algorithm to be the closest to the gold data by all five metrics. The average DSC/VPM/HD for algorithms A, B, and C were 0.81/84.1/1.6mm, 0.74/62.8/3.2mm, and 0.66/46.8/3.3mm, respectively.. 93% of model A structures were evaluated to be clinically useful. Conclusion The superior performance of the prototype was validated, even when trained with the same data. In addition to the challenges of perfecting the algorithms, the auto-segmentation results can differ when the same algorithm is trained at different institutions.
Purpose: In radiation therapy, X-ray dose must be precisely sculpted to the tumor, whilst simultaneously avoiding surrounding organs at risk. This requires modulation of X-ray intensity in space and/or time. Typically, this is achieved using a Multi Leaf Collimator (MLC) - a complex mechatronic device comprising over one hundred individually powered tungsten 'leaves' that move in or out of the radiation field as required. Here, an all-electronic X-ray collimation concept with no moving parts is presented, termed "SPHINX": Scanning Pencil-beam High-speed Intensity-modulated X-ray source. SPHINX utilizes a spatially distributed bremsstrahlung target and collimator array in conjunction with magnetic scanning of a high energy electron beam to generate a plurality of small X-ray "beamlets". Methods: A simulation framework was developed in Topas Monte Carlo incorporating a phase space electron source, transport through user defined magnetic fields, bremsstrahlung X-ray production, transport through a SPHINX collimator, and dose in water. This framework was completely parametric, meaning a simulation could be built and run for any supplied geometric parameters. This functionality was coupled with Bayesian optimization to find the best parameter set based on an objective function which included terms to maximize dose rate for a user defined beamlet width while constraining inter-channel cross talk and electron contamination. Designs for beamlet widths of 5, 7, and 10 mm2 were generated. Each optimization was run for 300 iterations and took approximately 40 hours on a 24 core computer. For the optimized seven-mm model, a simulation of all beamlets in water was carried out including a linear scanning magnet calibration simulation. Finally, a back-of-envelope dose rate formalism was developed and used to estimate dose rate under various conditions. Results: The optimized five-mm, seven-mm, and ten-mm models had beamlet widths of 5.1 mm, 7.2 mm, and 10.1 mm2 and dose rates of 3574 Gy/C, 6351 Gy/C and 10015 Gy/C respectively. The reduction in dose rate for smaller beamlet widths is a result of both increased collimation and source occlusion. For the simulation of all beamlets in water, the scanning magnet calibration reduced the offset between the collimator channels and beam centroids from 2.9+-1.9 mm to 0.01 +- 0.03mm. A slight reduction in dose rate of approximately 2% per degree of scanning angle was observed. Based on a back-of-envelope dose rate formalism, SPHINX in conjunction with next-generation linear accelerators has the potential to achieve substantially higher dose rates than conventional MLC based delivery, with delivery of an intensity modulated 100×100 mm2 field achievable in 0.9 to 10.6 s depending on the beamlet widths used. Conclusions: Bayesian optimization was coupled with Monte Carlo modelling to generate SPHINX geometries for various beamlet widths. A complete Monte Carlo simulation for one of these designs was developed, including electron beam transport of all beamlets through scanning magnets, X-ray production and collimation, and dose in water. These results demonstrate that SPHINX is a promising candidate for sculpting radiation dose with no moving parts, and has the potential to vastly improve both the speed and robustness of radiotherapy delivery. This article is protected by copyright. All rights reserved.
The International Organization for Medical Physics (IOMP) is the world's largest professional organization in the field of medical physics and has official non-governmental organization status with the World Health Organization (WHO) and the International Atomic Energy Agency (IAEA). IOMP is charged with a mission to advance medical physics practice worldwide by disseminating scientific and technical information, fostering the educational and professional development of medical physics and promoting the highest quality medical services for patients. IOMP's activities are directed towards the promotion of medical physics globally, improving patient care, and contributing to the benefit of healthcare to the society. Major organizational activities include but are not limited to scientific events, international collaborations, dissemination of information, education, training, and research. For nearly 60 years of existence, IOMP turned into a key factor not only in the field of medical physics, but also healthcare, and other related disciplines. IOMP is looking forward to future perspectives in international collaboration and enhancement of the professional skills, all directed towards enhancing patient benefit.
Knowledge-based planning solutions have brought significant improvements in treatment planning. However, the performance of a proton-specific knowledge-based planning model in creating knowledge-based plans (KBPs) with beam angles differing from those used to train the model remains unexplored. We used a previously validated RapidPlanPT model and scripting to create nine KBPs, one with default and eight with altered beam angles, for 10 recent oropharynx cancer patients. The altered-angle plans were compared against the default-angle ones in terms of grade 2 dysphagia and xerostomia normal tissue complication probability (NTCP), mean doses of several organs at risk, and dose homogeneity index (HI). As KBP could be suboptimal, a proof of principle automatic iterative optimizer (AIO) was added with the aim of reducing the plan NTCP. There were no statistically significant differences in NTCP or HI between default- and altered-angle KBPs, and the altered-angle plans showed a <1% reduction in NTCP. AIO was able to reduce the sum of grade 2 NTCPs in 66/90 cases with mean a reduction of 3.5 ± 1.8%. While the altered-angle plans saw greater benefit from AIO, both default- and altered-angle plans could be improved, indicating that the KBP model alone was not completely optimal to achieve the lowest NTCP. Overall, the data showed that the model was robust to the various beam arrangements within the range described in this analysis.
Memorial Sloan Kettering Cancer Center (MSKCC)Memorial Sloan Kettering Cancer Center (MSKCC) is considered to be one of the top cancer hospitals in the United States. The institution began as the New York Cancer HospitalNew York Cancer Hospital in 1884. In 1912, it became known as the Memorial Hospital for the Study of Cancer and Allied DiseasesMemorial Hospital for the Study of Cancer and Allied Diseases. With a donation of land from John D RockefellerJohn D Rockefeller, the hospital moved to its present location in 1939. In 1945, the industrialists Alfred SloanAlfred Sloan and Charles KetteringCharles Kettering of General MotorsGeneral Motors provided funds to establish the Sloan Kettering InstituteSloan Kettering Institute, a research arm for Memorial Hospital. The two entities were finally combined into MSKCC in the 1980s. The following three stories describe major technological developments and the evolution of interdisciplinary collaboration as they occurred at MSKCC from 1989 to 2012.
Purpose: To present recommendations for the use of imaging for evaluation and procedural guidance of brachytherapy for cervical cancer patients. Methods: An expert panel comprised of members of the Society of Abdominal Radiology Uterine and Ovarian Cancer Disease Focused Panel and the American Brachytherapy Society jointly assessed the existing literature and provide data-driven guidance on imaging protocol development, interpretation, and reporting. Results: Image-guidance during applicator implantation reduces rates of uterine perforation by the tandem. Postimplant images may be acquired with radiography, computed tomography (CT), or magnetic resonance imaging (MRI), and CT or MRI are preferred due to a decrease in severe complications. Pre-brachytherapy T2-weighted MRI may be used as a reference for contouring the high-risk clinical target volume (HR-CTV) when CT is used for treatment planning. Reference CT and MRI protocols are provided for reference. Conclusions: Image-guided brachytherapy in locally advanced cervical cancer is essential for optimal patient management. Various imaging modalities, including orthogonal radiographs, ultrasound, computed tomography, and magnetic resonance imaging, remain integral to the successful execution of image-guided brachytherapy.
Objectives: In this study, we propose a deep learning-based approach to predict Intensity-modulated radiation therapy (IMRT) quality assurance (QA) gamma passing rates using delivery fluence informed by log files. Methods: A total of 112 IMRT plans for chest cancers were planned and measured by portal dosimetry equipped on TrueBeam linac. The convolutional neural network (CNN) based learning model was trained using delivery fluence as inputs and gamma passing rates (GPRs) of 4 different criteria (3%/3 mm, 2%/3 mm, 3%/2 mm, and 2%/2 mm) as outputs. Model performance for both validation and test sets was assessed using mean absolute error (MAE), mean squared error (MSE), root MSE (RMSE), Spearman rank correlation coefficients (Sr), and Determination coefficient ( R ² ) between the measured and predicted GPR values. Results: In the test set, the MAE of the prediction model were 0.402, 0.511, 1.724, and 2.530, the MSE were 0.640, 0.986, 6.654, and 9.508, the RMSE were 0.800, 0.993, 2.580, and 3.083, the Sr were 0.643, 0.684, 0.821, and 0.824 ( P < .001) and the R ² were 0.4110, 0.4666, 0.6677, and 0.6769 for 3%/3 mm, 3%/2 mm, 2%/3 mm, and 2%/2 mm, respectively. The MAE and RMSE of the prediction model decreased with stricter gamma criteria while the Sr and R ² between measured and predicted GPR values increased. Conclusions: The CNN prediction model based on delivery fluence informed by log files could accurately predict IMRT QA passing rates for different gamma criteria. It could reduce QA workload and improve efficiency in pretreatment QA. Our results suggest that the CNN prediction model based on delivery fluence informed by log files may be a promising tool for the gamma evaluation of IMRT QA.
Introduction: The evolving treatment landscape for non-small-cell lung cancer (NSCLC) and complexities of regulations and reimbursement present challenges to community oncologists. Clinical pathways are tools to optimize care, but information on their value in the real world is limited. This retrospective study assessed treatment patterns and clinical outcomes in patients with Stage I–III NSCLC pre- and post-pathways implementation at Tennessee Oncology, a large, community-based oncology practice in the USA. Methods & Materials: Chart data were abstracted for adults diagnosed with Stage I–III NSCLC who received systemic treatment. Patients were divided into pre-pathways (treatment initiation 2014–2015) and post-pathways (treatment initiation 2016–2018) cohorts. Patient characteristics, treatment patterns and outcomes were summarized descriptively. Kaplan–Meier curves were used to assess time-dependent outcomes, and log-rank test was used to compare the cohorts. Results: 291 patients were included (Stage I–II: 38 pre-pathways, 55 post-pathways; Stage III: 105 pre-pathways, 93 post-pathways). Duration on first-line (1L) therapy was similar for Stage I–II patients pre- and post-pathways (median 1.9 months vs 2.1 months; p = 0.75), but increased for Stage III patients post-pathways (2.1 months vs 1.4 months pre-pathways; p < 0.01). Achievement of a complete or partial response with 1L therapy was similar post-pathways among Stage I–Stage –IIII patients (60.0% vs 55.2% pre-pathways), but increased for Stage III patients (56.0% vs 35.2% pre-pathways). Conclusion: Given that improvements in rates of treatment response post-pathways occurred only for patients diagnosed with Stage III NSCLC, among whom immunotherapy uptake increased post-pathways, such improvements may be attributable to evolving practices in cancer care, including advances in treatment and care delivery, rather than clinical pathways implementation. Further research is warranted to assess the impact of clinical pathways in the current treatment era, given that immunotherapy has now become the standard of care in NSCLC.
Whole-brain radiotherapy has been the standard palliative treatment for patients with brain metastases due to its effectiveness, availability, and ease of administration. Recent clinical trials have shown that limiting radiation dose to the hippocampus is associated with decreased cognitive toxicity. In this study, we updated an existing Knowledge Based Planning model to further reduce dose to the hippocampus and improve other dosimetric plan quality characteristics. Forty-two clinical cases were contoured according to guidelines. A new dosimetric scorecard was created as an objective measure for plan quality. The new Hippocampal Sparing Whole Brain Version 2 (HSWBv2) model adopted a complex recursive training process and was validated with five additional cases. HSWBv2 treatment plans were generated on the Varian HalcyonTM and TrueBeamTM systems and compared against plans generated from the existing (HSWBv1) model released in 2016. On the HalcyonTM platform, 42 cases were re-planned. Hippocampal D100% from HSWBv2 and HSWBv1 models had an average dose of 5.75 Gy and 6.46 Gy, respectively (p < 0.001). HSWBv2 model also achieved a hippocampal Dmean of 7.49 Gy, vs 8.10 Gy in HSWBv1 model (p < 0.001). Hippocampal D0.03CC from HSWBv2 model was 9.86 Gy, in contrast to 10.57 Gy in HSWBv1 (p < 0.001). For PTV_3000, D98% and D2% from HSWBv2 model were 28.27 Gy and 31.81 Gy, respectively, compared to 28.08 Gy (p = 0.020) and 32.66 Gy from HSWBv1 (p < 0.001). Among several other dosimetric quality improvements, there was a significant reduction in PTV_3000 V105% from 35.35% (HSWBv1) to 6.44% (HSWBv2) (p < 0.001). On 5 additional validation cases, dosimetric improvements were also observed on TrueBeamTM. In comparison to published data, the HSWBv2 model achieved higher quality hippocampal avoidance whole brain radiation therapy treatment plans through further reductions in hippocampal dose while improving target coverage and dose conformity/homogeneity. HSWBv2 model is shared publicly.
Purpose/Objective(s) Segmentation of head and neck (HN) organs at risk (OARs) is a laborious process. Here we introduce and validate a newly developed deep-learning-based auto-segmentation program and compare with a commercially available system, and the same system trained on internal data. Materials/Methods A total of 864 previously treated HN cancer patients were available to train and evaluate a prototype deep learning-based normal tissue 3D auto-segmentation algorithm. The algorithm is based on a fully convolutional network with U-Net and V-net features combined as the backbone of the network. A Dice loss function with the Adam optimizer was used in training the models with between 150-500 patients used per model. The OARs were delineated by a single experienced physician (gold data). A subset of 75 cases was withheld from training and used for validation. On those, we generated new OAR sets with three different deep-learning models and compared to the gold data: A) the prototype model trained with gold data, B) a commercial software package trained with the gold data (n=213), and C) the same commercial software with the model trained at another institution (n=589). The agreement between the gold data and auto-segmented structures was evaluated with Dice similarity coefficient (DSC) and voxel-penalty metric that penalizes each missing or extra pixel as a function of distance with forgiveness threshold distance. An ANOVA test with post hoc pair-wise analysis was performed to assess the differences in those metrics. The auto-segmented contours were also qualitatively evaluated by the physician on a scale of 0-5. Results The average DSC and voxel penalty metric scores for algorithms A, B, and C across all OARs in the 75 evaluation cases were 0.80/77.68, 0.74/62.75, and 0.66/45.26, respectively. The difference in mean DSC scores was statistically significant (p<0.05) for all 11 OARs where the data for all three algorithms were available. The A/B difference was significant in 6 OARs. Algorithm A scored the highest DSC and voxel penalty metric score in all OARs except for the pharyngeal constrictors. All OARs except for the pharyngeal constrictors showed DSC≥0.7 with algorithm A. For three structures the mean DSC was significantly different between the same algorithm trained at different institutions(B/C). From the qualitative evaluation by a blinded expert, 51 structures (20.2%) of model A were clinically acceptable without edits. The percentages of ‘clinically useful’ scores were largest in model A (95.2%) followed by model B (88.0%) and C (80.6%). Conclusion The prototype algorithm had improved performance compared to a commercial algorithm, even when trained on data from the same institution. Auto segmentation results can differ significantly when the same algorithm is trained on data from different institutions.
Background: Dose deposition characteristics of proton radiation can be advantageous over photons. Proton treatment planning however poses additional challenges for the planners. Proton therapy is usually delivered with only a small number of beam angles, and the quality of a proton treatment plan is largely determined by the beam angles employed. Finding the optimal beam angles for a proton treatment plan requires time and experience, motivating the investigation of automatic beam angle selection methods. Purpose: A deep learning-based approach to automatic beam angle selection is proposed for proton pencil-beam scanning treatment planning of liver lesions. Methods: We cast beam-angle selection as a multi-label classification problem. To account for angular boundary discontinuity, the underlying convolution neural network is trained with the proposed Circular Earth Mover's Distance based regularization and multi-label circular-smooth label technique. Furthermore, an analytical algorithm emulating proton treatment planners' clinical practice is employed in post-processing to improve the output of the model. Forty-nine patients that received proton liver treatments between 2017 and 2020 were randomly divided into training (n = 31), validation (n = 7), and test sets (n = 11). AI-selected beam angles were compared with those angles selected by human planners, and the dosimetric outcome was investigated by creating plans using knowledge-based treatment planning. Results: For 7 of the 11 cases in the test set, AI-selected beam angles agreed with those chosen by human planners to within 20 degrees (median angle difference = 10°; mean = 18.6°). Moreover, out of the total 22 beam angles predicted by the model, 15 (68%) were within 10 degrees of the human-selected angles. The high correlation in beam angles resulted in comparable dosimetric statistics between proton treatment plans generated using AI- and human-selected angles. For the cases with beam angle differences exceeding 20°, the dosimetric analysis showed similar plan quality although with different emphases on organ-at-risk sparing. Conclusions: This pilot study demonstrated the feasibility of a novel deep learning-based beam angle selection technique. Testing on liver cancer patients showed that the resulting plans were clinically viable with comparable dosimetric quality to those using human-selected beam angles. In tandem with auto-contouring and knowledge-based treatment planning tools, the proposed model could represent a pathway for nearly fully automated treatment planning in proton therapy. This article is protected by copyright. All rights reserved.
An anthropomorphic phantom has been developed by Varian Medical Systems for commissioning multileaf‐collimator (MLC), stereotactic radiosurgery (SRS) treatments on Varian TrueBeam and Edge linear accelerators. Northwest Medical Physics Center (NMPC) has collected end‐to‐end data on these machines, at six independent clinical sites, to establish baseline dosimetric and geometric commissioning criteria for SRS measurements with this phantom. The Varian phantom is designed to accommodate four interchangeable target cassettes, each designed for a specific quality assurance function. End‐to‐end measurements utilized the phantom to verify the coincidence of treatment isocenter with a hidden target in a Winston‐Lutz cassette after localization using cone‐beam computed tomography (CBCT). Dose delivery to single target (2 cm) and single‐isocenter, multitarget (2 and 1 cm) geometries was verified using ionization chamber and EBT3 film cassettes. A nominal dose of 16 Gy was prescribed for each plan using a site's standard beam geometry for SRS cases. Measurements were performed with three Millennium and three high‐definition MLC machines at beam energies of 6‐MV and 10‐MV flattening‐filter‐free energies. Each clinical site followed a standardized procedure for phantom simulation, treatment planning, quality assurance, and treatment delivery. All treatment planning and delivery was performed using ARIA oncology information system and Eclipse treatment planning software. The isocenter measurements and irradiated film were analyzed using DoseLab quality assurance software; gamma criteria of 3%/1 mm, 3%/0.5 mm, and 2%/1 mm were applied for film analysis. Based on the data acquired in this work, the recommended commissioning criteria for end‐to‐end SRS measurements with the Varian phantom are as follows: coincidence of treatment isocenter and CBCT‐aligned hidden target < 1 mm, agreement of measured chamber dose with calculated dose ≤ 5%, and film gamma passing > 90% for gamma criteria of 3%/1 mm after DoseLab auto‐registration shifts ≤ 1 mm in any direction.
Sentinel lymph node biopsy has led to an increase in the detection of isolated tumor cells (ITCs) in up to 10% of early stage endometrioid endometrial cancer patients. In addition, the risk of non-sentinel lymph node involvement is approximately 10% or lower in the patients with ITCs. In most studies, approximately 60%-70% of patients with ITCs either underwent completion lymphadenectomy or received adjuvant therapy. Therefore, though multiple studies have shown that the impact of ITCs on disease outcomes is favorable, the true impact of ITCs without additional treatment is not known. In this report we describe our philosophy of relying on extent of surgical nodal staging and presence or absence of adverse intra-uterine pathological factors at the time of adjuvant therapy decision making for endometrioid EC patients with ITCs.
Purpose: This study developed and evaluated a Fully Convolutional Network (FCN) for pediatric CT organ segmentation, and investigated the generalizability of the FCN across image heterogeneities such as CT scanner model protocols and patient age. We also evaluated the autosegmentation models as part of a software tool for patient-specific CT dose estimation. Methods: A collection of 359 pediatric CT datasets with expert organ contours were used for model development and evaluation. Autosegmentation models were trained for each organ using a modified FCN 3D V-Net. An independent test set of 60 patients was withheld for testing. To evaluate the impact of CT scanner model protocol and patient age heterogeneities, separate models were trained using a subset of scanner model protocols and pediatric age groups. Train and test sets were split to answer questions about the generalizability of pediatric FCN autosegmentation models to unseen age groups and scanner model protocols, as well as the merit of scanner model protocol or age-group-specific models. Finally, the organ contours resulting from the autosegmentation models were applied to patient-specific dose maps to evaluate the impact of segmentation errors on organ dose estimation. Results: Results demonstrate that the autosegmentation models generalize to CT scanner acquisition and reconstruction methods which were not present in the training dataset. While models are not equally generalizable across age groups, age-group-specific models do not hold any advantage over combining heterogeneous age groups into a single training set. Dice Similarity Coefficient (DSC) and Mean Surface Distance results are presented for 19 organ structures, for example median DSC of 0.52 (duodenum), 0.74 (pancreas), 0.92 (stomach), and 0.96 (heart). The FCN models achieve a mean dose error within 5% of expert segmentations for all 19 organs except for the spinal canal, where the mean error was 6.31%. Conclusions: Overall these results are promising for the adoption of FCN autosegmentation models for pediatric CT, including applications for patient-specific CT dose estimation. This article is protected by copyright. All rights reserved.
Institution pages aggregate content on ResearchGate related to an institution. The members listed on this page have self-identified as being affiliated with this institution. Publications listed on this page were identified by our algorithms as relating to this institution. This page was not created or approved by the institution. If you represent an institution and have questions about these pages or wish to report inaccurate content, you can contact us here.
282 members
Tobias Gass
  • Imaging Laboratory, Baden, Switzerland
Paul Baturin
  • Treatment Delivery and Imaging Systems
Magdalena Constantin
  • Oncology Systems
Mamdouh Abdel-Bary
  • Proton Therapy, Germany
Information
Address
Palo Alto, United States