A TPS integrated machine learning tool for predicting patient-specific quality assurance outcomes in volumetric-modulated arc therapy.

PURPOSE: Machine learning (ML) models have been demonstrated to be beneficial for optimizing the workload of patient-specific quality assurance (PSQA). Implementing them in clinical routine frequently requires third-party applications beyond the treatment planning system (TPS), slowing down the workflow. To address this issue, a PSQA outcomes predictive model was carefully selected and validated before being fully integrated into the TPS. MATERIALS AND METHODS: Nine ML algorithms were evaluated using cross-validation. The learning database was built by calculating complexity metrics (CM) and binarizing PSQA results into "pass"/"fail" classes for 1767 VMAT arcs. The predictive performance was evaluated using area under the ROC curve (AUROC), sensitivity, and specificity. The ML model was integrated into the TPS via a C# script. Script-guided reoptimization impact on PSQA and dosimetric results was evaluated on ten VMAT plans with "fail"-predicted arcs. Workload reduction potential was also assessed. RESULTS: The selected model exhibited an AUROC of 0.88, with a sensitivity and specificity exceeding 50 % and 90 %, respectively. The script-guided reoptimization of the ten evaluated plans led to an average improvement of 1.4 ± 0.9 percentage points in PSQA results, while preserving the quality of the dose distribution. A yearly savings of about 140 h with the use of the script was estimated. CONCLUSIONS: The proposed script is a valuable complementary tool for PSQA measurement. It was efficiently integrated into the clinical workflow to enhance PSQA outcomes and reduce PSQA workload by decreasing the risk of failing QA and thereby, the need for repeated replanning and measurements.

Implementation of volumetric-modulated arc therapy for locally advanced breast cancer patients: Dosimetric comparison with deliverability consideration of planning techniques and predictions of patient-specific QA results via supervised machine learning.

PURPOSE: The aim of this study was to implement a clinically deliverable VMAT planning technique dedicated to advanced breast cancer, and to predict failed QA using a machine learning (ML) model to optimize the QA workload. METHODS: For three planning methods (2A: 2-partial arc, 2AS: 2-partial arc with splitting, 4A: 4-partial arc), dosimetric results were compared with patient-specific QA performed with the electronic portal imaging device of the linac. A dataset was built with the pass/fail status of the plans and complexity metrics. It was divided into training and testing sets. An ML metamodel combining predictions from six base classifiers was trained on the training set to predict plans as 'pass' or 'fail'. The predictive performances were evaluated using the unseen data of the testing set. RESULTS: The dosimetric comparison highlighted that 4A was the highest dosimetric performant method but also the poorest performant in the QA process. 2AS spared the best heart, but provided the highest dose to the contralateral breast and lowest node coverage. 2A provides a dosimetric compromise between organ at risk sparing and PTV coverage with satisfactory QA results. The metamodel had a median predictive sensitivity of 73% and a median specificity of 91%. CONCLUSIONS: The 2A method was selected to calculate clinically deliverable VMAT plans; however, the 2AS method was maintained when the heart was of particular importance and breath-hold techniques were not applicable. The metamodel provides promising predictive performance, and it is intended to be improved as a larger dataset becomes available.

Exercise training improves radiotherapy efficiency in a murine model of prostate cancer.

Engaging in exercise while undergoing radiotherapy (RT) has been reported to be safe and achievable. The impact of exercise training (ET) on RT efficiency is however largely unknown. Our study aims to investigate the interactions between ET and RT on prostate cancer growth. Athymic mice received a subcutaneous injection of PPC-1 cells and were randomly assigned to either cancer control, cancer ET, cancer RT, or cancer RT combined with ET (CaRT-ET). Mice were sacrificed 24 days post-injection. All three intervention groups had reduced tumor size, the most important decrease being observed in CaRT-ET mice. Apoptotic marker cleaved caspase-3 was not modified by ET, but enhanced with RT. Importantly, this increase was the highest when the two strategies were combined. Furthermore, NK1.1 staining and gene expression of natural killer (NK) cell receptors Klrk1 and Il2rß were not affected by ET alone but were increased with RT, this effect being potentiated when combined with ET. Overall, our study shows that (a) ET enhances RT efficiency by potentiating NK cell infiltration, and (b) while ET alone and ET combined with RT both reduce tumor growth, the mechanisms mediating these effects are different.

A GATE/Geant4 Monte Carlo toolkit for surface dose calculation in VMAT breast cancer radiotherapy.

The accuracy of superficial dose calculations for breast cancer treatments with Volumetric Modulated Arc Therapy (VMAT) is of major importance. For target volumes close to the surface, the inverse dosimetric planning can lead to very high fluences in the build-up region to properly cover the volume to be treated. Various radiotherapy modalities are currently used in parallel with additional protocols to enable a better control on the dose delivery (bolus, target volume margins). One of the difficulties currently facing medical physicists is the lack of available tools to test the impact of these different solutions on the superficial dose distribution. We present a new open source toolkit to assist medical physicists in evaluating the 3D distributions of superficial dose in VMAT breast cancer treatments. This tool is based on the GATE Monte Carlo software, a Geant4 application dedicated to medical physics. A set of macros has been developed to simulate in an easy way a full VMAT plan from the information available in the DICOM-RT files (image, plan, structure and dose). The toolkit has been tested on a 6 MV Varian NovalisTx™ accelerator. The paper presents a precise comparison of 3D surface dose distributions from experimental measurements (EBT3 films), TPS (Varian Eclipse) and Monte Carlo simulation (GATE). The comparison made it possible to highlight both the TPS biases for the surface dose calculation and the good performances of the developed toolkit. The simulation of surface dose distributions on a real patient has also been performed to illustrate the potential clinical applications.

Comparison of Automated Atlas-Based Segmentation Software for Postoperative Prostate Cancer Radiotherapy.

Automated atlas-based segmentation (ABS) algorithms present the potential to reduce the variability in volume delineation. Several vendors offer software that are mainly used for cranial, head and neck, and prostate cases. The present study will compare the contours produced by a radiation oncologist to the contours computed by different automated ABS algorithms for prostate bed cases, including femoral heads, bladder, and rectum. Contour comparison was evaluated by different metrics such as volume ratio, Dice coefficient, and Hausdorff distance. Results depended on the volume of interest showed some discrepancies between the different software. Automatic contours could be a good starting point for the delineation of organs since efficient editing tools are provided by different vendors. It should become an important help in the next few years for organ at risk delineation.