ObjectiveThis work aims to use machine learning models to predict gamma passing rate of portal dosimetry quality assurance with log file derived features. This allows daily treatment monitoring for patients and reduce wear and tear on EPID detectors to save cost and prevent downtime.Methods578 VMAT trajectory log files selected from prostate, lung and spine SBRT were used in this work. Four machine learning models were explored to identify the best performing regression model for predicting gamma passing rate within each sub-site and the entire unstratified data. Predictors used in these models comprised of hand-crafted log file-derived features as well as modulation complexity score. Cross validation was used to evaluate the model performance in terms of R2 and RMSE.ResultUsing gamma passing rate of 1%/1mm criteria and entire dataset, LASSO regression has a R2 of 0.121 ± 0.005 and RMSE of 4.794 ± 0.013%, SVM regression has a R2 of 0.605 ± 0.036 and RMSE of 3.210 ± 0.145%, Random Forest regression has a R2 of 0.940 ± 0.019 and RMSE of 1.233 ± 0.197%. XGBoost regression has the best performance with a R2 and RMSE value of 0.981 ± 0.015 and 0.652 ± 0.276%, respectively.ConclusionLog file-derived features can predict gamma passing rate of portal dosimetry with an average error of less than 2% using the 1%/1mm criteria. This model can potentially be applied to predict the patient specific QA results for every treatment fraction.
Backgrounds: Respiratory gating is one of the motion management techniques that is used to deliver radiation dose to a tumor at a specific position under free breathing. However, due to the dynamic feedback process of this approach, regular equipment quality assurance (QA) and patient-specific QA checks need to be performed. This work proposes a new QA methodology using electronic portal imaging detector (EPID) to determine the target localization accuracy of phase gating. Methods: QA tools comprising 3D printed spherical tumor phantoms, programmable stages, and an EPID detector are characterized and assembled. Algorithms for predicting portal dose (PD) through moving phantoms are developed and verified using gamma analysis for two spherical tumor phantoms (2 cm and 4 cm), two different 6 MV volumetric modulated arc therapy plans, and two different gating windows (30%-70% and 40%-60%). Comparison between the two gating windows is then performed using the Wilcoxon signed-rank test. An optimizer routine, which is used to determine the optimal window, based on maximal gamma passing rate (GPR), was applied to an actual breathing curve and breathing plan. This was done to ascertain if our method yielded a similar result with the actual gating window. Results: High GPRs of more than 97% and 91% were observed when comparing the predicted PD with the measured PD in moving phantom at 2 mm/2% and 1 mm/1% levels, respectively. Analysis of gamma heatmaps shows an excellent agreement with the tumor phantom. The GPR of 40%-60% PD was significantly lower than that of the 30%-70% PD at the 1 mm/1% level (p = 0.0064). At the 2 mm/2% level, no significant differences were observed. The optimizer routine could accurately predict the center of the gating window to within a 10% range. Conclusion:We have successfully performed and verified a new method for QA with the use of a moving phantom with EPID for phase gating with real-time position management.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.