Machine learning for automated quality assurance in radiotherapy: A proof of principle using EPID data description

Abstract

identified one outlier cluster (0.34%) along Leaf offset Constancy (LoC) axis that coincided with TG-142 limits. Conclusion: Machine learning methods based on SVDD clustering are promising for developing automated QA tools and providing insights into their reliability and reproducibility.

Predicting gamma passing rates for portal dosimetry‐based IMRT QA using machine learning

“…Machine learning and deep learning methods have become powerful tools for radiotherapy QA, specifically in error detection and prevention, treatment machine QA, and patient‐specific QA. Recently El Naqa et al used machine learning methods based on support vector data description to detect anomalies in treatment machine performance using EPID . Granville et al used a linear support vector classifier that includes both treatment plan complexity and linac performance metrics to predict VMAT patient QA results measured using a diode array .…”

“…Recently El Naqa et al used machine learning methods based on support vector data description to detect anomalies in treatment machine performance using EPID. 39 Granville et al used a linear support vector classifier that includes both treatment plan complexity and linac performance metrics to predict VMAT patient QA results measured using a diode array. 18 Deep learning approaches based on convolution neural networks without the extraction of the plan complexity metrics have been reported to predict patient-specific QA results and could achieve comparable prediction performance as machine learning approaches.…”

Overview of artificial intelligence-based applications in radiotherapy: Recommendations for implementation and quality assurance

“…QA of linear accelerators is periodically performed to monitor longitudinal stability [135]. Measured data contain nonlinearity in a multidimensional space, making it difficult to interpret [136]. Due to developments seen in the delivery and monitoring systems, opportunities arise to complement with approaches such as Probabilistic Safety Assessment (PSA) [137] or risk analysis [21] to focus where AI can amplify detection levels and prediction accuracy of potential failure or deviation from intent.…”

The changing role of radiation oncology professionals in a world of AI – Just jobs lost – Or a solution to the under-provision of radiotherapy?

“…Automation and data mining can be used to optimize quality assurance schedules, and to auto-detect and identify errors/deviations. Upcoming error modes or machine breakdown may potentially even be predicted based on pattern recognition in retrospective measurements [24], and machine log files can be analyzed for detection of errors during delivery, such as MLC leaf positioning [25].…”