Predicting Gamma Passing Rate for Patient Specific Quality Assurance Using Machine and Deep Learning: A Review of Methodological Approaches
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Introduction: In radiation therapy using advanced techniques such as intensity modulated radiation therapy (IMRT) and volumetric‐arc radiation therapy (VMAT), patient specific quality assurance (QA) should be performed before treatment. The measured and planned dose distributions are commonly quantified by means of a gamma analysis (Gamma Passing Rate, GPR). However, patient‐specific QA procedures are requiring significant time and effort by the physicists. Various ML and DL models have shown promising prediction accuracy and a high potential as time‐efficient virtual QA tool. Purpose: In this paper, we review the ML and DL based models that were developed for patient specific IMRT and VMAT QA GPR predictions, as well as to identify perspective directions for future research in the field of virtual QA. Conclusion: The prediction of Gamma Passing Rates (GPR) using Machine Learning (ML) and Deep Learning (DL) is a rapidly evolving and highly promising field. This review demonstrates the evolution of methodological approaches, from the analysis of individual plan complexity metrics to the application of ensemble regression models, and further to sophisticated deep learning architectures. Research confirms that these predictive models can accurately identify plans at risk of verification failure, paving the way for a risk-based approach and a significant reduction in routine measurements. Key challenges for broader clinical integration remain, including ensuring model interpretability, overcoming class imbalance in datasets, improving model generalizability, and their integration into clinical workflows. Successfully addressing these challenges will enable the creation of intelligent decision-support systems capable of enhancing the efficiency, safety, and standardization of radiotherapy.
Fields of Study & Subjects
Advanced Radiotherapy Techniques
Artificial Intelligence in Healthcare and Education
Radiation Therapy and Dosimetry
Publication details
- Journal / source
- Medical Radiology and radiation safety
- Volume
- 71
- Issue
- 2
- Pages
- 147–152
- Publication date
- 1 March 2026
- Language
- EN