Home // AIHealth 2025, The Second International Conference on AI-Health // View article

Recent Advances in Machine Learning for Log File-Based PSQA for IMRT and VMAT

Authors:
Kellin De Jesus
Leon Dunn
David Thomas
Les Sztandera

Keywords: deep learning; machine learning; quality assurance; volumetric-arc radiation therapy; intensity-modulated radiation therapy

Abstract:
The paper addresses the critical need for a faster and more efficient approach to Patient-Specific Quality Assurance (PSQA) in radiation therapy. The accuracy of PSQA is crucial for the safety of radiation therapy, particularly with complex procedures like Intensity-Modulated Radiation Therapy (IMRT) and Volumetric-Arc Radiation Therapy (VMAT). Traditional phantom-based methods, while effective, are time-consuming and fail to account for patient-specific variability and real-time treatment adjustment. To address these limitations, alternative strategies leveraging trajectory log files—automatically recorded during treatment—have emerged as promising tools for PSQA. In recent years, the application of machine learning and deep learning algorithms to trajectory log files has been increasingly studied in literature. These algorithms have shown notable progress in predicting PSQA outcomes and detecting errors, though further development is required before they can be fully integrated into clinical practice. By surveying key studies, the paper highlights the potential of algorithms such as support vector machines, tree-based methods, and convolutional neural networks to enhance the efficiency and accuracy of log file-based PSQA. The findings underscore the promise of these techniques in replacing traditional methods while addressing current challenges to pave the way for clinical integration.

Pages: 12 to 18

Copyright: Copyright (c) IARIA, 2025

Publication date: March 9, 2025

Published in: conference

ISBN: 978-1-68558-247-0

Location: Lisbon, Portugal

Dates: from March 9, 2025 to March 13, 2025