Related Experiment Video
Updated: Dec 6, 2025
08:34
Proton Therapy Delivery and Its Clinical Application in Select Solid Tumor Malignancies
Published on: February 6, 2019
20.8K
A machine learning-based framework for delivery error prediction in proton pencil beam scanning using irradiation
Dominic Maes1, Stephen R Bowen2, Rajesh Regmi3
1Seattle Cancer Care Alliance Proton Therapy Center, 1570 N 115th St., Seattle, WA 98133, USA; Centre for Medical Radiation Physics, University of Wollongong, Wollongong, NSW 2500, Australia.
Summary
Machine learning accurately predicts proton pencil beam scanning delivery errors. This approach improves dose distribution accuracy in radiation therapy, enhancing treatment precision for patients.
Area of Science:
- Medical Physics
- Radiation Oncology
- Machine Learning Applications
Background:
- Proton pencil beam scanning (PBS) enables precise radiation delivery.
- Accurate prediction of delivery errors is crucial for optimizing treatment efficacy and safety.
Purpose of the Study:
- To investigate the efficacy of machine learning models in predicting delivery errors for PBS.
- To assess the potential of ML for improving dose prediction accuracy in proton therapy.
Main Methods:
- Trained three machine learning models using planned and delivered spot parameters from 20 prostate cancer patients.
- Employed K-fold cross-validation and Mean Absolute Error (MAE) for model selection.
- Validated the best model by predicting dose distribution for a test patient.
Main Results:
- The selected ML model achieved prediction error standard deviations of 0.22 mm (x) and 0.11 mm (y) for spot positions.
- Predicted dose distribution closely matched delivered dose, outperforming the planned distribution.
- Analysis showed ML significantly reduces discrepancies between planned and delivered proton therapy doses.
Conclusions:
- Machine learning models can accurately predict proton pencil beam scanning delivery errors.
- ML-driven dose prediction enhances the precision of proton therapy.
- This technology holds promise for improving patient outcomes in radiation oncology.
