Neural network-driven direct CBCT-based dose calculation for head-and-neck proton treatment planning.
Authors
Affiliations (5)
Affiliations (5)
- Paul Scherrer Institut PSI, Forschungsstrasse 111, Villigen, 5232, SWITZERLAND.
- Paul Scherrer Institute PSI, Forschungsstrasse 111, Villigen, 5232, SWITZERLAND.
- Paul Scherrer Institut PSI, Forschungsstrasse 111, Villigen, AG, 5232, SWITZERLAND.
- Center for Proton Therapy, Paul Scherrer Institute PSI, 5232 Villigen PSI, Villigen, AG, 5232, SWITZERLAND.
- Paul Scherrer Institut PSI, CH-5232 Villigen PSI, Villigen, AG, 5232, SWITZERLAND.
Abstract
Accurate dose calculation on cone beam computed tomography (CBCT) images is essential for modern proton treatment planning workflows, particularly when accounting for inter-fractional anatomical changes in adaptive treatment scenarios. Traditional CBCT-based dose calculation suffers from image quality limitations, requiring complex correction workflows. This study develops and validates a deep learning approach for direct proton dose calculation from CBCT images using extended Long Short-Term Memory (xLSTM) neural networks. 
Approach. A retrospective dataset of 40 head-and-neck cancer patients with paired planning CT and treatment CBCT images was used to train an xLSTM-based neural network (CBCT-NN). The architecture incorporates energy token encoding and beam's-eye-view sequence modelling to capture spatial dependencies in proton dose deposition patterns. Training utilized 82,500 paired beam configurations with Monte Carlo-generated ground truth doses. Validation was performed on 5 independent patients using gamma analysis, mean percentage dose error assessment, and dose-volume histogram comparison. The CBCT-NN achieved gamma pass rates of 95.1 ± 2.7% using 2mm/2% criteria. Mean percentage dose errors were 2.6 ± 1.4% in high-dose regions (>90% of max dose) and 5.9 ± 1.9% globally. Dose-volume histogram analysis showed excellent preservation of target coverage metrics (Clinical Target Volume V95% difference: -0.6 ± 1.1%) and organ-at-risk constraints (parotid mean dose difference: -0.5 ± 1.5%). Computation time is under 3 minutes without sacrificing Monte Carlo-level accuracy. This study demonstrates the proof-of-principle of direct CBCT-based proton dose calculation using xLSTM neural networks. The approach eliminates traditional correction workflows while achieving comparable accuracy and computational efficiency suitable for adaptive protocols.