Unsupervised deep learning model for fast energy layer pre-selection of delivery-efficient proton arc therapy plan optimization of nasopharyngeal carcinoma

2025-06-18

AI Summary

Energy-layer (EL) sequence optimization in proton arc therapy (PAT) for nasopharyngeal carcinoma is computationally intensive, hindering clinical deployment. Method: We propose an unsupervised deep learning framework for rapid EL preselection. A novel spot-count matrix representation encodes dose distributions, and a tri-objective UNet architecture—SPArcdl—is designed to jointly optimize target coverage, organ-at-risk (OAR) sparing, and EL switching time. Contribution/Results: We first demonstrate that constant EL sequences outperform conventional descending sequences. Compared with particle swarm optimization, our method improves the conformity index by 0.16, reduces the homogeneity index by 0.71, shortens total EL switching time by 38.4%, and lowers mean brainstem dose by 0.21 Gy. Each plan inference completes in under one second, achieving superior dosimetric quality and delivery efficiency simultaneously.

Technology Category

• Search and Optimization: Learning to Search
• Machine Learning: Optimization
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Objective. Proton arc therapy (PAT) is an emerging and promising modality in radiotherapy, offering several advantages over conventional intensitymodulated proton therapy (IMPT). However, identifying the optimal energy layer (EL) sequence remains computationally intensive due to the large number of possible energy layer transitions. This study proposes an unsupervised deep learning framework for fast and effective EL pre-selection, aiming to minimize energy layer switch time while preserving high plan quality. Approach. We introduce a novel data representation method, spot-count representation, which encodes the number of proton spots intersecting the target and organs at risk (OARs) in a matrix structured by sorted gantry angles and energy layers. This representation is the input of a UNet-based architecture, SPArcdl, which is trained to optimize a tri-objective function: maximizing target coverage, minimizing OAR exposure, and reducing energy switching time. The model is evaluated on 54 nasopharyngeal cancer cases, and its performance is benchmarked against plans generated by SPArcparticle swarm. Main results. SPArcdl produces EL pre-selection that significantly improves both plan quality and delivery efficiency. Compared to SPArc particle swarm, it enhances the conformity index by 0.16 (p<0.01), reduces the homogeneity index by 0.71 (p<0.01), shortens the energy switching time by 38.4% (p<0.01), and lowers the mean dose to brainstem by 0.21 (p<0.01). The results unintentionally reveal employing unchanged ELS is more time-wise efficient than descended ELS. SPArcdl's inference time is within 1 second. Significance. SPArcdl is a fast and effective tool for generating high-quality PAT plans by strategically pre-selecting energy layers to reduce delivery time while maintaining excellent dosimetric performance.

Problem

• Optimizing energy layer selection for proton arc therapy efficiency
• Reducing energy switching time while maintaining plan quality
• Improving dosimetric performance in nasopharyngeal carcinoma treatment

Innovation

• Unsupervised deep learning for energy layer pre-selection
• Spot-count representation for efficient proton arc therapy
• UNet-based SPArcdl optimizes target coverage and OAR exposure