Learning to Optimize Radiotherapy Plans via Fluence Maps Diffusion Model Generation and LSTM-based Optimization

TL;DR

This paper introduces a diffusion-based learning-to-optimize method for VMAT radiotherapy planning, enabling fast, flexible, and deliverable fluence map generation and refinement, reducing planning time and improving clinical outcomes.

Contribution

It presents a novel diffusion-driven L2O framework combining a fluence map diffusion model and LSTM-based optimization for efficient VMAT planning.

Findings

Improved planning efficiency over existing methods.

Enhanced flexibility and deliverability of treatment plans.

Validated on clinical prostate cancer cohorts.

Abstract

Volumetric Modulated Arc Therapy (VMAT) is a cornerstone of modern radiation therapy, enabling highly conformal tumor irradiation and healthy-tissue sparing. Yet, its planning solves inverse and nested optimization for multi-leaf collimators, monitor units and dose parameters, while enforcing their consistency to ensure mechanical deliverability. Nevertheless, this process often requires repeated re-optimization when treatment configurations change, resulting in substantial planning time per patient. To address these problems, we present a diffusion-driven Learning-to-Optimize (L2O) method for end-to-end VMAT planning. A distribution-matching distilled diffusion model learns a clinically feasible manifold of fluence maps, enabling their one-shot generation. On top of this, an LSTM-based L2O module learns gradient update dynamics to swiftly refine fluence maps toward prescribed dose objectives during inference. Experimental results on clinical and public prostate cancer cohorts demonstrate improved planning efficiency, flexibility, and machine deliverability over currently available end-to-end VMAT planners.