A Beam's Eye View to Fluence Maps 3D Network for Ultra Fast VMAT Radiotherapy Planning

2025-02-05

AI Summary

To address the complexity and computational inefficiency of conventional fluence map generation in ultra-rapid VMAT planning, this paper proposes a 3D deep neural network leveraging BEV (Bird’s Eye View) pre-alignment to directly predict full-arc fluence maps—spanning 180 control points—from 3D dose distributions. Crucially, BEV geometric projection is introduced to unify the spatial coordinate systems between input dose maps and output fluence maps. The network employs an end-to-end joint prediction strategy and is trained on synthetically generated clinical plans from Eclipse to enhance generalizability. The model achieves inference time under 20 ms, improves PSNR by approximately 8 dB over U-Net, and yields DVH metrics highly consistent with clinical plans; SSIM and PSNR both significantly surpass baseline methods. To our knowledge, this is the first approach enabling millisecond-level, high-fidelity, full-arc fluence map generation—establishing a new paradigm for real-time VMAT planning.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Planning, Routing, and Scheduling: Planning with Language Models
• Machine Learning: Deep Neural Architectures and Foundation Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Volumetric Modulated Arc Therapy (VMAT) revolutionizes cancer treatment by precisely delivering radiation while sparing healthy tissues. Fluence maps generation, crucial in VMAT planning, traditionally involves complex and iterative, and thus time consuming processes. These fluence maps are subsequently leveraged for leaf-sequence. The deep-learning approach presented in this article aims to expedite this by directly predicting fluence maps from patient data. We developed a 3D network which we trained in a supervised way using a combination of L1 and L2 losses, and RT plans generated by Eclipse and from the REQUITE dataset, taking the RT dose map as input and the fluence maps computed from the corresponding RT plans as target. Our network predicts jointly the 180 fluence maps corresponding to the 180 control points (CP) of single arc VMAT plans. In order to help the network, we pre-process the input dose by computing the projections of the 3D dose map to the beam's eye view (BEV) of the 180 CPs, in the same coordinate system as the fluence maps. We generated over 2000 VMAT plans using Eclipse to scale up the dataset size. Additionally, we evaluated various network architectures and analyzed the impact of increasing the dataset size. We are measuring the performance in the 2D fluence maps domain using image metrics (PSNR, SSIM), as well as in the 3D dose domain using the dose-volume histogram (DVH) on a validation dataset. The network inference, which does not include the data loading and processing, is less than 20ms. Using our proposed 3D network architecture as well as increasing the dataset size using Eclipse improved the fluence map reconstruction performance by approximately 8 dB in PSNR compared to a U-Net architecture trained on the original REQUITE dataset. The resulting DVHs are very close to the one of the input target dose.

Problem

• Accelerate VMAT fluence maps generation
• Predict fluence maps using deep learning
• Improve radiotherapy planning efficiency

Innovation

• Deep-learning predicts fluence maps directly
• 3D network trained with L1 and L2 losses
• Beam's eye view preprocessing enhances network input