发表机构
University of L’Aquila; Marbl Energy; Luiss Guido Carli（拉奎拉大学; Marbl Energy; 路易斯·吉多·卡利大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究优化质子治疗剂量去噪深度学习模型，用γ指数损失替代MSE提升临床指标，并通过L1结构化剪枝减少参数，实现高通过率与低内存占用。
AI 中文摘要
质子治疗通过布拉格峰相对于常规放疗具有剂量学优势，能够将剂量精确递送至肿瘤区域，同时保护周围组织。蒙特卡洛（MC）模拟提供了最准确的剂量计算，但对于临床工作流程而言计算成本过高。一种有前景的折中方案是使用减少的初级历史数量进行模拟，并通过深度学习去噪步骤恢复高质量的剂量分布。在本工作中，我们以近期文献中的三维多尺度U-Net作为案例研究，探讨了对于临床部署至关重要的两项改进。首先，我们将标准的均方误差（MSE）训练目标替换为可微分的基于γ指数的损失函数，并将其扩展至原生三维操作，证明直接优化临床评估指标在测试集上实现了99.47%的平均伽马通过率（2mm/2%）和95.67%的平均伽马通过率（1mm/1%），而MSE分别为92.10%和71.13%。其次，我们应用L1范数结构化滤波器剪枝以减少模型的内存占用，在部分剪枝下实现了13.3%的参数减少且性能损失可忽略，在微调后的全块剪枝下实现了20%的参数减少。
英文摘要
Proton therapy offers a dosimetric advantage over conventional radiotherapy through the Bragg peak, enabling precise dose delivery to the tumour while sparing surrounding tissue. Monte Carlo (MC) simulation provides the most accurate dose calculation but is computationally expensive for clinical workflows. A promising compromise is to simulate with a reduced number of primary histories and recover a high-quality dose distribution through a deep learning denoising step. In this work, we take a 3D multiscale U-Net from recent literature as a case study and investigate two improvements critical for clinical deployment. First, we replace the standard MSE training objective with a differentiable $γ$-index-based loss, extended to native 3D operation, and demonstrate that directly optimising the clinical evaluation metric yields a mean Gamma Passing Rate of 99.47\% at 2mm/2\% and 95.67\% at 1mm/1\% on the test set, compared to 92.10\% and 71.13\% for MSE. Second, we apply L1-norm structured filter pruning to reduce the model's memory footprint, achieving a 13.3\% parameter reduction with negligible performance loss under partial pruning, and 20\% reduction under full-block pruning after fine-tuning.
