轻量级3D U-Net实现脑肿瘤精准分割，效率更高
DALight-3D: A Lightweight 3D U-Net for Brain Tumor Segmentation from Multi-Modal MRI
- 采用深度可分离卷积与跨片注意力提升计算效率
- 50轮训练下Dice达0.727，参数仅222万
- 适合资源受限场景的医学图像分割应用
从多模态MRI自动分割脑肿瘤仍具挑战，因体积模型通常计算开销大。本文提出DALight-3D，一种紧凑的3D U-Net变体，融合深度可分离3D卷积、标识符条件归一化、跨片注意力和自适应跳跃融合。在医学分割十项赛任务01脑肿瘤基准上，与标准3D U-Net、注意力U-Net、残差3D U-Net及V-Net基线在相同优化设置下对比。50轮训练中，DALight-3D达到0.727平均Dice系数，参数量为222万；而残差3D U-Net为0.710 Dice，参数量320万。组件消融实验显示，移除任意模块（SepConv、标识符条件归一化、CSA或SSFB）均导致性能下降。结果表明，该方法在当前基准下实现了良好的精度-效率权衡。
原文摘要 · Abstract (English)
Automatic brain tumor segmentation from multi-modal MRI remains challenging because volumetric models often incur substantial computational cost. This paper presents DALight-3D, a compact 3D U-Net variant that combines depthwise separable 3D convolutions, identifier-conditioned normalization, cross-slice attention, and adaptive skip fusion. The method is evaluated on the Medical Segmentation Decathlon Task01 BrainTumour benchmark under matched optimization settings against standard 3D U-Net, Attention U-Net, Residual 3D U-Net, and V-Net baselines. In the reported 50-epoch comparison, DALight-3D achieves a mean Dice of 0.727 with 2.22M parameters, compared with 0.710 Dice and 3.20M parameters for Residual 3D U-Net. Component-wise ablations show consistent performance degradation when SepConv, identifier-conditioned normalization, CSA, or SSFB is removed. These results indicate that DALight-3D offers a favorable accuracy-efficiency trade-off within the present benchmark setting.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。