轻量级3D医学图像分割模型，精度高且计算开销小。
MLRU++: Multiscale Lightweight Residual UNETR++ with Attention for Efficient 3D Medical Image Segmentation
- 采用轻量化注意力模块与多尺度解码结构，提升特征表达能力。
- 在四个数据集上平均Dice达87.57%~93.00%，优于主流模型。
- 参数量和计算成本更低，适合临床部署与资源受限场景。
精准高效的医学图像分割对临床诊断至关重要，但受解剖变异性和体数据高计算需求挑战。现有混合CNN-Transformer架构虽性能优异，却引入显著复杂度。本文提出MLRU++，一种多尺度轻量残差UNETR++架构，兼顾分割精度与效率。核心创新包括：轻量通道与瓶颈注意力模块（LCBAM），以极低开销增强上下文特征编码；解码器中的多尺度瓶颈块（M2B），通过多分辨率特征聚合捕获细粒度细节。在四个公开基准数据集（Synapse、BTCV、ACDC、Decathlon Lung）上的实验表明，MLRU++达到先进性能，平均Dice分数分别为87.57%（Synapse）、93.00%（ACDC）和81.12%（Lung）。相比领先模型，其在Synapse和ACDC上分别提升5.38%和2.12%的Dice分数，同时显著降低参数量与计算成本。消融实验证实了各组件有效性。结果表明，MLRU++为3D医学图像分割提供了高效实用的解决方案。源代码见：https://github.com/1027865/MLRUPP
原文摘要 · Abstract (English)
Accurate and efficient medical image segmentation is crucial but challenging due to anatomical variability and high computational demands on volumetric data. Recent hybrid CNN-Transformer architectures achieve state-of-the-art results but add significant complexity. In this paper, we propose MLRU++, a Multiscale Lightweight Residual UNETR++ architecture designed to balance segmentation accuracy and computational efficiency. It introduces two key innovations: a Lightweight Channel and Bottleneck Attention Module (LCBAM) that enhances contextual feature encoding with minimal overhead, and a Multiscale Bottleneck Block (M2B) in the decoder that captures fine-grained details via multi-resolution feature aggregation. Experiments on four publicly available benchmark datasets (Synapse, BTCV, ACDC, and Decathlon Lung) demonstrate that MLRU++ achieves state-of-the-art performance, with average Dice scores of 87.57% (Synapse), 93.00% (ACDC), and 81.12% (Lung). Compared to existing leading models, MLRU++ improves Dice scores by 5.38% and 2.12% on Synapse and ACDC, respectively, while significantly reducing parameter count and computational cost. Ablation studies evaluating LCBAM and M2B further confirm the effectiveness of the proposed architectural components. Results suggest that MLRU++ offers a practical and high-performing solution for 3D medical image segmentation tasks. Source code is available at: https://github.com/1027865/MLRUPP
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。