发表机构
State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences; School of Artificial Intelligence, University of Chinese Academy of Sciences; Department of Radiology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College（中国科学院自动化研究所多模态人工智能系统全国重点实验室; 中国科学院大学人工智能学院; 中国医学科学院北京协和医学院北京协和医院放射科）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对全身扩散加权成像中多发性骨髓瘤病灶分割难题，提出两阶段框架：先由ADC图像生成骨骼ROI，再用解剖引导多模态U-Net（AMU-Net）融合ADC，实现高效分割，平均Dice达76.2%。
AI 中文摘要
全身扩散加权成像（WB-DWI）广泛用于多发性骨髓瘤（MM）评估，然而由于解剖结构界定有限以及骨髓高信号的低特异性，自动化病灶分割仍具挑战性。已有研究引入骨骼感兴趣区（ROI）信息和表观扩散系数（ADC）图以缓解这些模糊性，但实际限制依然存在。骨骼ROI构建通常依赖昂贵的人工标注、图像配准或专用骨骼模型，而ADC通常仅通过简单通道融合纳入，限制了其提供互补性结构和病灶判别线索的能力。为解决这些限制，我们提出一个用于WB-DWI上MM病灶分割的两阶段框架。在第一阶段，我们在无专用骨骼标签的情况下，从ADC图像训练一个骨骼ROI生成模型，为病灶分析提供高效且实用的解剖先验。在第二阶段，我们提出解剖引导多模态U-Net（AMU-Net），其以与临床病灶评估一致的方式利用ADC，而非将其视为通用辅助模态。大量实验证明了所提方法的有效性和实用性。它在评估方法中取得最佳总体性能，平均Dice得分为76.2%。
英文摘要
Whole-body diffusion-weighted imaging (WB-DWI) is widely used for multiple myeloma (MM) assessment, yet automated lesion segmentation remains challenging due to limited anatomical delineation and the low specificity of marrow hyperintensity. Existing studies have introduced bone region-of-interest (ROI) information and apparent diffusion coefficient (ADC) maps to mitigate these ambiguities, but practical limitations remain. Bone ROI construction often relies on costly manual annotation, image registration, or dedicated bone models, while ADC is usually incorporated only through simple channel fusion, limiting its ability to provide complementary structural and lesion-discriminative cues. To address these limitations, we propose a two-stage framework for MM lesion segmentation on WB-DWI. In the first stage, we train a bone ROI generation model from ADC images without dedicated bone labels, providing an efficient and practical anatomical prior for lesion analysis. In the second stage, we propose Anatomy-guided Multimodal U-Net (AMU-Net), which leverages ADC in a manner consistent with clinical lesion assessment rather than treating it as a generic auxiliary modality. Extensive experiments demonstrate the effectiveness and practicality of the proposed method. It achieves the best overall performance among the evaluated methods, with a mean Dice score of 76.2%.
