arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2609.18688cs.CVcs.AI

多模态影像中罕见病理检测的通用-专用混合专家模型

Generalist-Specialist Mixture-of-Experts for Rare Pathology Detection in Multimodal Imaging

  • Technical University of Munich (TUM)(慕尼黑工业大学)
  • TUM University Hospital(慕尼黑工业大学附属医院)
  • Hasso Plattner Institute for Digital Engineering(哈索·普拉特纳数字工程研究所)
  • University of Potsdam(波茨坦大学)
  • Imperial College London(伦敦帝国理工学院)

机构由 AI 辅助整理,请以论文原文为准。

Johannes Kaiser, Florian Braunmiller, Daniel Rückert, Georgios Kaissis

AI总结:

针对多模态医学影像中罕见病理检测,提出GS-MoE双分支架构,融合跨模态通用与模态特定专家,在RadImageNet上实现低患病率病理检测从0到最高+0.60 F1的突破,同时减少53%活跃参数。

AI中文摘要:

用于多模态医学影像的AI模型必须在模态特定专门化与跨模态共享表示之间取得平衡,而纯混合专家(MoE)架构目前无法满足这一权衡。基于专家的路由改善了域内学习,但可能牺牲跨模态信号,在我们的实验中,这些信号对罕见(低患病率)病理尤为重要。为解决此问题,我们提出了通用-专用混合专家(GS-MoE),一种双分支MoE架构,通过域约束特征融合将跨模态通用模型与不同的模态特定专家耦合。在RadImageNet(135万张图像,165种病理,三种模态)上,GS-MoE恢复了六种低患病率病理的检测,而所有基线在这些病理上的F1分数均为0,每类增益最高达+0.60 F1。同时,其整体性能甚至略超密集模型和仅专用专家的MoE聚合基线(MCC 0.770),且在推理时使用的活跃参数比最强密集模型少约53%。

英文摘要:

AI models for multimodal medical imaging must balance modality-specific specialization with cross-modal shared representations, a trade-off that pure Mixture-of-Experts (MoE) architectures currently fail to satisfy. Expert-based routing improves in-domain learning but may sacrifice cross-modal signals, which appear particularly important for rare (low-prevalence) pathologies in our experiments. To resolve this, we introduce Generalist-Specialist-MoE (GS-MoE), a two-branch (MoE) architecture that couples a cross-modal generalist model with distinct modality-specific specialists (experts) via domain-constrained feature fusion. On RadImageNet (1.35M images, 165 pathologies, three modalities), GS-MoE recovers detection of six low-prevalence pathologies on which every baseline scores F1 $=$ 0, with per-class gains up to +0.60 F1. It attains this while even slightly exceeding dense and specialist-only MoE aggregate baselines (MCC 0.770), while using ${\sim}53\%$ fewer active parameters at inference than the strongest investigated dense model.

↑