MedCRP-CL: 通过贝叶斯非参数语义模态发现实现连续医学图像分割
MedCRP-CL: Continual Medical Image Segmentation via Bayesian Nonparametric Semantic Modality Discovery
AI总结：
该研究提出MedCRP-CL框架，通过在线任务结构发现和结构感知的连续学习方法，解决医学图像分割在持续学习中的挑战，实现了73.3%的Dice得分和仅4.1%的遗忘率。
摘要：
医学图像分割在持续学习中面临根本性挑战：数据按顺序从异质源到来，但有效的持续学习需要发现哪些任务共享足够的结构以受益于联合学习。现有方法要么在所有任务上应用统一约束，导致任务冲突时发生灾难性遗忘，要么需要预定义的任务分组，无法预测未来任务多样性。我们引入MedCRP-CL框架，实现在线任务结构发现和结构感知的持续学习。利用中文餐厅过程（CRP），我们的方法从临床文本提示中动态推断任务分组，无需预定义聚类数量或访问未来任务。我们将发现的分组称为语义模态，因为它们通过整合解剖区域和病理背景捕捉更细粒度的结构。在发现的结构指导下，我们维护语义模态特定的LoRA适配器，通过内模态EWC正则化，确保在不同任务组之间参数隔离，同时促进相似组的知识转移。该框架也是无回放的，仅存储聚合统计信息而非原始患者数据。在16个医学分割任务和四种成像模态上的实验表明，MedCRP-CL实现了73.3%的Dice得分，仅4.1%的遗忘率，优于最佳基线8.0%，同时仅需6倍更少的参数。代码可在https://github.com/zygao930/MedCRP-CL获取。
英文摘要：
Medical image segmentation faces a fundamental challenge in continual learning: data arrives sequentially from heterogeneous sources, yet effective continual learning requires discovering which tasks share sufficient structure to benefit from joint learning. Existing methods either apply uniform constraints across all tasks, causing catastrophic forgetting when tasks conflict, or require predefined task groupings that cannot anticipate future task diversity. We introduce MedCRP-CL, a framework that performs online task structure discovery and structure-aware continual learning. Leveraging the Chinese Restaurant Process (CRP), our method dynamically infers task groupings from clinical text prompts as tasks arrive, without requiring predefined cluster counts or access to future tasks. We term these discovered groupings semantic modalities, as they capture finer-grained structure than physical imaging modalities by integrating anatomical region and pathological context. Guided by this discovered structure, we maintain semantic modality-specific LoRA adapters regularized by intra-modality EWC, ensuring parameter isolation across dissimilar task groups while facilitating knowledge transfer within similar ones. The framework is also replay-free, storing only aggregate statistics rather than raw patient data. Experiments on 16 medical segmentation tasks across four imaging modalities demonstrate that MedCRP-CL achieves 73.3% Dice score with only 4.1% forgetting, outperforming the best baseline by 8.0% while requiring 6$\times$ fewer parameters. Code is available at https://github.com/zygao930/MedCRP-CL.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
