GRACE:检索增强校正框架统一处理多器官医学影像分割纠错
Retrieval-Augmented Correction for Generalist Medical Image Segmentation
做医学影像分割的团队值得看看这篇。以前每个器官都要单独训一个纠错模型,GRACE 训一次就能跨器官复用,遇到新器官只需往知识库加病例、模型不用动,弱分割器的 DSC 能拉 0.35,代码也开源了。
EURECOM 团队在 arXiv 发布论文,提出检索增强分割校正框架 GRACE。该框架只需训练一次,即可替代此前“一个解剖结构训练一个校正器”的做法。评估覆盖 18 个公开数据集(CT、MRI、超声),测试 9 种精修配置和 7 个上游分割器。
技术机制上,GRACE 从一个由已验证历史病例构成的共享知识库中检索相似案例,作为形状先验来修正输入 mask。在弱分割器场景下,输入 DSC 为 0.386,GRACE 最多提升 +0.35 DSC;对输入 DSC 达 0.932 的强分割器,改动幅度很小,说明它主要在低质量预测上发力。leave-one-anatomy-out 实验还验证了免参数索引扩展:把未见器官的已验证病例加入知识库即可,精修器保持冻结。
作者指出,向训练集和索引中都缺失的成像模态迁移仍只能部分成功,且依赖具体解剖结构,是开放研究方向。团队开源了代码,位于 GitHub 的 robustml-eurecom/grace 仓库。
Retrieval-Augmented Correction for Generalist Medical Image Segmentation
Automated segmentation is routinely used as the first step in quantitative image-based clinical workflows, where predicted masks are subsequently reviewed and corrected by human operators. Automated correction of the residual structural errors is currently addressed one anatomy at a time, so the number of trained correctors scales with the number of segmentation targets. We present GRACE, a retrieval-augmented refiner that edits an input mask under context drawn from a shared knowledge base of verified prior cases. The refiner is trained once on a multi-organ corpus spanning compact organs, cavities, branching vessels, and tumour lesions. In the held-out dataset-transfer evaluation, three held-out datasets query a fixed knowledge base built from training data only. A separate leave-one-anatomy-out experiment tests parameter-free index extension: verified cases of an unseen solid organ enter the knowledge base, and the refiner stays frozen. A light shape-aware projection head on a frozen multimodal encoder maps embedding similarity to the shape agreement of the masks, so the knowledge base returns cases that the refiner can use as a shape prior. We evaluate GRACE on 18 public datasets spanning CT, MRI and ultrasound, using nine refinement configurations and seven upstream segmentors that span weak, mid, and strong operating regimes. We report each change with the DSC of its input mask. On the weak tier, the input DSC is 0.386 and GRACE adds up to +0.35 DSC. On a strong segmentor, the input DSC is 0.932 and the mean change is small. Transfer to an imaging modality that is absent from training and from the index is partial and depends on the anatomy, and remains an open research direction. Our results suggest retrieval-conditioned refinement is a practical alternative to anatomy-specific correction models. Code: https://github.com/robustml-eurecom/grace
本条情报归属 AITOP「论文」核心栏目,重点聚焦欧洲人工智能治理与技术生态。在跨区域产业观察坐标中,该动态直接关联「AI 开发者工具链、Coding Agent 与工程化实践」研究主线,反映出当前技术演进与商业落地的最新风向。涉及核心实体与技术议题包括:GRACE、医学影像、RAG、图像分割,读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。
信源核验说明:本文原始线索采编自「arXiv cs.AI」(@Vincenzo Marcianò, Xiaoming Zhang, Michela Antonelli, Sebastien Ourselin, Maria A Zuluaga),经 AITOP 多维评分管线结构化提炼。