从正常中识别异常：正常肾形态表征中的肾小球异常

中文摘要 AI 辅助

肾小球病理的细粒度评估必须区分正常肾小球与异常类型，如全球性和节段性肾小球硬化、废弃性、缺血性、凝固性、消失性和无小管肾小球。监督分类需要每个类别的标注样本，当亚型在训练队列中罕见或缺失时，这变得不切实际。一类异常检测通过建模正常数据并评分偏差提供了一种替代方案，从而能够检测出以前未见过的异常。我们使用Omni-Seg的冻结残差U-Net骨干网络，该网络预训练用于分割结构正常的肾脏原基，无需异常亚型标签。我们提出NoRDeC（正常参考检测与表征），一个结合马氏正常参考评分与逐层表征分析的框架，以确定肾小球病理是否以及在哪里被编码，空间聚合如何影响检测，以及异常是否以不同方式改变层间关系。使用来自两个机构的肾小球图像，我们评估骨干层和聚合策略，将NoRDeC与PaDiM和PatchCore进行比较，并使用中心核对齐（CKA）分析表征。第4层结合Center-70聚合实现了合并AUROC为$0.926\pm0.013$。NoRDeC在七个异常类别中的六个以及合并分析中达到了最高AUROC，而CKA表明亚型依赖的层间关系变化未被异常分数单独捕获。正常参考模型仅使用正常肾小球拟合；异常标签用于配置选择、评估和表征分析中的分组。这些结果表明，冻结的肾脏特征提取器可以支持肾小球异常的检测和表征级分析，而无需使用异常样本来拟合检测器。

英文摘要

Fine-grained evaluation of glomerular pathology must distinguish normal glomeruli from abnormalities such as global and segmental glomerulosclerosis, obsolescent, ischemic, solidified, disappearing, and atubular glomeruli. Supervised classification requires labeled examples of every category, which is impractical when subtypes are rare or absent from the training cohort. One-class anomaly detection offers an alternative by modeling normal data and scoring deviations, allowing previously unseen abnormalities to be detected. We use the frozen residual U-Net backbone of Omni-Seg, pretrained to segment structurally normal renal primitives without abnormal-subtype labels. We propose NoRDeC (Normal-Reference Detection and Characterization), a framework combining Mahalanobis normal-reference scoring with layer-wise representation analysis to determine whether and where glomerular pathology is encoded, how spatial aggregation affects detection, and whether abnormalities alter inter-layer relationships differently. Using glomerular images from two institutions, we evaluate backbone layers and aggregation strategies, compare NoRDeC with PaDiM and PatchCore, and analyze representations using centered kernel alignment (CKA). Layer 4 with Center-70 aggregation achieved a pooled AUROC of $0.926\pm0.013$. NoRDeC achieved the highest AUROC in six of seven abnormality categories and in the pooled analysis, while CKA suggested subtype-dependent changes in inter-layer relationships not captured by anomaly scores alone. The normal-reference model is fitted using only normal glomeruli; abnormality labels are used for configuration selection, evaluation, and grouping in the representation analysis. These results show that a frozen renal feature extractor can support both detection and representation-level characterization of glomerular abnormalities without using abnormal examples to fit the detector.

发表机构

• Cornell University（康奈尔大学）
• Cornell Tech（康奈尔科技校区）
• Sichuan University（四川大学）
• Johns Hopkins University（约翰斯·霍普金斯大学）
• Vanderbilt University（范德堡大学）
• University of Regensburg（雷根斯堡大学）
• New York University（纽约大学）
• Vanderbilt University Medical Center（范德堡大学医学中心）
• New York Medical College（纽约医学院）
• Weill Cornell Medicine（威尔·康奈尔医学院）

机构由 AI 辅助整理，请以论文原文为准。