自动分割膝关节软骨与骨组织，支持多中心研究。
LM-CartSeg: Automated Segmentation of Lateral and Medial Cartilage and Subchondral Bone for Radiomics Analysis
- 用两个3D nnU-Net模型加几何规则实现全自动分割。
- 后处理使误差从2.63mm降到0.36mm，DSC达0.91。
- 提取4650个特征，分类准确率最高达0.91，适合临床研究。
膝关节MRI影像组学需鲁棒且解剖意义明确的感兴趣区域（ROIs），同时涵盖软骨与软骨下骨。现有方法多依赖人工勾画，且极少报告质量控制（QC）。本文提出LM-CartSeg，一个全自动管道，可完成软骨/骨分割、内外侧（L/M）解剖分区及影像组学分析。基于SKM-TEA（138膝）和OAIZIB-CM（404膝）数据训练两个3D nnU-Net模型，测试时采用零样本预测并结合几何规则融合：连通域清理、物理空间中10mm软骨下骨带构建，以及基于主成分分析（PCA）与k-means的数据驱动胫骨内外侧分界。在OAIZIB-CM测试集（103膝）和SKI-10（100膝）上评估分割性能，使用体积与厚度特征进行质量控制。从10个ROI提取4,650个非形状影像组学特征，用于分析腔室间相似性、ROI大小依赖性，以及在OAIZIB-CM和临床Po-OA队列（185膝）上的骨关节炎（OA） vs 非OA分类。结果显示，后处理将宏观平均表面距离（ASSD）从2.63mm降至0.36mm，HD95从25.2mm降至3.35mm，DSC约0.91；零样本在SKI-10上DSC约为0.80。几何分界规则在不同数据集上表现稳定，而直接训练的L/M nnU-Net出现域依赖性侧向误判。每块ROI中仅6%-12%特征与体积或厚度强相关。基于影像组学的模型在OAIZIB-CM上达到最高AUC 0.91，Po-OA队列达0.83，显著优于仅依赖尺寸特征的模型。结论：LM-CartSeg生成自动、经质量控制的ROIs与具有区分能力的影像组学特征，超越简单形态测量，为多中心膝骨关节炎影像组学研究提供实用基础。代码已公开于https://github.com/jukieCheung/LM-CartSeg。
原文摘要 · Abstract (English)
Background and Objective: Radiomics of knee MRI requires robust, anatomically meaningful regions of interest (ROIs) that jointly capture cartilage and subchondral bone. Most existing work relies on manual ROIs and rarely reports quality control (QC). We present LM-CartSeg, a fully automatic pipeline for cartilage/bone segmentation, geometric lateral/medial (L/M) compartmentalization and radiomics analysis. Methods:Two 3D nnU-Net models were trained on SKM-TEA (138 knees) and OAIZIB-CM (404 knees). At test time, zero-shot predictions were fused and refined by simple geometric rules: connected-component cleaning,construction of 10mm subchondral bone bands in physical space, and a data-driven tibial L/M split based on PCA and $k$-means. Segmentation was evaluated on an OAIZIB-CM test set (103 knees) and on SKI-10 (100 knees). QC used volume and thickness signatures. From 10 ROIs we extracted 4,650 non-shape radiomic features to study inter-compartment similarity, dependence on ROI size, and OA vs. non-OA classification on OAIZIB-CM and a clinical Po-OA cohort (185 knees). Results: Post-processing improved macro ASSD on OAIZIB-CM from 2.63 to 0.36mm and HD95 from 25.2 to 3.35mm, with DSC approx 0.91; zero-shot DSC on SKI-10 was approx 0.80. The geometric L/M rule produced stable compartments across datasets, whereas a direct L/M nnU-Net showed domain-dependent side swaps. Only 6-12% of features per ROI were strongly correlated with volume or thickness. Radiomics-based models achieved AUC up to 0.91 (OAIZIB-CM) and 0.83 (Po-OA), clearly exceeding models restricted to size-linked features. Conclusions: LM-CartSeg yields automatic, QC'd ROIs and radiomic features that carry discriminative information beyond simple morphometry, providing a practical foundation for multi-centre knee OA radiomics studies. Code is available at https://github.com/jukieCheung/LM-CartSeg.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。