发表机构
Technical University of Denmark; Department of Applied Mathematics and Computer Science（丹麦技术大学; 应用数学与计算机科学系）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对医学成像中标记数据稀缺问题，提出COJEPA自监督框架，结合联合嵌入预测架构与对比损失，用于体积脑MRI。通过特定技术扩展到3D，在多任务中评估，取得了如最佳同卵双胞胎召回率等成果，证明组合目标能产生优质表示。
AI 中文摘要
自监督学习为医学成像提供了一种有吸引力的方法，因为标记数据稀缺且采集成本高。我们提出了COJEPA，这是一种用于体积脑MRI的自监督框架，它将联合嵌入预测架构（JEPA）与对比损失（CO）相结合，针对两个互补属性：局部预测性和全局可区分性。该模型在来自两个队列（HCP - YA和AABC，N = 2286，年龄22至90岁）的T1加权结构MRI上进行无标签训练，通过前景感知块掩码、分层卷积补丁嵌入和世界空间正弦位置编码将I - JEPA扩展到3D。我们在零样本双胞胎检索、脑肿瘤分割（BraTS 2024）和年龄回归（OpenBHB）中评估了所有三个目标。COJEPA在排名@1时实现了最佳的同卵双胞胎召回率（0.84），在OpenBHB 3.0T上实现了最佳的微调年龄平均绝对误差（2.55岁），并且在BraTS全肿瘤骰子系数上与CO匹配，表明组合目标产生了同时具有区分性和局部结构化的表示。
英文摘要
Self-supervised learning offers a compelling approach for medical imaging, where labeled data are scarce and acquisition costs are high. We present COJEPA, a self-supervised framework for volumetric brain MRI that combines a joint-embedding predictive architecture (JEPA) with a contrastive loss (CO), targeting two complementary properties: local predictivity and global discriminability. The model is trained without labels on T1-weighted structural MRI from two cohorts (HCP-YA and AABC, $N{=}2286$, ages 22 to 90), extending I-JEPA to 3D with foreground-aware block masking, a hierarchical convolutional patch embedding, and world-space sinusoidal positional encodings. We evaluate all three objectives across zero-shot twin retrieval, brain tumor segmentation (BraTS 2024), and age regression (OpenBHB). COJEPA achieves the best monozygotic twin recall at rank@1 (0.84), the best finetuning age MAE (2.55 years on OpenBHB 3.0T), and matches CO on BraTS whole-tumor Dice, demonstrating that the combined objective yields representations that are simultaneously discriminative and locally structured.
