面向多结构超低场（ULF）儿科脑部MRI分割的非对称配对标注学习
Asymmetric Paired-Annotation Learning for Multi-Structure ULF Pediatric Brain MRI Segmentation
浏览论文内容
中文总结 AI 辅助
针对0.064 T ULF儿科脑部MRI分割难题，提出基于nnU-Net的非对称监督策略AURA，利用两类非等效标注训练，在16例数据上取得优于基线的分割性能。
中文摘要 AI 辅助
便携式超低场（ULF）MRI可扩大儿科神经成像的可及性，但0.064 T下的分割仍具挑战性，原因包括解剖边界轮廓模糊、小结构可能仅部分可见，以及高场参考图像可能存在局部配准错误。LISA 2026挑战赛提供两种非等效标注，反映不同来源的解剖学证据：定义评分目标的高场衍生（HF）掩码，以及与ULF可见解剖结构对齐的低场编辑（LF）掩码。在本挑战赛报告中，我们描述了AURA——一种基于nnU-Net的非对称监督策略，该策略将这些标注视为不同的观测结果，而非可互换的真值。AURA以HF掩码为训练锚点，通过基于标签分歧、边界、预测不确定性、类别可靠性及训练阶段的有界可靠性门，整合LF掩码。在16例的开发拆分中，HF监督基线、AURA及其集成模型的Dice分数分别为0.7984、0.7950和0.7988，而集成模型的HD95为1.8892，ASSD为0.7855。这些结果为LISA 2026挑战赛内AURA提供了初步评估，并为在隐藏测试集及外部ULF队列上开展进一步评估提供了动力。我们的代码和预训练模型可在该https链接获取：A-nnU-Net-based-asymmetric-supervision-strategy
英文摘要
Portable ultra-low-field (ULF) MRI can expand access to pediatric neuroimaging, but segmentation at 0.064 T remains challenging because anatomical boundaries are weakly delineated, small structures may be only partially visible, and high-field references can be locally misregistered. The LISA 2026 Challenge provides two non-equivalent annotations reflecting different sources of anatomical evidence: a highfield-derived (HF) mask defining the scored target and a low-field-edited (LF) mask aligned with visible ULF anatomy. In this challenge report, we describe AURA, an nnU-Net-based asymmetric supervision strategy that treats these annotations as distinct observations rather than interchangeable ground truths. AURA anchors training to the HF mask and incorporates the LF mask through a bounded reliability gate based on label disagreement, boundaries, predictive uncertainty, class reliability, and training stage. On a 16-case development split, the HF-supervised baseline, AURA, and their ensemble achieved Dice scores of 0.7984, 0.7950, and 0.7988, respectively, while the ensemble achieved an HD95 of 1.8892 and an ASSD of 0.7855. These results provide a preliminary evaluation of AURA within the LISA 2026 Challenge and motivate further assessment on the hidden test set and external ULF cohorts. Our code and pretrained models are available at https://github.com/minhdang050806/ A-nnU-Net-based-asymmetric-supervision-strategy.
发表机构
- VinUniversity
- College of Engineering & Computer Science, VinUniversity（VinUniversity工程与计算机科学学院）
- Hanoi University of Science and Technology（河内科技大学）
- University of Information Technology, VNU-HCM（胡志明市国家大学信息技术大学）
- Carnegie Mellon University（卡内基梅隆大学）
- Northwestern University（西北大学）
- Center for Innovations in Health Sciences, VinUniversity（VinUniversity健康科学创新中心）
机构由 AI 辅助整理，请以论文原文为准。
