[论文解读] Self-Supervised Multi-Stage Domain Unlearning for White-Matter Lesion Segmentation

Domen Preložnik, Žiga Špiclin|dCOBISS.SI Digital Repository|Mar 4, 2026

Medical Image Segmentation Techniques被引用 0

一句话总结

提出了一种将自监督多阶段域取消学习（SSMSU）策略与 nnU-Net 相结合的方案，在没有目标域数据的情况下改善多扫描仪 FLAIR MRI 的白质病变分割的跨域性能。

ABSTRACT

Inter-scanner variability of magnetic resonance imaging has an adverse impact on the diagnostic and prognostic quality of the scans and necessitates the development of models robust to domain shift inflicted by the unseen scanner data. Review of recent advances in domain adaptation showed that efficacy of strategies involving modifications or constraints on the latent space appears to be contingent upon the level and/or depth of supervision during model training. In this paper, we therefore propose an unsupervised domain adaptation technique based on self-supervised multi-stage unlearning (SSMSU). Building upon the state-of-the-art segmentation framework nnU-Net, we employ deep supervision at deep encoder stages using domain classifier unlearning, applied sequentially across the deep stages to suppress domain-related latent features. Following self-configurable approach of the nnU-Net, the auxiliary feedback loop implements a self-supervised backpropagation schedule for the unlearning process, since continuous unlearning was found to have a detrimental effect on the main segmentation task. Experiments were carried out on four public datasets for benchmarking white-matter lesion segmentation methods. Five benchmark models and/or strategies, covering passive to active unsupervised domain adaptation, were tested. In comparison, the SSMSU demonstrated the advantage of unlearning by enhancing lesion sensitivity and limiting false detections, which resulted in higher overall segmentation quality in terms of segmentation overlap and relative lesion volume error. The proposed model inputs only the FLAIR modality, which simplifies preprocessing pipelines, eliminates the need for inter-modality registration errors and harmonization, which can introduce variability. Source code is available on https://github.com/Pubec/nnunetv2-unlearning.

研究动机与目标

• 解决在未见 MRI 扫描仪上 WML 分割性能下降的跨扫描变异性问题。
• 开发一种在无目标域数据情况下抑制域特征的无监督域适应方法。
• 将多阶段领域取消学习整合到具自我配置监 supervision 的 nnU-Net 框架中。
• 在四个公开的多扫描仪 WML 数据集上评估鲁棒性，并与最先进基线进行比较。

提出的方法

• 在每个编码器阶段（六个阶段）扩展 nnU-Net 的域分类器取消学习。
• 在来自编码器阶段 Ex 的特征上训练分类器 Cx 以预测扫描仪/域；使用基于 KL 散度的混淆损失来抑制域信息。
• 使用自监督学习–取消学习时间表，包含热身阶段，随后交替进行学习与取消学习。
• 将取消学习与分割损失平衡以保留任务性能；采用基于域分类器准确率的课程式耐心机制（UBA）。
• 输入仅限 FLAIR 模态以避免跨模态问题并简化预处理。

Figure 1. Comparison of MRI scans between different domains.

实验结果

研究问题

• RQ1 自监督的多阶段取消学习是否能抑制扫描仪特定特征并在没有目标域数据的情况下改善跨域 WML 分割？
• RQ2 取消学习深度（哪些编码器阶段）以及学习–取消学习比在看得见和看不见域上的分割性能有何影响？
• RQ3 相较于传统预处理流程，最小化预处理（仅 FLAIR、降低 Harmonization）对跨域泛化有何影响？

主要发现

• 与基线相比，SSMSU 提高了病变灵敏度（TPR/LTPR 更高）并降低了误检（RVE 更低），在若干指标上具有统计显著性。
• 在 unseen 的多扫描数据集（MSSEG、MSLJ、ISBI）上，SSMSU 的 DSC 及相关指标高于五种基线策略。
• 更深的编码器阶段取消学习可实现更强的域抑制和更好的泛化，而对所有阶段进行取消学习可能会降低性能。
• 在看得见数据上保持竞争力的 DSC，同时实现更强的跨域提升，但 LFDR 增加，表明存在少量额外的小型假阳性。
• 仅使用 FLAIR 输入简化预处理并增强对跨扫描变异性的鲁棒性。

Figure 2. (a) MRI preprocessing and (b) self-supervised multi-stage unlearning with nnU-Net.