# Self-Supervised Multi-Stage Domain Unlearning for White-Matter Lesion Segmentation

arXiv: https://arxiv.org/abs/2603.13328 · HTML: https://mlsanity.com/paper/2603.13328/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 54 references that are in the map.
- First posted on arXiv 4 Mar 2026.

## Builds on (most cited first, 17 of 17)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Domain-Adversarial Training of Neural Networks](https://mlsanity.com/paper/1505.07818/) (2015)
- [V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation](https://mlsanity.com/paper/1606.04797/) (2016)
- [Deep Learning and the Information Bottleneck Principle](https://mlsanity.com/paper/1503.02406/) (2015)
- [Deep Domain Confusion: Maximizing for Domain Invariance](https://mlsanity.com/paper/1412.3474/) (2014)
- [Underspecification Presents Challenges for Credibility in Modern Machine Learning](https://mlsanity.com/paper/2011.03395/) (2020)
- [Domain Adaptation for Medical Image Analysis: A Survey](https://mlsanity.com/paper/2102.09508/) (2021)
- [Efficient Multi-Scale 3D CNN with Fully Connected CRF for Accurate Brain Lesion Segmentation](https://mlsanity.com/paper/1603.05959/) (2016)
- [SynthSeg: Segmentation of brain MRI scans of any contrast and resolution without retraining](https://mlsanity.com/paper/2107.09559/) (2021)
- [MI-SegNet: Mutual Information-Based US Segmentation for Unseen Domain Generalization](https://mlsanity.com/paper/2303.12649/) (2023)
- [A Contrast-Adaptive Method for Simultaneous Whole-Brain and Lesion Segmentation in Multiple Sclerosis](https://mlsanity.com/paper/2005.05135/) (2020)
- [ImUnity: a generalizable VAE-GAN solution for multicenter MR image harmonization](https://mlsanity.com/paper/2109.06756/) (2021)
- [Learning Multi-Site Harmonization of Magnetic Resonance Images Without Traveling Human Phantoms](https://mlsanity.com/paper/2110.00041/) (2021)
- [Confidence intervals uncovered: Are we ready for real-world medical imaging AI?](https://mlsanity.com/paper/2409.17763/) (2024)
- [Learn to Ignore: Domain Adaptation for Multi-Site MRI Analysis](https://mlsanity.com/paper/2110.06803/) (2021)
- [MSCDA: Multi-level Semantic-guided Contrast Improves Unsupervised Domain Adaptation for Breast MRI Segmentation in Small Datasets](https://mlsanity.com/paper/2301.02554/) (2023)
- [SegHeD: Segmentation of Heterogeneous Data for Multiple Sclerosis Lesions with Anatomical Constraints](https://mlsanity.com/paper/2410.01766/) (2024)

## Abstract (arXiv)

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.