mlsanityData to 7 Oct 2026

SRPL-SFDA: SAM-Guided Reliable Pseudo-Labels for Source-Free Domain Adaptation in Medical Image Segmentation

arXiv 2506.09403 ↗First posted 11 Jun 2025cs.CVcited by 3

Not on the map. Why?

Abstract (arXiv)

Domain Adaptation (DA) is crucial for robust deployment of medical image segmentation models when applied to new clinical centers with significant domain shifts. Source-Free Domain Adaptation (SFDA) is appealing as it can deal with privacy concerns and access constraints on source-domain data during adaptation to target-domain data. However, SFDA faces challenges such as insufficient supervision in the target domain with unlabeled images. In this work, we propose a Segment Anything Model (SAM)-guided Reliable Pseudo-Labels method for SFDA (SRPL-SFDA) with three key components: 1) Test-Time Tri-branch Intensity Enhancement (T3IE) that not only improves quality of raw pseudo-labels in the target domain, but also leads to SAM-compatible inputs with three channels to better leverage SAM's zero-shot inference ability for refining the pseudo-labels; 2) A reliable pseudo-label selection module that rejects low-quality pseudo-labels based on Consistency of Multiple SAM Outputs (CMSO) under input perturbations with T3IE; and 3) A reliability-aware training procedure in the unlabeled target domain where reliable pseudo-labels are used for supervision and unreliable parts are regularized by entropy minimization. Experiments conducted on two multi-domain medical image segmentation datasets for fetal brain and the prostate respectively demonstrate that: 1) SRPL-SFDA effectively enhances pseudo-label quality in the unlabeled target domain, and improves SFDA performance by leveraging the reliability-aware training; 2) SRPL-SFDA outperformed state-of-the-art SFDA methods, and its performance is close to that of supervised training in the target domain. The code of this work is available online: https://github.com/HiLab-git/SRPL-SFDA.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 33 of its 64 references that are in the map. First posted on arXiv 11 Jun 2025.

Builds on

33 of 64 references.

References · 33

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  2. 2023Segment Anythingcited by 8,158
  3. 2020Contrastive Learning for Unpaired Image-to-Image Translationcited by 637
  4. 2018ADVENT: Adversarial Entropy Minimization for Domain Adaptation in Semantic Segmentationcited by 623
  5. 2023Medical SAM Adapter: Adapting Segment Anything Model for Medical Image Segmentationcited by 485
  6. 2021Domain Adaptation for Medical Image Analysis: A Surveycited by 351
  7. 2022Assembly101: A Large-Scale Multi-View Video Dataset for Understanding Procedural Activitiescited by 284
  8. 2018Aleatoric uncertainty estimation with test-time augmentation for medical image segmentation with convolutional neural networkscited by 278
  9. 2020Evaluating Prediction-Time Batch Normalization for Robustness under Covariate Shiftcited by 235
  10. 2023Segment Anything in Medical Imagescited by 226
  11. 2020Source Data-absent Unsupervised Domain Adaptation through Hypothesis Transfer and Labeling Transfercited by 173
  12. 2021Source-Free Domain Adaptation for Semantic Segmentationcited by 168
21 more
  1. 2020Unsupervised Bidirectional Cross-Modality Adaptation via Deeply Synergistic Image and Feature Alignment for Medical Image Segmentationcited by 134
  2. 2020MS-Net: Multi-Site Network for Improving Prostate Segmentation with Heterogeneous MRI Datacited by 121
  3. 2017DeepIGeoS: A Deep Interactive Geodesic Framework for Medical Image Segmentationcited by 110
  4. 2023A Comprehensive Survey on Source-free Domain Adaptationcited by 104
  5. 2020Test-Time Adaptable Neural Networks for Robust Medical Image Segmentationcited by 102
  6. 2017Deep MR to CT Synthesis using Unpaired Datacited by 86
  7. 20233DSAM-adapter: Holistic adaptation of SAM from 2D to 3D for promptable tumor segmentationcited by 70
  8. 2023Guiding Pseudo-labels with Uncertainty Estimation for Source-free Unsupervised Domain Adaptationcited by 63
  9. 2021Source-Free Domain Adaptive Fundus Image Segmentation with Denoised Pseudo-Labelingcited by 61
  10. 2021DINE: Domain Adaptation from Single and Multiple Black-box Predictorscited by 51
  11. 2024Unleashing the Potential of SAM for Medical Adaptation via Hierarchical Decodingcited by 48
  12. 2021Source-Free Domain Adaptation for Image Segmentationcited by 45
  13. 2021Deep Symmetric Adaptation Network for Cross-modality Medical Image Segmentationcited by 39
  14. 2022No More Strided Convolutions or Pooling: A New CNN Building Block for Low-Resolution Images and Small Objectscited by 26
  15. 2023UPL-SFDA: Uncertainty-aware Pseudo Label Guided Source-Free Domain Adaptation for Medical Image Segmentationcited by 21
  16. 2022Memory Consistent Unsupervised Off-the-Shelf Model Adaptation for Source-Relaxed Medical Image Segmentationcited by 15
  17. 2022PyMIC: A deep learning toolkit for annotation-efficient medical image segmentationcited by 14
  18. 2023Prior-guided Source-free Domain Adaptation for Human Pose Estimationcited by 13
  19. 2022Unsupervised Domain Adaptation for Segmentation with Black-box Source Modelcited by 7
  20. 2021CADA: Multi-scale Collaborative Adversarial Domain Adaptation for Unsupervised Optic Disc and Cup Segmentationcited by 5
  21. 2023Post-Deployment Adaptation with Access to Source Data via Federated Learning and Source-Target Remote Gradient Alignmentcited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 1; 2026: 2.

3 papers.

  1. 2025Uncertainty-Aware Extreme Point Tracing for Weakly Supervised Ultrasound Image Segmentationcited by 0
  2. 2026Tell2Adapt: A Unified Framework for Source Free Unsupervised Domain Adaptation via Vision Foundation Modelcited by 0
  3. 2026Dual-Foundation Models for Unsupervised Domain Adaptationcited by 0