Research map / Label-efficient and robust learning / Domain adaptation and generalization
FairDomain: Achieving Fairness in Cross-Domain Medical Image Segmentation and Classification
Yu Tian, Congcong Wen, Min Shi, Muhammad Muneeb Afzal, Hao Huang, Muhammad Osama Khan, Yan Luo, Yi Fang, Mengyu Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| apply domain shift from EnFace to SLO fundus (no adaptation) | TransUNet baseline · FairDomain-Segmentation, rim, gender | ES-Dice: 0.854 -> 0.485 | Hurt |
| add fair identity attention to DAFormer for domain adaptation (cup) | DAFormer · FairDomain-Segmentation, cup, gender | ES-Dice: 0.781 -> 0.802 | Improved |
| add fair identity attention to DAFormer for domain adaptation (rim) | DAFormer · FairDomain-Segmentation, rim, gender | ES-Dice: 0.344 -> 0.528 | Improved |
| add fair identity attention to DAFormer, compared with CBST on rim DA | DAFormer-FIA · FairDomain-Segmentation, rim, gender, DA | ES-Dice: CBST 0.693 vs DAFormer-FIA 0.528 | Hurt |
| add fair identity attention to DAFormer for domain generalization | DAFormer · FairDomain-Segmentation, cup, ethnicity | ES-Dice: 0.746 -> 0.785 | Improved |
| add fair identity attention to CDTrans for domain adaptation classification | CDTrans · FairDomain-Classification, gender | ES-AUC: 0.631 -> 0.633 | Improved |
| add fair identity attention to IRM for domain generalization | IRM · FairDomain-Classification, race | AUC overall: 0.672 -> 0.696 | Improved |
| add fair identity attention to IRM, Hispanic group | IRM · FairDomain-Classification, ethnicity, Hispanic group | AUC: 0.558 -> 0.580 | Improved |
| add FIA module to IRM | IRM · computational cost | training time per epoch (s): 149 -> 157 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DAFormer — DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation (CVPR 2022) · better on 7 of 7 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- FairSeg: A Large-Scale Medical Image Segmentation Dataset for Fairness Learning Using Segment Anything Model with Fair Error-Bound Scaling (ICLR 2024 · similarity 0.91)
- FairMedFM: Fairness Benchmarking for Medical Imaging Foundation Models (NeurIPS 2024 · similarity 0.90)
- The Boundaries of Fair AI in Medical Image Prognosis: A Causal Perspective (NeurIPS 2025 · similarity 0.89)
- FairCLIP: Harnessing Fairness in Vision-Language Learning (CVPR 2024 · similarity 0.89)
- FairTune: Optimizing Parameter Efficient Fine Tuning for Fairness in Medical Image Analysis (ICLR 2024 · similarity 0.89)
- Distribution-aware Fairness Learning in Medical Image Segmentation From A Control-Theoretic Perspective (ICML 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
