Research map / Label-efficient and robust learning / Domain adaptation and generalization
An Iterative Self-Learning Framework for Medical Domain Generalization
Zhenbang Wu, Huaxiu Yao, David Liebovitz, Jimeng Sun
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 |
|---|---|---|---|
| replace the best baseline MMLD with SLDG | Transformer · eICU readmission | AUPRC: 0.167 -> 0.186 | Improved |
| replace MMLD with SLDG | Transformer · MIMIC-IV readmission | AUPRC: 0.250 -> 0.274 | Improved |
| replace MMLD with SLDG | Transformer · MIMIC-IV mortality | AUPRC: 0.393 -> 0.416 | Improved |
| replace the Base model with SLDG | Transformer · eICU readmission | AUPRC: 0.104 -> 0.186 | Improved |
| replace hierarchical clustering with k-Means in SLDG (ablation) | SLDG · eICU readmission | AUPRC: 0.186 -> 0.148 | Hurt |
| replace hierarchical clustering with GMM in SLDG (ablation) | SLDG · eICU readmission | AUPRC: 0.186 -> 0.143 | Hurt |
| replace hierarchical clustering with k-Means (ablation) | SLDG · MIMIC-IV mortality | AUPRC: 0.416 -> 0.388 | Hurt |
| replace hierarchical clustering with GMM (ablation) | SLDG · MIMIC-IV readmission | AUROC: 0.690 -> 0.688 | Hurt |
| use domain IDs such as region and timestamp (DANN) instead of Base | Transformer · eICU readmission | AUPRC: 0.104 -> 0.135 | Improved |
| use MLDG with coarse domain partitions instead of Base | Transformer · eICU readmission | AUPRC: 0.104 -> 0.104 | No clear effect |
| increase the number of clusters (ablation) | SLDG · eICU readmission | AUPRC | Mixed |
| increase the number of iterations (ablation) | SLDG · eICU readmission | AUPRC | Mixed |
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.
- ManyDG — ManyDG: Many-domain Generalization for Healthcare Applications (ICLR 2023) · better on 8 of 8 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
- DearLLM: Enhancing Personalized Healthcare via Large Language Models-Deduced Feature Correlations (AAAI 2025 · similarity 0.84)
- TinyMIG: Transferring Generalization from Vision Foundation Models to Single-Domain Medical Imaging (ICML 2025 · similarity 0.84)
- Perturbating, Tuning, and Collaborating: Harnessing Vision Foundation Models for Single Domain Generalization on Medical Imaging (AAAI 2025 · similarity 0.84)
- Symmetric Self-Paced Learning for Domain Generalization (AAAI 2024 · similarity 0.84)
- Learning Generalized Medical Image Segmentation from Decoupled Feature Queries (AAAI 2024 · similarity 0.84)
- Domain Adaptive Diabetic Retinopathy Grading with Model Absence and Flowing Data (CVPR 2025 · similarity 0.83)
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.
