Research map / Label-efficient and robust learning / Semi-supervised learning
Alternate Diverse Teaching for Semi-supervised Medical Image Segmentation
Zhen Zhao, Zicheng Wang, Dian Yu, Longyue Wang, Yixuan Yuan, Luping Zhou
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 |
|---|---|---|---|
| add RPA alternate diverse updating with both teachers (ablation) | AD-MT · ACDC 5% labeled | mean Dice: 86.83 -> 87.88 | Improved |
| add Conflict-Combating Module (ablation) | AD-MT · ACDC 5% labeled | mean Dice: 87.88 -> 88.75 | Improved |
| use T1 (color augmentation) teacher rather than T2 (mix augmentation) (ablation) | AD-MT single teacher · ACDC 5% labeled | mean Dice: 86.22 -> 86.83 | Improved |
| CCM versus dropping conflicting predictions (ablation) | AD-MT · ACDC 5% labeled | Dice: 86.69 -> 88.75 | Improved |
| CCM versus entropy-based ensembling of teachers (ablation) | AD-MT · ACDC 5% labeled | Dice: 88.11 -> 88.75 | Improved |
| raise confidence threshold tau from 0.7 to 0.95 (ablation) | AD-MT · ACDC 5% labeled | Dice: 86.58 -> 88.75 | Improved |
| raise confidence threshold tau from 0.75 to 0.95 (ablation) | AD-MT · LA 5% labeled | Dice: 89.63 -> 86.94 | Hurt |
| use random rather than fixed switching periods for teacher updates (ablation) | AD-MT RPA · ACDC 5% labeled | Dice | Improved |
| AD-MT versus BCP at 20% labeled | AD-MT · Pancreas-NIH 20% labeled | Dice: 82.91 -> 82.61 | No clear effect |
| AD-MT versus BCP at 10% labeled | AD-MT · Pancreas-NIH 10% labeled | Dice: 73.83 -> 80.21 | Improved |
Closest papers by meaning
- DualFete: Revisiting Teacher-Student Interactions from a Feedback Perspective for Semi-supervised Medical Image Segmentation (AAAI 2026 · similarity 0.90)
- Steady Progress Beats Stagnation: Mutual Aid of Foundation and Conventional Models in Mixed Domain Semi-Supervised Medical Image Segmentation (CVPR 2025 · similarity 0.87)
- GapMatch: Bridging Instance and Model Perturbations for Enhanced Semi-Supervised Medical Image Segmentation (AAAI 2025 · similarity 0.87)
- Annotation Ambiguity Aware Semi-Supervised Medical Image Segmentation (CVPR 2025 · similarity 0.86)
- Dual-calibrated Co-training Framework for Personalized Federated Semi-Supervised Medical Image Segmentation (AAAI 2025 · similarity 0.86)
- SGTC: Semantic-Guided Triplet Co-training for Sparsely Annotated Semi-Supervised Medical Image Segmentation (AAAI 2025 · similarity 0.86)
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.
