Research map / Label-efficient and robust learning / Semi-supervised learning
PMT: Progressive Mean Teacher via Exploring Temporal Consistency for Semi-Supervised Medical Image Segmentation
Ning Gao, Sanping Zhou, Le Wang, Nanning Zheng
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 pseudo label filtering (PLF) with progressive design (ablation) | co-training of two VNets · LA (10% labeled) | Dice: 88.28 -> 89.90 | Improved |
| add discrepancy driven alignment (DDA) with progressive design (ablation) | co-training of two VNets · LA (10% labeled) | Dice: 88.28 -> 90.43 | Improved |
| add PLF on top of DDA (ablation) | PMT · LA (10% labeled) | Dice: 90.43 -> 90.43 (95HD 5.98 -> 5.49) | No clear effect |
| add mean teacher architecture to PLF+DDA (ablation) | PMT · LA (10% labeled) | Dice: 90.43 -> 90.81 | Improved |
| scale regularization strength lambda up to 100 for mean teacher term (ablation) | PMT · LA (10% labeled) | Dice: 90.81 -> 88.00 (lambda1 20, lambda2 100) | Hurt |
| vary lambda within a moderate range (ablation) | PMT · LA (10% labeled) | Dice: 89.66 to 90.81 | No clear effect |
| use PMT instead of BCP | semi-supervised segmentation · Pancreas-NIH (10% labeled) | Dice: 73.83 -> 81.00 | Improved |
| use PMT instead of BCP | semi-supervised segmentation · Pancreas-NIH (20% labeled) | 95HD: 6.43 -> 7.60 | Hurt |
| use PMT with 5% labeled data compared to MCF with 10% | semi-supervised segmentation · LA | Dice: 88.71 (MCF 10%) vs 89.47 (PMT 5%) | Improved |
Closest papers by meaning
- Medical Image Segmentation with Minimal Labeling Effort: How Far Can We Push the Limits? (AAAI 2026 · similarity 0.89)
- Rethinking Semi-Supervised Medical Image Segmentation: A Variance-Reduction Perspective (NeurIPS 2023 · similarity 0.89)
- Steady Progress Beats Stagnation: Mutual Aid of Foundation and Conventional Models in Mixed Domain Semi-Supervised Medical Image Segmentation (CVPR 2025 · similarity 0.89)
- SGTC: Semantic-Guided Triplet Co-training for Sparsely Annotated Semi-Supervised Medical Image Segmentation (AAAI 2025 · similarity 0.88)
- GapMatch: Bridging Instance and Model Perturbations for Enhanced Semi-Supervised Medical Image Segmentation (AAAI 2025 · similarity 0.88)
- Semi-supervised TEE Segmentation via Interacting with SAM Equipped with Noise-Resilient Prompting (AAAI 2024 · 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.
