Research map / Label-efficient and robust learning / Semi-supervised learning
Pseudo-Label Guided Contrastive Learning for Semi-Supervised Medical Image Segmentation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use pseudo-label-guided contrastive loss (PLGCL) (ablation) | Semi-supervised U-Net segmentation · ACDC, KiTS19, and CRAG (10% labeled) | DSC: 0.822 -> 0.891; 0.855 -> 0.919; 0.819 -> 0.882 | Helped |
| Add warm-up before PLGCL training (ablation) | Semi-supervised U-Net segmentation · ACDC, KiTS19, and CRAG (10% labeled) | DSC: 0.885 -> 0.891; 0.901 -> 0.919; 0.873 -> 0.882 | Helped |
| Add PLGCL and warm-up rather than using neither (ablation) | Semi-supervised U-Net segmentation · ACDC, KiTS19, and CRAG (10% labeled) | DSC: 0.799 -> 0.891; 0.831 -> 0.919; 0.813 -> 0.882 | Helped |
| Sample patches by entropy rather than cosine similarity (ablation) | PLGCL patch sampling · ACDC, 10% and 20% labeled | DSC: 0.820 -> 0.891; 0.832 -> 0.912 | Helped |
| Sample patches by entropy rather than class confidence (ablation) | PLGCL patch sampling · ACDC, 10% and 20% labeled | DSC: 0.873 -> 0.891; 0.877 -> 0.912 | Helped |
| Use the proposed method instead of LCLPL | Semi-supervised segmentation · ACDC, 10% labeled | DSC: 0.881 -> 0.891 | Helped |
| Use the proposed method instead of LCLPL | Semi-supervised segmentation · ACDC, 20% labeled | DSC: 0.905 -> 0.912 | Helped |
| Use the proposed method instead of GBDL | Semi-supervised segmentation · KiTS19, 2.50% labeled | DSC: 0.898 -> 0.905 | Helped |
| Use the proposed method instead of GBDL | Semi-supervised segmentation · KiTS19, 10% labeled | DSC: 0.911 -> 0.919 | Helped |
| Use the proposed method instead of Double-UA | Semi-supervised segmentation · CRAG, 10% labeled | DSC: 0.877 -> 0.882 | Helped |
| Use the proposed method instead of HCE | Semi-supervised segmentation · CRAG, 20% labeled | DSC: 0.885 -> 0.891 | Helped |
Closest papers by meaning
- DuSSS: Dual Semantic Similarity-Supervised Vision-Language Model for Semi-Supervised Medical Image Segmentation (AAAI 2025 · similarity 0.89)
- SGTC: Semantic-Guided Triplet Co-training for Sparsely Annotated Semi-Supervised Medical Image Segmentation (AAAI 2025 · similarity 0.89)
- Constructing and Exploring Intermediate Domains in Mixed Domain Semi-supervised Medical Image Segmentation (CVPR 2024 · similarity 0.88)
- DyCON: Dynamic Uncertainty-aware Consistency and Contrastive Learning for Semi-supervised Medical Image Segmentation (CVPR 2025 · similarity 0.88)
- Annotation Ambiguity Aware Semi-Supervised Medical Image Segmentation (CVPR 2025 · similarity 0.88)
- A Semantic Knowledge Complementarity based Decoupling Framework for Semi-supervised Class-imbalanced Medical Image Segmentation (CVPR 2025 · similarity 0.88)
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.