Research map / Label-efficient and robust learning / Semi-supervised learning
A Dual-space Patch-driven Complementary Learning Framework for Semi-supervised Multi-organ Segmentation
Baixi Liang, Shuohong Xia, Sihao Li, Yunyun Yang
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 Local Enhancement module (ablation) | Sup-Only baseline · Synapse (10% labeled) | Average Dice (%): 37.75 to increases by 6% | Improved |
| add patch-based contrastive loss (PBC) (ablation) | DPCL with Local Enhancement · Synapse (10% labeled) | Average Dice (%): further improves by 3.39% | Improved |
| add global-based contrastive loss (GBC) (ablation) | DPCL with Local Enhancement · Synapse (10% labeled) | Average Dice (%): 5.73% improvement | Improved |
| add Dual Normalization (ablation) | DPCL · Synapse (10% labeled) | Average Dice (%): drops by 5.83% without it | Improved |
| use combined uncertainty and class-frequency sampling instead of uniform (ablation) | DPCL · Synapse (10% labeled) | Average Dice (%): 65.57 -> 68.89 | Improved |
| use combined sampling instead of uncertainty only (ablation) | DPCL · Synapse (10% labeled) | Average Dice (%): 67.54 -> 68.89 | Improved |
| use combined sampling instead of class frequency only (ablation) | DPCL · Synapse (10% labeled) | Average Dice (%): 66.10 -> 68.89 | Improved |
| increase cropped patches n from 4 to 8 (ablation) | DPCL · Synapse (10% labeled) | Average Dice (%): 68.89 -> 67.34 | Hurt |
| reduce cropped patches n from 4 to 2 (ablation) | DPCL · Synapse (10% labeled) | Average Dice (%): 68.89 -> 67.91 | Hurt |
| increase crop size s from 64 to 128 (ablation) | DPCL · Synapse (10% labeled) | Average Dice (%): 68.89 -> 67.27 | Hurt |
| decrease crop size s from 64 to 16 (ablation) | DPCL · Synapse (10% labeled) | Average Dice (%): 68.89 -> 66.23 | Hurt |
| add PBC and GBC contrastive losses (ablation) | DPCL feature space · Synapse (10% labeled) | Silhouette Score: 3.44% improvement | Improved |
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.
- FixMatch — FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence (NeurIPS 2020) · better on 3 of 3 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
- A Semantic Knowledge Complementarity based Decoupling Framework for Semi-supervised Class-imbalanced Medical Image Segmentation (CVPR 2025 · similarity 0.90)
- Neighbor Does Matter: Density-Aware Contrastive Learning for Medical Semi-supervised Segmentation (AAAI 2025 · similarity 0.88)
- DyCON: Dynamic Uncertainty-aware Consistency and Contrastive Learning for Semi-supervised Medical Image Segmentation (CVPR 2025 · similarity 0.88)
- SGTC: Semantic-Guided Triplet Co-training for Sparsely Annotated Semi-Supervised Medical Image Segmentation (AAAI 2025 · similarity 0.87)
- Pancakes: Consistent Multi-Protocol Image Segmentation Across Biomedical Domains (NeurIPS 2025 · similarity 0.87)
- Unleashing the Potential of SAM for Medical Adaptation via Hierarchical Decoding (CVPR 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.
