Research map / Detection and segmentation / Weakly supervised detection and segmentation
Collaborative Learning of Semi-Supervised Segmentation and Classification for Medical Images
Check 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 |
|---|---|---|---|
| add semi-supervised lesion attention and adversarial training (ablation) | Lesion grading model · IDRID | accuracy: 0.9016 -> 0.9134 | Helped |
| add lesion-mask information from the pretrained segmenter (ablation) | DR grading model · IDRID | accuracy: 0.8458 -> 0.8725 | Helped |
| add semi-supervised lesion attention (ablation) | DR grading model · EyePACS | Kappa: 0.8445 -> 0.8617 | Helped |
| add adversarial training (ablation) | semi-supervised lesion segmentation · IDRID haemorrhages | AUC PR: 0.6812 -> 0.6936 | Helped |
| use Xception modules instead of normal convolution (ablation) | multi-lesion segmentation · IDRID hard exudates | AUC PR: 0.8263 -> 0.8455 | Helped |
| add semi-supervised learning using lesion attention pseudo-masks (ablation) | Xception-based segmentation model · IDRID soft exudates | AUC PR: 0.7161 -> 0.7337 | Helped |
| add adversarial training to semi-supervised segmentation (ablation) | Xception-based segmentation model · IDRID soft exudates | AUC PR: 0.7337 -> 0.7407 | Helped |
| use the collaborative lesion-attention method instead of AFN | DR grading · EyePACS | Kappa: 0.859 -> 0.872 | Helped |
| use the collaborative lesion-attention method instead of the strongest listed challenge result | microaneurysm segmentation · IDRID | AUC PR: 0.5017 -> 0.4960 | Hurt |
| use the collaborative lesion-attention method instead of the strongest listed challenge result | hard exudate segmentation · IDRID | AUC PR: 0.885 -> 0.8872 | Helped |
Closest papers by meaning
- HACDR-Net: Heterogeneous-Aware Convolutional Network for Diabetic Retinopathy Multi-Lesion Segmentation (AAAI 2024 · similarity 0.88)
- Deep Multi-Task Learning for Diabetic Retinopathy Grading in Fundus Images (AAAI 2021 · similarity 0.88)
- Combinatorial CNN-Transformer Learning with Manifold Constraints for Semi-supervised Medical Image Segmentation (AAAI 2024 · similarity 0.86)
- Unleashing the Potential of Vision-Language Pre-Training for 3D Zero-Shot Lesion Segmentation via Mask-Attribute Alignment (ICLR 2025 · similarity 0.85)
- Prototype-Based Image Prompting for Weakly Supervised Histopathological Image Segmentation (CVPR 2025 · similarity 0.85)
- Enhancing SAM with Efficient Prompting and Preference Optimization for Semi-supervised Medical Image Segmentation (CVPR 2025 · similarity 0.85)
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.