Research map / Label-efficient and robust learning / Domain adaptation and generalization
Minimizing Labeling Cost for Nuclei Instance Segmentation and Classification with Cross-domain Images and Weak Labels
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 |
|---|---|---|---|
| Add cyclic pseudo-label adaptation (ablation) | Hover-Net UDA · PanNuke test (CoNSep source) | Dice: 0.729 -> 0.740 | Helped |
| Add point-level weak supervision (ablation) | Hover-Net UDA · PanNuke test (CoNSep source) | AJI: 0.516 -> 0.523 | Helped |
| Add image-level weak supervision (ablation) | Hover-Net UDA · PanNuke test (CoNSep source) | Neo and Epithelial F1: 0.380 -> 0.392 | Helped |
| Add hard-transfer sample alignment using point labels (ablation) | Hover-Net WDA: point · PanNuke test (CoNSep source) | Detection F1: 0.710 -> 0.735 | Helped |
| Add hard-transfer sample alignment using point labels (ablation) | Hover-Net WDA: point · PanNuke test (CoNSep source) | AJI: 0.523 -> 0.534 | Helped |
| Add image-level supervision instead of point-level supervision (ablation) | Hover-Net WDA · PanNuke test (CoNSep source) | Detection F1: 0.700 -> 0.702 | No clear effect |
| Use image-level rather than point-level supervision (ablation) | Hover-Net WDA · PanNuke test (CoNSep source) | Dice: 0.738 vs 0.743 | Mixed |
| Replace point-level supervision with image-level supervision | Hover-Net WDA · PanNuke test (CoNSep source) | Detection F1: 0.735 -> 0.702 | Hurt |
| Add point-level supervision and hard-transfer alignment | Hover-Net UDA · PanNuke test (CoNSep source) | Neo and Epithelial F1: 0.381 -> 0.462 | Helped |
| Add point-level supervision and hard-transfer alignment | Hover-Net UDA · PanNuke test (CoNSep source) | SQ: 0.753 -> 0.741 | Hurt |
Closest papers by meaning
- Are nuclear masks all you need for improved out-of-domain generalisation? A closer look at cancer classification in histopathology (NeurIPS 2024 · similarity 0.88)
- UniCell: Universal Cell Nucleus Classification via Prompt Learning (AAAI 2024 · similarity 0.87)
- Incremental Nuclei Segmentation from Histopathological Images via Future-class Awareness and Compatibility-inspired Distillation (CVPR 2024 · similarity 0.86)
- Prototype-Based Image Prompting for Weakly Supervised Histopathological Image Segmentation (CVPR 2025 · similarity 0.86)
- Enhancing Instance-Level Image Classification with Set-Level Labels (ICLR 2024 · similarity 0.85)
- Mudslide: A Universal Nuclear Instance Segmentation Method (CVPR 2024 · 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.