Research map / Detection and segmentation / Weakly supervised detection and segmentation
Multiple Instance Learning Framework with Masked Hard Instance Mining for Gigapixel Histopathology Image Analysis
Wenhao Tang, Sheng Huang, Heng Fang, Fengtao Zhou, Bo Liu, Qingshan Liu
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 Randomly High Score Masking (RHSM) (ablation) | TransMIL · CAMELYON AUC, ResNet-50, TransMIL | AUC: 91.2 -> 92.4 | Improved |
| add RHSM to PLIP-feature TransMIL (ablation) | TransMIL · CAMELYON, PLIP | AUC: 94.6 -> 93.9 | Hurt |
| add Global Recycle Network (GRN) and consistency loss on top of RHSM (ablation) | TransMIL · CAMELYON AUC, ResNet-50, TransMIL | AUC: 92.4 -> 93.5 | Improved |
| add Consistency Loss on top of RHSM+GRN (ablation) | TransMIL · CAMELYON, ResNet-50 | AUC: 92.4 -> 93.5 | Improved |
| use class-aware instance probability instead of attention (ablation) | TransMIL · CAMELYON AUC, ResNet-50, TransMIL | AUC: 92.4 -> 93.5 | Improved |
| use class-aware instance probability instead of attention (ablation) | AB-MIL · TCGA-NSCLC ResNet-50 | AUC: 96.1 -> 96.2 | Improved |
| use initialized plus momentum teacher instead of student copy (ablation) | AB-MIL · TCGA-NSCLC ResNet-50 | AUC: 94.1 -> 96.2 | Improved |
| use momentum teacher instead of student-copy teacher (ablation) | TransMIL · CAMELYON ResNet-50 | AUC: 91.7 -> 92.5 | Improved |
| add pretrained initialization plus momentum to teacher (ablation) | TransMIL · CAMELYON AUC, ResNet-50, TransMIL | AUC: 92.5 -> 93.5 | Improved |
| add random score masking (RSM) to RHSM+CL (ablation) | TransMIL · CAMELYON AUC, ResNet-50, TransMIL | AUC: 92.6 -> 93.0 | Improved |
| add GRN to RSM (ablation) | TransMIL · CAMELYON AUC, ResNet-50, TransMIL | AUC: 93.0 -> 93.5 | Improved |
| use RSM instead of LSM (ablation) | TransMIL · CAMELYON AUC, ResNet-50, TransMIL | AUC: 92.5 -> 93.5 | Improved |
Closest papers by meaning
- SI-MIL: Taming Deep MIL for Self-Interpretability in Gigapixel Histopathology (CVPR 2024 · similarity 0.90)
- Fast and Accurate Gigapixel Pathological Image Classification with Hierarchical Distillation Multi-Instance Learning (CVPR 2025 · similarity 0.89)
- xMIL: Insightful Explanations for Multiple Instance Learning in Histopathology (NeurIPS 2024 · similarity 0.88)
- Dynamic Policy-Driven Adaptive Multi-Instance Learning for Whole Slide Image Classification (CVPR 2024 · similarity 0.88)
- Promptable Representation Distribution Learning and Data Augmentation for Gigapixel Histopathology WSI Analysis (AAAI 2025 · similarity 0.88)
- CaMIL: Causal Multiple Instance Learning for Whole Slide Image Classification (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.
