Research map / Label-efficient and robust learning / Recognition and long-tail
TopoSlide: Topologically-Informed Histopathology Whole Slide Image Representation Learning
Shahira Abousamra, Asmita Sood, Sylvia Plevritis
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 |
|---|---|---|---|
| train slide encoder with topology-aware objective instead of TITAN-V objective on same cohort | TopoSlide vs TITAN-V (TCGA-LUAD) · DHMC-LUAD retrieval K=1 | majority-vote macro F1: 41.570 -> 51.388 | Improved |
| use TopoSlide instead of GigaPath trained on the same cohort | TopoSlide vs GigaPath (TCGA-LUAD) · DHMC-LUAD retrieval K=1 | majority-vote macro F1: 25.420 -> 51.388 | Improved |
| small-data topology-aware training instead of large-scale foundation models | TopoSlide · DHMC-LUAD retrieval K=1 | majority-vote macro F1: 44.740 (TITAN) -> 51.388 | Improved |
| use TopoSlide for cluster-composition retrieval | TopoSlide · DHMC-LUAD K=5 | RMSE of cluster proportions: 8.82 (CONCH) -> 7.91 | Improved |
| train TopoSlide on CPTAC-LUAD and test cross-cohort retrieval at K=3 | TopoSlide · DHMC-LUAD retrieval K=3 | majority-vote macro F1: TITAN 41.401 vs TopoSlide (CPTAC) 34.642 | No clear effect |
| use TopoSlide for survival prediction | TopoSlide · TCGA-LUAD DSS | C-index: 0.646 (GigaPath) vs 0.643 | No clear effect |
| use TopoSlide trained on CPTAC-LUAD for survival on CPTAC-LUAD | TopoSlide · CPTAC-LUAD PFI | C-index: 0.636 (TITAN) vs 0.524 | Hurt |
| use TopoSlide for BRCA histologic subtype retrieval | TopoSlide · TCGA-BRCA retrieval | balanced accuracy | Mixed |
| use TopoSlide for BRCA survival vs mean pooling | TopoSlide · TCGA-BRCA DSS | C-index: 0.711 (Virchow mean pool) vs 0.690 | Hurt |
| use TopoSlide for gene mutation prediction | TopoSlide · TCGA-LUAD mean AUROC | AUROC: 0.632 (CONCH) vs 0.615 / 0.621 | No clear effect |
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.
- TITAN — TITAN: Query-Token based Domain Adaptive Adversarial Learning (ICCV 2025) · better on 3 of 4 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
- Dynamic Graph Representation with Knowledge-aware Attention for Histopathology Whole Slide Image Analysis (CVPR 2024 · similarity 0.90)
- MATCH: Multi-faceted Adaptive Topo-Consistency for Semi-Supervised Histopathology Segmentation (NeurIPS 2025 · similarity 0.89)
- Rotation-Agnostic Image Representation Learning for Digital Pathology (CVPR 2024 · similarity 0.88)
- TopoCellGen: Generating Histopathology Cell Topology with a Diffusion Model (CVPR 2025 · similarity 0.88)
- Sequential Attention-based Sampling for Histopathological Analysis (NeurIPS 2025 · similarity 0.88)
- Unsupervised Foundation Model-Agnostic Slide-Level Representation Learning (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.
