Research map / Label-efficient and robust learning / Recognition and long-tail
DPsurv: Dual-Prototype Evidential Fusion for Uncertainty-Aware and Interpretable Whole Slide Image Survival Prediction
Yucheng Xing, Ling Huang, Jingying Ma, Ruping Hong, Jiangdong Qiu, Pei Liu, Kai He, Huazhu Fu, Mengling Feng
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 GMM slide embedding to foundation-model features (ablation) | DPsurv · TCGA mean | C-index: 0.621 -> 0.658 | Improved |
| add GMM slide embedding (ablation) | DPsurv · TCGA mean | IBLL: 3.223 -> 1.574 | Improved |
| add GRFN evidential framework (ablation) | DPsurv · TCGA mean | IBS: 0.585 -> 0.331 | Improved |
| add GRFN evidential framework (ablation) | DPsurv · TCGA mean | IBLL: 1.574 -> 0.894 | Improved |
| add GRFN evidential framework (ablation) | DPsurv · TCGA mean | C-index: 0.658 -> 0.663 | No clear effect |
| add component prototypes without evidence mixture (ablation) | DPsurv · TCGA mean | C-index: 0.663 -> 0.538 | Hurt |
| add evidence mixture without component prototypes (ablation) | DPsurv · TCGA mean | IBS: 0.331 -> 0.220 | Improved |
| add evidence mixture to CP (ablation) | DPsurv · TCGA mean | C-index: 0.538 -> 0.687 | Improved |
| add CP to GRFN+EM (full model) (ablation) | DPsurv · TCGA mean | C-index: 0.645 -> 0.687 | Improved |
| add CP to GRFN+EM (full model) (ablation) | DPsurv · TCGA mean | IBLL: 0.720 -> 0.604 | Improved |
| use CONCH instead of UNI2-h features (ablation) | DPsurv · UCEC | C-index: 0.766 -> 0.780 | Improved |
| vary number of component prototypes K 2 to 6 (ablation) | DPsurv · KIRC | C-index: 0.712--0.732 | 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.
- TransMIL — TransMIL: Transformer based Correlated Multiple Instance Learning for Whole Slide Image Classification (NeurIPS 2021) · better on 1 of 1 shared result
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
- Leveraging Tumor Heterogeneity: Heterogeneous Graph Representation Learning for Cancer Survival Prediction in Whole Slide Images (NeurIPS 2024 · similarity 0.88)
- Interpretable Vision-Language Survival Analysis with Ordinal Inductive Bias for Computational Pathology (ICLR 2025 · similarity 0.87)
- Multimodal Prototyping for cancer survival prediction (ICML 2024 · similarity 0.87)
- Prototypical Information Bottlenecking and Disentangling for Multimodal Cancer Survival Prediction (ICLR 2024 · similarity 0.85)
- Modeling Dense Multimodal Interactions Between Biological Pathways and Histology for Survival Prediction (CVPR 2024 · similarity 0.85)
- From Representation Space to Prognostic Insights: Whole Slide Image Generation with Hierarchical Diffusion Model for Survival Prediction (AAAI 2025 · similarity 0.84)
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.
