Research map / Vision-language and multimodal / Cross-modal learning
Structural Prognostic Event Modeling for Multimodal Cancer Survival Analysis
Yilan Zhang, Li Nanbo, Changchun Yang, Jürgen Schmidhuber, Xin Gao
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 |
|---|---|---|---|
| full SlotSPE vs vanilla slot baseline (ablation) | SlotSPE · TCGA 10 cohorts overall | C-index: 0.687 -> 0.721 | Improved |
| add selective slot activation (ablation) | SlotSPE · TCGA overall | C-index: 0.699 -> 0.721 | Improved |
| add slots regularization reconstruction (ablation) | SlotSPE · TCGA overall | C-index: 0.704 -> 0.721 | Improved |
| add cross-modal reconstruction with biological prior (ablation) | SlotSPE · TCGA overall | C-index: 0.696 -> 0.721 | Improved |
| add iterative cross-attention (ablation) | SlotSPE · TCGA overall | C-index: 0.699 -> 0.721 | Improved |
| slot-based multimodal model vs prototype methods PIBD/MMP | SlotSPE · TCGA overall | C-index: 0.676 / 0.669 -> 0.721 | Improved |
| remove genomics at test time (use cross-modal reconstruction imputation) | SlotSPE · TCGA overall | C-index: 0.721 -> 0.704 | Mixed |
| multimodal vs unimodal models | multimodal baselines · TCGA overall | C-index: e.g. Porpoise 0.676, PIBD 0.676, MMP 0.669 vs ABMIL 0.681 | Mixed |
| reduce slots to 8 per modality (ablation) | SlotSPE · TCGA overall | C-index: 0.721 -> 0.706 | Hurt |
| fewer slot-attention iterations (T=3 vs 10) (ablation) | SlotSPE · TCGA overall | C-index: 0.721 -> 0.704 | Hurt |
Closest papers by meaning
- Modeling Dense Multimodal Interactions Between Biological Pathways and Histology for Survival Prediction (CVPR 2024 · similarity 0.86)
- Multimodal Prototyping for cancer survival prediction (ICML 2024 · similarity 0.86)
- CA-MLIF: Cross-Attention and Multimodal Low-Rank Interaction Fusion Framework for Tumor Prognostic Prediction (AAAI 2025 · similarity 0.86)
- Tumor Micro-environment Interactions Guided Graph Learning for Survival Analysis of Human Cancers from Whole-slide Pathological Images (CVPR 2024 · similarity 0.84)
- Leveraging Tumor Heterogeneity: Heterogeneous Graph Representation Learning for Cancer Survival Prediction in Whole Slide Images (NeurIPS 2024 · similarity 0.84)
- Prototypical Information Bottlenecking and Disentangling for Multimodal Cancer Survival Prediction (ICLR 2024 · 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.
