HySurvPred: Multimodal Hyperbolic Embedding with Angle-Aware Hierarchical Contrastive Learning and Uncertainty Constraints for Survival Prediction

2025-03-18

AI Summary

Current cancer survival prediction methods suffer from three key limitations: (1) modeling multimodal hierarchical structures—e.g., patch-level relationships in histopathology images or gene-to-pathway hierarchies—in Euclidean space fails to capture their intrinsic non-linear geometric properties; (2) discretizing continuous survival times disregards their ordinal nature and differentiability; and (3) discarding censored samples neglects their inherent uncertainty. To address these, we propose the first hyperbolic-space multimodal survival prediction framework. It learns modality-specific hierarchical embeddings in hyperbolic space, introduces an angle-aware ranking contrastive loss to jointly optimize continuous-time ordinal relationships, and incorporates a censoring-conditioned uncertainty constraint to enable full-sample training. Evaluated on five benchmark datasets, our method consistently outperforms state-of-the-art approaches. Code is publicly available.

Technology Category

• Machine Learning: Multimodal Learning
• Reasoning under Uncertainty: Relational Probabilistic Models
• Computer Vision: Multi-modal Vision

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Multimodal learning that integrates histopathology images and genomic data holds great promise for cancer survival prediction. However, existing methods face key limitations: 1) They rely on multimodal mapping and metrics in Euclidean space, which cannot fully capture the hierarchical structures in histopathology (among patches from different resolutions) and genomics data (from genes to pathways). 2) They discretize survival time into independent risk intervals, which ignores its continuous and ordinal nature and fails to achieve effective optimization. 3) They treat censorship as a binary indicator, excluding censored samples from model optimization and not making full use of them. To address these challenges, we propose HySurvPred, a novel framework for survival prediction that integrates three key modules: Multimodal Hyperbolic Mapping (MHM), Angle-aware Ranking-based Contrastive Loss (ARCL) and Censor-Conditioned Uncertainty Constraint (CUC). Instead of relying on Euclidean space, we design the MHM module to explore the inherent hierarchical structures within each modality in hyperbolic space. To better integrate multimodal features in hyperbolic space, we introduce the ARCL module, which uses ranking-based contrastive learning to preserve the ordinal nature of survival time, along with the CUC module to fully explore the censored data. Extensive experiments demonstrate that our method outperforms state-of-the-art methods on five benchmark datasets. The source code is to be released.

Problem

• Captures hierarchical structures in histopathology and genomics data
• Preserves continuous and ordinal nature of survival time
• Effectively utilizes censored data in model optimization

Innovation

• Uses hyperbolic space for hierarchical data mapping
• Implements angle-aware ranking-based contrastive learning
• Incorporates censored data with uncertainty constraints