Multimodal Cancer Survival Analysis via Hypergraph Learning with Cross-Modality Rebalance
This work addresses cancer survival prediction for medical applications, offering a novel approach to handle multimodal data imbalance, though it is incremental in improving existing methods.
The paper tackled the problem of multimodal cancer survival prediction by addressing information loss in pathology images and modality imbalance between pathology and genomics, resulting in a model that outperforms advanced methods by over 3.4% in C-Index performance on five TCGA datasets.
Multimodal pathology-genomic analysis has become increasingly prominent in cancer survival prediction. However, existing studies mainly utilize multi-instance learning to aggregate patch-level features, neglecting the information loss of contextual and hierarchical details within pathology images. Furthermore, the disparity in data granularity and dimensionality between pathology and genomics leads to a significant modality imbalance. The high spatial resolution inherent in pathology data renders it a dominant role while overshadowing genomics in multimodal integration. In this paper, we propose a multimodal survival prediction framework that incorporates hypergraph learning to effectively capture both contextual and hierarchical details from pathology images. Moreover, it employs a modality rebalance mechanism and an interactive alignment fusion strategy to dynamically reweight the contributions of the two modalities, thereby mitigating the pathology-genomics imbalance. Quantitative and qualitative experiments are conducted on five TCGA datasets, demonstrating that our model outperforms advanced methods by over 3.4\% in C-Index performance.