ProtoPathway: Biologically Structured Prototype-Pathway Fusion for Multimodal Cancer Survival Prediction

Researchers introduced ProtoPathway, an interpretable multimodal model for cancer survival prediction.

It combines whole-slide pathology images with transcriptomics using learned morphological prototypes and a pathway-aware graph neural network.

Cross-modal attention links genes, pathways, and tissue regions to improve both prediction and interpretability.

Across five TCGA cohorts, ProtoPathway achieved competitive or better survival performance with lower compute and stronger biological grounding.