An Uncertainty-Guided Evidential Deep Learning Framework for Reliability-Aware Multimodal Fusion in Cancer Prognosis
TL;DR
REM-Fuse is introduced, an evidential deep learning framework in which per-source Dirichlet uncertainty adaptively weights each source through dual-channel weighting, asymmetric cross-scale enhancement, and Dempster–Shafer-inspired evidence accumulation that provides a compact reliability-aware fusion strategy for cancer prognosis.
Abstract
Background: Reliability-aware integration of heterogeneous data sources remains a fundamental challenge in multimodal deep learning: prevailing fusion strategies assume uniform reliability across sources and instances, limiting their responsiveness to data-dependent trustworthiness. Methods: We introduce REM-Fuse (Reliability-aware Evidential Multimodal Fusion), an evidential deep learning (EDL) framework in which per-source Dirichlet uncertainty adaptively weights each source through dual-channel weighting, asymmetric cross-scale enhancement, and Dempster–Shafer-inspired evidence accumulation. As a case study for cancer prognosis, REM-Fuse integrates multi-scale histopathology (10×, 20×) and RNA-seq on TCGA-BRCA (n = 831) via five-fold cross-validation with subtype- and stage-stratified analyses. Results: REM-Fuse attained a concordance index of 0.715 and a 60-month time-dependent AUC of 0.729, indicating moderate discrimination and significant risk separation (log-rank p < 0.001). Adaptive source weights and per-patient uncertainty varied significantly across molecular subtypes (Kruskal–Wallis p = 0.010 and p = 0.007), indicating patient-specific rather than fixed multimodal integration. Conclusions: REM-Fuse provides a compact reliability-aware fusion strategy for cancer prognosis, although external validation is needed before broader clinical or cross-cohort generalization.