Brim

Overview

This study introduces an interpretable bridged multimodal fusion model (Brim) designed to predict prognostic outcomes in pan-cancer patients. Addressing the challenge of incomplete multimodal data in real-world clinical settings, Brim integrates histopathology, genomics, and transcriptomics. The model facilitates more precise prognosis predictions, even in the absence of complete molecular features. Validated across 12 cancer types, Brim demonstrates optimal performance in both complete and missing modality scenarios. This work presents a clinically applicable medical multimodal fusion model, offering the potential to reduce the healthcare burden on cancer patients and enhance clinical decision-making for practitioners.