Fusing imaging and metabolic modeling via multimodal deep learning in ovarian cancer (opens in new tab)

By integrating patient-specific metabolic models, CT imaging, and transcriptomics, this study introduces a previously unexplored multimodal deep-learning framework for ovarian cancer survival prediction. The proposed approach improves the accuracy of survival prediction while crucially providing mechanistic insights. The interpretation of the model also uncovers multi-omic biomarkers for high-risk patients.