Perspective on multi-modal modeling for biomarker discovery in oncology

Multi-modal data fusion of routinely collected biomedical data including raw data generation, data preprocessing & labeling, feature extraction and embedding representation. 

Development of effective multimodal fusion approaches is becoming increasingly important as a single modality might not be consistent and sufficient to capture the heterogeneity of complex diseases to tailor medical care and improve personalized medicine. 

In this perspective, published in Nature Machine Intelligence, the Gevaert lab covers current challenges and reflects on opportunities through deep learning to tackle data sparsity and scarcity, multimodal interpretability and standardization of datasets.