Transforming Precision Medicine: Dynamic Learning and Prediction of Disease Progression in Massive, Diverse, and Multimodal Cohorts

Enter the text here that is the new abstract information for your application. This section must be no longer than 30 lines of text. Modern healthcare data, e.g., electronic medical records (EHRs) and biobanks, provide massive multi-level and multi-scale information over a long period. These datasets offer a unique opportunity to quantify disease progression and the associated time-varying risk factors. However, existing statistical algorithms and tools that can effectively analyze exposure trajectories and disease onset at this scale and complexity lag behind. This proposal aims to fill these gaps by delivering novel statistical methods, computational algorithms, and user-friendly software to study individual health trajectories relevant to disease onset and progression for biobank scale real-world data. It is motivated by our team’s recent experiences analyzing the UK Biobank, VA EHRs, NIH All of US, and other complex data sources. The proposed work will push forward several frontiers in biostatistics, optimization, and clinical medicine (i.e., personalized diabetes management, dynamic prediction, and early detection of disease progression).