Our goal is to develop algorithms that will enable early diagnosis, disease severity assessment, and personalization and adaptability of the therapy. To this end, we combine neural and physiological signal processing, machine learning and neurophysiology, and healthcare data analytics. Our two main thrusts are 1) to understand how the nervous system senses the state and affects the function of the immune, metabolic and cardiopulmonary systems, in order to develop neuromodulation devices that are able to diagnose and treat various diseases and conditions by interacting with the nervous system and 2) to combine multiple healthcare data modalities (EHR, continuous vitals, imaging, unstructured notes) with machine learning methods to develop clinical predictive and diagnostic models.