The Journey from AI to Bedside: Translating Academic Research to Serving Patients Using Ray

In the current Healthcare Data Revolution, physicians are about to get a bigger and more powerful stethoscope. Natural language, telemetry data, labs, effectiveness of drug treatment in clinical practice encapsulates the precise patterns of the patients presentation, trajectories of the disease and also the effectiveness of intervention by the very physicians at the bedside. Risk stratification is the mainstay for how diagnostic testing, treatment and triage is directed. We employ data from the patients directly from the community and create models for continuous daily risk stratification of a new disease to better direct physician intervention and explore a large scale use case in India. RLlib api provides a powerful framework to incorporate these new models for both tailored management of patients with high risk disease and expanding the functionalities of healthcare software.