The TemPredict initiative was spearheaded at UCSF in 2020, to bring together experts in a variety of disciplines, including machine learning and epidemiology, to forecast COVID-19 cases, and track the progression and spread of the virus. The initiative received seed money from the health tech company, Ōura, who also provided the wearable technology (i.e., the Ōura ring), to collect the personalized health data (e.g., body temperature, heart rate, etc.) of TemPredict study participants. The initiative has continued to receive funding from sources such as the Department of Defense (DoD) and the Medical Technology Enterprise Consortium. 

We contributed our expertise in data management, modeling, integration, and ingestion, to build out a large-scale data pipeline and end-to-end cyberinfrastructure that supports the real-time collection and management of physiological data. Our goal is to develop a sustainable framework that enables personalized solutions and insights to be provided to users and their physicians, such that COVID-19 cases can be predicted and addressed as early as possible. One of the key components of this framework is an easily navigable UI, such that domain experts can access advanced computational and data resources via the Kubernetes-based Nautilus portal at UCSD implemented on the backend.

The framework we have developed can be extended and scaled to different areas of application, including menstruation tracking (to provide insights into conditions such as menorrhagia).