Home  /  Areas to Support  /  COVID-19 Research  /  Social Distancing 2.0: Tools for Infection Risk Assessment Based on Combining Location Data with Wearable Sensor Data

# Social Distancing 2.0: Tools for Infection Risk Assessment Based on Combining Location Data with Wearable Sensor Data

Chiara Daraio
Professor of Mechanical Engineering and Applied Physics

Azita Emami
Andrew and Peggy Cherng Professor of Electrical Engineering and Medical Engineering; Investigator, Heritage Medical Research Institute; Executive Officer for Electrical Engineering

Tapio Schneider
Theodore Y. Wu Professor of Environmental Science and Engineering; Jet Propulsion Laboratory Senior Research Scientist

Caltech researchers have imagined a way to achieve targeted social distancing with voluntary participation from only a small fraction of the population that improves automatically with wider participation and added information. They propose a scalable platform that provides users with individualized and predictive measures of infection risk, based on location data from cell phones and monitoring of core body temperature with inexpensive, wearable devices. Social distancing measures, where applied consistently, are successfully inhibiting the spread of COVID-19. However, the blanket restrictions come at an immense economic cost. This platform is aimed at developing optimal social distancing in the least restrictive way possible. It builds on the multidisciplinary expertise of Chiara Dario, who develops new materials for highly accurate temperature sensing; Azita Emami, who designs compact integrated circuits for wearable and implantable medical devices; and Tapio Schneider, who develops risk assessment algorithms and software. The core idea is to use results from network theory to calculate the likelihood of a virus to reach a node (individual) on a graph whose edges (links) are established by users that are in close contact for a brief period. Available location data from cell phones and other mobile devices can be used to establish links between individuals, and statistical algorithms can be used to calculate the likelihood of infection at each node. A first estimate can be calculated from location data alone, paired with knowledge of geographic variations of infection risk. The added information, which augments the platform's capability, can come from self-reporting of symptoms and from inexpensive, wearable sensors that measure and automatically report core body temperatures. Within 6 months, the researchers propose to prototype this platform, which combines search and automated learning algorithms with accurate wearable temperature sensors. The sensors are millimeter-thin, Band-Aid-like patches that will wirelessly connect to cellphones. The platform will provide individual users with risk assessment tools regarding self-isolation or visiting certain locations. It would learn automatically as the user base grows over time and more data about infections is provided, leading to steadily improving risk assessments.

Philanthropic goal: $300,000 Suggested minimum gift:$1,000

Give Now