Srinivasan Receives Funding from Google to Advance Contact Tracing Methods

connected bubbles of people with masks
Fri Sep 18, 2020

A University of Maryland expert in algorithms and high-performance computing has been funded by Google to develop computational techniques to improve contact tracing methods in the wake of COVID-19, research that could ultimately provide fundamental new insights into how epidemics can be controlled.

Aravind Srinivasan, a Distinguished University Professor of computer science with an appointment in the University of Maryland Institute for Advanced Computer Studies, is one of the four principal investigators of the $140K project. He is joined by researchers from Princeton University and the University of Virginia who, like Srinivasan, have more than 25 years of experience in computational epidemiology.

The money comes from Google’s $8.5 million investment in data analytics and artificial intelligence to better understand the impact of COVID-19 on communities–especially vulnerable populations and healthcare workers.

The researchers’ primary objective is to develop algorithms that prioritize who to test and when. These new algorithmic techniques will balance testing rates, infection rates and social distancing data.

Then, the team will use simulations and machine learning techniques to test and refine the algorithms, improving their efficacy. This approach is unique, say the researchers, because the simulations will prove the accuracy of their theories.

One of Srinivasan’s main goals is to develop an algorithm that will be able to identify a “superspreader” faster than current contact tracing methods allow. A superspreader is an infected individual who is coming into contact with a large number of people, explains Srinivasan, like a bus driver or nurse, for example.

Amplification testing is another project component, in which the researchers will study existing data of social interactions to increase the combined power of testing and social distancing.

Privacy is significant challenge for this project, say the researchers, due to security risks and the sensitivity of the medical data. Their goal is to guarantee differential privacy so that in the event of a data leak, individuals’ information would be indistinguishable.

Srinivasan’s team hopes to not only contribute to the curbing of COVID-19, but to also provide a blueprint for the future.

“Unfortunately, this is not going to be the last pandemic,” he says. “Our goal is to develop theory to future-proof our systems and be prepared for the next one.”

–Story by Maria Herd