Ju Sun Co-Led Project Using AI Technology for COVID-19 Diagnosis

Department of Computer Science & Engineering assistant professor Ju Sun co-led a project that examines artificial intelligence (AI)’s effectiveness in diagnosing COVID-19. Working with Christopher Tignanelli, an associate professor in the University of Minnesota Medical School, their work has been recently published in the Journal of the American Medical Informatics Association.

Sun and Tignanelli introduce federated learning, a machine learning technique, to the diagnosis of COVID-19 in chest x-rays. Federated learning allows the use of a particular algorithm on various decentralized datasets while protecting the integrity of sensitive medical information. This process both improves the accuracy in diagnosis and eliminates some of the biases that are unavoidable in smaller, more specific data sets.

“We’re proud to be among the first teams implementing and further refining federated learning in real-world healthcare settings,” Sun said. “Data is the oil for modern AI, and federated learning makes the perfect oil refinery to advance AI for healthcare.” 

Federated learning has the potential to improve a number of different areas of healthcare, including image and text analysis, diagnostic tools, and accelerated drug discovery.

This project was a collaboration between the U of M, M Health Fairview, Emory University, Indiana University School of Medicine and University of Florida. There was also supportive funding provided by Cisco.

Read the full story on the U of M Medical School website with a research brief from the College of Science and Engineering.