Dr. Purkayastha has received $826,795 grant from the National Science Foundation, entitled “FW-HTF-RM: Measuring learning gains in man-machine assemblage when augmenting radiology work with artificial intelligence.”
Purkayastha is the principal investigator with co-PIs Elizabeth Krupinski of Emory University, Joshua Danish of the IU School of Education, and Judy Gichoya, M.D., formerly of Indiana University. Gichoya received her M.S. in health informatics from the School of Informatics and Computing and M.D. in radiology from IU School of Medicine before receiving a fellowship at the Dotter Department of Interventional Radiology at Oregon Health and Science University.
The three-year grant will look at artificial intelligence (AI) in a unique and novel way: as part of a man-machine assemblage. In the past, these would have been trained as separate entities—much like a self-driving car and driver—rather than learning and training together at the same time.
The project will develop radiologist-AI assemblage, such that they will work on routine radiology work tasks together in partnership for decision support. These assemblages will adapt to diverse human capabilities ranging from first-year residents to expert radiologist staff, to create “assemblage learning,” where machine learning and human learning happened together, and generated knowledge shared between a swarm of such assemblages.