Dr James Zou is an Assistant Professor of Biomedical Data Science and Computer Science at Stanford University. He works on making machine learning more reliable, human-compatible and statistically rigorous. He received his Ph.D from Harvard in 2014, and was at one time a member of Microsoft Research, a Gates Scholar at Cambridge and a Simons fellow at U.C. Berkeley.

He joined Stanford in 2016 and his research is supported by the Sloan Fellowship and the Google and Tencent AI awards.

He recently published a very high quality paper looking at using EchoNet, an AI model to help interpret echocardiogram scans that was published in Nature medicine. It showed that AI saved clinician time and had as good scan interpretation as echocardiogram sonographers. One of the key factors that made this paper so effective was the fact that it was randomized , blinded and prospective. Factors that often are not present in medical papers but are crucial to make the evidence of what they show robust. We discussed his work on EchoNet and similar projects, what factors he considers are important in making effective medical AI papers and where he sees these models being deployed in the future.

You can find Dr James Zou at:
Twitter: @james_y_zou
Website: https: www.james-zou.com

You can find me at:
Twitter: @Dr_AlexDavidson
Website: www.alexdavidson.me

Support the Show.