Paul Cohen, PhD (SignatureDx)

Abstract

Noninvasive prenatal testing is becoming the norm, and a huge business, but it is limited.  We are developing noninvasive screens for a variety of maternal and fetal pathologies.  Given one or more samples of blood plasma from pregnant women, three broad  statistical problems arise:  Classify the plasma, determine what’s in the plasma, and characterize changes in subsequent samples. 

Bio

Paul Cohen is a Professor of Computer Science in the School of Computing and Information at Pitt.  He is also CTO of SignatureDx, which develops noninvasive screens for fetal phenotyping and silent diseases. Paul was the Founding Dean of the School in 2017, having returned to academia after a four-year stint at the Defense Advanced Research Projects Agency (DARPA).  Paul developed and managed programs in AI-assisted modeling at DARPA, specifically modeling cancer pathways and food insecurity, as well as a program on communicating with computers.  Paul has held faculty and leadership positions at the University of Massachusetts, the University of Southern California’s Information Sciences Institute, and the University of Arizona. He is a Fellow of the Association for the Advancement of Artificial Intelligence.