Courosh Mehanian “Using AI to Tackle Global Health Problems”

Abstract: The machine learning group at Global Health Labs harnesses the power of AI to tackle global health problems. Through an extensive network of field partners, governmental and non-governmental agencies, and universities, the team has curated or obtained extensive datasets for training data-hungry machine learning models. The team has developed algorithms for automated detection and counting of malaria parasites in blood smears; automated detection of precancerous lesions from smartphone cervix images; and automated detection of childhood pneumonia with ultrasound. Overviews of these three projects will be given and lessons learned will be discussed.

Bio: After earning a Ph.D. in physics from Cornell University, Courosh Mehanian changed course to pursue a career doing research in the areas of machine learning and computer vision. He has made contributions in neural networks for early visual processing, automatic target recognition, semiconductor quality control, automated pap-smear interpretation, and automated digital pathology for drug safety testing. Prior to joining Global Health Labs, he held positions at Boston University, MIT Lincoln Laboratory, Cytyc, KLA-Tencor, and Charles River Laboratories.