Adverse drug events (ADEs) are a grave problem facing medical care. The Food
and Drug Administration places ADEs as the fourth-leading cause of mortality
in the U.S., where they harm more than two million people and cost $136
billion in additional care each year. Researchers have responded by
developing computational versions of epidemiological study designs and
analyses, but machine learning techniques have not yet been widely applied.
Popular Science, in a March 31, 2014 article, features the research of graduate student Sean Andrist on gaze aversion and Nao robots, part of a strategy to make interactions with robots seem more natural.