Augmenting psychiatric diagnosis with data
The latest "virtual therapist" -- charged with helping diagnose depression -- comes from psychologist Albert Rizzo and computer scientist Louis-Philippe Morency at the University of Southern California. Their system of cameras and microphones analyzes a person's gestures, facial expressions, body habitus, and vocalization -- taking in 30 data points per second -- then compares them to those of typical depressed and non-depressed people.
The therapist persona was crafted carefully. "Everything has been thought of," Morency told NPR's Alix Spiegel on today's Morning Edition. "We have recorded more than 200 'uh-huhs' ... a simple 'uh-huh' and a silence -- if they are done the right way -- can be extremely powerful. So we spent a lot of time on these little details."
Those details, though, are rendered with an eerie ethereality.
SimSensei's poise seems to be constantly set at 100, "uh-huh" registering as "I'm flawless, and I am judging you." Maybe they should've given her a wandering eye or something. Still, the idea is promising. Morency says they look at the tool like a physician might look at a blood test; part of the big picture that shapes a diagnosis.