Using artificial intelligence to diagnose PTSD 

Scientists trying to develop an objective diagnostic test for PTSD — which is typically identified via interviews and surveys — say that a speech-based algorithm correctly classified patients with and without PTSD almost 90 percent of the time in a preliminary study. The research team compared the voices of 52 veterans with PTSD and 77 veterans without the condition and used artificial intelligence to sort through the tape and pick out ways to decipher differences in the voices. The voice recognition tool honed in on a number of speech features, determining that PTSD was more associated with slower, more monotonous speech. The researchers say the tool needs further refinement and validation before it could be put into regular use.