In Harvard partnership, WPI professor used AI to predict suicidal behavior

Photo | Grant Welker Worcester Polytechnic Institute

Write a Comment

By Isabel Tehan

Dmitry Korkin, a Worcester Polytechnic Institute professor of computer science, led a team of researchers at WPI and Harvard Medical School-affiliated McLean Hospital on a three-year study on use of artificial intelligence to better predict suicide risk in women suffering from certain trauma-related disorders.

The researchers worked to develop an algorithm to identify patients at greatest risk for suicidal thoughts and behaviors, according to a Wednesday press release from WPI. Using data from 123 female patients at McLean Hospital in Belmont, the artificial intelligence was able to accurately predict suicidal history.

“This research represents a significant advance towards our objective of identifying patients at greatest risk of suicide and directing effective therapies at those individuals,” Jean King, a neuroscientist and the dean of arts and sciences at WPI, said in the press release.

The research team hopes the technology can be used to screen for warning signs of suicidal behaviors. It represents a breakthrough in the use of artificial intelligence in the realm of suicidal ideation, according to WPI.

The work was initiated and received funding from the Julia Kasparian Fund for Neuroscience Research, created by Harry Kasparian, a 1973 WPI graduate, in memory of his daughter who died by suicide at 23 in 2016.

The research team intends to continue use of the models identified and apply them to other groups susceptible to suicidal behavior.