Artificial intelligence may be able to detect suicidal tendencies based on how your brain processes certain words, a new study reveals.
The current system for finding out if someone is suicidal is to straight up ask them if they are feeling suicidal, but often people lie.
“Nearly 80% of patients who die by suicide deny suicidal ideation in their last contact with a mental health care professional,” according to the study, published in the journal Nature Human Behavior.
Marcel Just, a professor of psychology at Carnegie Mellon and director of the Center for Cognitive Brain Imaging, taught a computer program to recognize and identify emotions and hoped the technology could find ways to prevent suicidal behavior.
The study identified suicide as the second-leading cause of death among young adults, validating the need to find ways to be more alert for suicidal tendencies in teens. The researchers decided to see if brains scans could determine if a person was suicidal or not, using a functional magnetic resonance imaging machine and Just’s program to see what words triggered which responses.
Thirty-four participants, 17 who were suicidal and 17 who said they were not, were tested. They read 30 words that were positive, like “carefree” or “good,” negative like “cruelty” or “shame,” and words related to death like “overdose” or “hopeless” while the machine tracked how their brains lit up on the scans.
The machine had an accuracy of 91%, correctly identifying 15 of the 17 suicidal participants and 16 of the 17 non-suicidal controls.
The program could be an early warning sign of self-destructive thinking.
“We can look at the neural signature and see how it’s changed, see what this person is thinking, whether it’s unusual,” Just told the Daily Beast.
The program was also able to tell between subjects who had engaged in suicidal thinking and those who had actually attempted suicide. It found that those who had attempted suicide responded to death-related words with less sadness, while those who had just contemplated it responded to death- and sadness-related words with more anger.
It won’t completely prevent suicide but is a major step in understanding what Just called “thought disorders.”
“This isn’t a wild pie in the sky idea,” Just said. “We can use machine learning to figure out the physicality of thought. We can help people.”