Scientists say machine learning is up to 93 percent accurate in identifying a suicidal person based on their responses to interview questions.

The algorithm was described in a study published in the journal Suicide and Life-Threatening Behavior. Researchers were able to use the tool to classify patients as being suicidal, mentally ill but not suicidal, or neither.

“These computational approaches provide novel opportunities to apply technological innovations in suicide care and prevention, and it surely is needed,” study author John Pestian said in a press release. “When you look around healthcare facilities, you see tremendous support from technology, but not so much for those who care for mental illness. Only now are our algorithms capable of supporting those caregivers.”

In the study, researchers recruited 379 patients between October 2013 and March 2015 from emergency departments and inpatient and outpatient centers at three locations. The sample included people who were suicidal, diagnosed with a mental illness but not suicidal, and a control group that did not demonstrate either behavior.

The technology assessed verbal and non-verbal language from interviews with patients, measuring their responses to questions including “do you have any hope?”, “are you angry?”, and “does it hurt emotionally?”

Scientists noted patients that were either mentally ill or suicidal expressed more anger and emotional pain. Subjects in the control group were found to be more likely to laugh and less likely to sigh during interviews. The team is confident the approach can be used to supplement suicide prevention strategies.

“This methodology easily can be extended to schools, shelters, youth clubs, juvenile justice centers, and community centers, where earlier identification may help to reduce suicide attempts and deaths,” Pestian added.