One of science fiction writer Isaac Asimov's famous “Three Laws of Robotics” was that a machine could never harm a human. Software developers have taken that ball and run with it way past anything even Asimov could have conceived: Using a person’s spoken or written words, robots can now identify with great accuracy whether that person is suicidal, mentally ill but not suicidal, or neither.
A new study out of Cincinnati Children’s Hospital Medical Center shows that computer technology known as machine learning is up to 93 percent accurate in correctly classifying a suicidal person and 85 percent accurate in identifying a person who is suicidal, has a mental illness but is not suicidal, or neither. John Pestian, PhD, professor in the divisions of Biomedical Informatics and Psychiatry at the Medical Center and lead author, believes that these results provide strong evidence for using advanced technology as a decision-support tool to help clinicians and caregivers identify and prevent suicidal behavior.
“These computational approaches provide novel opportunities to apply technological innovations in suicide care and prevention, and it surely is needed,” says Pestian. “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. 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.”
The researchers enlisted 379 patients in the study between October 2013 and March 2015 from emergency departments and inpatient and outpatient centers at three sites. These patients were a mix of those who were suicidal, had been diagnosed as mentally ill and not suicidal, or neither. This last served as the control group.
Each subject completed standardized behavioral rating scales and participated in a semi-structured interview answering five open-ended questions to stimulate conversation. Typical questions included “Do you have hope?” “Are you angry?” and “Does it hurt emotionally?”
After extracting and analyzing verbal and non-verbal language from the data, Pestian and his team then used machine learning algorithms to classify the patients into one of the three groups. The results showed that machine learning algorithms can tell the differences between the groups with up to 93 percent accuracy. Besides better accuracy, another added benefit was that the control patients tended to laugh more during interviews, sigh less, and express less anger, less emotional pain and more hope.
The study is published in the journal Suicide and Life-Threatening Behavior.