Suicidal risk is very difficult to accurately predict and is often a troubling concern for mental health professionals and caretakers alike. Suicide is one of the leading causes of preventable death in young people, and if detected early, suicidal intent can be managed effectively.
Marcel Just of Carnegie Mellon University and his colleagues developed an approach to identifying suicidal intent by analyzing the changes in how brains represented certain concepts such as death, cruelty, and trouble.
They used machine learning algorithms to identify the representation of concepts related to suicide. This gave them an unique opportunity to view how suicidal individuals thought about suicide, self-harm and related concepts.
What is central to this new study is that we can tell whether someone is considering suicide by the way that they are thinking about the death-related topics,” – Just, the D.O. Hebb University Professor of Psychology in CMU’s Dietrich College of Humanities and Social Sciences.
10 death-related words, 10 words relating to positive concepts (e.g. carefree) and 10 words related to negative ideas (e.g. trouble) were presented to two groups of 17 people with known suicidal tendencies and 17 “normal” individuals.
The machine-learning algorithm was applied to six word-concepts that best discriminated between the two groups as the participants thought about each one while in the brain scan.
The six-word concepts were were death, cruelty, trouble, carefree, good and praise. Based on the brain activation patterns of these six concepts, their algorithm identified ( with 91 percent accuracy ) whether a participant was from the control or suicidal group.
The researchers also used an archive of neural signatures for emotions (particularly sadness, shame, anger, and pride) to measure the strength an emotion that was evoked in a participant’s brain by the six discriminating concepts. The algorithm was also able to able to accurately predict which group the participant belonged to (suicidal vs normal)
The reality of a brain scan which can predict suicidality?
This study is obvious proof of concept and further research with an increased sample size will determine if this will become a future consumer possibility.
Advances in neuroimaging and machine learning, reducing costs and large-scale applicability are key factors that would predict further developments. The researchers hope that these findings will translate to real-world technology that saves lives.
Machine learning of neural representations of suicide and emotion concepts identifies suicidal youth, Nature Human Behaviour (2017). www.nature.com/articles/s41562-017-0234-y
A list of suicide helplines in India here