Contributing to the fact that suicide is the second-leading cause of death among young adults in the US is the difficulty psychologists face in accurately identifying suicide risk in patients. Following some compelling research suggesting brain inflammation is linked to suicidal thoughts, a team of scientists has now used a machine-learning algorithm and brain imaging to accurately identify whether a person is considering suicide.
This latest study looked at 34 subjects (17 with suicidal tendencies and 17 neurotypical subjects) and recorded their brain activity using functional magnetic resonance imaging (fMRI) while they were presented with three lists of 10 words. One list comprised death-related words (eg. "death", "fatal"), one list was filled with positive concepts ("carefree", "praise") and the final list related to negative ideas ("trouble", "boredom").
The researchers' algorithm examined the brain scans and was able to correctly distinguish the suicidal group from the control with an accuracy rate of 91 percent. Honing in even further on the brain imaging the researchers were able to identify the members of the suicidal group who had actually attempted to take their own lives versus those who had only thought about it.
The research team has previously developed a library of what they are calling "neural signatures." These are brain images that act like fingerprints for certain emotional responses, most specifically sadness, shame, anger and pride. The new study revealed that those with suicidal tendencies have different emotional responses to certain concepts when compared to neurotypical subjects.
"People with suicidal thoughts experience different emotions when they think about some of the test concepts," explains Marcel Just from Carnegie Mellon University. "For example, the concept of 'death' evoked more shame and more sadness in the group that thought about suicide. This extra bit of understanding may suggest an avenue to treatment that attempts to change the emotional response to certain concepts."
Of course, this study is based on a very small sample size and follows on from some very specific assumptions regarding maps of brain activity and emotions. It's unclear how broadly these results could apply to any practical outcome, but the researchers are confident that more general testing could help them develop a method that has clinical applications.
"Further testing of this approach in a larger sample will determine its generality and its ability to predict future suicidal behavior, and could give clinicians in the future a way to identify, monitor and perhaps intervene with the altered and often distorted thinking that so often characterizes seriously suicidal individuals," says the University of Pittsburgh's David Brent.
Research into identifying suicidal individuals is a major focus at the moment. As well as the new understanding into possible inflammatory foundations of depression and the discovery of a gene that increases risk of suicide, another team earlier this year developed an algorithm that can pore through a patient's medical records and predict future suicide attempts almost 85 percent of the time.
This latest study is another piece of evidence suggesting those individuals that are more prone to suicide can possibly be objectively identified either though a blood test, or in this instance, a brain scan.
The study was published in the journal Nature Human Behavior.
Source: Carnegie Mellon University