SciTech

CMU and Pitt researchers use fMRI to prevent suicide

Credit: Isabelle Vincent/ Credit: Isabelle Vincent/

One reason suicide is so unfortunately difficult to predict and prevent is because of its unpredictability. In some cases, suicidal people choose to conceal their struggles and scars, so on the outside they may seem perfectly fine. When they take their own life, they leave behind a trail of unanswered questions and regrets — “Why didn’t I see this coming? What could I have done to help them?” The consistency in high suicide rates is such a dark and alarming phenomenon that would be easy to take down if we could know when someone in depression turns their thoughts into action.

Researchers at Carnegie Mellon University and the University of Pittsburgh may have taken us a step closer to suicide prevention. On Oct. 30, Professor Marcel Just from CMU and Professor David Brent from Pitt announced that they have developed an approach to identify individuals with suicidal thoughts by analyzing brain activity when they think of concepts like death or cruelty. Just and Brent used fMRI to monitor the neural responses of 34 subjects — 17 suicidal, 17 control subjects — to a list of 30 words. The list contained words with positive, negative, and suicidal connotations, and the researchers asked each participant to think for three seconds about each word. They then recorded cerebral blood flow to infer which parts of the brain were activated. An algorithm was then used to differentiate whether or not the participant had suicidal thoughts.

Just and Brent were able to find that the algorithm successfully classifies a participant as “suicidal” 91 percent of the time. They found that six words were especially good predictors that activated certain areas of the brain much more than other words: death, cruelty, trouble, carefree, good, and praise. They then focused on the six words and used a similar approach to test if the algorithm could identify the participants who had experience of a suicide attempt or self-harm. They were able to find that the algorithm could accurately find the participants who had attempted to take their lives with 94 percent accuracy.

These results open the window for so many new potential applications in psychiatry. According to Professor Brent, “the most immediate need is to apply these findings to a much larger sample and then use it to predict future suicide attempts. It 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.”

Not only do the results of this research help clinicians identify individuals with suicidal thoughts, but they also suggest that new methods of treatment and therapy for those in depression. The study showed that the concept of "death" evoked relatively more shame and sadness in the group of participants that had thought about suicide. This information can easily be used by psychiatrists to gear their therapy to tailor the specific emotional needs of their patients.

However, this does not mean that the technique is perfect. According to science magazine WIRED, fMRI scans have certain shortcomings that suggest they may not be perfectly accurate indicators of brain activity. Especially with a relatively small sample size of 34 participants, it is difficult to guarantee that the algorithm will be successful for every single individual with suicidal thoughts.

It’s also extremely crucial not to confuse correlation with causation. Just because two things occur at the same time doesn’t prove that one causes the other — in this case, more testing would be necessary to indisputably prove that the suicidal thoughts caused distinct brain activity. According to Professor Just, “comparing groups of individuals with or without suicide risk isn’t the same thing as holding up a brain scan and assigning its owner a likelihood of going through with it.”

But the results of their study definitely point us in the right direction. A study published in the Journal of Clinical Psychiatry showed that nearly 80 percent of the people who committed suicide in the hospital or immediately after discharge denied thinking about suicide to their assigned medical professional. This serves to show how reluctant people with dark thoughts may be to share their emotional states, and how the need for a method to scientifically and methodically predict suicide is growing. Clearly, it’s better to know the brain is doing something than knowing nothing at all.