Center for Cognitive Brain Imaging

at Carnegie Mellon University

CBS News May 2, 2019, 10:57 AM

Brain scans show promise in spotting suicidal thoughts

Carnegie Mellon psychology professor Dr. Marcel Just has spent decades developing a sophisticated system to try to read our minds. Now, he's making a breakthrough into suicidal thinking.

"We found that there were certain concepts that were altered in people who thought about suicide," Just told CBS News' Nikki Battiste. "And our method was able to detect those differences and analyze exactly what the differences consist of."

His method begins with a sophisticated piece of equipment: a functional MRI. With fMRI you see a picture of the brain's activity. It tracks neurons firing in the brain as a person thinks about an object or a concept. In other words, every thought or idea in our brain has its own specific pattern. What Just found is that those patterns are similar across many people — unless you have suicidal thoughts.

"That's what psychiatric illnesses do. They change the way you think. And this method can, I think, identify those changes," he said.

When comparing two scans of people's brains thinking about the word "death," you get different results in a person who has thought about suicide and one who has not. In the scan imagery, red represents activation of the brain related to self-thinking. Among people who have attempted suicide, there's even more of the dark red.

Dr. Marcel Just with an fMRI brain scan.
Dr. Marcel Just with an fMRI brain scan.

Thirty-four people took part in Dr. Just's latest study. Half of them never had suicidal thoughts, but the other 17 had, and over half of those 17 had even attempted suicide in the past. The participants were asked to think about 30 different concepts like "carefree," "praise" and "death" while their brain activity was tracked in the fMRI. The results were second-by-second snapshots of brain patterns that were then analyzed by a complex computer server, storing hundreds of thoughts by hundreds of people, with the power to search for unique patterns.

"We can tell whether someone's feeling anger or happiness or sadness. We can tell what number a person's thinking," Just said. "We can tell what topic a person is reading a paragraph about."

Just's method was 90 percent accurate in determining who had past suicidal thoughts and attempts. Dr. Joshua Gordon, director of the National Institute of Mental Health, thinks his work could lead to more effective treatment.

"If we could understand the neurobiology underlying the drive to harm one's self then we would be able to design better therapies aimed at reducing that desire," Gordon said. "The treatments that we have now don't work all that well."

Just said it's "slightly optimistic but not impossible" that in 10 or 20 years patients might be able to go to the doctor's office and find out whether or not they have a mental illness or are suicidal.

The National Institute of Mental Health recently awarded Just and his team a $3.8 million grant to continue their research. They have already proven their method works to identify autism spectrum disorder. In the future, Just hopes to apply his approach to other psychiatric disorders like depression, schizophrenia, and bipolar disorder.

© 2019 CBS Interactive Inc. All Rights Reserved.