People with suicidal thoughts stand no chance of hiding from psychologists who might miss out on crucial signs. This is after scientists launched a study in which they’re training a computer program to detect people with suicidal thoughts.
The method deployed by the study will heavily rely on brain scans and it will at one point be used in diagnosing mental health conditions.
It’s unfortunate that approximately one million people worldwide die by suicide every year, moreover, predicting suicide remains difficult since many people feel uncomfortable talking about the problem.
A study that was published today in the journal of Nature Communications, researchers studied the brain activity of two groups of adults - one had suicidal thoughts and one didn’t - while they thought about words such as “evil” or “praise.”
The scientists fed the data on the algorithm that was able to predict persons that had suicidal thoughts with an accuracy level of 91%. The algorithm was also able to predict whether the individuals had attempted suicide with an accuracy level of 94%.
Medical test ought to be perfect, but that’s not the case with the algorithm. The scan may not be widely used since brain scans are expensive. However, one of the psychologist at Carnegie Mellon University, Marcel Just, revealed that it’s also important to have the additional method in place.
The study involved thirty-four volunteers: 17 of them had suicidal thoughts and 17 without. The volunteers read 30 words that were either positive, negative, or related to death and thought about the meanings while undergoing a type of brain scan called MRI.
When we think about certain subjects, the neurons fire in a certain way. The neurons might fire in a specific pattern when a word such as “cat” or “dog” is mentioned.
The scientist found that responses to words such as “trouble”, “death”, “carefree,” “good,” “praise,” and “cruelty” had such huge differences among the two groups of participants. The results were then assigned to a machine-learning algorithm with the exception of one person.
For every word, they told the program which neural activation patterns came from which group. Then, they gave them the missing person’s results and asked the algorithm to predict which group the person belonged to. The machine got it right 91 percent of the time. In a second experiment, scientists used the same methods to teach an algorithm to distinguish people who had attempted suicide from those who hadn’t, this time with 94 percent accuracy.
One of the scientists acknowledges that the small number of participants is a limitation of today’s research. He also believes that in the future the algorithm could be used to diagnose people with suicidal thoughts, or even to check whether treatments for psychiatric disorders are working. More research will have to be done with more volunteers to enhance the accuracy of the algorithm and also diagnose if they have specific psychiatric problems.