AI Has Learned to Spot Suicidal Tendencies

Visualization of the brain and processing of data using artificial intelligence will prevent many cases of suicide, scientists believe. The study was published in the journal Nature Human Behavior.

Annually 800 thousand people commit suicide and often it becomes a surprise for relatives and acquaintances of a suicide. The researchers suggested that his brain can tell about a person's willingness to commit suicide. They invited 34 volunteers, half of whom had suicidal tendencies and some of them had already attempted to commit suicide earlier, and the second half had no such inclinations. Participants underwent brain scans with fMRI, during which 10 words related to suicide ("lifelessness", "despair", etc.), 10 words with a positive shade (for example, "carelessness") and 10 - with a negative ("problems", etc.).

The brain reaction of potential suicides was the most different from the response of the control group in the words "death", "cruelty", "problem", "carelessness", "good" and "praise". Based on these data, the machine-learning algorithm was able to recognize possible suicides in 91% of cases, gazeta.ru reports.

According to the researchers, further work with larger samples will allow us to better assess the accuracy of the method - perhaps, it will actually allow us to identify suicidal tendencies in time and thus prevent suicide.

However, other experts are not so optimistic - in addition to a small sample, they also note the high cost of the method, which will make it available only in selected cases. In addition, while it is impossible to say for sure how much the revealed features of the reaction of the brain really correspond to what happens to the human psyche.

© All rights reserved. Citing to www.ict.az is necessary upon using news