New Study Highlights Suicide Prevention Using AI Now Possible

We have all heard the news about how suicide rates are rising in the United States. But did you know that in the US, every 16.2 minutes one person commits suicide? It’s also one of the leading causes of deaths amongst adults and is one of the most difficult to find out beforehand.

To help reduce suicides, scientists and psychologists have been trying for decades to find ways to predict those who have the highest risk of committing suicide. Unfortunately, rarely do the patients tell their doctors about their suicidal thoughts. Almost 80% of those who have committed suicide did not let their therapist know of their intentions before taking their own lives. According to statistics, suicide is the second leading cause for 24 to 35-year-olds while it’s the third leading cause of deaths for 15 to 24-year-olds. Unfortunately, due to our weakened social networking and support, more and more middle-aged Americans are also turning to suicide.

Sucidal Thoughts

Now, we may finally be seeing a new technology using AI (Artificial Intelligence) that could reliably scan patients to find suicidal behaviors and thoughts.

New Breakthrough In Identifying Suicidal Thoughts And Behaviors:

A few days ago, in the journal Nature Human Behavior, a report was published that may be a major milestone in preventing suicides through using AI. This study used machine learning and neural imaging to find out if the brains of those who have suicidal thoughts are triggered differently to positive and negative words regarding life and death.

According to Matthew Nock who is a clinical psychologist at Harvard University and co-author of this report, “We can predict with a pretty surprising degree of accuracy who’s had thoughts of suicide and who hasn’t—and even among those with thoughts of suicide, who has made an attempt and who hasn’t.” While the study conducted for the report is sized with only a few participants, the results that the researchers have managed to obtain have significant, according to Barry Horwitz who is the chief of the Brain Imaging and Modeling Section at the National Institute on Deafness and Other Communication Disorders.

Suicidal Thoughts And Behaviors

Horwitz went on to say that it’s impressive that the AI managed to differentiate correctly, the participants who had previously made attempts at suicide. The reason for Horwitz’s remarks is due to how challenging psychiatrists find it to make that distinction successfully. He also said that “It’s hard to imagine any other method or risk factor allowing you to make that kind of distinction.”

Other Studies On Using Technology To Detect Suicide:

This is not the first time technology has been used to detect the risk of suicide.  Earlier this year, another study was conducted using machine learning of pervious health records to find suicide risk with 80-90% accuracy. While this report was encouraging, this new study has managed to find the possible biological marker that causes suicidal thinking. A cognitive neuroscientist at Carnegie Mellon University and the report’s co-author Marcel Just said this, “It’s not just a reported behavior, we get the actual thoughts they have about suicide, and we see how they’re altered.”

What Nock and Just managed to do is incorporate two different lines of research to produce a technique to identify the biological markers for suicidal thoughts and behavior. The most straightforward example of this technique would be to use implicit association tests to pair words related to suicide, death, and life to works that connect to the person like ‘like me’ or ‘not like me.’ They used this technique repeatedly replicate the results consistently.

How Was The Study Conducted?

The study used a fMRI (functional magnetic resonance imaging) to find the patterns that correspond to particular thoughts, which is the technique Just has named neuro-semantics. This technique works by the recognition of concepts, not words. What Just has managed to do is collect the recognizable patterns of various thoughts and emotions. These emotions include anger and jealousy, were captured by using method actors in a separate study by Just.

Thought Scans

The research pool had 17 subjects who had tried to commit suicide before with 17 others being the control subjects. Each subject was asked to be scanned in the fMRI and asked to think about some words related to suicide along with specific positive and negative feelings and thoughts. This questioning resulted in the most impactful thoughts to be trouble, death, cruelty, good, praise and carefree. After the data was collected, the researchers asked the AI to find the differences between the 34 subjects. The AI accurately identified 15 out of 17 subjects who had tried to kill themselves before while 16 out of the 17 control subjects were accurately identified. This meant that the accuracy of the AI had been 91% for this test, which is very promising considering the rise in suicides around the world.

Is This New Technology Useful To Prevent Future Suicides?

While the other experts and suicidologist note that this research into the neural signatures being used to identify suicide risk has a bright future, it’s still in its first steps and needs to be further validated and solidified before we can use it for the public. Alexis May who is a clinical psychologist at the University of Utah and not part of the study says “These results leave open many questions about the mechanisms at work and the clinical implications given the intensive nature of the task and the expense of fMRI.”


To this statement, both Nock and Just agree with their desire to start off with replicating the study with success. After which will come the time to see if the technique is clinically viable since having every person take a fMRI scan is not going to be an option.  Nock went on to say “If you can get good prediction using a three-minute behavioral test, why spend a thousand dollars to put someone in a scanner? What we’re doing is seeing if these approaches give us a different piece of the puzzle.”

The conclusion to take from this new research is that it is possible to find suicidal tendencies in people using this technique from Nock and Just. However, this may not be available for our use until more tests and research is conducted.

And If you find yourself feeling suicidal thoughts, please get the help you need and deserve.

New Study Highlights Suicide Prevention Using AI Now Possible
Rate this

About Jessica J 240 Articles
Jessica is a health and nutrition expert. Jessica has over 10 years of experience in the health & beauty industry. She is the chief editor for this website & is currently studying medicine. Apart from her love for health, nutrition & beauty, she loves researching. Loves taking a dip in the pool whenever she can. Follow me on Google+ & Pinterest.