Artificial intelligence to predict anxiety disorders

Thanks to artificial intelligence, researchers have been able to identify, in adolescents, the warning signs of the onset of anxiety disorders in early adulthood.

A new study, conducted by scientists from the National Institute of Health and Medical Research (Inserm), the National Center for Scientific Research (CNRS), the École Normale Supérieure and the University of Paris-Saclay , highlighted the predictive power of artificial intelligence (AI) in the context of psychiatric illnesses and more specifically anxiety disorders. The results, published in the journal Molecular Psychiatry indicate that AI is able to detect early warning signs of anxiety that would occur in young adults as early as adolescence.

Three signs identified

To arrive at this result, more than 2,000 adolescents of both sexes completed online questionnaires on their state of psychological health at the age of 14, 18 and 23 years. “The follow-up over time of the volunteers made it possible to measure the evolution of the diagnosis of anxiety”, explains Inserm. Then, a statistical learning study, which relies on an artificial intelligence algorithm, made it possible to determine whether some of the answers formulated in adolescence (14 years old) had an impact on the diagnosis of anxiety disorders at adult age (18-23 years). The researchers then highlighted three predictive signs: neuroticism, i.e. “a persistent tendency to feel negative emotions (fear, sadness, embarrassment, anger, guilt, disgust), poor control of impulses, and maladjustment to stress” ; the despair “associated with a low score for responses to questionnaires assessing optimism and self-confidence” and the emotional symptoms that “cover responses to questionnaires indicating symptoms such as ‘headache/stomach ache’, ‘a lot of worries, often worried’, ‘often unhappy, downcast or tearful’, ‘nervous in new situations, easily loses confidence ‘, ‘is easily frightened'”.

Using MRI to go further

In addition to AI, the researchers also relied on magnetic resonance imaging (MRI) to observe the brains of volunteers. As its development involves a change in the volume of different brain regions at the time of adolescence, the scientists sought to identify a possible modification at the level of gray matter. “While imagery did not improve the predictive performance of anxiety disorders as a whole compared to data from questionnaires alone, it could nevertheless make it possible to determine more precisely the type of anxiety disorder towards which a person is subject to change”explains the Institute.

“Our study reveals for the first time that it is possible to predict in an individualized way, from adolescence, the appearance of future anxiety disorders, notes Jean-Luc Martinot, co-author of the study, director of research at Inserm and child psychiatrist. These identified predictors or warning signs could make it possible to detect people at risk earlier and offer them an appropriate and personalized intervention, while limiting the progression of these pathologies and their consequences on daily life. »

21% of adults affected by anxiety disorders

It is estimated that 21% of adults are affected at least once in their life by anxiety disorders. The person who suffers from it feels “a strong and lasting anxiety unrelated to any real danger or threat, which interferes with normal functioning and daily activities”, says Inserm. These disorders tend to appear or consolidate in early adulthood, although they can sometimes begin in childhood or adolescence. “Thus, better identification in these age groups would prevent a worsening of symptoms over the course of life”considers the Institute.

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.