A new artificial intelligence – AI – is reportedly capable of detecting any signs of depression and anxiety in the speech of children. The machine learning algorithm possesses the potential for coming up with an effective way of diagnosing conditions that are otherwise considered difficult to ascertain.
According to studies, every one in five children is suffering from depression and anxiety. However, children that are under the age of eight cannot use proper articulation for expressing their emotions, and therefore it makes it difficult for the adults to diagnose the problem.
Ellen McGinnis, a clinical psychologist at the University of Vermont Medical Center’s Vermont Center for Children, Youth and Families and lead author of the study, said, ‘We need quick, objective tests to catch kids when they are suffering. The majority of kids under eight are undiagnosed.’
The key to treating mental health disorders is an early diagnosis. If the disorder is caught and treated while the brain cells are undergoing development, there is a higher chance of positive outcomes in the children. The very same children are subject to a higher risk of substance abuse and suicide if the disorder is left untreated for later.
The team of researchers tested the new AIL by having a total of 71 children between the ages of three and eight through the Trier-Social Stress Task. This protocol is designed for the sake of inducing stress that is psychological. The kids were told that they had to improvise a three-minute story and that they would be judged on how interesting the story was. During the test, they were only provided with neutral or negative feedback. Ellen McGinnis says, ‘The task is designed to be stressful, and to put them in the mindset that someone was judging them.’
The team then made use of a machine learning algorithm – AI – for analyzing the audio recordings of each kid’s story. The AI’s results actually matched the ones that were received from structured clinical interviews and parents’ questionnaire.
University of Vermont biomedical engineer and study senior author Ryan McGinnis said, ‘The machine learning algorithm was able to identify children with a diagnosis of an internalizing disorder with 80% accuracy, and in most cases that compared really well to the accuracy of the parent checklist. And it did it in just a few seconds.’
The next step for the team is to develop the machine learning algorithm further so that it is able to become a universal screening tool ready for clinical use. It might find its application as a smartphone app that would be able to record and analyze the results right away.
The Journal of Biomedical and Health Informatics has published this research.