AI Can Diagnose Depression in Speech Patterns of Kids

The research report published in the Journal of Biomedical and Health Informatics supports the unique idea of developing artificial intelligence (AI) system that can detect anxiety and depression signs in the speech patterns of children. According to scientists, the tool is efficient enough to provide the simplest and most effective way to diagnose conditions that are difficult to spot and is often overlooked in young people.

Known as “internalizing disorders”, this tool will help to analyze the issue, as children under the age of eight usually cannot express their emotional suffering. Ellen McGinnis, the clinical psychologist from the University of Vermont in the U.S. said that it is important to have quick, objective tests to understand the kids’ suffering as about one in five children suffer from anxiety and depression. Early diagnosis of these issues can help children respond well to treatment as their brains are still in the developing stage. Late detection can create a greater risk of substance abuse and suicide among them in later life.

Researchers have been looking for ways to use AI and machine learning to make diagnosis faster. The standard diagnosis includes a 60-90 minute semi-structured interview along with a trained clinician, whereas, the Trier-Social Stress Task will cause feelings of stress and anxiety to diagnose it quickly and more appropriately. The test also includes a clinically structured interview and parent questionnaire, to determine the internalized disorders in children.

The researchers used a machine learning algorithm to analyze the statistical features of the audio recordings of the kid’s story and relate them to the diagnosis process. The algorithm is able to identify children with a diagnosis with 80% accuracy.

McGinnis said that the future plan will be to develop the speech analysis algorithm tool for clinical use, probably via a smartphone app, which could record and show results instantly. The voice analysis can be combined with the motion analysis with battery-driven technology-assisted diagnostic tools to help identify children at risk of anxiety and depression.