Scientists using artificial intelligence to detect” post-traumatic stress disorder " in children

  

Diagnosing post-traumatic stress disorder (PTSD) in children is a major challenge due to their limited ability to communicate and poor emotional awareness, and to meet these challenges, a research team at the University of South Florida has developed an artificial intelligence system that analyzes the child's facial movements during interviews, to detect expressive patterns associated with the disorder, while maintaining the participants ' complete privacy. How the system works

A research team has developed an artificial intelligence system that could be a cost-effective tool for doctors to diagnose PTSD in children and adolescents, with the possibility of tracking the improvement of their condition over time. The research team is led by Dr. Alison Salloum, a professor at the School of Social Work at the University of South Florida, and Dr. Sean Canavan, an associate professor at the Bellini School of Artificial Intelligence, Cybersecurity, and Computer Science. The system does not rely on the videos recorded during the interviews with the children as they are, but rather on certain non-identifying facial signals, such as eye movement, mouth position, and head position. This approach ensures the concealment of any data that may reveal the identity of the child, focusing only on the subtle movements of the facial muscles that reflect the emotional state. The study on testing the system, published in the Journal of Pattern Recognition Letters, showed that children with PTSD showed characteristic emotional expressions that could be accurately monitored. It was also found that these expressions were more pronounced during their sessions with specialists compared to sessions conducted by parents, which is consistent with psychological research indicating that children tend to express emotions more in front of specialists than parents because of shyness or to avoid facing painful feelings. Dr Alison Salloum says: "I noticed during the interviews of traumatized children that their faces reveal their suffering even when they don't talk much. “ Hence the idea of taking advantage of artificial intelligence to monitor these signals scientifically,” he said. “What distinguishes our system is that it does not use raw video at all, but is limited to anonymous facial movement data, taking into account the fact that the child is talking with one of his parents or with a specialist,” said Dr. Sean Canavan.

Study results and future applications

The research team analyzed data taken from 18 sessions, which included more than 100 minutes of video per child, and about 185 thousand frames per video. Artificial intelligence algorithms were able to identify fine motor patterns associated with the disorder. The researchers hope to expand the scope of the study to investigate the influence of factors such as age, gender, and cultural background, especially in young children whose diagnosis is often based on parental observations. They also believe that this technology can be used in the future to provide instant feedback to doctors during treatment sessions, helping them make more accurate decisions and improve the quality of psychological care.

Promising prospects

Although the system is still in its early stages, the research team believes that its future applications will be wide-ranging. The majority of the children involved in the study had complex psychological conditions, such as depression, ADHD, or anxiety, which makes the system suitable for diagnosing various conditions with high accuracy. Dr. Sean Canavan says: "obtaining accurate and reliable data from this age group is rare, and we are proud that we have been able to develop a system that adheres to ethical standards and preserves the privacy of children “ If this system proves effective in larger trials, it could change the way PTSD is diagnosed in children in the future, he said. This is where artificial intelligence can provide important support, not to replace doctors, but to enhance their tools; in the future, the system can be used to provide immediate feedback to doctors during treatment sessions and monitor the extent of improvement without having to repeat psychologically exhausting interviews for the child, says Dr. Alison Salloum.