Post-traumatic stress disorder (PTSD) in children has long been challenging to diagnose due to limited communication skills and emotional awareness. Researchers at the University of South Florida (USF) have developed an innovative AI system utilizing facial recognition technology to identify PTSD through subtle facial expressions. Led by Alison Salloum, an expert in PTSD diagnosis, and Shaun Canavan from the AI and Cybersecurity department, this system prioritizes patient privacy by blurring identities and analyzing physical cues like head position and facial gestures. Trained on over 100 minutes of video per child, the AI detects emotional responses effectively, providing real-time feedback for therapists. The study aims to maintain ethical standards while highlighting the potential for this technology in diagnosing other mental health conditions such as anxiety, depression, and ADHD. Continuous efforts are underway to remove biases tied to gender, culture, and age, ensuring reliability in diagnosing pre-schoolers. This AI solution supplements, rather than replaces, traditional clinical practices.
Source link
