AI-Based Digital Therapeutics for Adolescent Mental Health Management and Disaster Response

This study focuses on the development and evaluation of an AI-based digital therapeutic prototype for adolescent mental health management and disaster response. The system integrates real-time monitoring, AI-driven conversation analysis, personalized psychological treatment programs, and multimodal data analysis. An algorithm was developed to detect gaslighting and verbal abuse using a BERT-based classification model, achieving 85% accuracy in gaslighting detection and 87% accuracy in verbal abuse detection. Additionally, a psychological disaster-recovery support module was included, which demonstrated a 30% improvement in users’ stress reduction rates during simulated disaster scenarios. This study demonstrates that digital therapeutic approaches can significantly contribute to early intervention in adolescent mental health issues. Additionally, these approaches provide effective support during disasters. The developed prototype demonstrates the potential of AI and digital technology to innovate mental health management and disaster response strategies.