Developing a Multi-Modal Edge-AI Framework for Continuous Infant Monitoring: Predicting Mental Health Outcomes

The evolution of Edge-AI technologies has created new opportunities in pediatric healthcare, allowing for real-time monitoring of infants while maintaining privacy. This research introduces an innovative multi-modal Edge-AI framework that combines video, audio, and physiological data to anticipate potential mental health issues in infants. The proposed system processes information locally on edge devices, minimizing latency, enhancing privacy, and enabling continuous monitoring in both clinical and home settings. By employing lightweight AI models for on-device processing, the system promotes early identification of neurodevelopmental challenges and encourages timely interventions. This approach aims to shift healthcare from a reactive stance to a preventive one, ultimately aiming to foster long-term enhancements in mental health. The paper outlines the system's architecture, techniques for optimizing AI models, and prospective applications in pediatric healthcare environments.