A Smart Depression Assessment Approach Based on Remote Health Monitoring System

Depression is a prevalent mental disorder with significant public health implications, necessitating continuous monitoring and management due to its recurrent nature. Traditional healthcare information systems fall short in providing continuous monitoring required for effective depression management. Smart health monitoring (SHM), utilizing mobile devices and wearables, offers a solution by enabling remote continuous tracking of patients. However, current SHM approaches primarily focus on absolute levels of behavior, neglecting the importance of relative changes. Our study proposes a deep learning-based smart depression assessment approach that captures the relative changes to habitual behavior and the relative changes across temporal dimensions in a high-dimensional space. The evaluation using real-world data from patients with depression shows that our approach outperforms existing models, providing an effective tool for depression assessment. This study contributes to intelligent depression management by leveraging SHM data, facilitating the understanding of the patient’s condition and provision of timely interventions.