Developing Personalized Algorithms for Sensing Mental Health Symptoms in Daily Life

The integration of artificial intelligence (AI) and pervasive computing offers new ways to sense mental health symptoms and deliver real-time interventions via mobile devices. This study explores personalized versus generalized machine learning models for detecting individual and family mental health symptoms using data from smartphones and smartwatches collected on the Colliga app. Over 60 days, data from 35 families resulted in approximately 14 million data points from 52 data streams. Findings showed that the personalized models outperformed the generalized models. Model performance varied significantly based on individual factors and symptom profiles, highlighting the need for tailored approaches. These results suggest that successful implementation of passive sensing AI technologies for mental health interventions requires considering each user’s unique characteristics. Further research is needed to refine the models, address data stream heterogeneity, and develop scalable systems for effective personalized mental health interventions.