Medicine and Health
Personalized prediction of negative affect in individuals with serious mental illness followed using long-term multimodal mobile phenotyping
C. A. Webb, B. Ren, et al.
Smartphones and wearables can detect real-time spikes in negative emotions by passively tracking behavior. In a year-long study of 68 adults with mood or psychotic disorders, a personalized ensemble machine learning model predicted states like irritability and loneliness (AUCs 0.72–0.79), with GPS location features most predictive and substantial individual variability observed. These results point to smartphone-triggered, timely emotional interventions. Research conducted by Christian A. Webb, Boyu Ren, Habiballah Rahimi-Eichi, Bryce W. Gillis, Yoonho Chung, and Justin T. Baker.
Related Publications
Explore these studies to deepen your understanding of the subject.
