Applications of Technology in Mental Health Assessment and Treatment

A pilot study led by Christian A. Webb, Ph.D., a Young Investigator at the Brain & Behavior Research Foundation (BBRF), explored the use of smartphones and AI tools like ChatGPT to enhance the assessment of anhedonia in depressed adolescents. This study involved 38 participants, ages 13–18, undergoing behavioral activation (BA) therapy for anhedonia, where they participated in weekly therapy sessions and provided daily reports on their mood and activities. Passive data collection via smartphone sensors was employed to track movement and activity levels, while ChatGPT analyzed text entries for emotional insights.

The findings revealed a positive correlation between the increases in participants' engagement, as measured by smartphone data, and their self-reported activation levels. This suggests that real-time tracking and feedback could bolster therapeutic outcomes. The researchers concluded that AI tools can extract meaningful emotional insights from everyday language, aiding therapists in monitoring patient progress.

In another study, researchers developed the SenseToKnow app, designed to screen toddlers for autism spectrum disorder (ASD) during well-child visits. The app utilizes video-based analysis and machine learning to enhance early detection of autism, with an accuracy of 88% for sensitivity and 81% for specificity in distinguishing between autistic and neurotypical behaviors. Combining its results with existing screening tools significantly improved diagnostic accuracy, underscoring the potential of integrating novel technologies in clinical settings.

Additionally, wearable technology, specifically Fitbit devices, has shown promise in predicting mood shifts in bipolar disorder (BD). A study led by Jessica M. Lipschitz, Ph.D., demonstrated that passive monitoring of physical activity and sleep patterns using Fitbit data could effectively predict clinically significant mood changes with an accuracy of 89.1% for manic symptoms and 80.1% for depressive episodes. This research highlights how digital phenotyping can facilitate timely interventions for individuals with BD.

Overall, these advancements emphasize the transformative potential of integrating digital technologies in mental health care, enhancing both diagnostic processes and treatment efficacy.