My Experiences: Leveraging Digital Technologies to Better Understand Mental Health

Overview

Mental health disorders are one of the most challenging chronic conditions to identify, treat and manage. This is largely due to the fact that diagnoses are almost entirely based on the patient's recall of current and past subjective experiences of symptoms; and then further interpreted by a healthcare professional introducing multiple layers of information biases in the formulation of a diagnosis. Accordingly, mental health conditions remain prevalent with high rates of misdiagnosis, inappropriate treatment and delayed intervention. In light of the heterogeneity across and within mental health conditions, a personalized interventional approach holds merit, yet the tools to effectively track an individual's day to day objective and subjective experience needed to achieve an individualized care approach have not until recently existed. Digital technologies such as passive and active sensing from smartphones and from wearable devices are shedding light on the capabilities of tracking new objective measures of health that could translate to key symptoms of mental health conditions. 'Multimodal data' approaches are those that attempt to translate a variety of electrical signals from digital devices to relevant health outcomes. The combination of digital devices to detect multimodal measures of mental health symptoms offers a unique opportunity to take a ground up approach in understanding the fluidity of mental health symptoms occurring at the individual level that might lend insight into new phenotypes of mental health illnesses that could have a physiological underpinning. The Study Investigators aim to characterize the multiplexing and fluid nature of mental health symptoms across individuals experiencing mental health symptoms and conditions using digital tools (i.e., wearables and mobile apps) and additional context information collected from virtual study support calls. The Investigators hope to know how objective measures from sensor data translate to core symptoms, episodes and flares across mental health disorders, and develop new (or new applications of) machine learning anomaly detection approaches and determine whether anomalies in expected symptom portraits can be reliably detected and enhanced by the addition of objectively measured data.

Conditions