Wearable Technology for Mental Health

Our goal is to improve mental health outcomes and enhance the well-being of individuals grappling with mental disorders. In pursuit of this mission, we explore signals from wearable technology, as an omnipresent and invaluable data source for in-depth digital phenotyping of psychiatric diseases. Recognizing that conventional methodologies employed in mental health research may fall short of encapsulating the intricacies of real-life experiences, we embrace the potential of wearable device data. We envision a continuous stream of information that holds the key to unveiling a nuanced understanding of psychiatric conditions. This not only opens avenues for uncovering subtle patterns of human behaviour but also sets the stage for targeted and timely interventions. In our approach, we leverage both locally collected and publicly available datasets, conducting a comprehensive analysis of various sensors. Our focus lies in harnessing the predictive capabilities of these sensors to gauge disease severity. By intertwining diverse data sources, we aim to construct a robust foundation for predictive modelling that not only contributes to the scientific discourse but also translates into tangible improvements in mental health outcomes.

Research Project

We aim to shed light on the contributions of diverse sensors in the digital phenotyping of psychiatric disorders, paving the way for informed directions in future studies in this field. Drawing inspiration from the success of the ReMAP app, our focus has been on scrutinizing the predictive capabilities of various algorithms in estimating depression severity. Encouraged by promising outcomes, we have extended our investigation to assess the predictive power of these algorithms across different dimensions of depressive symptomatology. Looking ahead, our overarching goal is to establish a robust scientific foundation that not only refines our understanding of mental health but also serves as a catalyst for targeted interventions driven by algorithmic insights derived from digital phenotyping data. In the pursuit of this long-term vision, we envision conducting comprehensive cost-benefit analyses to unravel the policy-level implications associated with deploying various interventions aligned with these predictive algorithms. This forward-thinking approach not only positions our research at the forefront of innovation but also underscores our commitment to translating scientific advancements into tangible, impactful outcomes for mental health intervention strategies.