Smartwatches and Psychiatry: A New Perspective

Using data collected from over 5,000 adolescents aged 9 to 14, the researchers developed artificial intelligence (AI) models to predict psychiatric conditions such as ADHD and anxiety. The data, sourced from the Adolescent Brain Cognitive Development Study, included heart rate, physical activity levels, calorie expenditure, step counts, and sleep metrics. 

Prof. Mark Gerstein explained: “In traditional psychiatry, a doctor will assess your symptoms and you’ll either be diagnosed with an illness or won’t. But in this study, we focused on processing the wearable data in a way that could both be leveraged to predict illnesses more comprehensively, and to better connect them to underlying genetic factors”. 

The researchers introduced the concept of the "digital phenotype," describing traits measured through devices like smartwatches. This approach enables continuous tracking of physical and behavioral patterns over time, offering a detailed view of psychiatric illnesses. 

Jason Liu, a member of Gerstein’s lab and co-lead author of the study, says: “One advantage of doing this is that we can use the digital phenotype almost as a diagnostic tool or a biomarker, and also bridge the gap between disease and genetics.” 

The team identified that heart rate was particularly significant for ADHD prediction, while sleep quality and stages were more relevant for anxiety diagnosis. Additionally, wearable data could help differentiate between subtypes of conditions like ADHD. For example, distinctions between inattention and hyperactivity might improve treatment strategies, noted Beatrice Borsari, a postdoctoral associate and co-lead author. 

One groundbreaking aspect of this research was the integration of smartwatch data with genetic analysis. Using multivariate statistical tools, the team explored the relationship between digital phenotypes and genotypes. Professor Diego Garrido Martín explained:   

“Our methodology has made it possible, for the first time, to simultaneously analyze the relationship between genetics and the different measures provided by smartwatches”. 

The study identified 37 genes associated with ADHD by examining how genetic mutations influenced smartwatch data. Interestingly, when focusing solely on ADHD diagnoses, no specific genetic links were found, emphasizing the importance of continuous data collection. 

These findings suggest that wearable sensors could enhance the understanding and treatment of psychiatric disorders. While the study concentrated on ADHD and anxiety, researchers believe the methodology could extend to conditions like neurodegenerative diseases. By integrating quantitative behavioral measurements with genetic analysis, future diagnostics might move beyond traditional symptom-based methods. 

“This method holds great promise for addressing long-standing challenges in psychiatry and may ultimately reshape the way we understand the genetics and symptom structure of psychiatric disorders”, concluded Walter Roberts, assistant professor of psychiatry at Yale School of Medicine. 

MEDICA-tradefair.com; Source: Universidad de Barcelona