New Delhi, March 18 A study has proposed a scalable and accessible framework for analyzing data from wearable devices, such as smartwatches, to detect early signs of diabetes.
Scientists from Google Research in the US predicted insulin resistance among 1,165 participants using data collected from smartwatches, along with demographic and routine blood biomarker information, including fasting glucose and lipid profile.
Participants with insulin resistance have a higher risk of diabetes, cardiovascular disease, hyperlipidemia, and hypertension, the authors said in the study published in the Nature journal.
Experiments showed that fasting glucose alone is not sufficient for estimating insulin resistance, highlighting the importance of lifestyle factors, they said.
"In this study, we present a method for predicting insulin resistance (IR) using signals derived from a consumer smartwatch, demographics, and routinely measured blood biomarkers. This method has the potential to be scaled to millions of people, and to enable widespread identification of IR," the authors wrote.
"We assembled a large cohort (n=1,165) with a combined set of data from wearable devices, along with demographics and blood biomarkers, and a ground-truth measure of IR," they said.
The team also developed a large language model called 'IR agent' that combines the assessment model's results with lifestyle and biomarker data to provide holistic insights into one's metabolic health and diabetes risk, and offers personalized recommendations.
"This work establishes a scalable, accessible framework for early detection of metabolic risk, which could enable timely lifestyle interventions to prevent progression to type-2 diabetes," the authors said.
In a 'News and Views' article published in the Nature journal, Christopher M Hartshorn from the US' National Institutes of Health (NIH), who was not involved in the study, said that rather than a snapshot, this study offers "something closer to a 'movie' of (one's) metabolic health".
Continuously collected data by smartwatches can capture fluctuations in activity, sleep, and heart function over time that reflect cumulative demands of metabolic regulation, he said.
"By drawing on continuous signals from daily life, the authors' approach highlights physiological strain that is invisible to episodic testing," Hartshorn said.
Identifying insulin resistance -- a key sign of diabetes -- could possibly enable simpler interventions and, ultimately, reduce the downstream burden of metabolic disease, the author said.