Technical University of Denmark researchers develop AI tool for home diabetes risk assessment

Researchers from the Technical University of Denmark and Italian physician Umberto Maggiore have developed MEDWACS, an online AI-driven tool designed to estimate the risk of prediabetes and undiagnosed type 2 diabetes using seven personal data points.

Validated in a study published in the Journal of Clinical Epidemiology, the tool proved highly effective, performing as well as or better than conventional diagnostic methods. The research utilized three decades of health data from the National Health and Nutrition Examination Survey (NHANES) in the United States and underwent external validation in populations from the U.S. and South Korea.

Users can assess their risk at home using common items like a scale, measuring tape, and blood pressure monitor. The seven required parameters are age, sex, body mass index (BMI), waist circumference, systolic blood pressure, arm circumference, and thigh length.

Investigators noted that thigh length is a significant indicator, as it may reflect early childhood nutrition and muscle mass, both of which are crucial for blood glucose regulation. Postdoctoral researcher Daniel Yoo stated, “A test online could motivate those who need it to seek medical attention.”