2025/12/03 by Henriques, Ian, Lynda Elhassar, Sarvesh Relekar +23
Engineering · Health Professions · Medicine · #Artificial Intelligence in Healthcare #Artificial neural network #COVID-19 diagnosis using AI #Calibration #Computers and Society (cs.CY) #Diabetes mellitus #Disease #FOS: Computer and information sciences #Machine Learning (cs.LG) #Non-Invasive Vital Sign Monitoring #Scalability #Type 2 diabetes #Wearable computer #Wearable technology
paper · pdf · doi:10.48550/arxiv.2512.03471
published in arXiv (Cornell University) (Cornell University)
openalex publication_date 2025/12/03 · openalex created_date 2025/12/05 · openalex updated_date 2026/07/28
The global rise in type 2 diabetes underscores the need for scalable and cost-effective screening methods. Current diagnosis requires biochemical assays, which are invasive and costly. Advances in consumer wearables have enabled early explorations of machine learning-based disease detection, but prior studies were limited to controlled settings. We present SweetDeep, a compact neural network trained on physiological and demographic data from 285 (diabetic and non-diabetic) participants in the EU and MENA regions, collected using Samsung Galaxy Watch 7 devices in free-living conditions over six days. Each participant contributed multiple 2-minute sensor recordings per day, totaling approximately 20 recordings per individual. Despite comprising fewer than 3,000 parameters, SweetDeep achieves 82.5% patient-level accuracy (82.1% macro-F1, 79.7% sensitivity, 84.6% specificity) under three-fold cross-validation, with an expected calibration error of 5.5%. Allowing the model to abstain on less than 10% of low-confidence patient predictions yields an accuracy of 84.5% on the remaining patients. These findings demonstrate that combining engineered features with lightweight architectures can support accurate, rapid, and generalizable detection of type 2 diabetes in real-world wearable settings.