<ul><li>The global prevalence of diabetes, particularly type 2 diabetes mellitus (T2DM), is rapidly increasing and poses significant health and economic challenges.</li><li>T2DM not only disrupts blood glucose regulation but also damages vital organs, leading to substantial morbidity and mortality.</li><li>A novel unsupervised framework using Non-negative Matrix Factorization (NMF) and statistical techniques has been proposed to identify individuals at risk of developing T2DM.</li><li>This method leverages data-driven insights from comorbidity and medication usage to estimate T2DM risk in undiagnosed individuals, offering an interpretable and scalable solution for timely interventions.</li></ul>

Unsupervised Latent Pattern Analysis for Estimating Type 2 Diabetes Risk in Undiagnosed Populations

Discover more