The highest AIRE-DM risk quartile predicted incident type 2 diabetes with hazard ratios of 4.75 in BIDMC, 7.52 in UK Biobank, and 3.96 in ELSA-Brasil compared to the lowest quartile.
Cohort (n=268,882)
Yes
Does an AI-enhanced electrocardiography model (AIRE-DM) predict incident type 2 diabetes and detect prevalent disease?
An AI-enhanced ECG model (AIRE-DM) can detect prevalent and predict incident type 2 diabetes, offering a potential tool for opportunistic cardiometabolic screening that performs comparably or superior to standard clinical risk scores and HbA1c.
Hazard Ratio: 4.75
Abstract Background A significant proportion of type 2 diabetes cases remain undiagnosed despite screening advances, carrying substantial cardiometabolic risk. Artificial intelligence-enhanced electrocardiography (AI-ECG) detects subtle ECG changes in subclinical disease, potentially enabling opportunistic screening. Methods We developed AIRE-DM, a convolutional neural network with discrete-time survival loss, for diagnosis of prevalent and prediction of incident type 2 diabetes. It was trained on 1,163,401 ECGs from 189,537 individuals from Beth Israel Deaconess Medical Center (BIDMC) and externally validated in UK Biobank (N = 65,606) and ELSA-Brasil (N = 13,739). Results AIRE-DM demonstrated moderate discrimination for prevalent type 2 diabetes (AUC: BIDMC 0.724, UK Biobank 0.733, ELSA-Brasil 0.706) and incident type 2 diabetes (C-index: BIDMC 0.667, UKB 0.688, ELSA-Brasil 0.625). The highest AIRE-DM risk quartile had elevated incident diabetes risk versus the lowest (HR: BIDMC 4.75, UKB 7.52, ELSA-Brasil 3.96). AIRE-DM was non-inferior to the ADA Diabetes Risk Test in BIDMC, with improved predictive accuracy when combined. In normoglycaemic patients, AIRE-DM was superior to HbA1c for predicting incident diabetes in BIDMC and non-inferior in ELSA-Brasil. The highest risk quartile reached 5% cumulative T2DM incidence 5.4 years (BIDMC) and 4.8 years (ELSA-Brasil) earlier than the lowest risk quartile, after adjusting for HbA1c, age and sex. Phenome- and genome-wide association studies revealed biologically plausible associations with glucose regulation, cardiac morphology, diastolic dysfunction, arterial stiffness, and lipid metabolism. Conclusion AIRE-DM detects prevalent type 2 diabetes and predicts incident disease, uniquely identifying high-risk individuals within the normoglycaemic range. Combined with clinical scores or biomarkers, it enhances risk stratification, enabling earlier intervention.
Pastika et al. (Tue,) conducted a cohort in Type 2 Diabetes Mellitus (n=268,882). AIRE-DM (Artificial intelligence-enhanced electrocardiography) vs. Lowest AIRE-DM risk quartile was evaluated on Incident type 2 diabetes (HR 4.75). The highest AIRE-DM risk quartile predicted incident type 2 diabetes with hazard ratios of 4.75 in BIDMC, 7.52 in UK Biobank, and 3.96 in ELSA-Brasil compared to the lowest quartile.