{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"submitter":["Koga D"],"funding":["Moonshot Research and Development Program"],"pagination":["415"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC12604153"],"repository":["biostudies-literature"],"omics_type":["Unknown"],"volume":["24(1)"],"pubmed_abstract":["<h4>Background</h4>Early detection of prediabetes is crucial for diabetes prevention, yet it remains challenging due to its asymptomatic nature and low screening rates. This study aimed to develop and rigorously validate artificial intelligence (AI) models to identify individuals with prediabetes solely using electrocardiograms (ECGs).<h4>Methods</h4>We defined prediabetes/diabetes based on fasting plasma glucose ≥ 110 mg/dL, hemoglobin A<sub>1c</sub> ≥ 6.0%, or ongoing diabetes treatment. From a primary cohort of 16,766 health checkup records, 269 ECG features were extracted to develop a novel AI model. The final model was subsequently evaluated using an internal held-out test dataset and an independent external validation cohort (n = 2,456). SHAP (SHapley Additive exPlanations) was appli"],"journal":["Cardiovascular diabetology"],"pubmed_title":["Artificial intelligence identifies individuals with prediabetes using single-lead electrocardiograms."],"pmcid":["PMC12604153"],"funding_grant_id":["JPMJMS2023"],"pubmed_authors":["Horino M","Koga D","Yamada T","Komiya C","Shimizu H","Okazaki R","Aoki J","Ikeda K","Ohno S","Tsujimoto K","Takeuchi A","Katagiri H","Murakami M","Hara K","Kaneda R","Ishii R"],"additional_accession":[]},"is_claimable":false,"name":"Artificial intelligence identifies individuals with prediabetes using single-lead electrocardiograms.","description":"<h4>Background</h4>Early detection of prediabetes is crucial for diabetes prevention, yet it remains challenging due to its asymptomatic nature and low screening rates. This study aimed to develop and rigorously validate artificial intelligence (AI) models to identify individuals with prediabetes solely using electrocardiograms (ECGs).<h4>Methods</h4>We defined prediabetes/diabetes based on fasting plasma glucose ≥ 110 mg/dL, hemoglobin A<sub>1c</sub> ≥ 6.0%, or ongoing diabetes treatment. From a primary cohort of 16,766 health checkup records, 269 ECG features were extracted to develop a novel AI model. The final model was subsequently evaluated using an internal held-out test dataset and an independent external validation cohort (n = 2,456). SHAP (SHapley Additive exPlanations) was appli","dates":{"release":"2025-01-01T00:00:00Z","publication":"2025 Nov","modification":"2026-06-05T15:07:53.408Z","creation":"2026-05-18T03:13:07.344Z"},"accession":"S-EPMC12604153","cross_references":{"pubmed":["41214697"],"doi":["10.1186/s12933-025-02982-4"]}}