Deep Learning Algorithm for Management of Diabetes Mellitus via Electrocardiogram-Based Glycated Hemoglobin (ECG-HbA1c): A Retrospective Cohort Study

Abstract

Background: glycated hemoglobin (HbA1c) provides information on diabetes mellitus (DM) management. Electrocardiography (ECG) is a noninvasive test of cardiac activity that has been determined to be related to DM and its complications. This study developed a deep learning model (DLM) to estimate HbA1c via ECG. Methods: there were 104,823 ECGs with corresponding HbA1c or fasting glucose which were utilized to train a DLM for calculating ECG-HbA1c. Next, 1539 cases from outpatient departments and health examination centers provided 2190 ECGs for initial validation, and another 3293 cases with their first ECGs were employed to analyze its contributions to DM management. The primary analysis was used to distinguish patients with and without mild to severe DM, and the secondary analysis was to explore the predictive value of ECG-HbA1c for future complications, which included all-cause mortality, new-onset chronic kidney disease (CKD), and new-onset heart failure (HF). Results: we used a gender/age-matching strategy to train a DLM to achieve the best AUCs of 0.8255 with a sensitivity of 71.9% and specificity of 77.7% in a follow-up cohort with correlation of 0.496 and mean absolute errors of 1.230. The stratified analysis shows that DM presented in patients with fewer comorbidities was significantly more likely to be detected by ECG-HbA1c. Patients with higher ECG-HbA1c under the same Lab-HbA1c exhibited worse physical conditions. Of interest, ECG-HbA1c may contribute to the mortality (gender/age adjusted hazard ratio (HR): 1.53, 95% conference interval (CI): 1.08–2.17), new-onset CKD (HR: 1.56, 95% CI: 1.30–1.87), and new-onset HF (HR: 1.51, 95% CI: 1.13–2.01) independently of Lab-HbA1c. An additional impact of ECG-HbA1c on the risk of all-cause mortality (C-index: 0.831 to 0.835, p < 0.05), new-onset CKD (C-index: 0.735 to 0.745, p < 0.01), and new-onset HF (C-index: 0.793 to 0.796, p < 0.05) were observed in full adjustment models. Conclusion: the ECG-HbA1c could be considered as a novel biomarker for screening DM and predicting the progression of DM and its complications.

Electrocardiogram-Based Heart Age Estimation by a Deep Learning Model Provides More Information on the Incidence of Cardiovascular Disorders

“…We used the corresponding International Classification of Diseases, Ninth Revision and Tenth Revision (ICD-9 and ICD-10) to define certain CVD-related outcomes. The detail codes were described as previously study ( 10 , 13 , 20 , 27 ). For each CVD-related outcome, patients with ICD-9 or ICD-10 with corresponding diagnosis codes before the first-exam ECG collected in the physical examination center were excluded.…”

“…The presented performance in the validation cohort was only evaluated once, and the estimated ECG-age in the follow-up cohort was used to predict the CVD-related outcomes. This system is configured to visualize the basis for the AI predictions using class activation mappings (CAMs) and attention mechanism ( 10 , 11 , 13 , 20 , 29 ).…”

“…A deep learning model (DLM) is a technique used to learn useful features and provide an opportunity to speed up the process of converting unstructured data for analysis, which can also provide better accuracy in ECG interpretation ( 7 ). Previous studies have also developed a series of ECG-based DLMs on arrhythmia ( 8 ), acute myocardial infarction ( 9 , 10 ), aortic dissection ( 11 ), dyskalemia ( 12 – 14 ), left ventricular dysfunction ( 15 , 16 ), mitral regurgitation ( 17 ), aortic stenosis ( 18 ), glycemic profile ( 19 , 20 ), etc. Moreover, the ECGs can even be used to predict the atrial fibrillation after a month ( 21 ).…”

Artificial Intelligence-Enabled Electrocardiography Predicts Left Ventricular Dysfunction and Future Cardiovascular Outcomes: A Retrospective Analysis

“…This strategy was to ensure a balanced gender distribution in each batch and additionally consider the weight of age computed on the prevalence of 20 equidistant intervals in the development cohort. We trained the above three DLMs to compare their performance, similar to a previous study [ 31 ]. The network with the highest Pearson’s correlation (r) between estimated ECG-EF and ECHO-EF in the validation cohort was used.…”

“…With the rapid progression of deep learning models (DLMs) on ECG [ 19 ], these models have expanded to multiple applications and achieved human-level performance, effectively detecting cardiac diseases with large annotated ECG datasets, including arrhythmia detection [ 20 ], dyskalemia [ 21 , 22 , 23 ], myocardial infarction [ 24 , 25 , 26 ], aortic dissection [ 27 ], thyrotoxic periodic paralysis [ 28 ], and digoxin toxicity [ 29 ]. Interestingly, current studies have started to use DLM to interpret chronic changes in ECGs, such as anemia [ 30 ], diabetes [ 31 ], conduction abnormality [ 32 ], future atrial fibrillation [ 33 ], and mortality prediction [ 34 ].…”