Abstract：Objective To assess the accuracy of wearable electrocardiogram ( ECG) monitoring device for detecting arrhythmias ( sinus rhythm, premature ventricular contraction premature atrial contraction, and atrial fibrillation), and heart rate in a clinical. Methods A total of 321 volunteers (875 ECG records) participated in this study, including 146 males ( 531 ECG records) and 175 females ( 344 ECG records). A wearable ECG monitoring device was employed to record ECGs for detecting arrhythmias and heart rate. The detection results were compared with physicians clinical diagnostic results based on 12-lead ECG machine (regraded as the gold standard for diagnosing arrhythmias and detecting heart rate). Accuracy and consistency evaluation of the two detection methods was conducted in all the subjects, including accuracy, sensitivity, specificity, F1 score, area under curve (AUC) of the receiver operating characteristic curve, Bland-Altman consistency analysis, mean absolute error, and correlogram. Results Regarding arrhythmia identification, wearable ECG monitoring device exhibits impressive performance metrics, with an AUC of 0. 98, accuracy of 0. 98, sensitivity of 0. 96, specificity of 0. 99, and an F1 score of 0. 96. In terms of heart rate monitoring, a robust correlation between the wearable ECG monitoring equipment and the ECG machine is evident, with a substantial total correlation coefficient of 0. 84 (P< 0. 01). Bland-Altman consistency analysis reveals that the two methods display a mean difference of - 0. 97 times per minute, a mean standard deviation of 8. 02 times per minute, and a 95% confidence interval ranging from -16. 69 to 14. 75 within an acceptable level. Conclusion Wearable ECG monitoring device can accurately detect arrhythmias and heart rate.
