Predictive characteristics of electrocardiogram for atrial fibrillation based on ECG-ViEW database
Zhang Fengzhi
Chen Shuning
Xu Zongzheng
Wang Yue
Li Yuan
Wang Deguo
Abstract：Objective To discuss the effect of surface electrocardiograms(ECG)characteristics in predicting atrial fibrillation(AF)by applying public ECG-ViEW database,and provide a basis for developing computer-assisted AF prediction.Methods ECG of sinus rhythm was screened in AF patients from ECG data in public ECG-ViEW database,and differences in ECG parameters were compared between AF patients and non-AF patients.The risk factors for AF in ECG parameters were analyzed by using Logistic regression analysis.The specificity and sensitivity of ECG parameters in predicting AF risk were determined by using receiver operating characteristic(ROC)curve analysis.Results Among 181 560 patients,there were 3431(1.9％)with AF(AF group),and there were 178129 without AF(non-AF group,98.1％).The heart rate was faster,and PR interval,QRS wave duration and QTc interval were significantly higher in AF group than those in non-AF group.Hodges-corrected QTc interval(95％CI:0.449～0.855,P=0.004),abnormal P wave axis(left deviation:95％CI:0.402～0.524,P＜0.001;right deviation:95％CI:0.630～0.786,P＜0.001),and widened QRS-T angle(95％CI:3.165～3.871,P＜0.001)were independent risk factors for AF.The results of ROC curve analysis showed that,after including age and gender and in combination of abnormal P wave axis,prolonged QTc-Hodges and widened QRS-T angle for predicting AF,sensitivity 0.705,and specificity was 0.686(AUC=0.739,95％CI:0.731～0.748,P＜0.001).Conclusion The ECG characteristics including abnormal P wave axis,prolonged QTc-Hodges and widened QRS-T angle are independent risk factors for AF,which can be used for AF risk screen in population.
Keywords：QT intervalAtrial fibrillationElectrocardiogramQTc interval
Publication Date：2025-02-20
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：5( 141-144,167 )