Abstract 4144205: Diagnostic accuracy of artificial intelligence-based electrocardiogram algorithm to predict heart failure with reduced ejection fraction: a systematic review and meta-analysis
INTRODUCTION: AI-based EKG has shown good accuracy for diagnosing heart failure. However, due to the heterogeneity of studies regarding cutoff points, its precision for specifically detecting heart failure with reduced ejection fraction (LVEF <40%) is not yet well established. Research question: What it is the sensitivity and specificity of artificial-based electrocardiogram to diagnose heart failure with low ejection fraction(cut-off of 40%) Aims: We conducted a meta-analysis and systematic review to evaluate the accuracy of artificial intelligence electrocardiograms in predicting an ejection fraction below 40%. METHODS: We searched PubMed, Embase and Cochrane Library for studies evaluating the performance of AI EKGs in diagnosing HFrEF. We computed true positives, true negatives, false positives and false negatives events to estimate pooled sensitivity, specificity and area under the curve, using R software version 4.3.1, under a random-effects model. RESULTS: We identified 8 studies,including 136151 patients with a paired artificial intelligence enabled electrocardiogram with an echocardiography. 9349(6.8%) patients had an ejection fraction below 40% according to the echocardiogram. The AI-ECG data yielded areas under the receiver operator of, sensitivity of 0.861(0.815-0.897), and specificity of 0.874(0.834-0.905) , and area under the curve of 0.929(0.876-0.949). Mean/median age ranged from 60±9 to 67.8±14.4 years. CONCLUSIONS: In this systematic review and meta-analysis, the use of electrocardiogram-based artificial intelligence models demonstrated high sensitivity and specificity for the diagnosis of heart failure with an ejection fraction below 40%