AI-based ECG algorithms detect HFpEF with 85% sensitivity, 82% specificity, and 0.88 AUC, demonstrating high accuracy for diagnosis.
Does artificial intelligence-based electrocardiogram (AI-ECG) accurately diagnose heart failure with preserved ejection fraction (HFpEF)?
AI-ECG algorithms demonstrate high diagnostic accuracy (AUC 0.88) for detecting HFpEF, highlighting their potential as an accessible screening tool.
Absolute Event Rate: 0% vs 0%
Abstract Background Heart failure with preserved ejection fraction (HFpEF) is a highly prevalent syndrome associated with substantial morbidity and mortality. Accurate diagnosis is often challenging and costly, requiring special imaging techniques and clinical expertise, which may lead to delayed treatment. Recently, artificial intelligence-based electrocardiogram (AI-ECG) algorithms have emerged to make the diagnosis of HFpEF less expensive and more accessible. However there is a lack of data regarding their overall accuracy of diagnosis. Purpose We aim to conduct a systematic review and meta-analysis to evaluate the diagnostic performance of AI-ECG algorithms in detecting heart failure with HFpEF. Methods We searched PubMed, Embase, and Cochrane databases for studies evaluating AI-ECGs in diagnosing HFpEF. We extracted true positives, true negatives, false positives, and false negatives events to estimate pooled sensitivity, specificity, and area under the curve. Statistical analysis was performed using RStudio version 4.2.2 under a random-effects model. Heterogeneity was assessed with I² statistics. Results We included 4 studies comprising 172,275 patients (66.53% with HFpEF), in which an artificial intelligence-based electrocardiogram was evaluated alongside echocardiography, with a left ventricular ejection fraction of ≥50%. The patients’ mean age ranged from 56.7 ± 15 to 66 ± 10 years and 58.77% were male. The AI-based ECG demonstrated a sensitivity of 0.85 (95% CI: 0.61 to 0.95) - Fig. 1A, a specificity of 0.82 (95% CI: 0.71 to 0.89) - Fig. 1B, and an area under the receiver operating characteristic curve of 0.88 (95% CI: 0.85–0.92) - Fig. 2. Conclusion Our meta-analysis revealed that AI-ECG algorithms exhibit high sensitivity, specificity, and excellent accuracy in diagnosing HFpEF, highlighting their potential as an effective tool for routine screening and early diagnosis. Nevertheless, larger-scale prospective studies are necessary to confirm our observations and establish optimal practices for integrating this tool into various clinical settings.Sensitivity(A) and Specificity(B) of AI sROC curve of AI-ECG predicting HFpEF
David et al. (Sat,) reported a other. AI-based ECG algorithms detect HFpEF with 85% sensitivity, 82% specificity, and 0.88 AUC, demonstrating high accuracy for diagnosis.