König Sebastian, Hohenstein Sven, Nitsche Anne, Pellissier Vincent, Leiner Johannes, Stellmacher Lars, Hindricks Gerhard, Bollmann Andreas
Department of Electrophysiology, Heart Center Leipzig at University of Leipzig, Strümpellstr. 39, 04289 Leipzig, Germany.
Helios Health Institute, Real World Evidence & Health Technology Assessment, Schwanebecker Chaussee 50, 13125 Berlin, Germany.
Eur Heart J Digit Health. 2023 Dec 20;5(2):144-151. doi: 10.1093/ehjdh/ztad081. eCollection 2024 Mar.
The diagnostic application of artificial intelligence (AI)-based models to detect cardiovascular diseases from electrocardiograms (ECGs) evolves, and promising results were reported. However, external validation is not available for all published algorithms. The aim of this study was to validate an existing algorithm for the detection of left ventricular systolic dysfunction (LVSD) from 12-lead ECGs.
Patients with digitalized data pairs of 12-lead ECGs and echocardiography (at intervals of ≤7 days) were retrospectively selected from the Heart Center Leipzig ECG and electronic medical records databases. A previously developed AI-based model was applied to ECGs and calculated probabilities for LVSD. The area under the receiver operating characteristic curve (AUROC) was computed overall and in cohorts stratified for baseline and ECG characteristics. Repeated echocardiography studies recorded ≥3 months after index diagnostics were used for follow-up (FU) analysis. At baseline, 42 291 ECG-echocardiography pairs were analysed, and AUROC for LVSD detection was 0.88. Sensitivity and specificity were 82% and 77% for the optimal LVSD probability cut-off based on Youden's J. AUROCs were lower in ECG subgroups with tachycardia, atrial fibrillation, and wide QRS complexes. In patients without LVSD at baseline and available FU, model-generated high probability for LVSD was associated with a four-fold increased risk of developing LVSD during FU.
We provide the external validation of an existing AI-based ECG-analysing model for the detection of LVSD with robust performance metrics. The association of false positive LVSD screenings at baseline with a deterioration of ventricular function during FU deserves a further evaluation in prospective trials.
基于人工智能（AI）的模型用于从心电图（ECG）检测心血管疾病的诊断应用不断发展，已有一些 promising 结果报道。然而，并非所有已发表的算法都有外部验证。本研究的目的是验证一种现有的从 12 导联 ECG 检测左心室收缩功能障碍（LVSD）的算法。
从莱比锡心脏中心 ECG 和电子病历数据库中回顾性选择有数字化 12 导联 ECG 和超声心动图数据对（间隔≤7 天）的患者。将先前开发的基于 AI 的模型应用于 ECG 并计算 LVSD 的概率。在总体以及根据基线和 ECG 特征分层的队列中计算受试者工作特征曲线下面积（AUROC）。在索引诊断后≥3 个月记录的重复超声心动图研究用于随访（FU）分析。在基线时，分析了 42291 对 ECG - 超声心动图数据对，LVSD 检测的 AUROC 为 0.88。基于约登指数（Youden's J）的最佳 LVSD 概率截断值的敏感性和特异性分别为 82%和 77%。在伴有心动过速、心房颤动和宽 QRS 波群的 ECG 亚组中，AUROC 较低。在基线时无 LVSD 且有可用 FU 的患者中，模型生成的 LVSD 高概率与 FU 期间发生 LVSD 的风险增加四倍相关。
我们对现有的基于 AI 的 ECG 分析模型进行了外部验证，以检测 LVSD，其性能指标稳健。基线时 LVSD 假阳性筛查与 FU 期间心室功能恶化之间的关联值得在前瞻性试验中进一步评估。
