Cardiovascular Division and Cardiovascular Institute, Stanford University, CA, USA; CoMMLab and Electronic Engineering Department, Universitat de Valencia, VA, Spain.
Cardiovascular Division and Cardiovascular Institute, Stanford University, CA, USA.
Comput Biol Med. 2022 Jun;145:105451. doi: 10.1016/j.compbiomed.2022.105451. Epub 2022 Apr 1.
Automatic detection of atrial fibrillation (AF) by cardiac devices is increasingly common yet suboptimally groups AF, flutter or tachycardia (AT) together as 'high rate events'. This may delay or misdirect therapy.
We hypothesized that deep learning (DL) can accurately classify AF from AT by revealing electrogram (EGM) signatures.
We studied 86 patients in whom the diagnosis of AF or AT was established at electrophysiological study (25 female, 65 ± 11 years). Custom DL architectures were trained to identify AF using N = 29,340 unipolar and N = 23,760 bipolar EGM segments. We compared DL to traditional classifiers based on rate or regularity. We explained DL using computer models to assess the impact of controlled variations in shape, rate and timing on AF/AT classification in 246,067 EGMs reconstructed from clinical data.
DL identified AF with AUC of 0.97 ± 0.04 (unipolar) and 0.92 ± 0.09 (bipolar). Rule-based classifiers misclassified ∼10-12% of cases. DL classification was explained by regularity in EGM shape (13%) or timing (26%), and rate (60%; p < 0.001), and also by a set of unipolar EGM shapes that classified as AF independent of rate or regularity. Overall, the optimal AF 'fingerprint' comprised these specific EGM shapes, >15% timing variation, <0.48 correlation in beat-to-beat EGM shapes and CL < 190 ms (p < 0.001).
Deep learning of intracardiac EGMs can identify AF or AT via signatures of rate, regularity in timing or shape, and specific EGM shapes. Future work should examine if these signatures differ between different clinical subpopulations with AF.
心脏设备自动检测心房颤动（AF）的情况越来越常见，但将 AF、扑动或心动过速（AT）组合为“高心率事件”的方式并不理想，这可能会延迟或误导治疗。
我们假设深度学习（DL）可以通过揭示心电图（EGM）特征来准确地对 AF 与 AT 进行分类。
我们研究了 86 名在电生理研究中确诊为 AF 或 AT 的患者（女性 25 名，65±11 岁）。使用 N=29340 个单极和 N=23760 个双极 EGM 段，训练定制的 DL 架构来识别 AF。我们将 DL 与基于速率或规则性的传统分类器进行了比较。我们使用计算机模型来解释 DL，以评估在 246067 个从临床数据重建的 EGM 中，形状、速率和时间的受控变化对 AF/AT 分类的影响。
DL 对单极和双极 EGM 的 AF 识别的 AUC 分别为 0.97±0.04 和 0.92±0.09。基于规则的分类器将约 10-12%的病例分类错误。DL 分类可由 EGM 形状（13%）或时间（26%）、速率（60%；p<0.001）的规则性来解释，也可由一组与速率或规则性无关但可分类为 AF 的单极 EGM 形状来解释。总的来说，最佳的 AF“指纹”包括这些特定的 EGM 形状、＞15%的时间变化、＜0.48 的连续 EGM 形状之间的相关性以及 CL＜190ms（p＜0.001）。
通过对心内 EGM 的速率、时间或形状规则性以及特定 EGM 形状的特征进行深度学习，可识别 AF 或 AT。未来的工作应该检查这些特征是否在不同的 AF 临床亚群之间存在差异。
