Department of Medicine (M.I.A., A.J.R., J.A.B.Z., T.B., P.C., P.J.W., S.M.N.), Stanford University.
Department of Computer Science (F.A., P.B., M.Z.), Stanford University.
Circ Arrhythm Electrophysiol. 2020 Aug;13(8):e008160. doi: 10.1161/CIRCEP.119.008160. Epub 2020 Jul 6.
Advances in ablation for atrial fibrillation (AF) continue to be hindered by ambiguities in mapping, even between experts. We hypothesized that convolutional neural networks (CNN) may enable objective analysis of intracardiac activation in AF, which could be applied clinically if CNN classifications could also be explained.
We performed panoramic recording of bi-atrial electrical signals in AF. We used the Hilbert-transform to produce 175 000 image grids in 35 patients, labeled for rotational activation by experts who showed consistency but with variability (kappa [κ]=0.79). In each patient, ablation terminated AF. A CNN was developed and trained on 100 000 AF image grids, validated on 25 000 grids, then tested on a separate 50 000 grids.
In the separate test cohort (50 000 grids), CNN reproducibly classified AF image grids into those with/without rotational sites with 95.0% accuracy (CI, 94.8%-95.2%). This accuracy exceeded that of support vector machines, traditional linear discriminant, and k-nearest neighbor statistical analyses. To probe the CNN, we applied gradient-weighted class activation mapping which revealed that the decision logic closely mimicked rules used by experts (C statistic 0.96).
CNNs improved the classification of intracardiac AF maps compared with other analyses and agreed with expert evaluation. Novel explainability analyses revealed that the CNN operated using a decision logic similar to rules used by experts, even though these rules were not provided in training. We thus describe a scaleable platform for robust comparisons of complex AF data from multiple systems, which may provide immediate clinical utility to guide ablation. Registration: URL: https://www.clinicaltrials.gov; Unique identifier: NCT02997254. Graphic Abstract: A graphic abstract is available for this article.
尽管专家之间存在共识，但心房颤动（AF）消融领域的进展仍因标测的不明确性而受阻。我们假设卷积神经网络（CNN）可实现对 AF 心内激活的客观分析，如果 CNN 分类也可解释，那么该方法即可应用于临床。
我们对 AF 中的双心房电信号进行全景记录。我们使用希尔伯特变换在 35 例患者中产生了 175000 个图像网格，这些网格由专家进行旋转激活标记，专家们的一致性较好（κ[κ]=0.79），但存在一定变异性。在每个患者中，消融均终止了 AF。我们在 100000 个 AF 图像网格上开发和训练了一个 CNN，并在 25000 个网格上进行了验证，然后在另外 50000 个网格上进行了测试。
在独立的测试队列（50000 个网格）中，CNN 可重复性地以 95.0%的准确率（置信区间，94.8%-95.2%）将 AF 图像网格分类为存在/不存在旋转部位。该准确率超过了支持向量机、传统线性判别和 k-最近邻统计分析。为了探究 CNN，我们应用了梯度加权类激活映射，结果表明，决策逻辑与专家使用的规则非常相似（C 统计量为 0.96）。
CNN 提高了心内 AF 图的分类准确率，优于其他分析方法，与专家评估结果一致。新颖的可解释性分析表明，尽管在训练过程中未提供这些规则，但 CNN 是使用与专家使用的规则相似的决策逻辑进行操作的。因此，我们描述了一种可扩展的平台，用于对来自多个系统的复杂 AF 数据进行稳健比较，这可能为指导消融提供即时的临床实用价值。注册：网址：https://www.clinicaltrials.gov；唯一标识符：NCT02997254。图表摘要：本文提供了图表摘要。
