Prediction of the Presence of Ventricular Fibrillation From a Brugada Electrocardiogram Using Artificial Intelligence

Prediction of the Presence of Ventricular Fibrillation From a Brugada Electrocardiogram Using Artificial Intelligence
复制标题

DOI:
10.1253/circj.cj-22-0496
复制
发表时间:
2023-07-01
影响因子:
3.3
通讯作者:
Sasano, Tetsuo
Sasano, Tetsuo
中科院分区:
医学3区
文献类型:
--
作者:
Nakamura, Tomofumi;Aiba, Takeshi;Sasano, Tetsuo

文献摘要

被引文献

相似文献

背景:Brugada 综合征是心源性猝死 (SCD) 的潜在原因,其特征是独特的心电图,但并非所有 A Brugada 心电图患者都会发生 SCD。在这项研究中,我们试图检查人工智能 (AI) 模型是否可以通过 Brugada 心电图预测之前或未来的心室颤动 (VF) 发作。方法和结果:我们使用卷积神经网络开发了一种支持人工智能的算法。从 157 名疑似 Brugada 综合征患者中获得 2,053 份心电图,并将数据集分为 5 个数据集进行交叉验证。在基于心电图的评估中,精确度、召回率和 F1 分数分别为 0.79 & PLUSMN;0.09、0.73 & PLUSMN;0.09 和 0.75 & PLUSMN;0.09。受试者工作特征曲线下的平均面积 (AUROC) 为 0.81 & PLUSMN; 0.09。根据每位患者的评估,AUROC 为 0.80 & PLUSMN;0.07。该模型预测 VF 存在的精度为 0.93 & PLUSMN;0.02,召回率为 0.77 & PLUSMN;0.14,F1 分数为 0.81 & PLUSMN;0.11。阴性预测值为0.94±0.11,阳性预测值为0.44±0.29。结论:这项概念验证研究表明,人工智能算法可以预测心室颤动的存在,并具有良好的性能。这意味着人工智能模型可以检测到人类无法检测到的微妙心电图变化。
Background: Brugada syndrome is a potential cause of sudden cardiac death (SCD) and is characterized by a distinct ECG, but not all patients with A Brugada ECG develop SCD. In this study we sought to examine if an artificial intelligence (AI) model can predict a previous or future ventricular fibrillation (VF) episode from a Brugada ECG.Methods and Results: We developed an AI-enabled algorithm using a convolutional neural network. From 157 patients with sus-pected Brugada syndrome, 2,053 ECGs were obtained, and the dataset was divided into 5 datasets for cross-validation. In the ECG-based evaluation, the precision, recall, and F1 score were 0.79 & PLUSMN;0.09, 0.73 & PLUSMN;0.09, and 0.75 & PLUSMN;0.09, respectively. The average area under the receiver-operating characteristic curve (AUROC) was 0.81 & PLUSMN; 0.09. On per-patient evaluation, the AUROC was 0.80 & PLUSMN;0.07. This model predicted the presence of VF with a precision of 0.93 & PLUSMN;0.02, recall of 0.77 & PLUSMN;0.14, and F1 score of 0.81 & PLUSMN;0.11. The negative predictive value was 0.94 & PLUSMN;0.11 while its positive predictive value was 0.44 & PLUSMN;0.29. Conclusions: This proof-of-concept study showed that an AI-enabled algorithm can predict the presence of VF with a substantial performance. It implies that the AI model may detect a subtle ECG change that is undetectable by humans.