The Effectiveness of a Deep Learning Model to Detect Left Ventricular Systolic Dysfunction from Electrocardiograms
The Effectiveness of a Deep Learning Model to Detect Left Ventricular Systolic Dysfunction from Electrocardiograms
复制标题
DOI：
10.1536/ihj.21-407
复制
发表时间：
2021-11-01
影响因子：
1.5
通讯作者：
Komuro, Issei
中科院分区：
文献类型：
--
作者：
Katsushika, Susumu;Kodera, Satoshi;Komuro, Issei
深度学习模型可以应用于心电图（ECGs）来检测左心室（LV）功能障碍。我们假设应用深度学习模型可以提高心内科医生从心电图预测左室功能障碍的诊断准确性。我们获得了2015年1月至2019年12月期间接受超声心动图检查的患者的37,103对心电图和超声心动图数据记录。我们训练了一个卷积神经网络，使用23,801个心电图数据集来识别左室功能障碍（射血分数< 40%）患者的数据记录。在独立的7,196张心电图上进行测试时，我们发现受试者工作特征曲线下的面积为0.945（95%置信区间：0.936-0.954）。当7名心脏病学家从7196张心电图的测试数据集中随机选择50张心电图时，他们预测左室功能障碍的准确率为78.0%±6.0%。参考模型的输出，心脏病医生的准确率提高到88.0% +/- 3.7%，表明模型支持显著提高了心脏病医生的诊断准确率（P = 0.02）。灵敏度图显示，该模型在检测心电图上的左室功能障碍时主要关注QRS复合体。我们开发了一种深度学习模型，可以高精度地检测心电图上的左室功能障碍。此外，我们证明了深度学习模型的支持可以帮助心脏科医生识别心电图上的左室功能障碍。
Deep learning models can be applied to electrocardiograms (ECGs) to detect left ventricular (LV) dysfunc-tion. We hypothesized that applying a deep learning model may improve the diagnostic accuracy of cardiolo-gists in predicting LV dysfunction from ECGs. We acquired 37,103 paired ECG and echocardiography data re-cords of patients who underwent echocardiography between January 2015 and December 2019. We trained a convolutional neural network to identify the data records of patients with LV dysfunction (ejection fraction < 40%) using a dataset of 23,801 ECGs. When tested on an independent set of 7,196 ECGs, we found the area under the receiver operating characteristic curve was 0.945 (95% confidence interval: 0.936-0.954). When 7 car-diologists interpreted 50 randomly selected ECGs from the test dataset of 7,196 ECGs, their accuracy for pre-dicting LV dysfunction was 78.0% +/- 6.0%. By referring to the model's output, the cardiologist accuracy im -proved to 88.0% +/- 3.7%, which indicates that model support significantly improved the cardiologist diagnostic accuracy (P = 0.02). A sensitivity map demonstrated that the model focused on the QRS complex when detect-ing LV dysfunction on ECGs. We developed a deep learning model that can detect LV dysfunction on ECGs with high accuracy. Furthermore, we demonstrated that support from a deep learning model can help cardiolo-gists to identify LV dysfunction on ECGs.