Fujiki Goro, Kodera Satoshi, Setoguchi Naoto, Tanabe Kengo, Miyaji Kotaro, Kushida Shunichi, Saji Mike, Nanasato Mamoru, Maki Hisataka, Fujita Hideo, Kato Nahoko, Watanabe Hiroyuki, Suzuki Minami, Takahashi Masao, Sawada Naoko, Ando Jiro, Sato Masataka, Sawano Shinnosuke, Katsushika Susumu, Shinohara Hiroki, Takeda Norifumi, Fujiu Katsuhito, Akazawa Hiroshi, Morita Hiroyuki, Komuro Issei
Department of Cardiovascular Medicine, The University of Tokyo Hospital, Tokyo, Japan.
Division of Cardiology, Mitsui Memorial Hospital, Tokyo, Japan.
JACC Asia. 2024 Dec 10;5(1):88-98. doi: 10.1016/j.jacasi.2024.10.012. eCollection 2025 Jan.
Heart failure should be diagnosed as early as possible. Although deep learning models can predict one or more echocardiographic findings from electrocardiograms (ECGs), such analyses are not comprehensive.
This study aimed to develop a deep learning model for comprehensive prediction of echocardiographic findings from ECGs.
We obtained 229,439 paired ECG and echocardiography data sets from 8 centers. Six centers contributed to model development and 2 to external validation. We identified 12 echocardiographic findings related to left-sided cardiac abnormalities, valvular heart diseases, and right-sided cardiac abnormalities. These findings were predicted using convolutional neural networks, and a composite label was analyzed using logistic regression. A positive composite label indicated positivity in any of the 12 findings.
For the composite findings label, the area under the receiver-operating characteristic curve was 0.80 (95% CI: 0.80-0.81) on hold-out validation and 0.78 (95% CI: 0.78-0.79) on external validation. The composite findings label applying logistic regression had an area under the receiver-operating characteristic curve of 0.80 (95% CI: 0.80-0.81) with accuracy of 73.8% (95% CI: 73.2-74.4), sensitivity of 81.1% (95% CI: 80.5-81.8), and specificity of 60.7% (95% CI: 59.6-61.8).
We have developed convolutional neural network models that predict a wide range of echocardiographic findings, including left-sided cardiac abnormalities, valvular heart diseases, and right-sided cardiac abnormalities from ECGs and created a model to predict a composite findings label by logistic regression analysis. This model has potential to serve as an adjunct for early diagnosis and treatment of previously undetected cardiac disease.
心力衰竭应尽早诊断。虽然深度学习模型可以从心电图（ECG）预测一项或多项超声心动图检查结果，但此类分析并不全面。
本研究旨在开发一种深度学习模型，用于从心电图全面预测超声心动图检查结果。
我们从8个中心获取了229439对心电图和超声心动图数据集。6个中心参与模型开发，2个中心参与外部验证。我们确定了12项与左心异常、心脏瓣膜病和右心异常相关的超声心动图检查结果。使用卷积神经网络预测这些结果，并通过逻辑回归分析复合标签。复合标签呈阳性表明12项结果中的任何一项呈阳性。
对于复合结果标签，在保留验证中，受试者工作特征曲线下面积为0.80（95%CI：0.80 - 0.81），在外部验证中为0.78（95%CI：0.78 - 0.79）。应用逻辑回归的复合结果标签在受试者工作特征曲线下面积为0.80（95%CI：0.80 - 0.81），准确率为73.8%（95%CI：73.2 - 74.4），灵敏度为81.1%（95%CI：80.5 - 81.8），特异性为60.7%（95%CI：59.6 - 61.8）。
我们开发了卷积神经网络模型，可从心电图预测广泛的超声心动图检查结果，包括左心异常、心脏瓣膜病和右心异常，并通过逻辑回归分析创建了一个预测复合结果标签的模型。该模型有潜力作为早期诊断和治疗先前未被发现的心脏病的辅助手段。
