Rogers Albert J, Bhatia Neal K, Bandyopadhyay Sabyasachi, Tooley James, Ansari Rayan, Thakkar Vyom, Xu Justin, Soto Jessica Torres, Tung Jagteshwar S, Alhusseini Mahmood I, Clopton Paul, Sameni Reza, Clifford Gari D, Hughes J Weston, Ashley Euan A, Perez Marco V, Zaharia Matei, Narayan Sanjiv M
Department of Medicine, Stanford University School of Medicine, Stanford, CA, USA.
Cardiovascular Institute, Stanford University School of Medicine, Stanford, CA, USA.
NPJ Digit Med. 2025 Jan 11;8(1):21. doi: 10.1038/s41746-024-01407-y.
Cardiac wall motion abnormalities (WMA) are strong predictors of mortality, but current screening methods using Q waves from electrocardiograms (ECGs) have limited accuracy and vary across racial and ethnic groups. This study aimed to identify novel ECG features using deep learning to enhance WMA detection, referencing echocardiography as the gold standard. We collected ECG and echocardiogram data from 35,210 patients in California and labeled WMA using unstructured language parsing of echocardiographic reports. A deep neural network (ECG-WMA-Net) was trained and outperformed both expert ECG interpretation and Q-wave indices, achieving an AUROC of 0.781 (CI: 0.762-0.799). The model was externally validated in a diverse cohort from Georgia (n = 2338), with an AUC of 0.723 (CI: 0.685-0.757). Explainability analysis revealed significant contributions from QRS and T-wave regions. This deep learning approach improves WMA screening accuracy, potentially addressing physiological differences not captured by standard ECG-based methods.
心脏壁运动异常（WMA）是死亡率的有力预测指标，但目前使用心电图（ECG）Q波的筛查方法准确性有限，且在不同种族和族裔群体中存在差异。本研究旨在利用深度学习识别新的心电图特征，以加强WMA检测，并将超声心动图作为金标准进行参考。我们收集了加利福尼亚州35210名患者的心电图和超声心动图数据，并通过对超声心动图报告进行非结构化语言解析来标记WMA。训练了一个深度神经网络（ECG-WMA-Net），其表现优于专家心电图解读和Q波指标，曲线下面积（AUROC）达到0.781（置信区间：0.762 - 0.799）。该模型在佐治亚州的一个多样化队列（n = 2338）中进行了外部验证，AUC为0.723（置信区间：0.685 - 0.757）。可解释性分析揭示了QRS和T波区域的显著贡献。这种深度学习方法提高了WMA筛查的准确性，有可能解决基于标准心电图方法未捕捉到的生理差异问题。
