Medical Big Data Research Center, Beijing Key Laboratory for Precision Medicine of Chronic Heart Failure, Key Laboratory of Ministry of Industry and Information Technology of Biomedical Engineering and Translational Medicine, Chinese PLA General Hospital, Beijing, China.
BioMind Technology, Zhongguancun Medical Engineering Center, Beijing, China.
JACC Cardiovasc Imaging. 2022 Apr;15(4):551-563. doi: 10.1016/j.jcmg.2021.08.015. Epub 2021 Nov 17.
This study sought to develop a deep learning (DL) framework to automatically analyze echocardiographic videos for the presence of valvular heart diseases (VHDs).
Although advances in DL have been applied to the interpretation of echocardiograms, such techniques have not been reported for interpretation of color Doppler videos for diagnosing VHDs.
The authors developed a 3-stage DL framework for automatic screening of echocardiographic videos for mitral stenosis (MS), mitral regurgitation (MR), aortic stenosis (AS), and aortic regurgitation (AR) that classifies echocardiographic views, detects the presence of VHDs, and, when present, quantifies key metrics related to VHD severities. The algorithm was trained (n = 1,335), validated (n = 311), and tested (n = 434) using retrospectively selected studies from 5 hospitals. A prospectively collected set of 1,374 consecutive echocardiograms served as a real-world test data set.
Disease classification accuracy was high, with areas under the curve of 0.99 (95% CI: 0.97-0.99) for MS; 0.88 (95% CI: 0.86-0.90) for MR; 0.97 (95% CI: 0.95-0.99) for AS; and 0.90 (95% CI: 0.88-0.92) for AR in the prospective test data set. The limits of agreement (LOA) between the DL algorithm and physician estimates of metrics of valve lesion severities compared to the LOAs between 2 experienced physicians spanned from -0.60 to 0.77 cm vs -0.48 to 0.44 cm for MV area; from -0.27 to 0.25 vs -0.23 to 0.08 for MR jet area/left atrial area; from -0.86 to 0.52 m/s vs -0.48 to 0.54 m/s for peak aortic valve blood flow velocity (V); from -10.6 to 9.5 mm Hg vs -10.2 to 4.9 mm Hg for average peak aortic valve gradient; and from -0.39 to 0.32 vs -0.31 to 0.32 for AR jet width/left ventricular outflow tract diameter.
The proposed deep learning algorithm has the potential to automate and increase efficiency of the clinical workflow for screening echocardiographic images for the presence of VHDs and for quantifying metrics of disease severity.
本研究旨在开发一种深度学习（DL）框架，以自动分析超声心动图视频中是否存在瓣膜性心脏病（VHD）。
尽管 DL 的进步已应用于超声心动图的解释，但尚未有报道称其可用于解释彩色多普勒视频以诊断 VHD。
作者开发了一个 3 阶段的深度学习框架，用于自动筛选超声心动图视频，以筛查二尖瓣狭窄（MS）、二尖瓣反流（MR）、主动脉瓣狭窄（AS）和主动脉瓣反流（AR），该框架可对超声心动图视图进行分类、检测 VHD 的存在，以及在存在 VHD 时，量化与 VHD 严重程度相关的关键指标。该算法使用来自 5 家医院的回顾性研究进行了训练（n=1335）、验证（n=311）和测试（n=434）。一个前瞻性收集的 1374 例连续超声心动图数据集用作真实世界的测试数据集。
疾病分类准确率较高，在前瞻性测试数据集中，MS 的曲线下面积为 0.99（95%CI：0.97-0.99）；MR 为 0.88（95%CI：0.86-0.90）；AS 为 0.97（95%CI：0.95-0.99）；AR 为 0.90（95%CI：0.88-0.92）。与两名经验丰富的医生之间的 LOA 相比，DL 算法和医生对瓣膜病变严重程度的度量的 LOA 范围从 MV 面积的-0.60 到 0.77cm 到-0.48 到 0.44cm；从 MR 射流面积/左心房面积的-0.27 到 0.25 到-0.23 到 0.08；从峰值主动脉瓣血流速度（V）的-0.86 到 0.52m/s 到-0.48 到 0.54m/s；从平均峰值主动脉瓣梯度的-10.6 到 9.5mmHg 到-10.2 到 4.9mmHg；从 AR 射流宽度/左心室流出道直径的-0.39 到 0.32 到-0.31 到 0.32。
所提出的深度学习算法有可能实现自动化，并提高筛查超声心动图图像中 VHD 存在和量化疾病严重程度指标的临床工作流程的效率。
