Masutani Evan M, Chandrupatla Rahul S, Wang Shuo, Zocchi Chiara, Hahn Lewis D, Horowitz Michael, Jacobs Kathleen, Kligerman Seth, Raimondi Francesca, Patel Amit, Hsiao Albert
From the Departments of Bioengineering (E.M.M.) and Radiology (R.S.C., L.D.H., M.H., K.J., S.K., A.H.), University of California, San Diego, 9300 Campus Point Dr, MC 0841, La Jolla, CA 92037-0841; Department of Medicine, University of Virginia, Charlottesville, Va (S.W., A.P.); and Meyer Children's Hospital IRCCS, Cardiac Imaging Unit, Pediatric Cardiology, University of Florence, Florence, Italy (C.Z., F.R.).
Radiol Cardiothorac Imaging. 2023 May 11;5(3):e220202. doi: 10.1148/ryct.220202. eCollection 2023 Jun.
To assess the feasibility of a newly developed algorithm, called (DLSS), to infer myocardial velocity from cine steady-state free precession (SSFP) images and detect wall motion abnormalities in patients with ischemic heart disease.
In this retrospective study, DLSS was developed by using a data set of 223 cardiac MRI examinations including cine SSFP images and four-dimensional flow velocity data (November 2017 to May 2021). To establish normal ranges, segmental strain was measured in 40 individuals (mean age, 41 years ± 17 [SD]; 30 men) without cardiac disease. Then, DLSS performance in the detection of wall motion abnormalities was assessed in a separate group of patients with coronary artery disease, and these findings were compared with consensus results of four independent cardiothoracic radiologists (ground truth). Algorithm performance was evaluated by using receiver operating characteristic curve analysis.
Median peak segmental radial strain in individuals with normal cardiac MRI findings was 38% (IQR: 30%-48%). Among patients with ischemic heart disease (846 segments in 53 patients; mean age, 61 years ± 12; 41 men), the Cohen κ among four cardiothoracic readers for detecting wall motion abnormalities was 0.60-0.78. DLSS achieved an area under the receiver operating characteristic curve of 0.90. Using a fixed 30% threshold for abnormal peak radial strain, the algorithm achieved a sensitivity, specificity, and accuracy of 86%, 85%, and 86%, respectively.
The deep learning algorithm had comparable performance with subspecialty radiologists in inferring myocardial velocity from cine SSFP images and identifying myocardial wall motion abnormalities at rest in patients with ischemic heart disease. Neural Networks, Cardiac, MR Imaging, Ischemia/Infarction © RSNA, 2023.
评估一种新开发的名为深度学习心肌速度与应变（DLSS）的算法从电影稳态自由进动（SSFP）图像推断心肌速度并检测缺血性心脏病患者壁运动异常的可行性。
在这项回顾性研究中，利用包含电影SSFP图像和四维流速数据的223例心脏MRI检查数据集（2017年11月至2021年5月）开发了DLSS。为确定正常范围，在40名无心脏病个体（平均年龄41岁±17[标准差]；30名男性）中测量节段应变。然后，在另一组冠心病患者中评估DLSS检测壁运动异常的性能，并将这些结果与四位独立心胸放射科医生的一致结果（金标准）进行比较。通过使用受试者操作特征曲线分析评估算法性能。
心脏MRI检查结果正常的个体中节段径向峰值应变的中位数为38%（四分位间距：30%-48%）。在缺血性心脏病患者（53例患者中的846个节段；平均年龄61岁±12；41名男性）中，四位心胸阅片者检测壁运动异常的Cohen κ值为0.60-0.78。DLSS在受试者操作特征曲线下的面积为0.90。使用固定的30%异常径向峰值应变阈值，该算法的敏感性、特异性和准确性分别为86%、85%和86%。
在从电影SSFP图像推断心肌速度以及识别缺血性心脏病患者静息时心肌壁运动异常方面，深度学习算法与专科放射科医生的表现相当。神经网络、心脏、磁共振成像、缺血/梗死 ©RSNA，2023年
