Multi-channel deep learning model-based myocardial spatial-temporal morphology feature on cardiac MRI cine images diagnoses the cause of LVH

作者全名："Diao, Kaiyue; Liang, Hong-qing; Yin, Hong-kun; Yuan, Ming-jing; Gu, Min; Yu, Peng-xin; He, Sen; Sun, Jiayu; Song, Bin; Li, Kang; He, Yong"

作者地址："[Diao, Kaiyue; Sun, Jiayu; Song, Bin] Sichuan Univ, West China Hosp, Dept Radiol, Chengdu, Sichuan, Peoples R China; [Liang, Hong-qing] Army Med Univ, Third Mil Med Univ, Affiliated Hosp 1, Dept Radiol,Southwest Hosp, Chongqing, Peoples R China; [Yin, Hong-kun; Yu, Peng-xin] Infervis Med Technol Co Ltd, Inst Adv Res, Beijing, Peoples R China; [Yuan, Ming-jing] Chongqing Med Univ, Yongchuan Hosp, Dept Radiol, Chongqing, Peoples R China; [Gu, Min] Univ Chinese Acad Sci, Chongqing Gen Hosp, Dept Radiol, Chongqing, Peoples R China; [He, Sen; He, Yong] Sichuan Univ, Dept Cardiol, West China Hosp, 37 Guo Xue Xiang, Chengdu 610041, Sichuan, Peoples R China; [Song, Bin] Sanya Municipal Peoples Hosp, Dept Radiol, Sanya, Hainan, Peoples R China; [Li, Kang] Sichuan Univ, West China Hosp, West China Biomed Big Data Ctr, Medx Ctr Informat, 37 Guo Xue Xiang, Chengdu 610041, Sichuan, Peoples R China; [Li, Kang] Sichuan Univ, Medx Ctr Informat, Chengdu, Peoples R China"

通信作者："He, Y (通讯作者)，Sichuan Univ, Dept Cardiol, West China Hosp, 37 Guo Xue Xiang, Chengdu 610041, Sichuan, Peoples R China.; Li, K (通讯作者)，Sichuan Univ, West China Hosp, West China Biomed Big Data Ctr, Medx Ctr Informat, 37 Guo Xue Xiang, Chengdu 610041, Sichuan, Peoples R China.; Li, K (通讯作者)，Sichuan Univ, Medx Ctr Informat, Chengdu, Peoples R China."

来源：INSIGHTS INTO IMAGING

ESI学科分类：CLINICAL MEDICINE

WOS号：WOS:000975384000001

JCR分区：Q1

影响因子：4.1

年份：2023

卷号：14

期号：1

开始页：　

结束页：　

文献类型：Article

关键词：Cardiac cine MRI; Left ventricular hypertrophy; Case prediction; Deep learning

摘要："BackgroundTo develop a fully automatic framework for the diagnosis of cause for left ventricular hypertrophy (LVH) via cardiac cine images.MethodsA total of 302 LVH patients with cine MRI images were recruited as the primary cohort. Another 53 LVH patients prospectively collected or from multi-centers were used as the external test dataset. Different models based on the cardiac regions (Model 1), segmented ventricle (Model 2) and ventricle mask (Model 3) were constructed. The diagnostic performance was accessed by the confusion matrix with respect to overall accuracy. The capability of the predictive models for binary classification of cardiac amyloidosis (CA), hypertrophic cardiomyopathy (HCM) or hypertensive heart disease (HHD) were also evaluated. Additionally, the diagnostic performance of best Model was compared with that of 7 radiologists/cardiologists.ResultsModel 3 showed the best performance with an overall classification accuracy up to 77.4% in the external test datasets. On the subtasks for identifying CA, HCM or HHD only, Model 3 also achieved the best performance with AUCs yielding 0.895-0.980, 0.879-0.984 and 0.848-0.983 in the validation, internal test and external test datasets, respectively. The deep learning model showed non-inferior diagnostic capability to the cardiovascular imaging expert and outperformed other radiologists/cardiologists.ConclusionThe combined model based on the mask of left ventricular segmented from multi-sequences cine MR images shows favorable and robust performance in diagnosing the cause of left ventricular hypertrophy, which could be served as a noninvasive tool and help clinical decision."

基金机构：　

基金资助正文：