影像组学对房颤患者左心房与左心耳连接处心肌的特征分析

Analysis of the characteristics of the myocardium at the junction of the left atrium and the left atrial appendage in patients with atrial fibrillation by radiomics based on CTA images

目的探索应用影像组学方法提取房颤患者与动脉粥样硬化患者左心房与左心耳连接处心肌的影像组学特征的可能性。方法回顾性收集2019年1月至2019年8月84例房颤患者和84例冠脉动脉硬化患者的心脏CTA图像,输入影像组学分析平台,在75%扫描时相勾画左心房与左心耳连接处心肌,利用影像组学Lasso降维分析方法提取影像组学特征。结果房颤患者与冠状动脉粥样硬化患者左心房与左心耳连接处心肌的特征性影像组学标签为:一阶特征:灰度最大值、灰度变异度、峰度;纹理特征:灰度区域大小矩阵变异度,灰度级长均匀性归一化,灰度相关矩阵均匀性归一化。结论影像组学Lasso降维分析方法能够提取出房颤与动脉粥样硬化患者的左心房与左心耳连接处心肌的相关影像特征,CTA结合影像组学的方法有可能有助于评价房颤患者的心肌重塑性改变及预后分析。

Objective To explore the possibility of applying the method of image histology to extract the characteristics of left atrium and left atrial appendage in patients with atrial fibrillation and atherosclerosis.Methods The CTA images of 84 patients with atrial fibrillation and 84 patients with coronary arteriosclerosis from January 2019 to August 2019 in our hospital were collected retrospectively.The myocardial at the junction of left atrium and left atrial appendage was delineated in 75%of the scans,and the features of image histology were extracted by lasso dimension reduction analysis.Results The characteristic image group labels of the myocardium at the junction of left atrium and left atrial appendage in patients with atrial fibrillation and coronary atherosclerosis were as follows:first-order features:gray maximum,gray variation,kurtosis;texture features:gray area size matrix variation,gray level uniformity normalization,gray-scale correlation matrix uniformity normalization.Conclusion Lasso dimension reduction analysis can extract the related image features of the left atrium and left atrial appendage in patients with atrial fibrillation and atherosclerosis.CTA combined with image histology may be helpful to evaluate the myocardial remodeling and prognosis in patients with atrial fibrillation.