基于计算机断层扫描血管造影图像的计算流体力学在主动脉瓣狭窄严重程度分级上的初步应用  

Application of computational fluid dynamics based on computed tomography angiography in aortic stenosis grading

作　　者：祝雨雷 陈锐 汪晶晶 杜志成 费洪文[5] 刘辉 俞祝良[1,3] ZHU Yu-lei;CHEN Rui;WANG Jlng-jing;DU Zhi-cheng;FEI Hong-wen;LIU Hui;YU Zhu- liang(School of Medicine, South China University of Technology, Guangzhou 510006, China;Department of Ra- diology, Guangdong General Hospital , Guangdong Academy of Medical Sciences , Guangzhou 510080, China;College of Automation Science and Technology, South China University of Technology, Guangzhou 510080, China;Department of Public Health Medical Statistics and Epidemiology, Health Information Re- search Center, Key Laboratory of Information Science of Guangdong Province, Sun Yat-sen University, Guang- zhou 510080, China;Department of Cardiology, Guangdong Cardiovascular Institute, Guangdong General Hospital, Guangdong Academy of Medical Sciences, Guangzhou 510080, China)

机构地区：[1]华南理工大学医学院,广州510000 [2]广东省人民医院(广东省医学科学院)放射科,广州510080 [3]华南理工大学自动化与工程学院,广州510000 [4]中山大学公共卫生医学统计与流行病学系,卫生信息研究中心,广东省信息学重点实验室,广州510080 [5]广东省心血管病研究所心内科,广东省人民医院(广东省医学科学院),广州510080

出　　处：《岭南心血管病杂志》2018年第2期200-204,248,共6页South China Journal of Cardiovascular Diseases

摘　　要：目的建立一种基于多层螺旋计算机断层扫描(computed tomography,CT)血管造影(multi-detector com-puted tomography angiography,MDCTA)的计算流体力学(computational fluid dynamics,CFD)方法,无创性地获得主动脉瓣狭窄(aortic stenosis,AS)患者的峰值跨瓣膜压差(peak systolic pressure gradient,PSPG),并以此为基础对AS严重程度进行分级,为AS患者狭窄严重程度无创分级提供新的途径。方法本研究纳入30例AS患者,包括轻度狭窄组、中度狭窄组和重度狭窄组各10例。所有患者在完成经胸超声心动图和MDCTA检查后2周内完成心导管检查。由MDCTA原始数据重建获得CFD仿真模型,仿真计算得到的PSPG与作为"金标准"的心导管测量得到的PSPG进行一致性分析,分级效果使用多分类Logistic回归和Kappa检验来分析。结果仿真获得的PSPG与心导管测量的PSPG具有高度的一致性(R2=0.98,P<0.001;mean diff=-0.257mmHg,1mmHg=0.133kPa)。多分类Logistic回归结果显示CFD方法进行AS分级准确率达86.67%,基于CFD获得的PSPG与心导管测量PSPG在AS严重程度分级上具有高度一致性(к=0.80,P<0.001)。结论基于MDCTA原始数据,利用CFD方法,可以无创、准确地获得AS患者的PSPG,并能准确进行AS严重程度分级,为AS严重程度的无创性分级提供了新的途径。Objectives To provide a method based on computational fluid dynamics（CFD）model derived from multi-detector computed tomography angiography（MDCTA）raw data to obtain the peak systolic pressure gradient（PSPG）noninvasively and grade the patients with aortic stenosis（AS）. Methods Thirty patients with AS who underwent bothtransthoracic echocardiography（TTE）and MDCTA within two weeks prior to cardiac catheterization were enrolled inthis study,including 10 mild AS patients,10 moderate AS patients and 10 severe As patients. CFD models were created from MDCTA raw data. Consistency test was performed between simulated PSPG and PSPG measured bycatheterization. Performance of classification was analyzed by multinomial Logistic regression and Kappa test. Results Simulated PSPG showed an excellent agreement with PSPG from catheterization（R2=0.98,P〈0.001;mean diff=-0.257mmHg;1mmHg=0.133kPa）. Multinomial Logistic regression showed that CFD method had a high accuracy（86.67%）. The high к value（к=0.80,P〈0.001）in Kappa test indicted that classification of CFD was excellent agreedwith catheterization. Conclusions CFD scheme based on MDCTA raw data is an accurate and convenient method inobtaining PSPG from AS patients. The preliminary results from our study demonstrate the capability in combiningclinical imaging data by novel CFD tools and provide a noninvasive approach for grading of AS patients.

关 键 词：主动脉瓣狭窄 多层螺旋计算机断层扫描计算机血管造影 计算流体力学 

分 类 号：R542.52[医药卫生—心血管疾病]