基于超声纹理匹配技术无创分析易损斑块生物力学特性并预测斑块破裂的实验研究

Cardiovascular disease is the leading cause of death in urban and rural areas in China, and the morbdity is rising year by year,which bring heavy burden to the country and people. The rupture of vulnerable plaque is the main cause of acute cardiovascular events. However, there is a lack of noninvasive and accurate diagnostic methods. In this project, we propose a texture matching technique based on ultrasonic image, which has been preliminarily proved to be accurate in calculating the elastic modulus of the vessel wall. This project will further perform vivo animal experiment, and the value of texture matching technique for noninvasively analyzing the mechanical characteristics of vulnerable plaques will be explored through a comparative study of image and pathologic methods. Our research can be detailed as: Firstly, the high resolution ultrasound imaging method is proposed to be studied, the image segmentation method and algorithm are expected to be optimized and then the calculation accuracy improved; Secondly, the vulnerable plaque animal model will be constructed, and the feasibility and accuracy of texture matching technique for analyzing the mechanical characteristics of vulnerable plaques are to be confirmed; Thirdly, we will trigger the plaque rupture using drug, and then study the stress-strain conditions of plaque rupture. The successful implementation of this project is expected to provide a novel method and idea for noninvasive and accurate assessment of vulnerable plaques.

心血管疾病是我国城市和乡村人群的首要死亡原因，且发病率逐年上升，给国家和人民带来沉重负担。易损斑块的破裂是急性心血管事件的主要病因，然而目前临床缺乏无创、准确的诊断方法。本项目提出一种基于超声图像的纹理匹配技术，前期研究已初步证实其计算血管壁弹性模量的准确性。拟通过活体动物实验，以影像与病理学对照分析，研究易损斑块力学特征与组织成分的关系，并进一步分析易损斑块破裂的力学条件，评价纹理匹配技术预测易损斑块破裂及破裂部位的可行性。具体包括：1）研究高分辨超声成像方法，优化精确图像分割方式和算法，提高纹理匹配技术的计算精度；2）构建易损斑块动物模型，以病理学验证纹理匹配技术分析易损斑块力学特征、诊断易损斑块的可行性和准确性；3）药物触发斑块破裂，研究易损斑块破裂的应力-应变条件。本项目的成功实施有望为易损斑块的无创、准确评估提供一种新方法、新思路。

心血管疾病是我国城市和乡村人群的首要死亡原因，易损斑块的破裂是急性心血管事件的主要病因，然而目前临床缺乏无创、准确的诊断方法。我们前期研究初步验证了纹理匹配技术计算血管壁弹性模量的准确性。因而，我们提出应用无创、基于高分辨超声图像的纹理匹配技术分析易损斑块的力学特征，利用互相关算法获得斑块内部的弹性模量值并以不同颜色标定，分析斑块易损性与弹性模量值的关系。. 本项目首先进行了纹理匹配技术的高分辨超声成像方法学研究，建立最佳超声成像条件，并通过优化精确图像分割方式和算法，提高计算精度。体外仿体实验以PVA水凝胶制备两种和三种组织成分的斑块仿体，模仿人体血液循环搭建体外循环实验系统，以纹理匹配技术计算斑块仿体的弹性模量分布，结果显示软组织成分、硬组织成分和周围组织成分的弹性模量测量值与弹性压缩测试的结果一致性好。活体动物实验通过ApoE基因敲除小鼠和新西兰大白兔建立腹主动脉斑块模型，斑块的弹性模量分布与组织染色结果匹配，脂质和纤维组织成分分别对应低和高弹性模量区域。临床试验采集颈动脉斑块致管腔重度狭窄的手术患者数据，结果提示斑块内弹性成分的分布与超声回声特征相符，与病理染色结果较一致。经过验证，超声纹理匹配技术可用于分析斑块的生物力学特性，有望为临床提供一种新的、无创诊断斑块易损性的方法。该研究部分结果发表在IEEE Access（2020，共同一作）。