【作者】 胡宏伟；

【导师】 孙正；

【作者基本信息】 华北电力大学 ， 电子与通信工程（专业学位）， 2015， 硕士

【摘要】 血管内超声(intravascular ultrasound,IVUS)是分析血管形态和检测粥样硬化斑块的常用介入影像手段。它利用心导管顶端的微型高频超声探头获取血管腔横断面图像,具有较好的探测深度,但是空间分辨率较低。血管内光学相干断层成像(intravascular optical coherence tomography,IV-OCT)是一种新的血管内断层成像方法,结合光学技术与超灵敏探测器获取血管腔横断面图像,空间分辨率高,但其组织穿透力较弱。本文目的在于将同一段血管的IV-OCT与IVUS图像数据融合起来,充分发挥IVUS成像的强组织穿透力和IV-OCT成像高分辨率的优势,获得对血管壁以及粥样硬化斑块更为全面的描述,为冠心病的计算机辅助诊断和制定介入治疗计划等提供依据。图像配准是图像融合的重要前提,本文根据IVUS和IV-OCT图像的特点分别采用基于特征点和基于血管腔轮廓两种方法完成图像配准。前者需要从IV-OCT和IVUS图像中提取标记点,后者需要提取血管腔内膜的轮廓,并以此为依据进行检索得到待配准的IVUS图像,之后估计轮廓的配准参数并对IVUS图像进行相应变换实现配准。完成图像配准之后,本文分别采用基于小波变换的融合算法和基于Bandelet变换的融合算法,对配准后的IVUS和IV-OCT图像进行融合。采用临床采集的图像数据分别从主观评价和客观评价两方面对比融合结果,分析方法的优缺点。更多还原

【Abstract】 Intravascular ultrasound(IVUS) is a unique imaging tool to analyze the morphological vessel structures and make decisions about plaque presence. It utilizes a tiny ultrasound probe mounted on the tip of the cardiac catheter to obtain cross-sectional images with better detection depth, but lower spatial resolution. Intravascular optical coherence tomography(IV-OCT) is a new intravascular imaging modality. Similar to IVUS, it combines optical technology with ultra-sensitive detector to obtain cross-sectional images of the vessel lumen, with high resolution but poor tissue penetration.This study aims to fuse IV-OCT and IVUS image data acquired from a vessel segment to obtain comprehensive accessment and description of the vessel wall and atherosclerotic plaques. The advantages of strong tissue penetration of IVUS and high resolution of IV-OCT imaging are combined to assist the computer-aided diagnosis and interventional treatment planning of coronary arterial diseases.Image registration is an important prerequisite for image fusion. In this dissertation, two registration methods based on feature points and contour feature are presented. The former needs to detect landmarkers, such as bifurcation and calcification, and register the images according to them. The latter needs to extract vessel lumen contours from IV-OCT and IVUS images, and register both images based on lumen contours. Then, IVUS and IV-OCT images are fused based on wavelet transform algorithm and Bandelet fusion algorithm. The experimental results with clinically acquired image are subjectively and objectively evaluated. Also, the advantages and limits of both methods are discussed.更多还原