基于梯度矢量流snake与多尺度分析的血管内超声图像轮廓提取

提取血管内超声(IVUS)图像的血管内腔轮廓对冠状动脉疾病的诊断有积极意义。本文提出了一种基于梯度矢量流(GVF)snake与多尺度分析的轮廓提取新方法。针对传统GVF snake的轮廓初始化、抑制噪声和伪像干扰两个难点,本文方法在如下两个方面进行了改进:第一,利用序列图像的特性产生时间方差图并得到初始轮廓,从而实现轮廓提取过程的全自动化;第二,引入多尺度分析,使GVF snake在离散小波变换得到的多尺度图中形变,从而增强了算法的鲁棒性。对仿真图像和实际IVUS图像的实验表明,该方法在轮廓提取的精度方面优于传统的GVF snake方法。

Contour detection from IVUS images based on gradient vector flow snakes and multiscale analysis

The detection of luminal borders (contours) from intravascular ultrasound (IVUS) images is helpful for the diagnosis of coronary artery diseases.A novel scheme for the contour detection is proposed based on gradient vector flow (GVF) snakes and multiscale analysis.To solve two difficulties of the traditional GVF snake,i.e.the contour initialization and the suppression of noise and artifact interference,there are two improvements in our proposed scheme.First,the procedure is made into full automation,with adopting characteristics of image sequences to yield the temporal variance image and detect an initial contour.Secondly,to enhance the robustness of the algorithm,the multiscale analysis is employed in the scheme,namely the GVF snake evolves in the muhiscale images generated from the discrete wavelet transform.The proposed scheme is verified on both synthetic images and real IVUS images.Results show that this scheme is superior to the traditional GVF snake in terms of the boundary localization.