A novel automatic algorithm for the segmentation of the lumen of the carotid artery in ultrasound B-mode images

Abstract

A new algorithm is proposed for the identification and segmentation of the lumen and bifurcation boundaries of the carotid artery in 2D longitudinal ultrasound B-mode images. It uses the hipoechogenic characteristics defining the lumen of the carotid for its identification and echogenic characteristics for the identification of the bifurcation. The input image is preprocessed with the application of an anisotropic diffusion filter for speckle removal, and morphologic operators for the detection of the artery. This information is then used for the definition of two initial contours, one corresponding to the lumen and the other to the bifurcation boundaries, for the posterior application of the Chan-Vese level set model. A set of longitudinal B-mode images of the CCA was acquired, using a GE Healthcare Vivide ultrasound system, with 256 gray levels. All these images include a part of the CCA and the bifurcation that separates the CCA into the ICA and ECA. In order to achieve robustness in our acquisitions, with the highest contrast and lowest speckle noise levels as possible, the parameter settings of the scanner were different for each acquisition according to the associated image characteristics. We were able to successfully apply a carotid segmentation technique based on cervical ultrasonography. The main advantage of our segmentation method relied on the automatic identification of the carotid lumen, overcoming the limitations of traditional methods.