Vessel attachment nodule segmentation using integrated active contour model based on fuzzy speed function and shape–intensity joint Bhattacharya distance

Abstract

Because of similar intensity in adjacent regions, intensity inhomogeneity and fuzzy boundary, vessel attachment nodule segmentation is a difficult task. In this paper, a novel integrated active contour model based on fuzzy speed function and shape–intensity joint Bhattacharya distance is proposed for vessel attachment nodule segmentation. First, the regularization component of the energy function is formulated according to the fuzzy speed function. The fuzzy speed function is calculated by the degree of membership based on intensity information and shape information. Second, the data component of the energy function is formulated according to the shape–intensity joint Bhattacharya distance between probability density of local region and probability density of the internal and external contour curve. To calculate the estimated probability density, intensity information and shape information are utilized. The proposed integrated active contour model is experimented on both the synthetic and clinical vessel attachment nodules for evaluating its performance. Experimental results demonstrate that the proposed integrated active contour model can accurately segment vessel attachment nodules.