It is valuable to segment intravascular ultrasound (IVUS) images of atherosclerotic plaques during the diagnosis of atherosclerotic diseases. Aiming at two problems of traditional IVUS segmentation methods

i.e. the contour initialization and robustness

a novel scheme is proposed to segment IVUS images of plaques based on the active contour model and the contourlet multiresolution analysis. With the contourlet transform

the original image is decomposed into lowpass components and bandpass directional subbands. The template matching is adopted in lowpass components to yield initial contours of the lumen border and the media-adventitia border. The anisotropic diffusion is then utilized in bandpass directional subbands to suppress the noise as well as preserve vascular boundaries

and the contour evolution in the boundary vector field is used to obtain final contours. The method is evaluated via 100 simulated images and 120 real images

and results demonstrate that it can automatically and accurately extract two vascular borders. For real IVUS images

it outperforms the traditional active contour model by 3.04 pixels in terms of the mean distance error

and 6.30% in terms of the percentage of area differences.