Segmentation in Ultrasonic B-Mode Images of Healthy Carotid Arteries Using Mixtures of Nakagami Distributions and Stochastic Optimization

Abstract

The goal of this work is to perform a segmentation of the intimamedia thickness (IMT) of carotid arteries in view of computing various dynamical properties of that tissue, such as the elasticity distribution (elastogram). The echogenicity of a region of interest comprising the intima-media layers, the lumen, and the adventitia in an ultrasonic B-mode image is modeled by a mixture of three Nakagami distributions. In a first step, we compute the maximum a posteriori estimator of the proposed model, using the expectation maximization (EM) algorithm. We then compute the optimal segmentation based on the estimated distributions as well as a statistical prior for disease-free IMT using a variant of the exploration/selection (ES) algorithm. Convergence of the ES algorithm to the optimal solution is assured asymptotically and is independent of the initial solution. In particular, our method is well suited to a semi-automatic context that requires minimal manual initialization. Tests of the proposed method on 30 sequences of ultrasonic B-mode images of presumably disease-free control subjects are reported. They suggest that the semi-automatic segmentations obtained by the proposed method are within the variability of the manual segmentations of two experts.

Segmentation of Skin Lesions in 2-D and 3-D Ultrasound Images Using a Spatially Coherent Generalized Rayleigh Mixture Model

“…Hence, the proposed α-Rayleigh mixture model can be interpreted as a transformation of the symmetric α-stable mixture model studied in [37]. Finally, akin to [17], [19], [22], [23], it is interesting to mention that the model (4) uses a simplified image representation based on regions and does not describe the boundaries between tissues explicitly.…”

“…One can think of using the EM algorithm [46] that has received much attention for mixture problems (see [19], [22] for applications to ultrasound images). However, EM algorithms have many known shortcomings.…”

“…Rayleigh-mixture models were extended to tissues with generalized Rayleigh statistics by Destrempes et al [22], who proposed a carotid artery segmentation method based on a Nakagami mixture and a deformable model. As in [19], the estimation of the mixture parameters was achieved using an EM algorithm under the assumption that observations are independent.…”

Medical Decision Support System of Common Carotid Artery Using Image Mining and Dual Snake Segmentation Procedure

“…The many training samples are needed in order to obtain a satisfactory shape prior for a particular class, when the intra-class variability of the object shapes is high 5 . To deal this problem, the shape prior has an elegant variational formulation that allows local adaptations of the parameters associated with a trained shape prior.…”

An automatic 2D CAD algorithm for the segmentation of the IMT in ultrasound carotid artery images

“…They include techniques based on edgedetection [1][2][3], dynamic programming [4], active contours (snakes) [5] and probabilistic approaches [6]. The majority of these techniques were built on the assumption that the two interfaces bordering the IMT are easily distinguishable and in general, they require a certain degree of user interaction.…”