Wall position and thickness estimation from sequences of echocardiographic images

Abstract

This paper presents a new method for endocardial (inner) and epicardial (outer) contour estimation from sequences of echocardiographic images. The framework herein introduced is ne-tuned for parasternal short axis views at the papillary muscle level. The underlying model is probabilistic it captures the relevant features of the image generation physical mechanisms and of the heart morphology. Contour sequences are assumed two-dimensional noncausal rst order Markov random processes each variable has a spatial index and a temporal index. The image pixels are modelled as Rayleigh distributed random variables with means depending on their positions (inside endocardium, between endocardium and pericardium, or outside pericardium). The complete probabilistic model is built under the Bayesian framework. As estimation criterion the maximum a posteriori (MAP) is adopted. To solve the optimization problem one is led to (joint estimation of contours and distributions' parameters), we introduce an algorithm herein named iterative multigrid dynamic programming (IMDP). It is a fully data driven scheme with no ad-hoc parameters. The method is implemented on an ordinary workstation, leading to computation times compatible with operational use. Experiments with simulated and real images are presented.Key words: Echocardiography, Contour Estimation, Bayesian, Dynamic Programming. 1Refererence 1] is a short version of this work.

Deformable templates for tracking and analysis of intravascular ultrasound sequences

“…Moreover speckle noise or high correlated noise, due to the acquisition process and caused by tissue microstructures, can be modeled by a Rayleigh distribution [5].…”

Evaluation of Cardiac Ultrasound Data by Bayesian Probability Maps

“…Our model makes a dependency assumption of neighboring pixels via total variation. A similar approach is employed in [6][7][8][9][10], where Markov random field (MRF) regularization is used. Like our model, in [7,[9][10][11] a Bayesian framework is used, although the construction of the posterior density function is different.…”

Automatic contour estimation in fetal ultrasound images

“…Moreover, being a coherent imaging technique, echography is characterized by the strong presence of speckle noise, which follows a Rayleigh distribution and can not be modelled as being independent nor additive [1]. All these perturbations make conventional contour estimation and segmentation techniques, usually based on local information, inappropriate for these images [5]. Although several special purpose schemes have been proposed for cardiac echography (see [5] and references therein), very few attempts have been made at automatizing the analysis of fetal images [3,9].…”

“…Conventional snakes have several drawbacks (such as the strict use of local data) which have stimulated a great amount of research [2,4,7,11,12,13,16]. Although most limitations of the original formulation have been successfully addressed, only specialpurpose approaches have been able to deal with ultrasound images [5,12]. Moreover, active contour models (or their probabilistic reformulations [5,7,16]) require careful tuning of several parameters, such as those controlling the trade-off between smoothness/robustness and estimation accuracy.…”