In this paper we propose a Bayesian approach for describing the position distribution of the endocardium in cardiac ultrasound image sequences. The problem is formulated using a latent variable model, which represents the inside and outside of the endocardium, for which the posterior density is estimated. As the Rayleigh distribution has been previously shown to be a suitable model for blood and tissue in cardiac ultrasound image, we start our construction by assuming a Rayleigh mixture model and estimate its parameters by expectation maximization. The model is refined by incorporating priors for spatial and temporal smoothness, in the form of total variation, preferred shapes and position, by using the principal components and location distribution of manually segmented training shapes. The posterior density is sampled by a Gibbs method to estimate the expected latent variable image which we call the Bayesian Probability Map, since it describes the probability of pixels being classified as either heart tissue or within the endocardium. Our experiments showed promising results indicating the usefulness of the Bayesian Probability Maps for the clinician since, instead of producing a single segmenting curve, it highlights the uncertain areas and suggests possible segmentations.