Využití pravděpodobnostních modelů při analýze CT snímků

Abstract

Medical images are often covered with noise, for example, which devalues these images. Probabilistic modelling then represents one of the approaches to image analysis. In this Bachelor thesis we will try to build a model of the distribution of areas of the CT image of the abdominal aorta, with which we will then try to find the optimal segmentation of such an image. We will use the principles of Bayesian statistics and Markov Chain Monte Carlo simulation algorithms, which are well suited for such tasks. To properly grasp these concepts and the task itself, the basic concepts of measure theory, probability theory, Markov chains, Markov random fields, measure theory and then an explanation of the principles of Monte Carlo methods are introduced.