Metody pro registraci medicínských obrazů – laboratorní úloha

Abstract

The bachelor thesis is focused on the processing of real medical image data. The first part of the thesis deals with the process of image acquisition and processing, image segmentation methods and especially image registration, registration methods and applications. The registration process, various aspects of the classification of registration methods, basic types of transformations, interpolations and similarity metrics are defined in the thesis. The second part of the thesis describes the implementation of selected registration methods on real dataset. The aim of the bachelor thesis is to find the optimal parameters of the transformation. The first implementation is rotation based registration. The task of this type of registration is to determine the optimal rotation angle of the floating image. The second output of the thesis is to find the optimal combination of input parameters for registration based on intensity. The third output are laboratory tasks used for educational purposes. Laboratory tasks are based on tasks in the practical part of the work. Testing of the methods took place in the MATLAB R2018b by MathWorks. The method results were processed in Excel.