Detection of vascular system in chest tomosynthesis screens

OData support
Supervisor:
Dr. Horváth Gábor
Department of Measurement and Information Systems

The purpose of this thesis is to create a vessel enhancement filter, which is applied on digital tomosynthesis (DTS) images. The filter is used to detect abnormalities in the chest area. The enhancement of the vessels is important, because in many cases it is hard to distinguish the blob-like abnormalities and some type of vessels. Therefore, it is crucial to separate these structures.

The method is developed on phantom images, because the enhancement of the vessels is easier on these screens. Hence, if the method does not give promising results on them, it means that it will not work on the real DTS images. The study will present the methods and algorithms which were used to detect the vasculature. After that, we will try to judge the detections of a blob-like structure finder program. In fact these phantom images represents only the vasculature, and they do not contain abnormalities. So we have to prove, that the detections are false positive.

Since the phantoms do not contain abnormalities we use phantom images equipped with artificial blob-like structures too. So we can examine the results of our filter applied on these type of images.

At the end of the thesis we also test the results of the created vessel enhancement filter on real DTS images.

Downloads

Please sign in to download the files of this thesis.