Декомпозиционный метод и подсистема повышения качества визуализации анатомических и патологических структур на цифровых маммограммах

Abstract

This dissertation was submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in technical sciences for specialty 05.11.17 – Biological and Medical Devices and Systems – at the National Technical University "Kharkov Polytechnic Institute", Kharkov, 2107. This work is dedicated to solving important scientific and technical tasks to improve the quality of visualization of anatomical and pathological structures on low-contrast gray-scale images of mammary glands (MG) using new method and a subsystem for visualizing mammographic images. A mathematical model of the MG image in a mammogram based on the original image decomposition into components was developed. The model further served as a foundation for the development of a decomposition method to improve the image quality of the MG. Next, the parameters of the method for improving the quality of IMRI-MAM mammogram visualization were optimized. As a result of this work, a new method of multi-criteria image quality assessment method, based on subjective and objective mammographic image characteristics, was developed. The results of processing digital mammograms with the IMRI-MAM method are presented. The structure of the decision making support system to be used in mammalogy is presented. A software to be used to visualize mammographic images was developed. The subsystems were verified on real medical data.