Boiko D. A decomposition method and a subsystem that can be used to improve the quality of visualization of anatomical and pathological structures in digital mammograms

The object of study - the process of visualization of anatomical and pathological structures on mammograms. The aim - to improve the quality of visualization of anatomical and pathological structures on low-contrast gray-scale images of mammary glands using decomposition method and a subsystem for visualizing mammographic images. Methods - methods of digital processing of two-dimensional signals; methods of expert evaluations; theory of decision support in conditions of uncertainty. Theoretical and practical results - development a subsystem for visualizing the anatomical and pathological structures on digital mammograms, which is based on the proposed methods for improving and assessing the quality of low-contrast grayscale images, which allows improving the quality of women mammographic examination and time reducing of their study. Novelty - first developed a mathematical model of the image of the mammary gland structure on a digital mammogram, which based on the layerwise decomposition of the original image into components (images containing the structure and density of tissues) for digital processing of each of them and their subsequent formulation, that has allowed to develop a method of increasing the quality of imaging anatomical and pathological structures, taking into account the features of low-contrast halftone images of the mammary gland; first developed a decomposition method which consists in an adaptive filtering of the image, alignment gradient changes breast background image while improving the contrast of displayed anatomic structures; further developed a method of the multicriteria quality assessment of anatomical and pathological structures on digital mammograms in conditions of uncertainty, which, in contrast to the known, includes the development of scoring scales of individual criteria of subjective and objective characteristics of a mammographic image, it takes into account the features of the analyzed images and provides fast and accurate image quality expert evaluation. The degree of implementation - "Radmir" Company (Kharkiv), State institution "Institute of Medical Radiology by S.P. Grigorieva of National Academy of Medical Sciences of Ukraine" (Kharkiv), Department of Computer Science and Programming of National Technical University "Kharkiv Polytechnic Institute" (Kharkiv). Scope - the field of biomedical engineering.