Кафедри
Постійне посилання на розділhttps://repository.kpi.kharkov.ua/handle/KhPI-Press/35393
Переглянути
3 результатів
Результати пошуку
Документ Developing an informational model of instrumental examination(2019) Filatova, A. E.; Povoroznyuk, A. I.; Gavrylenko, Svitlana; Fahs, MohamadThis document presents the development of an informational model of instrumental examination of a patient using the data flow diagrams (DFD). The developed informational model of the instrumental examination is presented in the form of a context diagram, its decomposition, and the decomposition of the subsystems «Registration and analysis of biomedical signals and images with locally concentrated features» and «Diagnostics». Taking into account the proposed information model, UML diagrams of the activity of the biomedical decision support system based on morphological analysis of biomedical signals and images with locally concentrated features, of the module for morphological analysis of an electrocardiogram, and of the module for improving the quality of visualization of biological objects on radiological images based on the IMRI method are developed.Документ Method of automatic determination of the heart’s electrical axis in cardiological decision support systems(Одеський національний політехнічний університет, 2021) Filatova, A. E.; Fahs, MohamadThe work is devoted to solving the scientific and practical problem of automating the heart’s electrical axis calculation to im-prove the quality of morphological analysis of biomedical signals with locally concentrated features in cardiological decision support systems, which in turn reduces the likelihood of medical errors. The work shows that existing methods for in the determining the electrical axis of the heart require morphological analysis of an electrocardiogram. The method is based on determining the integral signal in the frontal plane from all limb leads, taking into account the lead angle in the hexaxial reference system. In graphic form in polar coordinates, the integral electrocardiological signal is a figure, predominantly elongated along the axis, the direction’n of which corresponds to the heart’s electrical axis. The position of the heart’s electrical axis is calculated as the angle between the axis of standard lead I and the vector, the end of which is at the center of mass of the locus of the points the farthest away from the reference point. Cluster analysis is used to find the most distant points from the reference point. The proposed method for of calculating the heart’s electrical axis makes it possible not to carry out a preliminary morphological analysis of an electrocardiogram. To implement the method proposed in the article, a program was written in the Matlab language, which is connected as a dynamic link library to the cardiological decision support system “TREDEX telephone” operating as part of the medical diagnostic complex “TREDEX” manu-factured by “Company TREDEX” LLC, Kharkiv. Verification of the results was carried out using a database of electrocardiograms, which were recorded using a transtelephone digital 12-channel electrocardiological complex “Telecard”, which is part of the medical diagnostic complex “TREDEX”, and deciphered by cardiologists of the communal non-profit enterprise of the Kharkiv Regional Council “Center for Emergency Medical aid and disaster medicine”. Comparison of the results of calculating the heart’s electrical axis according to electrocardiograms by a doctor and automatically using the proposed method showed that in the overwhelming majority of cases the decisions made coincide. At the same time, cardiologists make mistakes, and errors are made during automatic calculation using the proposed method. The paper explains the reasons for these errors.Документ Development of method of matched morphological filtering of biomedical signals and images(Allerton Press, 2019) Povoroznyuk, A. I.; Filatova, A. E.; Zakovorotniy, A. Yu.; Shehna, Kh.Formalized approach to the analysis of biomedical signals and images with locally concentrated features is developed on the basis of matched morphological filtering taking into account the useful signal models that allowed generalizing the existing methods of digital processing and analysis of biomedical signals and images with locally concentrated features. The proposed matched morphological filter has been adapted to solve such problems as localization of the searched structural elements on biomedical signals with locally concentrated features, estimation of the irregular background aimed at the visualization quality improving of biological objects on X-ray biomedical images, pathologic structures selection on mammogram. The efficiency of the proposed methods of matched morphological filtration of biomedical signals and images with locally concentrated features is proved by experiments.