The article discusses the possibilities for the development of automated processing of mammographic images based on the principles of a systematic approach. A structural diagram of an intelligent system for processing and analyzing mammographic images is proposed, which contains three main modules: a module for forming cascading windows, a module for combining cascading windows, and a module for classification and decision making. The software was developed in MATLAB 2018b environment. The possibilities of functioning of each module are considered. Experimental testing of the software of the intelligent system for the classification of breast radiographs according to the classes “no area of interest” or “area of interest” was carried out. A criterion for evaluating the results of classification of mammographic images is selected. Experiments on control samples showed diagnostic efficiency for the classes of radiographs “no region of interest” - “region of interest” not less than 90%.
The purpose of the research is to create new informative Internet technologies that would provide creators of intellectual research systems with available information for constructing training and control samples. An information and diagnostic system based on the formation of a unified structured knowledge base, which is intended for the analysis and classification of X-ray images, is proposed. Research methods for the developed information technology for the synthesis of X-ray classifiers are “the Interior”. The Interior can be presented as a computer program, an autonomous Interior agent, which allows interactively carrying out research on meta-analysis of the classifiers efficiency. The function of the Interior is to build an X-ray classifier with the connection of many users with different databases. The result of the research is the formation of a unified structured knowledge base. The basis of the Interior is a DBMS containing: image files, files of marks of morphological formations, files of personal data, files of meta-analysis results, and files of classifiers. Integration into interiors-associates allows you to build a distributed system of intelligent agents. Based on this system, it becomes possible to form classifiers for global use, as well as to carry out a meta-analysis of the diagnostic procedures effectiveness.
The essence of the article is to develop a method for the formation of descriptors intended for neural networking classifiers of medical risks based on the analysis of transition processes in the biomaterial in the in vivo experiment. The essence of the proposed method is to form the test effects of the probing current on the anatomical areas with anomalous electrical conductivity and obtaining the amplitude-phase-frequency characteristics of the impedance of the biomaterial to which the test effect was carried out. The coordinates of the Cowla’s graphic were used as descriptors. Cowla’s graphic was obtained on the basis of the conversion of Carson counts of the transition process in a quadrupole, the element of which is the impedance of the studied biomaterial. A sequence of unipolar rectangular pulses was supplied to the inlet of the quadrupole. It is shown that the linear model of the biomaterial impedance allows you to obtain descriptors based on its amplitude-phase-frequency characteristics, taking into account the dissipative properties of the biomaterial. The receipt of the Cowla’s graphic model with its dissipative properties, allows you to build medical risk classifiers for socially significant diseases.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.