Since the outbreak of the pandemic in 2019, Covid-19 has become one of the most important topics in the field of medicine. This disease, caused by the SARSCoV-2 virus, can lead to serious respiratory diseases and other complications. They can even lead to death. In recent years, the number of Covid-19 cases around the world has increased significantly, resulting in the need for rapid and effective diagnosis of the disease. Currently, the use of deep learning in medical diagnostics is becoming more and more common. It provides the high diagnostic efficacy that scientists, doctors and patients care about. During the Covid-19 diagnostic procedure, most clinicians order images from Xray and CT to be taken from patients. It is the analysis of these images that gives a full diagnosis. In this article, we will discuss the use of deep neural networks in the diagnosis of Covid-19, especially using chest images taken from X-ray and CT.
The relevance of the study of the agrarian market of Europe, and in particular of Ukraine, is connected with the growing competition of states on world food markets. Trade regulation does not produce the expected results, does not reduce the level of monopolization by transnational corporations of many industry and regional food markets. The problems of population dissatisfaction and ensuring the independence of states are intensifying, which, in turn, increases the role of rational protectionism of the domestic food market. The results of research on this problem will expand the possibilities of scientific approaches to create large-scale research and educational structures capable of participating in various areas of activity to ensure food security. Since the agricultural sector is still one of the most important sectors of the economy in many countries, providing employment and being the main source of income for significant sections of their population, it is not surprising that most of them are interested in the introduction of new technologies and the adaptation of agriculture to development. Changes in the scale of the world economy in recent years have significantly increased interest in investment tasks, which is evidenced by the intensification of trading in the shares of large and medium-sized international companies and, accordingly, causes a rapid increase in their values. As a special case in investment theory, the problem of decision-making regarding the formation and optimization of an investment portfolio is considered. This task is in the field of attention of both large investment companies and private investors, since, choosing among possible alternatives for the distribution of capital investments within the market of financial assets, the investor will get different results. As a result, it is necessary to understand the amount of income received during the period of ownership of the investment portfolio. In the presented work, using the C# programming language, a model is proposed that can be applied in the formation of a portfolio of securities, which allows potential investors to independently assess the effectiveness of the investment set by comparing the growth dynamics of shares available on the financial market. It is known that most of the information an investor encounters is tabular in nature, and according to the methodology of scientific knowledge, a person better perceives visualized ways of presenting information. The model proposed in the work uses visualization tools built into the software product, which presents available tabulated information in a structured graphic form.
In the work, a study of the design and technological direction of creating a dual-functional non-contact pressure and temperature sensor based on silicon-cholesteric crystal systems was carried out. An optically active environment based on a supramolecular spiral structure is a sensitive element for forming the optical response of a microelectronic label sensor during non-contact monitoring of the state of a physical object using a spectrometer. The optical response of the device provides maximum intensity in the case of interference that corresponds to the Wulff–Bragg condition. The functionality of the sensor is ensured by the conditions for avoiding high electromagnetic interference during optical identification. A hardware and software modules have been created for non-contact monitoring of the state of physical objects, the main part of which is graphical user interface that corresponds to the MVVM architectural design pattern. Visualization of the spectral intensity using the software module was provided by conversion to the RGB model. The algorithm of procedure for calculating the color rendering index is given. In a wide range of light waves, the main parameters' color temperature was defined.
Application of intelligent expert systems for structuring and automation of laboratory activities in the educational process has been analysed. It is shown that combination of teaching materials with cross-references and transitions by the means of expert systems allows educators to make optimal decisions on time, set priorities and prepare more effective laboratory classes. The implementation of expert systems for classification sections of academic disciplines in both engineering and technical sciences and humanities has been discussed. An algorithm for making decisions by an expert system, where the main emphasis is given to laboratory tasks and work with laboratory equipment, is proposed. It is shown how to develop a new expert system which can help university educators prepare remote laboratory work, and also to use of virtual simulators or effective practical training. The proposed expert systems can be used for classifying thematic sections of a wide range of disciplines, including natural sciences, engineering, and humanities. They can be used for preparing new courses and training new educators in different areas. Thus expert systems are described as high-scale software units without restrictions on the depth of the question tree and the number of logical branches of the classifier.