The purpose of research is development of a method for classifying the functional state of the respiratory system based on the analysis of the variability of slow waves of the VLF range for diagnosing the functional state of the respiratory system.

Methods. The proposed method is based on multimodal analysis of the rhythms of the cardiorespiratory system. To select slow waves of the second order, the wavelet transform of monitoring cardiac signals is used. The detector of slow waves of the second order is built on the basis of the Fourier analysis of each line of the wavelet plane belonging to the region of the breathing rhythm. This allows you to build a hierarchical structure of "weak" classifiers with their subsequent strengthening. When generating descriptors for such classifiers, the signals of slow waves reflecting variations in the breathing rhythm are extracted from the monitoring cardiosignal by means of exploratory analysis in the frequency range of the breathing rhythm and subsequent wavelet analysis in the frequency range corresponding to the frequency range of the breathing rhythm determined as a result of the exploratory analysis. The components of the relevant strings of the wavelet plane are used to calculate the descriptors of the trained neural network, which makes a decision on assigning the current state of the respiratory system to the tested state.

Results. The studies have shown that the functional state of the cardiorespiratory system is characterized by the dynamics of slow waves of the first and second order, which are associated both with the systemic rhythms of the autonomic nervous system and with the systemic rhythms of the central nervous system. In the course of experimental studies on the control sample, a comparative analysis of two methods of classification of the functional state of the respiratory system was carried out: the proposed one and the radiological one. The proposed research method is superior to radiological in specificity and somewhat inferior in sensitivity, which makes it possible to recommend them for clinical practice.

Conclusion. In the course of the study, it was revealed that for the identification of descriptors, on the basis of which classifiers of the functional state of the respiratory system are built, the correlation of wavelet-planes of the electrocardiosignal with the respiratory system using Fourier analysis can be used.

The purpose of research is to develop a web-resource for a dental clinic with the ability to make online appointments for clients as part of the implementation of the national program "Digital Economy of the Russian Federation" in the Kursk region, projects for the introduction of digital technologies and platform solutions in the economy, social sphere, including healthcare, based on solutions included in the "base of effective cases" of the organization "Digital Economy".

Methods. The web resource of a dental clinic with an online appointment system consists of two main modules: a web service for the patient, as well as an administrator panel developed for the clinic website. When creating a software-information system, the PHP language was used to build the system logic, HTML, CSS, JS and the Bootstrap framework to model the interface. The web resource can be installed on a personal computer using the cross-platform distribution kit Apache "XAMPP version from 7.3.x", the required hard disk space is at least 100 MB, the HTTP server is loaded. The software product is intended for use by a wide range of people involved in the reception of clients, in the future it can be used not only for work in dental clinics and private medical centers.

Results. The following features are implemented in the developed software product for the user: menu navigation, making an appointment, leaving feedback, the ability to write a letter, authorization for the user and administrator, adding a client to an appointment, deleting a client from an appointment, updating client or appointment data, adding a user , deleting a user, updating access rights, logging out of an account.

Conclusion. In the course of the work, a web-resource for a dental clinic was developed and implemented with the ability to make online appointments for clients. As a result of the tests, it was found that the software system fully meets the functional requirements.

Purpose of research. Wireless sensor networks are promising direction in the telemetric monitoring field of industrial objects. The most important limitation in their design and deployment is the amount of transmitted information via wireless communication channels. Various methods are used to reduce it. One of them is the method of the quasistationary measurement data aggregation in sensor nodes. This article is devoted to the current issue of the influence of the quasi-stationary measurement data aggregation in sensor nodes on the amount of transmitted information and the total power consumption of a wireless sensor network. The aim of the study is to evaluate the changes in the amount of transmitted information via wireless communication channels and energy gain obtained from the quasistationary measurement data aggregation in sensor nodes.

Methods. For a theoretical and practical assessment of the influence of the quasi-stationary measurement data aggregation in sensor nodes on the amount of transmitted information via wireless communication channels and the total energy consumption of a wireless sensor network, the methods of expert analysis, statistical assessments theory, decision-making information theory, rapid prototyping online service "Circuito.io" and the integrated development environment "Arduino IDE" were used.

Results. The assessment of the influence of the quasi-stationary measurement data aggregation in sensor nodes on the amount of transmitted information and power consumption of a wireless sensor network is carried out, the character of relation between main aggregation values was defined.

Conclusion. In the research the assessment of the influence of the quasi-stationary measurement data aggregation in sensor nodes on the changes of transmitted information via wireless communication channels and energy gain is carried out. In the experiment the basic measuring data amount was reduced on 9,8-18,7% and the decrease in energy consumption was the 7-15%.

The purpose of research is to develop an intelligent information system that allows to implement the process of melanoma diagnosis based on a convolutional neural network.

Methods. The basis for the formation of an intelligent diagnostic information system is the use of convolutional neural networks based on binary classification. During the study, a set of 21,000 images was analyzed, where 10,500 images were an image of a melanoma (malignant formation), and 10,500 images of a nevus (benign formation).As a result of the research, a module containing a convolutional neural network has been developed, which accepts informative signs of the analyzed image as input and forms a response consisting of a set of probabilistic characteristics of the selected pathological formation belonging to one of the classes of formations - melanoma or nevus. To ensure an increase in the level of classification accuracy, the convolutional neural network Xceptionwas used. The last layers of the neural network used were retrained, the network was regularized and the method of excluding neurons was implemented in order to reduce the loss function.

Results. Based on the processing of pathological objects in the images, a set of input information features designed for an intelligent recognition system has been formed. The use of convolutional neural network allowed to establish the accuracy of classification of pathological objects - 94.59%. The maximum value of the loss function during the research reached 18.79%. A module was formed representing a Linux container with an interface for interaction via the HTTP protocol. Based on the analysis of the input image, the module generates a response consisting of a probabilistic characteristic of the belonging of the studied object to one of the studied classes - melanoma and nevus.

Conclusion. The results of the sequential formation of a convolutional neural network by analyzing changes in key indicators (loss function and model accuracy) during the change in the size of the network layers are presented.

The purpose of research is develop a method for efficient processing of an optical image on a digital computer.

Methodology. Two sequentially applied methods are presented for representing a digital image as continuously approaching the original signal values. The quantized light intensity is considered as grouped data. The maximum likelihood method on grouped data allows you to get the best estimate of the original analog value of the light intensity for each photosensitive element of the sensor. The light intensity measured by each photosensitive element is considered as a double integral of the original optical signal over the area defined by the element aperture. Interpolation by atomic functions can use the integrated data to obtain an approximation to the original optical signal.

Results. An urgent problem in vision systems is the loss of information when converting an optical signal into a digital form. Therefore, the problem of efficient digital image processing is relevant. It is necessary to develop a method, to investigate it by conducting statistical modeling, to draw conclusions. In order to evaluate the effectiveness of the developed method, statistical modeling was carried out, after which a comparative analysis of the results obtained for the developed method and the method of moments was carried out. On the whole, the results of the experiment agree with the theoretical calculations.

Conclusion. The use of interpolation, an interpolation polynomial, which is obtained as a result of the proposed method for representing an optical image, allows you to obtain additional information about the optical signal and the optical system used. In turn, this makes it possible to reduce the requirements for expensive optical equipment and reduce material costs or increase the quality characteristics of the resulting optical image without increasing the performance of the imaging equipment.

The purpose of research is to study a way to accelerate the processes of object recognition in images using hardware video accelerators, which include kernels CUDA and TENSOR.

Methods. The method of software interaction with hardware computing means of a video accelerator with CUDA and RT cores using the Python programming language was used; Docker containers from NVIDIA GPU Cloud (NGC), OpenCV to run the feed from the camera and TensorRT to speed up the output of the data stream, the single shot detection network with InceptionV2 as the backbone. To investigate the performance and accelerate the recognition processes, an environment is designed that interacts with equipment and a number of libraries for rendering images. Within this environment, object recognition processes take place, using various computing cores CUDA and TENSOR, as well as various algorithms and accuracy classes.

Results. The use of TENSOR kernels accelerates the recognition process when using FP16 accuracy, as well as when using the combined FP16 and FP32 accuracy. Using single precision INT8 shows significantly better performance when using TensorRT on an accelerator with TENSOR cores.

Conclusion. The capabilities of hardware video accelerators have been investigated when they are used for object recognition tasks. A study of the performance of the CUDA and TENSOR kernels when interacting with the TensorRT engine has been carried out. A software solution for the interaction of video accelerators with CUDA and TENSOR cores with the TensorRT engine is proposed. This solution demonstrates high rates of object recognition speed with INT8 accuracy and slightly lower performance when using FP32 accuracy.

The purpose of research is to improve the methodology for calculating the cost of launching space rockets for launching satellites for remote sensing of the earth's surface into orbit.

Methods. The article provides a systematic analysis of the characteristics of modern domestic space rockets (RKN) of various classes known in the open literature and approaches to calculating the cost of their launch in the interests of launching satellites for remote sensing of the earth's surface into orbit.

Results. The capabilities of domestic ILVs of light, medium and heavy classes for launching payloads into various types of orbits are analyzed. The relevance of creating a promising medium-class ILV capable of launching a payload weighing from 8 to 17 tons with a competitive unit cost of its launch into low Earth orbit is shown. It is noted that the unit cost of launching the payload is directly related to the cost of launching the ILV.

Conclusion. In the proposed method for calculating the cost of launching the ILV, the shortcomings inherent in already existing methods are excluded. This technique can be used both to develop algorithms, models and methods for calculating the cost of launches of IL Vs of various classes, and to assess the effectiveness of their application of IL Vs for their intended purpose.

The purpose of research is to reduce the computational and resource costs associated with performing procedures on board unmanned aerial vehicles for determining and evaluating the level of parameters of uncoordinated deviations from a given trajectory based on the results of image processing of the underlying surface.

Methods. Methods of photogrammetric processing of overlapping aerospace images and approximation of trigonometric functions were used to mathematically describe the process of determining and evaluating the level of uncoordinated deviations of an unmanned aerial vehicle from a given trajectory.

Results. Functional dependences between the parameters of deviations of unmanned aerial vehicles from a given trajectory and changes in parallaxes of overlapping images of the underlying surface caused by these deviations are obtained. A procedure for calculating parallax changes based on the approximation of the obtained functional dependencies has been developed. Quantitative estimates of approximation errors, as well as the joint and separate influence of deviations on the magnitude of changes in the parallaxes of overlapping images are performed.

Conclusion. The difference in the values of the longitudinal parallaxes of adjacent pairs of overlapping images and the difference from zero of their transverse parallaxes serve as a criterion for the presence of uncoordinated deviations of the flight route from the specified trajectory in altitude and direction of flight. The developed model is based on the use of the underlying surface for the analysis and processing of current images and provides automatic detection and assessment of the level of these deviations with a relative error of 0.9%. Unlike the well-known ones, the proposed approach is based on simple computational procedures, which significantly reduces the level of computational and resource costs.

The purpose of research is to develop a method for predicting the occurrence and development of thrombotic complications (thrombotic precedents) provoked by the action of a new coronavirus infection (COVID-19) on the human body, which allows improving therapeutic and diagnostic measures for patients with this pathology.

Methods. The methodology of synthesis of hybrid fuzzy decision rules was chosen as the basic mathematical apparatus, which proved itself well in the process of solving problems with a fuzzy description of the classes under study with a data structure similar to the problem being solved in the work.

Results. In the course of the research, mathematical models for predicting the occurrence and development of thrombotic precedents were synthesized. Expert evaluation and mathematical modeling have shown that confidence in the correct decision-making on the prognosis of the occurrence and development of the studied class of thrombotic complications exceeds 0.9. The paper presents fuzzy mathematical models for predicting the occurrence and development of thrombotic precedents in people with confirmed coronavirus infection, for which the leading risk factor is secondary antiphospholipid syndrome with the occurrence of microangiopathy.

Conclusion. In the course of the conducted studies, the expediency of using the results obtained in the practice of such doctors as immunologists, infectious disease specialists, pulmonologists, cardiologists and cardiovascular surgeons was shown.

The purpose of research is to analyze and simulate a software system in the form of a virtual machine with multithreading support. Modern ideas about threads and multithreading are given, types of multithreading, options for interaction between threads are considered. A conceptual model of a software system in the form of a virtual machine with support for message transfer between threads is proposed.

Methods. Threads are models of computing processes with shared and separate memory. There are two ways of thread switching: cooperative and preemptive. In the case of cooperative multithreading, control transfer planning is assigned to the programmer.

Results. As a result of computer modeling, a software system was developed that implements a virtual machine. The overall computing process is divided into the required number of threads. The main thread can create child threads and terminate them if necessary. A virtual machine allows you to exchange messages in threads.

Conclusion. The resulting software system can be used to solve various tasks that require the division of the computational process into subtasks. The virtual machine provides portability and cross-platform compatibility of the system. The exchange of messages between threads allows you to solve the problems of race states, deadlocks and synchronization.