The purpose of research is to define the requirements and develop the concept and architecture of the multiservice system “Client medical web service”.

Methods. The requirements to the system "Client medical web-service" are based on the survey of 500 respondents about their preferences in the field of using personal health information. The answers were analyzed using the methods of mathematical statistics.

Results. The results of the survey revealed the main requirements for the development of multiservice systems – the presence of personal involvement of the patient, the usability of the system and the speed of information transfer. Based on the requirements the two-component system "Client medical web-service" was developed on the basis of the map-scheme proposed by us. The system architecture included a system of interconnected modules whose information is transmitted via HTTPS-protocol with optimal caching and storage of medical information. The proposed system will enable secure information sharing between the healthcare facility and the patient by providing the parties with access level management. The integrated approach to information storage is realized by the API server operation, providing a high level of cloud storage data security and its implementation within remote access.

Conclusion. The proposed system "Client medical web-service" offers prospects in improving the quality of services for providing medical information, its exchange and storage. Patient participation in completing their health information, managing its access levels and disseminating it when necessary will achieve a high result, as the information will be open to the patient and thus increase their share in self-monitoring of treatment and prescription fulfillment. Practical implementation of the proposed system will provide improved disease management dynamics while maintaining data security and not violating confidentiality principles.

The purpose of the research is to develop a software and information system for automating the assessment of cognitive abilities of employees and candidates using scientifically validated testing methodologies in HR departments.

Methods. To implement the software system, the C# programming language was used with the ASP.NET Core framework, along with JavaScript and the React library. For storing data on users, testing results, and reports, the MSSQL database was utilized. The following scientifically validated methodologies were employed: "Schulte Tables" (G. Schulte) to measure attention concentration, "Pair Word Memorization Method" (H. Ebbinghaus) to assess shortterm memory, "Raven's Progressive Matrices" (J. Raven) to analyze logical thinking, "Holmes and Rahe Stress Scale" (T. Holmes, R. Rahe) to evaluate stress resilience, and "Belbin's Methodology" (R. M. Belbin) to determine team roles. The system uses test results as input data, which are subsequently analyzed to generate reports and provide recommendations.

Results. During the development, a software and information system was created to automate the process of assessing cognitive abilities of employees and candidates. Interfaces for users and HR specialists were implemented, providing access to tests, results, and analytics. Testing on a sample of company employees demonstrated a diagnostic accuracy level of 93%. The system offers personalized and group analysis capabilities, including generating recommendations for skill development and building summary reports, making it an effective tool for HR departments. The testing results showed that the developed software fully meets functional requirements and is ready for use.

Conclusion. The developed software and information system enables the automation of cognitive testing processes for employees and candidates, providing HR departments with a convenient tool for evaluating key skills and developing development plans. The system ensures high diagnostic accuracy and saves specialists’ time through automated data processing. Future development prospects include expanding functionality, adding new testing methodologies, implementing adaptive algorithms for dynamically adjusting task complexity, and integrating with HRM systems to optimize personnel management processes.

The purpose of research is to analyze the role and significance of distributed computing systems (DCS) in shaping and developing key areas of modern digital infrastructure, as well as to identify the prospects and challenges related to their further integration into technological ecosystems.

Methods. The materials used include statistical data and analytical reports from authoritative sources (Statista, DBEngines, MarketsandMarkets, etc.), as well as technical specifications of frameworks and systems (Apache Hadoop, Cassandra, IBM Summit , etc.). The methodology involves comparative analysis, generalization of practical case studies, and forecasting based on technological development trends.

Results. It has been established that DCS form the foundation of cloud computing, Big Data, IoT, high-performance computing (HPC), and blockchain technologies. Key technological trends have been identified: integration with artificial intelligence, the growth of edge and fog computing, the development of quantum distributed architectures, and trusted computing. Noted risks include management complexity, cybersecurity vulnerabilities, scalability challenges, and legal issues.

Conclusion. Distributed computing systems play a crucial role in digital transformation. Their implementation ensures fault tolerance, scalability, and high performance of IT services. The future of DCS lies in AI integration, automated management, and adaptation to emerging computing models. Sustainable development of DCS requires advances in security, regulatory frameworks, and optimization of network infrastructure.

The purpose of the research is to to develop an intelligent decision support system for physicians based on automated analysis of dermatoscopic images using machine learning algorithms, designed for early diagnostics and detection of malignant skin neoplasms. The development of an intelligent system for supporting physicians' decision making for both specialized specialists and general practitioners and nursing staff performing primary examination of patients with skin neoplasms is a research relevant area.

Methods. The intelligent system architecture for supporting doctors' decision-making based on the analysis of dermatoscopic images is proposed. The configuration used is a network approach based on the client-server mode. The client is a web application implementing the doctor's personal account functionality. This server hosts a cloud infrastructure that collects, stores and analyzes dermatoscopic images, and also maintains a report on the nosological group of skin lesions. In the analyzing process dermatoscopic images, machine learning methods are used based on the neural network’s usage with the virtual transformer architecture and a formed set of dermatoscopic images.

Results. The developed intelligent system for supporting physician decision-making has been practically implemented and tested in clinical conditions. It is characterized by accuracy values exceeding 93% for the Accuracy indicator and 89% – F-measure at the training stage and more than 89% (Accuracy indicator) during medical examinations. The obtained values of experimental assessments made it possible to formulate recommendations for integrating the developed intelligent system for supporting physician decision-making into the work processes of medical institutions.

Conclusion. The developed system provides automated image analysis, metadata structuring, visualization of model predictions and the possibility of expert marking and can be used not only by specialized doctors during medical examinations and studies, but also by general practitioners and mid-level medical personnel during screening examinations, mobile preventive appointments and medical examinations.

The purpose of research. The purpose of this scientific work is to conduct a comprehensive theoretical and analytical review of modern machine learning algorithms used to solve the problems of dynamic route planning for mobile robots. The main focus is on a comparative assessment of the effectiveness of various learning paradigms – reinforcement learning, teacher-based learning, and hybrid approaches – in a changing and uncertain environment where rapid adaptation, learnability, and algorithm stability are important.

Methods. The study is based on an analysis of more than 40 peer-reviewed scientific publications selected from leading international academic databases for the period from 2020 to 2024. A structured methodology was used, including descriptive, comparative, and analytical approaches. The main evaluation criteria were: convergence rate; computational efficiency; generalization ability; noise tolerance; adaptability to real-time and stable behavior in changing conditions.

Results. It is shown that tabular algorithms provide basic navigation functionality, but they do not scale for complex tasks. Deep models have a high degree of adaptability and efficiency. Teaching with a teacher demonstrates accuracy in the presence of expert data, but is vulnerable to the accumulation of errors. Hybrid architectures combining graph neural networks and symbolic modeling achieve the best interpretability and stability in an unstable environment.

Conclusion. The results obtained form a reliable theoretical basis for the selection and application of autonomous navigation algorithms. The comparative analysis highlights the value of flexible, scalable, and explicable models in intelligent robotics systems of a new generation.

The purpose of the research is analysis of key features of signal representation in a small-base polarization measuring system.

Methods. The methods of probability theory, mathematical statistics, statistical radio engineering and computational mathematics were used in the scientific research. The design features of the small-base polarization measuring system were taken into account. As a result, a number of equations were obtained that allow calculating the output voltages for the receiving antennas of both positions. These relationships are valid in the case when the designs of both positions are similar and include an antenna for receiving and transmitting signals of the same linear polarization and a transceiver. For emitted and received signals of horizontal and vertical polarization, orthogonal to the polarization of the first position, their inherent features are determined. The studies made it possible to formulate an algorithm for calculating the phase shifts of reflected signals for antennas of the small-base polarization measuring system. The proposed algorithm is valid for horizontal and vertical polarization.

Results: the independence of the received wave amplitude value of reflected signals from the design feature of the small-base polarization measuring system was substantiated; the dependence of the energy indicators of received signals on the effective scattering surface of the object (the values σv.v, σv.g, σg.v, σg.g) for horizontal and vertical polarization was established; an algorithm for calculating the phase shifts of reflected signals for antennas of the small-base polarization measuring system was formulated; the factors influencing the technical characteristics of the equipment of the small-base polarization measuring system on the energy indicators of received signals were investigated.

Conclusion. The scientific article proposes a method for calculating the phases of the phase center of the small-base polarization measuring system, which allows implementing the internal problem of the theory of antenna systems. The proposed method allows tying the phase center of the small-base polarization measuring system to an arbitrary conditional point within the base, including the centers of the antennas of any of the two positions. 

The purpose of the research is to develop a dermatoscopic images containing high-quality labeling of clinically significant signs of skin neoplasms of the Russian population skin phototypes, intended for early diagnostics and detection of malignant skin neoplasms. The formation and implementation of sets of dermatoscopic images in automated systems and approaches to the early malignant skin neoplasms detection during medical examinations of patients is a relevant research area.

Methods. An approach to the formation of a dermatoscopic images data set with high-quality labeling of clinically significant features is proposed. The basis of the formed data set is dermatoscopic skin neoplasmsimages with confirmed diagnoses, including using clinical research methods, according to the existing nosology of the dermatovenereological profile patients of the Russian Federation population by dermatologists and oncologists. A distinctive feature of the developed data set, in addition to belonging to the skin phototype of the Russian population, is the high-quality labeling of clinically significant features, which allows the developed set to be used in methods and algorithms of machine learning and pattern recognition.

Results. The generated data set of dermatoscopic images contains 657 dermatoscopic images, accompanied by extended metadata and preliminary clinical conclusions, of melanocytic (melanoma and nevus) and non-melanocytic (squamous cell carcinoma, dermatofibroma, vascular lesions, keratosis, etc.) neoplasms. This data set is based on the distribution both by age criterion and by the affiliation and course systemic nature of the disease in patients.

Conclusion. The practical focus of the developed dermatoscopic images data set with high-quality marking of clinically significant features allows the use of the generated images both in decision support systems for doctors in medical practice and in systems based on the machine learning methods and algorithms usage for the early malignant skin neoplasms diagnosis.

The purpose of the research is to study the concept of a "smart environment" and design a video surveillance system to address the issue of timely removal of municipal solid waste (MSW) by a regional operator. The function of these technologies is to monitor the environment in real time 24 hours a day, seven days a week. The implementation of a "smart environment" will bring benefits and improve the quality of life of the urban community. The purpose of the "smart environment" is to manage and create a sustainable environment that ensures the maximum use of technologies for the benefit of the population.

Methods. The research methods are based on the Decree of the Government of the Russian Federation of 20.05.2022 No. 913 "on approval of the regulation on the federal state information system for accounting of municipal solid waste", research and software solutions in terms of automation of waste removal. The use of video surveillance systems to monitor the situation in the "smart environment" of the region allows you to analyze the controlled parameters in real time.

Results. A "smart environment" system was modeled for the task of timely notification of the need for removal (MSW). The use of video surveillance analytics in the city of Kursk made it possible to obtain real-time data and visualize environmental conditions in the monitoring zone, which allowed the regional operator to make decisions based on the processed data.

Conclusion. The implementation of the modern concept of a "smart environment" for the tasks of monitoring the removal of solid municipal waste is currently a relevant area. The organization of the work of the regional operator for the handling of solid municipal waste is a process that requires modern information support. To implement the process of building a system of "smart environments", infocommunication kits and data transmission systems are used to track the level of filling of containers with solid municipal waste.

The purpose of the research is comparison of machine learning-based approaches (deep learning) and classical methods for mass spectrum annotation in big data conditions, as well as identification of the optimal scenario for their integration.

Methods. The study is based on the PXD004452 dataset containing 2,5 million unique peptides. An interaction scheme based on Python/TensorFlow/PyTorch has been developed, which provides parallel processing of peptide spectra on a GPU cluster. The following steps were used: filtering of the top 150 peaks by intensity; generation of theoretical B-/Y-ions, taking into account modifications; prediction of peptides (PepNet – convolutional+recurrent network; Tidesearch – index-shifting strategy). Metrics: number of matches, delta mass, Levenshtein distance, ROC curves, error distribution.

Results. PepNet requires significant computational resources, while the prediction quality is inferior to Tide-search, especially for long peptides and modifications (~average match: 4,2 pi vs 9,7; p < 0,001). However, PepNet performs better in those spectra where relevant sequences are missing in the database search, demonstrating an important ability to identify novel peptides. Levenshtein distance distribution: ~30% is a complete match (0); ~52% is a small deviation (1-5); the rest is significant discrepancies (>5).

Conclusions. The deep learning (PepNet) method shows promise, but without integration with database search, it is inferior in accuracy. A hybrid architecture is proposed: pep-tagging via PepNet, followed by refinement and verification via database search. Such a big data pipeline will combine the discovery of new peptides (de novo) and high identification reliability (database search).

Purpose of research. The widespread use of temperature measurement devices with resistive sensors connected via three-wire or four-wire configurations in industrial process control systems leads to increased complexity of cabling networks and switching components, and significantly limits the number of measurement channels per analog input module. A two-wire connection scheme, which minimizes these drawbacks, introduces significant measurement errors due to additional and unstable resistance from the connecting wires. This work investigates a method proposed by the authors for reducing the error introduced by the connecting line, based on estimating resistance from the integration of the capacitor discharge transient process across the sensor. The study evaluates the impact of quantization effects, noise, and integration intervals of the measurement circuit’s response on the accuracy of resistance determination.

Methods. The assessment of the factors was carried out through simulation in MATLAB and experimental validation of the proposed solution on a prototype model.

Results. A significant advantage of trapezoidal integration has been demonstrated, allowing for measurement errors more than 300 times smaller under experimental conditions compared to the left rectangle method. The optimal integration time was determined, which minimizes the influence of quantization noise and interference on measurement error across the sensor's resistance range. It was found that the optimal integration time is proportional to the resistance value of the sensor, and deviations from the optimal value by up to 50% result in only a minor increase in estimation error.

Conclusion. The measurement error of the resistive sensor's resistance using the proposed method under identical conditions is comparable to that of the method based on integrating the entire capacitor discharge transient process. The maximum error under experimental conditions does not exceed 0.12% when measuring a resistance of 2 kΩ, and is significantly lower for other values. At the same time, the proposed solution allows for a substantial increase in speed due to reduced integration time, and a reduction in equipment size by eliminating the need for a complex switch matrix in scanning systems.

The purpose of the research is to improve the quality of risk assessment for vibration-induced stress syndrome (VIS) and associated neurotic disorders in individuals exposed to hand-held vibrating tools by using hybrid models that combine traditional occupational pathology criteria with indicators of the adaptation level of target organs based on fuzzy intelligent technologies.

Methods. Given that the class of problems under study is poorly formalized and has an ambiguous description of the data structure, fuzzy decision logic was chosen as the basic mathematical framework, specifically a methodology for synthesizing hybrid fuzzy decision rules. This methodology was used to develop hybrid models for assessing the risk of VIS and associated neurotic disorders. These models, along with traditional modern medical criteria, include indicators characterizing the adaptive potential of the body as a whole and the adaptive potential of target organs.

Results. To address the practical challenges of assessing the risk of localized hand vibration disease and associated neurotic disorders, a decision management algorithm was developed for the corresponding decision support system, improving the quality of medical care for the studied patient population. Given the significant use of expert evaluation methods in the synthesis of decision-making models and the general recommendations of the selected methodology, three methods were implemented to validate the performance of the resulting decision rule: expert evaluation, expert modeling of a control sample, and statistical testing on representative control samples.

Conclusion. Expert evaluation and expert modeling of control samples revealed that the proposed method enables a 10–20% improvement in the quality of decisions compared to models that do not use indicators characterizing the body's adaptive potential and the level of adaptation of target organs. The same results were obtained during statistical tests conducted according to the rules generally accepted in recognition theory.

The purpose of the research is to develop a decision support system for gynecologists, based on synthesized prognostic and diagnostic decision rules, which will be used in the diagnosis and prognosis of postpartum endometritis.

Methods. The following tools were used in the development of the decision support system: medical statistics, expert system development methods, decision theory, and pattern recognition methods. Such tools for statistical data analysis as Excel and Statistica were used to process and analyze medical data, as well as to verify crucial diagnostic rules. To assess the risk of postpartum endometritis and its diagnosis, 31 signs are included, ranked according to their degree of informativeness. These data were used as the basis for the development of an algorithm for predicting the risk of developing postpartum endometritis in a decision support system.

Results. The application of the developed diagnostic decision rules on clinically representative material showed a diagnostic effectiveness of 0,96±0,02. The developed expert system can be effectively applied in clinical conditions. It is also possible to use this expert system in the educational process when training medical professionals.

Conclusion. Consideration of risk factors, integration of data from various sources, the use of prognostic models and the formation of individual recommendations for treatment and prevention are all key aspects that should be taken into account when developing such a system. The introduction of such a system into clinical practice can significantly improve the quality of diagnosis and treatment of postpartum endometritis, reduce the risk of complications and improve patient outcomes.

The purpose of the research is to develop and systemically analyze a comprehensive mathematical model of tuberculosis epidemiology in Russia, taking into account multidrug resistance (MDR-TB) and quarantine measures, to formalize the dynamics of infection and provide information support for management decisions.

Methods. The methodology of systems analysis was used. A deterministic mathematical model (TB-SEIRZ-Q) described by a system of nonlinear ordinary differential equations was developed. The model expands classical approaches by introducing a latent stage, stratification of infected people by sensitivity to treatment and bacterial excretion, as well as separate quarantine groups. An analysis of the stability of the model was carried out, the basic reproductive number (R₀) was calculated using the next-generation method. The parameters were identified based on official data for Russia. Numerical modeling of the epidemic dynamics and sensitivity analysis of key parameters were performed.

Results. The TB-SEIRZ-Q model was obtained that adequately describes the specifics of TB. The estimated basic reproduction number R₀ ≈ 2,258, indicating instability of the disease-free state and the transition of the system to endemic equilibrium. The results of numerical modeling demonstrate high correspondence to the real data on TB incidence in Russia for 2018-2023 (R² = 0,92). Sensitivity analysis revealed the key role of infection transmission and isolation rates in the R0 value. Increasing the isolation efficiency to 0,5 reduces R₀ below 1 (to 0,95), providing the possibility of eliminating the epidemic.

Conclusion. The developed TB-SEIRZ-Q model is an effective tool for systemic analysis of the tuberculosis epidemic in Russia. It formalizes the infection dynamics taking into account MDR-TB and quarantine measures, as well as an information basis for assessing and optimizing epidemic management strategies. The model allows predicting the development of the situation and quantifying the impact of various interventions, such as strengthening quarantine measures.

The purpose of the research. The COVID-19 pandemic has shown that mathematical modeling has become important in the management of infectious diseases. The relevance of the study lies in understanding the dynamics of the spread of COVID-19 using mathematical modeling methods that play a key role in developing control strategies. Infectionspecific models make it possible to analyze patterns, predict trajectories, and evaluate the effect of measures, including quarantine, social distancing, and vaccination.

The purpose of the research is to develop and analyze an improved SEIRD model using a hybrid numerical method designed to improve the accuracy of forecasting the occurrence and development of pandemic waves and assessing the impact of sanitary measures.

Methods. The research objectives include building a new SEIRD model as an extension of the classic SIR model with the addition of additional categories – "Exposed", "Recovered" and "Dead". To implement the proposed categories, the following methods were applied: explicit Euler method, fourth – order Runge – Kutta and adaptive Runge-Kutta schemes to increase reliability. Methodologically, the SEIRD system is solved using a hybrid numerical scheme combining the advantages of classical and adaptive methods, which made it possible to obtain accurate simulations and assess the impact of interventions.

Results. The results showed that the proposed refined SEIRD model provides reliable forecasts of the occurrence and development of pandemic waves.

Conclusion. An analysis of the results shows that a 10% increase in the number of infections signals the beginning of a new wave that requires adjustments to the parameters and rapid response of public health services, as well as the implementation of rapid sanitary and epidemiological measures. The SEIRD model with hybrid methods reflects the dynamics of COVID-19, and can also be adapted to model future epidemics.

The purpose of the research is to improve the quality of predicting coronary heart disease by using decision-making models together with a set of generally accepted risk factors that characterize the functioning of the protective mechanisms of the cardiovascular system.

Methods. At the preliminary stage of the study, an exploratory analysis revealed that the "high risk" class of coronary heart disease has significantly overlapping boundaries in relation to alternative classes. In these conditions, specialists focused on solving poorly formalized problems recommend using the theory of fuzzy decision-making logic, and in particular, the methodology for synthesizing hybrid vague decision rules developed at Southwestern State University. At the same stage of the research, the composition of informative signs was determined, which included signs traditionally used in medical practice, indicators of the degree of ischemic damage to the brain and heart, indicators characterizing the functioning of the antioxidant system, the energy imbalance of the "cardiac" acupuncture points and characteristics of the level of protection of the cardiovascular system.

Results. The paper provides a mathematical model for predicting coronary artery disease using a system of traditional predictors for medical practice in combination with blocks of signs describing the degree of ischemic damage to the heart and brain, the functioning of the antioxidant defense system, the energy imbalance of BAT "associated" with heart disease, and the characteristics of the level of protection of the cardiovascular system.vascular system.

Conclusion. The conducted studies have shown that in order to improve the quality of forecasting, it is advisable to combine the following in appropriate decisive rules: predictors of traditional medicine; indicators characterizing the degree of ischemic damage to the brain and heart; indicators characterizing the functioning of the antioxidant system; energy imbalance of "cardiac" acupuncture points; characteristics of the level of cardiovascular protection. It was shown that the quality of forecasting using the models obtained in the work increases by 10–15% compared with models that do not use indicators of the body's level of protection.