The purpose of the research is predicting the progression of primary glaucoma by discriminant method.

Methods. The study involved 141 patients with stage I primary glaucoma and 145 patients with stage II primary glaucoma. On the second day, in the morning, 2 ml of blood was taken, the analysis of which, counting the number of blood cells studied, was performed by the automated Sysmex XN-A1 system (Japan) and the calculation of the analyzed indices. Statistical processing of the obtained data was carried out on a computer and included the use of a discriminant method to create mathematical models that provide 95% error-free differentiation (classification) of patients with stage I and II primary glaucoma.

Results. A comparative assessment of the indices of weak immune inflammation among patients with the progression of the first glaucoma relative to patients without the progression of primary glaucoma revealed significant differences in all the indices under consideration. In accordance with the highest informative value of the SIRI and AISI indexes, the following model was developed by discriminant analysis: Y1 = –12,561 + 0,69X₁ + 0,08X₂, and for patients without progression of primary glaucoma, the model Y₂ = –4,768 + 0,085X₁ + 0,024X₂. It was found that discriminant models mistakenly assigned 13,25% of patients with primary glaucoma progression to the group of patients without primary glaucoma progression. However, discriminant models were classified into the group of patients without primary glaucoma progression, 12,81% with primary glaucoma progression.

Conclusion. The use of the discriminant method and the SIRI and AISI indexes allowed us to develop a discriminant model that makes it possible to differentiate patients with primary glaucoma progression from patients without primary glaucoma progression with the required quality.

The purpose of the research is to develop a software and hardware complex that allows optimizing the control of a robotic device based on electroencephalogram signals based on the recognition of control commands, regardless of the principle of user interaction with a computer.

Methods. Presented is a software and hardware complex capable of functioning on the basis of both synchronous and asynchronous paradigms of the "brain-computer" interface, designed for introduction into various control loops of complex human-machine systems. The hardware includes an electroencephalography signal recording board as well as a neuroharnithra with sensor electrodes. The Cyton electroencephalography signal registration board is a key element of the hardware of the developed software and hardware complex. This is an 8-channel device compatible with the OpenBCI platform, designed for high-precision registration of brain biopotentials, as well as signals of heart and muscle activity. Software developed in Python allows you to record electroencephalography signals; preprocessing electroencephalography signals when implementing synchronous and asynchronous interfaces; classify electroencephalography data; generate control signals to the robotic device.

Results. To perform the task of classifying electroencephalography data, a model of a fully connected artificial neural network, which is a multilayer per-septron (Multi-Layer Perceptron or MLP), was developed. In synchronous mode, the P300 has a potential recognition accuracy of ~ 60%. In asynchronous mode, the accuracy of the binary classification of motor images (imaginary movements of the left and right hands) was 65%. The developed complex provides the ability to work with operators without a preliminary training stage.

Conclusion. The developed complex opens up prospects for introducing the brain-computer interface into critical technological processes where high reliability of human-eyelid machine interaction is required. Further development of the system can be aimed at improving classification accuracy, expanding the number of recognizable commands and optimizing the hardware for mobile applications.

The purpose of the research is analyze the possibility of using questionnaires: Clinical COPD Questionnaire, COPD Assessment Test and modified Medical Research Council Dyspnoea Scale - for remote assessment of the health status of patients with chronic obstructive pulmonary disease and their integration into a software package developed for the purpose of dynamic monitoring in an outpatient setting.

Methods. The software system was developed using the Python programming language. Its architecture includes a real-time automated scoring module, a personal data de-identification system, and a local database. To ensure offline functionality, an offline version of the system was developed with a graphical user interface based on the Tkinter library. The algorithm incorporates mathematical models for interpreting the scales in accordance with international standards.

Results. A functional software system was developed that enables the collection and structured storage of data on symptoms, functional status, and mental health status of patients. The system allows for the creation and maintenance of long-term archives of parameters to assess disease progression. Internal testing confirms 100% reproducibility of calculations and stability of data transfer. API integration ensures potential compatibility with state medical information systems.

Conclusion. The developed software system aligns with the principles of lean healthcare and can reduce specialist physician time expenditure by 20–30% through the automation of routine calculations. Implementation of the system into outpatient practice facilitates a transition to a proactive model of COPD management, thereby enabling early detection of exacerbations and reducing the rate of rehospitalizations. Adaptation to Russian regulatory requirements positions the system as a promising solution for national telemedicine.

The purpose of the research is to improve the effectiveness of upper limb motor function rehabilitation by synchronizing real and imagined motor activity with the servo drives of the units and mechanisms that provide technical support for the rehabilitation process.

Methods. A prototype biotechnical system for managing the rehabilitation of upper limb motor function disorders was developed. This system incorporates EEG, EMG, and dynamometer analysis modules into the control loop’s forward loop, and a virtual reality control module into the biofeedback loop, enabling adaptation of local rehabilitation procedure parameters to the patient’s current functional state. An algorithm for controlling upper limb rehabilitation was developed, featuring monitoring of an indirect biotechnical resonance criterion through distance analysis in a cluster space constructed using dynamometer readings and muscle fatigue coordinates, enabling global rehabilitation procedure parameters to be controlled with a granularity of one patch.

Results. Experimental studies of a biotechnical system monitoring the indirect criterion of biotechnical resonance were conducted. These studies demonstrated that the results obtained through biotechnical resonance monitoring exceed those of the established rehabilitation method by 6% on the MBI scale, 20% on the MFT scale, 13% on the muscle strength scale, 23% on the ARAT scale, 12% on the FMA scale, and 13% on the Ashworth scale.

Conclusion. This article presents a solution to the problem of synchronizing the motor activity of human upper limbs and assistive robotic devices. The results of the study can be used to develop biotechnical rehabilitation systems for motor disorders with rehabilitation procedure control through biotechnical resonance monitoring.

The purpose of the research is to develop criteria for detecting and classifying overloads occurring in multi-machine computing systems based on network traffic data based on an analysis of its regular trend and subsequent probabilistic interpretation of the overload state.

Methods. The wavelet regression processing of the time series is used as a method for detecting the regular trend of network traffic, which ensures the suppression of high-frequency disturbances and the preservation of the regular component. To identify overload intervals, a Bayesian pulse signal detector is used, which generates a smoothed pulse sequence of a posteriori probabilities of exceeding the normal mode. To quantify each overload episode, a parametric pulse model based on Gaussian approximation is introduced, which makes it possible to associate overload with three measurable characteristics: amplitude, duration, and integral area. Additionally, normalized parameter indicators relative to normal mode statistics and an integral deviation criterion are proposed, which provides an assessment of the severity of the overload condition.

Results. Criteria and rules for classifying overloads into five types are obtained: impulse, background, progressive, periodic and attacking. The results of simulation modeling on data with different load modes showed the efficiency of the proposed approach: overload episodes are consistently highlighted, and the typing corresponds to the expected structure of scenarios. The features of the probabilistic detector’s response to a smoothly increasing load, which can manifest itself as a sequence of local pulses, are revealed.

Conclusion. The developed technique provides a transition from visual interpretation of the Bayesian detector output signal to quantification and typification of congestion and can be used in tasks of automated monitoring of network traffic and support for quality-of-service management solutions in multi-machine computing systems.

The purpose of the research are to study the movement of the distal end of the ureteroscope during retrograde intrarenal surgery, to develop an algorithm for autonomous movement of the distal end of the ureteroscope from the initial position to a given point, to conduct a kinematic analysis of the movement of the distal end of the ureteroscope and to solve the inverse kinematics problem.

Methods. Methods of mathematical modeling of dynamic systems, control theory, and theory of mechanisms and machines were used to solve these problems. When creating software products, the mathematical package Matlab is used. A method has been developed for planning the trajectory of the distal end of the ureteroscope based on the analysis of magnetic resonance imaging and computed tomography images. When the distal end of the ureteroscope moves along the resulting trajectory, the polynomial law of motion of the actuating links is applied.

Results. In the course of the study, the main tasks solved by the surgeon during surgery using retrograde intrarenal surgery were considered, and a strategy for autonomous control of the ureteroscope stand was formulated. A method has been developed for planning the trajectory of the distal end of the ureteroscope based on the analysis and processing of magnetic resonance imaging and computed tomography images. An algorithm has been developed for the autonomous movement of the distal end of the ureteroscope from the initial position to a given point. A kinematic analysis of the movement of the distal end of the ureteroscope has been performed. To ensure movement along the planned trajectory, the inverse kinematic problem has been solved, which makes it possible to establish a relationship between the position and orientation of the distal end of the ureteroscope and control actions.

Conclusion. The proposed algorithms for autonomous motion control of the distal end of the urethra, created on the basis of mathematical motion modeling, will be used in the development of a robot for performing operations using retrograde intrarenal surgery in an autonomous mode, which will allow performing operations using retrograde intrarenal surgery much easier and more accurately than now. Also in this paper, a strategy for autonomous management of the ureteroscope medical stand was formulated.

The purpose of the research  is improving the accuracy and speed of real-time recognition of vehicle registration plates through the use of a hybrid method adapted to Russian standards and resistant to environmental conditions.

Methods. This paper proposes and tests a hybrid method for recognizing vehicle registration plates based on the YOLOv12 – EasyOCR pipeline architecture. A unique training dataset was synthesized to ensure maximum variability. A comparative analysis of detectors (YOLOv10, YOLOv12, RT-DETR) was conducted, confirming the feasibility of using the YOLOv12 model, which demonstrated the best performance on the most important metrics (Precision = = 0,99, Recall = 0,98, and mAP_0,5:0,95 = 0,91). A multivariate image preprocessing pipeline for OCR was developed, performing sequential transformations of the cropped image, including adaptive scaling, contrast enhancement, noise reduction, and morphological operations. To minimize typical OCR errors, a limited set of acceptable characters was used, and an algorithm for checking compliance with the Russian license plate format was implemented. Results. The OCR performance was assessed by comparing the recognized text with the reference text using quantitative metrics such as Character Recognition Rate (CRR) and Plate Recognition Rate (PRR), which is the industry equivalent of Word Recognition Rate (WRR). A series of experiments were conducted on a test dataset. Experimental validation of the proposed hybrid method for recognizing vehicle license plates confirmed its high efficiency (quantitative metric values of mAP_0,5:0,95 = 0,91, CRR = 99,0 %, PRR = 98,5 %), demonstrating the method’s ability to ensure robust recognition under real-world conditions.

Conclusion. The obtained results correspond to the upper limit of the performance range of modern automatic recognition systems, including commercial solutions. This demonstrates the high competitiveness of the proposed method.

The purpose of the research is to develop a clinically oriented methodology for training, validating, and selecting a neural network model for pneumothorax segmentation on chest radiographs under severe class imbalance and the need for controlled limitation of false positive alerts in clinical decision support systems.

Methods. We propose a two-loop segmentation quality assessment framework comprising a technical loop for monitoring training convergence using the loss function values and raw metrics, and a clinical loop for evaluating binarized masks after a fixed post-processing pipeline. For each training epoch, a binarization threshold is swept; an operating point is deemed acceptable if predefined clinical constraints are satisfied. The final epoch is then selected as the one that maximizes Dice index on positive cases among all acceptable solutions. The segmenter is implemented using a UNet++ architecture and trained via a two-stage scheme with increasing spatial resolution, using a hybrid loss function.

Results. On clinically representative data SIIM-ACR development set, configuration selected by the protocol achieved Dice index on positive cases = 0,753 (95% CI 0,734–0,771) with false positive rate = 0,177 (95% CI 0,158– 0,197) and image-level sensitivity = 0,948, and substantially outperformed the baseline configuration.

Conclusion. The proposed two-loop validation methodology and constraint-based epoch selection within the admissible region provide reproducible control of the trade-off between segmentation quality on pathological images and the false alarm rate on non-pathological images, reducing the risk of selecting a suboptimal model when relying solely on minimum loss. The results support the use of the developed protocol in designing segmentation modules for decision support systems in radiological imaging tasks.

The purposes of the reseach are to develop and justify a medical decision support system that uses an algorithm for classifying maxillary sinus pathologies in digital dia-fanoscopy based on the ResNet-50 convolutional neural network to increase the accuracy of detecting pathologies (sinusitis and cystic change) in the context of population health screening.

Methods. The features of pathologies were considered, their forms were revealed, which are most often found in rhinological patients. A classifier of maxillary sinus pathologies was developed into three classes based on the results of the analysis of diaphanograms recorded using a digital diaphanoscopy software and hardware complex using the ResNet-50 convolutional neural network. The work proposes algorithms for a system for supporting medical decisionmaking.

Results. A software tool has been developed for classifying the state of maxillary sinus tissues using the ResNet-50 convolutional neural network algorithm, which implements algorithms for learning and classifying maxillary sinus pathologies. The developed medical decision support system was tested. The following accuracy indicators were obtained when differentiating the condition of the maxillary sinuses into three classes: "absence of pathology" – sensitivity 0,94, specificity 0,96, accuracy 0,95; "sinusitis" – sensitivity 0,9, specificity 0,95 and accuracy 0,88; "cystic change" – sensitivity 0,86, specificity 0,96 and accuracy 0,88.

Conclusion. The proposed medical decision support system provides more accurate diagnostics than existing economically available methods. In addition, the proposed solution is able to differentiate the types of pathological changes on diaphanograms in contrast to the previously proposed approaches in digital diaphanoscopy. In this way, the developed technology for optical diagnostics of the condition of the maxillary sinuses, together with the classification model developed on the basis of the SNS in the form of a system for supporting the adoption of medical decisions, can be used both as part of population health screening (for early detection of pathology), and as part of the diagnosis and control of the effectiveness of the prescribed therapy in otolaryngology, ENT oncology and dentistry, as well as in telemedicine.

The purpose of the research is a comparative analysis of static PHP code analysis tools to assess their effectiveness in identifying errors, potential vulnerabilities, and typing problems in the early stages of web application development. Particular attention is paid to determining the practical applicability of the solutions under consideration in projects of various scales, as well as their impact on improving software quality, reducing the number of defects in the code and minimizing security risks.

Methods. The study uses a comparative analysis method based on testing tools on a set of typical scenarios reflecting common errors and vulnerabilities of PHP applications. The tools were evaluated according to the criteria of completeness of error detection, accuracy of diagnosis, flexibility of rule configuration, ease of integration into the development process, productivity and resource consumption. Additionally, an analysis of documentation and configuration options was performed. All experiments were carried out repeatedly to ensure the statistical reliability of the results.

Results. The study revealed differences in the depth of analysis, the rigor of type checking, and the mechanisms for configuring rules. It has been found that the tools exhibit varying sensitivity to logical errors, type inconsistencies, and potentially unsafe designs. Their strengths and weaknesses have been identified in the context of use in small and large projects.

Conclusion. The results confirm the effectiveness of using static analysis tools as a means of improving the quality and security of PHP code. The choice of a specific solution should be based on the requirements of the project, the level of rigor of the analysis and the specifics of the development process. Regular use of such tools can significantly reduce the risk of defects and vulnerabilities, increasing the reliability and stability of web applications.

Purpose of research. With the increasing share of generation from solar plants in distribution networks, the requirements for the quality of supplied electricity are growing, as the instability of solar generation directly affects the network load and the voltage parameters of the electricity supplied to the consumer. The purpose of the research is experimental validation of methods for forecasting the output of a PV plant for the next day with hourly planning, which makes it possible to link the magnitude of the forecast error with the electric power quality indicators according to GOST 32144-2013.

Methods. To solve this problem, a comparative study of three approaches was carried out: ARIMA/ARIMAX models, the Random Forest ensemble, and the hybrid STL+RF scheme. In all cases, the influence of meteorological factors on the accuracy of solar generation forecasting was analyzed.

Results. It was found that the inclusion of weather data significantly improves the forecast quality compared to models using only the history of consumption. The best results were shown by the hybrid STL+RF model with meteorological features, which provided an optimal compromise between forecast accuracy and stability. This assessment shows that in the tasks of network operation management, the use of hybrid models is more justified than the use of statistical models alone.

Conclusion. The proposed approach can be used as a basis for early warning systems for the risks of electric power quality deterioration in networks with a high share of RES. The practical value of the work lies in linking the results of solar generation forecasting with the regulatory requirements of GOST 32144-2013 and the tasks of dispatch control.

The purpose of the research  is to study the characteristics of students’ brain activity while performing intellectual tasks under low (individual assignment) and high (exam) stress conditions based on the analysis of electroencephalogram spectral characteristics.

Methods. Thirty-two students (18 women, 14 men; average age 19,4 ± 1,2 years) participated in the study. Electroencephalogram recordings were conducted using a portable 4-channel BrainBit neural interface (Neurotech LLC, Russia) in the Fp1, Fp2, O1, and O2 positions. Two conditions were compared: completing an individual assignment and taking an oral exam. The relative power of theta, alpha and beta rhythms was calculated. Statistical analysis was performed using the Wilcoxon signed-rank test for related samples (p < 0,05).

Results. When transitioning from the individual assignment to the exam, a significant decrease in the alpha rhythm (from 42,3 to 36,5 %; Z = –4,12, p < 0,001, r = 0,73) and a significant increase in the beta rhythm (from 28,5 to 36,1 %; Z = –4,37, p < 0,001, r = 0,77) were observed. The theta rhythm demonstrated a moderate decrease (from 29,2 to 27,4 %; Z = –2,04, p = 0,041, r = 0,36).

Conclusion. The exam situation causes students’ neurodynamics to shift toward an activation profile: relaxation (α) decreases, while cognitive activation and concentration (β) increase. The obtained data confirm the feasibility of using quantitative electroencephalogram analysis as an objective tool for assessing the functional state of students in real-world educational settings.

Purpose of research. The objectives of the study are to identify key risk factors and create a mathematical model for predicting treatment outcomes for knee injuries in patients with diabetes mellitus.

Methods. To solve this problem, a set of methods was used to take into account many factors affecting the risk of knee injury in patients with diabetes mellitus. These include descriptive statistics for sample characteristics (such as mean, standard deviation, median, quartiles), correlation analysis of Spearman and Pearson coefficients to identify relationships between variables, as well as logistic regression and machine learning methods (Random Forest and SVM) for building and evaluating predictive models. This approach provides an in-depth study of the relationships between clinical and functional indicators, which makes it possible to create accurate and reliable predictive models. As a result, this contributes to improved risk assessment and the development of individual preventive approaches, as well as personalized therapies, which significantly increases the effectiveness of treatment.

Results. A correlation was found between the level of HbA1c and the frequency of infectious complications (r = 0,68, p < 0,01). A mathematical model was developed that allows predicting an unfavorable treatment outcome with an accuracy of 89 % (AUC = 0,89). The prognostic model identified key risk factors for adverse treatment outcomes, including HbA1c levels exceeding 7,5 %, BMI over 30 kg/m² and age over 60 years.

Conclusion. In this work the predictive model works with an accuracy of 86 %, which is a good diagnostic indicator and a risk assessment tool for solving the important task of medical forecasting and prevention of complications.

The purpose of the research is to construct and research a biotechnical control system using the Hill model for describing muscle dynamics, using the wrist joint as an example. The model is based on a new mathematical model capable of adequately describing muscle and joint dynamics during motion in three planes, including the sequential achievement of several predetermined angular positions, cross-couplings, and the physiological limitations of the wrist joint.

Methods. To study the considered biotechnical system, a mathematical model is proposed that includes the dynamics of the joint in the planes of flexion, extension, abduction, and supination, which consists of a block of joint dynamics, blocks of muscle dynamics, and a block of criss-cross connections that take into account the relationship of movements in different planes. The dynamics of muscles is represented by three pairs of "agonist antagonist" muscles implemented on the basis of the three-component Hill model. Control actions are formed using PI regulators for angles and angular velocities in each control channel (in each plane of motion).

Results. Computational experiments were performed where the wrist joint was sequentially moved to three predetermined angular positions. The resulting transient responses for angles and angular velocities demonstrated the system’s stability and the precise achievement of all target positions. Analysis of muscle lengths and tendon tension forces revealed the coordinated action of muscle pairs: the control torque is generated by changing the ratio of agonist and antagonist muscle forces.

Conclusion. The obtained results confirm the adequacy of the developed mathematical model for describing the dynamics of a biotechnical system using the wrist joint and muscle-tendon complex as an example and demonstrate the fundamental feasibility of using the proposed approach for modeling biotechnical systems.

Purpose of research. Development and testing of a specialized automated system for recording, structured storage, and analysis of ultra-high-resolution electrocardiogram data obtained during experiments modeling myocardial ischemia in laboratory animals using a new ultra-high-resolution electrocardiography method is the purpose of research.

Methods. To address this problem, an automated system consisting of recording, storage, and analysis modules was implemented. The recording module was specifically designed to store UHR ECS and includes the ability to visualize the experiment progress to monitor the condition of the biological sample. The experimental data storage module consists of a document-oriented database and specialized software that enables systematization and retrieval of heterogeneous information based on specified criteria. The automatic analysis module sequentially executes software modules specifically designed for analyzing UHR ECS.

Results. The introduction of a developed system for storing and analyzing a large amount of ultra-high resolution electro-cardio signal data obtained during experiments made it possible to significantly optimize research processes. The approach used made it possible to speed up the completion of routine operations by automating the data processing cycle from recording and system optimization to analysis. The absence of manual assignment of metadata, as well as subsequent validation, eliminates unstructured information storage and human errors, allowing previously unavailable meta-analysis.

Conclusion. The system ensures reliable storage and secure access to large volumes of heterogeneous data, improving the reproducibility, quality, and effectiveness of studies on the early diagnosis of myocardial ischemia using ultra-high-resolution electrocardiography.