The accurate segmentation of lung regions in computed tomography (CT) scans is critical for the automated analysis of lung diseases such as chronic obstructive pulmonary disease (COPD) and COVID-19. This paper focuses on enhancing the accuracy of U-Net segmentation models through a robust preprocessing pipeline. The pipeline includes CT image normalization, binarization to extract lung regions, and morphological operations to remove artifacts. Additionally, the proposed method applies region-of-interest (ROI) filtering to isolate lung areas effectively. The dataset preprocessing significantly improves segmentation quality by providing clean and consistent input data for the U-Net model. Experimental results demonstrate that the Intersection over Union (IoU) and Dice coefficient exceeded 0.95 on training datasets. This work highlights the importance of preprocessing as a standalone step for optimizing deep learning-based medical image analysis.

This study examines the application and implementation of energy-efficient measures and strategies for use in buildings aimed at achieving the goal of virtually zero energy consumption by buildings. In particular, the aim of the study is to introduce integrated solar devices into the building, functioning as a functional component of the structure for heat storage, in order to increase the flexibility of the structure while maintaining comfortable conditions inside. This article discusses an experimental comparison of flat and tubular solar collectors in the southern and northern regions of Kazakhstan. Compared to glass (S = 3), a tubular solar collector has a coefficient of thermal conductivity that is two times lower for single-layer elements and 2.5 times lower for double-layer elements. This means that less heat is lost. A flat solar collector was also developed, a heat-insulating translucent double-glazed window with reduced pressure, and the coolant is made of thin-walled corrugated stainless-steel pipe. In this collector, there is a constant circulation of heat, which increases the efficiency of heat transfer by eliminating additional partitions between the panel and thermal insulation. A calculation was carried out that allowed us to determine the productivity and useful gain for two collectors. The annual increase in the useful energy of flat solar collectors is about 12780000 kJ/hour and 20100000 kJ/hour for tubular solar collectors; therefore, the annual increase in the useful energy of evacuated tubular collectors is 20% more than that of flat plate collectors in the northern region.

The aim of the paper is to design and develop an information system for assessing student health to form a set of measures to prevent disease using artificial intelligence algorithms. Within the framework of this scientific problem, a whole range of tasks related to the analysis of medical data, identification of patterns, search for patterns, and visualization of results using artificial intelligence methods is solved. This study is particularly relevant for Kazakhstan’s educational institutions. The authors of the study expect to form a scientific and technological basis for promoting the idea and principles of “Healthy University - Healthy nation” among educational organizations of the Republic of Kazakhstan. The information system with visualization of the results of aggregation and analysis of data on health indicators and indicators related to health can be used by various groups of internal and external stakeholders as a regularly updated information resource for the development of various social or medical support programs.

The work aims to develop a methodology for data collection and mining algorithms for an information system for assessing students’ health. Within the framework of this scientific problem, several problems are solved related to analysis, search for patterns, visualization of results, and a combination of medical, social, and academic data using artificial intelligence methods. This study will be especially targeted at Kazakh educational institutions. An information system with visualization of the results of aggregation and analysis of data on health-related indicators can be used by various groups of internal and external stakeholders as a regularly updated information resource for the development of various social or medical support programs. The work studied and compared machine learning algorithms based on specially collected data on consultations of medical specialists. In addition, statistical analysis was carried out.

Modern life's ubiquitous component of stress has a significant impact on many facets of human existence. This article presents the development of a wearable device integrated with internet of things (IoT) technology, aiming to identify and quantify stress levels in real-time. This technology provides a possible means of improving stress assessment, enabling prompt treatments and individualized stress management techniques. ESP32-PICO computation platform was used as part of wearable stress monitor. The developed wearable monitor also includes a high-sensitivity pulse oximeter and heart-rate sensor (MAX30102) and galvanic skin response (GSR) sensors to acquire physiological signals associated with stress status. The wearable monitor device delivers data to the firebase platform via Wi-Fi. The benefits and prospective uses of the IoT-enabled wearable device are also covered in the article. It demonstrates the mobile wearable monitor adaptability in a variety of scenarios, such as offices, classrooms, and healthcare facilities, where stress management is vita and required for activity optimization. Continuous monitoring capabilities allow users to learn about their stress levels and take proactive self-care measures. During the validation experiments, the accuracy of measurement capabilities of the developed wearable monitor were evaluated reduced errors of heart rate and respiratory rate being observed.