Aisulu Ataniyazova

Accurate and automated segmentation of ischemic stroke lesions from magnetic resonance imaging (MRI) data is crucial for rapid clinical diagnosis, treatment planning, and patient prognosis. This study presents a comprehensive evaluation of various deep learning architectures for segmenting acute and subacute ischemic lesions using the multicenter ISLES 2022 dataset. We compare traditional 3D U-Net models, attention mechanisms, and modern hybrid architectures, specifically Swin-UNETR (Swin Transformer-based UNet TRansformer). Our experiments utilize multimodal MRI data, including diffusion-weighted images (DWI) and apparent diffusion coefficient (ADC) maps. Results demonstrate that the Swin-UNETR architecture, combined with a specialized pad or crop preprocessing pipeline, advanced data augmentation using the TorchIO library, and a combined Dice–Cross Entropy (DiceCE) loss function, provides competitive performance. The best Swin-UNETR model achieved a Dice of 0.717, precision of 0.793, recall of 0.708, and HD95 of 12.24 mm on the validation set, with a loss of 0.26. The baseline 3D U-Net achieved a comparable Dice of 0.726, while Swin-UNETR showed higher precision, recall, and lower HD95. This study highlights the effectiveness of integrating hierarchical transformers with convolutional decoders for extracting both local and global contextual information in complex medical imaging tasks.

Ischemic stroke remains a leading cause of death and disability. Accurate segmentation of affected areas on brain CT scans plays a crucial role in timely diagnosis and treatment decisions. This study examines an ensemble approach to ischemic stroke foci segmentation on non-contrast CT scans using three 3D U-Net models. CT scans of 50 patients with confirmed acute ischemic stroke (provided by the International Tomography Center, SB RAS) were used as input data. Experiments demonstrated that the 3D U-Net ensemble outperforms each of the individual models in segmentation quality, achieving a Dice coefficient of 0.7008. Analysis of the results demonstrates more accurate delineation of pathological areas and a reduction in segmentation errors. The proposed approach can find application in automated diagnosis of stroke and other diseases requiring accurate segmentation of medical images.

Ischemic stroke is one of the leading causes of mortality and disability. Accurate segmentation of damaged regions in brain CT images is critical for timely diagnosis and clinical decision-making. In this study, an ensemble approach is proposed, combining SE-UNETR and Swin UNETR transformer models via weighted voting. The Dice coefficient was used for evaluation, measuring the overlap between predicted lesion regions and reference annotations. Unlike single-model approaches, ensemble neural network methods provide higher reliability and segmentation accuracy by integrating predictions from multiple architectures. Three-dimensional CT scans of 98 patients with acute ischemic stroke, provided by the International Tomography Center of the Siberian Branch of the Russian Academy of Sciences, were used. The results demonstrated that the proposed ensemble outperforms individual models. The average Dice coefficient was 0.7983, indicating the high effectiveness of the method in segmenting ischemic lesions. Analysis showed that the ensemble approach more accurately delineates lesion boundaries in brain CT images and reduces segmentation errors. The proposed method can be applied not only to stroke but also to other pathologies requiring precise medical image analysis in automated diagnostic systems.

In this paper, the influence of a perpendicular structure in front of parallel buildings on the stabilization of pollutants is investigated, taking into account the porous medium. And to test the model, numerical simulation was carried out formulated using the Navier–Stokes governing equations....

The increasing frequency of extreme hydrological events highlights the critical importance of reliable dam safety assessment and sediment management in large river systems. Three-dimensional numerical modeling of overflow and sediment transport processes is essential for ensuring the reliability of hydraulic structures and understanding the dynamics of reservoir systems. This study focuses on the Shardara Reservoir, located in the Turkestan region of Kazakhstan. The primary objective is to develop a high-resolution 3D numerical model to assess potential variations in water discharge and to analyze the hydrodynamic behavior of the reservoir, including the evaluation of dam breach risks. To simulate multiphase flows, the Volume of Fluid (VOF) method was employed in conjunction with the Pressure-Implicit with Splitting of Operators (PISO) algorithm to solve the Navier–Stokes equations, ensuring computational stability and accuracy under complex flow conditions. Unmanned Aerial Vehicles (UAV) were used to acquire high-resolution topographic data and conduct detailed monitoring of the hydraulic infrastructure. The integration of real-world topography significantly improved the accuracy of boundary conditions and hydrological parameters, enabling precise identification of high-risk zones. The proposed modeling approach is novel in its incorporation of actual terrain features and engineered structures for simulating overflow and dam breach scenarios within this regional context. Numerical experiments demonstrated how changes in discharge affect flow velocity and sediment transport dynamics. The assessment of erosion and sediment transport processes was conducted using a weakly coupled approach, in which flow velocity fields and hydrodynamic regime characteristics were first computed, followed by the analysis of potential erosion and deposition zones based on the Froude criteria. The simulation results showed good agreement with experimental and observational data, confirming the reliability and predictive capability of the developed model.

The duration of continuous operation of unmanned aerial vehicles (UAVs) is significantly limited by the resource of their onboard batteries. This article presents a comprehensive system that provides a fully autonomous drone landing and subsequent charging of its battery, designed to eliminate this limitation. The proposed solution includes an intelligent charging station equipped with a protective canopy and a multimodal guidance system, as well as a specially designed quadcopter optimized for close integration with this station. The station's architecture combines a vision system (an annular multi-camera installation with a 360° view and active LED markers), wireless communication channels and a high-performance real-time computing module that offloads the drone's onboard systems by transferring resource-intensive computer vision and control tasks to the station. The system implements a multi-stage landing algorithm: GPS-based global navigation is combined with precise visual guidance using convolutional neural networks and fiducial markers.; Deep reinforcement learning is used to plan the approach trajectory under the canopy; at the final stage, an adaptive PID controller is used for positioning with an accuracy of several centimeters. A magnetic docking mechanism is provided for secure fixation of the drone during charging, and security protocols provide the possibility of emergency interruption of the process if necessary. During the simulation and testing of the prototype, the system successfully performed autonomous landing in various conditions (including at night and in moderate winds), demonstrating high positioning accuracy and stable operation. The results obtained indicate that the developed autonomous landing and charging system can significantly increase the duration of UAV flights, while reducing the need for operator participation.

Soil salinization poses a significant threat to agricultural productivity and environmental sustainability, particularly in arid and semi-arid regions. This study presents a comprehensive spatiotemporal analysis of soil salinity dynamics in irrigated lands of Alakol District, Zhetisu Region, Kazakhstan, using multi-temporal Sentinel-2 satellite imagery and the Normalized Difference Salinity Index (NDSI). The analysis covered the 2024 growing season, from March to November, with one cloud-free image selected for each month. NDSI values were calculated monthly and classified into four salinity categories: non-saline, slightly saline, moderately saline, and highly saline. Field sampling at 31 locations provided electrical conductivity (EC) data for validation, enabling comparison between surface reflectance-based salinity estimates and ground measurements. The results demonstrated pronounced seasonal trends: NDSI values were lowest in spring due to leaching by precipitation and early irrigation, gradually increasing through summer as evaporation concentrated salts at the surface, and fluctuating in autumn depending on rainfall and drainage conditions. Spatially, fields situated in topographic depressions or near Lake Alakol exhibited the highest salinity levels, whereas upland areas remained relatively unaffected. Notably, no fields exceeded the moderate salinity threshold, indicating that while salinization is present, it remains in early stages. The NDSI approach proved effective for surface salinity detection, capturing both temporal fluctuations and spatial heterogeneity. These findings underscore the utility of remote sensing for operational salinity monitoring and highlight the importance of continuous observation to inform timely land management interventions. This study offers actionable insights for sustainable agriculture, particularly in tailoring irrigation and drainage strategies to mitigate salinity risks across vulnerable farmlands in Central Asia.

Sign languages are an extremely important means of communication in many cases, especially for deaf and hard of hearing people. But the same gesture can convey different meanings in different countries, so many different sign languages have been developed all over the world. In this study, a convolutional neural network (CNN) model was developed based on an ensemble method containing the ResNet-50 and VGG-19 architectures, which will be able to classify the Kazakh sign language (KSL) consisting of 42 Kazakh alphabet signs (classes). A proprietary data set of 57,708 images for 42 signs of the KSL has been formed. The ensemble model was compared with ResNet-50 and VGG-19 by evaluation metrics such as accuracy, precision, recall, f1-measure, and loss function. The recognition accuracy of the ensemble method reached 95.7%, exceeding the performance of ResNet-50 and VGG-19. The developed method was also tested on test data, where 35 out of 42 gestures were recognized completely correctly. The reliability of the proposed approach and the classification results obtained by using preprocessing methods and data augmentation techniques to expand the data set was confirmed by a computational experiment.