Classification of pathologies on digital chest radiographs using machine learning methods

International Journal of Electrical and Computer Engineering (IJECE)
Vol. 14, No. 2, April 2024, pp. 1899~1905
ISSN: 2088-8708, DOI: 10.11591/ijece.v14i2.pp1899-1905  1899
Journal homepage: http://ijece.iaescore.com
Classification of pathologies on digital chest radiographs using
machine learning methods
Murat Aitimov1
, Ainur Shekerbek2
, Igor Pestunov3
, Galitdin Bakanov4
, Aiymkhan Ostayeva5
,
Gulzat Ziyatbekova6
, Saule Mediyeva7
, Gulmira Omarova2
1
Kyzylorda Regional Branch at the Academy of Public Administration under the President of the Republic of Kazakhstan, Kyzylorda,
Republic of Kazakhstan
2
Department of Information Systems, Faculty of Information Technology, L.N. Gumilyov Eurasian National University, Astana,
3
Federal Research Center for Information and Computational Technologies, Novosibirsk, Russia
4
Faculty of Natural Sciences, Akhmet Yassawi International Kazakh-Turkish University, Turkestan, Republic of Kazakhstan
5
Educational Program Informatics and Information and Communication Technologies, Korkyt Ata Kyzylorda University, Kyzylorda,
6
Department of Information Systems, Faculty of Information Technology, Al-Farabi Kazakh National University, Almaty, Republic of
Kazakhstan
7
Department of Language Development Center, Karaganda State Medical University, Karaganda, Republic of Kazakhstan
Article Info ABSTRACT
Article history:
Received Sep 13, 2023
Revised Dec 1, 2023
Accepted Dec 13, 2023
This article is devoted to the research and development of methods for
classifying pathologies on digital chest radiographs using two different
machine learning approaches: the eXtreme gradient boosting (XGBoost)
algorithm and the deep convolutional neural network residual network
(ResNet50). The goal of the study is to develop effective and accurate
methods for automatically classifying various pathologies detected on chest
X-rays. The study collected an extensive dataset of digital chest radiographs,
including a variety of clinical cases and different classes of pathology.
Developed and trained machine learning models based on the XGBoost
algorithm and the ResNet50 convolutional neural network using pre-
processed images. The performance and accuracy of both models were
assessed on test data using quality metrics and a comparative analysis of the
results was carried out. The expected results of the article are high accuracy
and reliability of methods for classifying pathologies on chest radiographs,
as well as an understanding of their effectiveness in the context of clinical
practice. These results may have significant implications for improving the
diagnosis and care of patients with chest diseases, as well as promoting the
development of automated decision support systems in radiology.
Keywords:
eXtreme gradient boosting
Machine learning
Medical imaging texture
Pathology
Residual network
X-rays
This is an open access article under the CC BY-SA license.
Corresponding Author:
Ainur Shekerbek
Department of Information Systems, Faculty Information Technology, L. N. Gumilyov Eurasian National
University
010000 Astana, Republic of Kazakhstan
Email: shekerbek80@mail.ru
1. INTRODUCTION
Modern medical care is faced with the need to effectively and accurately detect pathologies on chest
X-rays. Digital radiographs provide extensive information about the state of the respiratory and cardiovascular
systems, but their analysis requires high qualifications and time on the part of medical specialists [1]. In this
work, machine learning methods became a key tool to automate and improve the process of classifying
pathologies in radiographs. One of the key challenges in this field is the automatic classification of pathologies

 ISSN: 2088-8708
Int J Elec & Comp Eng, Vol. 14, No. 2, April 2024: 1899-1905
1900
in chest X-rays [2]–[4] with high accuracy and efficiency. Correct and rapid diagnosis of chest diseases such as
lung cancer, tuberculosis, and other pathologies is critical for successful treatment and increasing patient
survival. Machine learning techniques such as the eXtreme gradient boosting (XGBoost) algorithm and deep
neural networks provide new opportunities for the automatic diagnosis and classification of these diseases.
In this paper, we present the results of a study on pathology classification in digital chest
radiographs using two different machine learning methods: XGBoost [5]–[7] and the deep convolutional
neural network residual network (ResNet50) [8]–[10]. We compare the performance and accuracy of both
models and consider their advantages and limitations in the context of clinical practice. The main objective of
this work is to find out which method is best suited to classify pathologies on chest radiographs, taking into
account accuracy, sensitivity, and specificity. The results obtained will contribute to the development of
automated diagnostic and disease monitoring systems, improving the quality of medical care and speeding up
the decision-making process in clinical practice.
Koohbanani et al. [11] proposes a self-supervised convolutional neural network (CNN) structure
that allows the use of unlabeled data to explore generalizable and domain-invariant representations in
pathology images. The proposed structure, called Self-Path, uses multi-task learning, where the main task is
tissue classification, and the pre-text tasks are various self-supervision tasks with labels inherent in the input
images. They present novel pathology-specific self-monitoring tasks that use contextual, layered, and
semantic features in pathology images for semi-supervised learning and domain adaptation. Azizi et al. [12]
examines the effectiveness of self-guided learning as a pre-learning strategy for medical image classification.
Conducted experiments on two different tasks: classifying dermatological conditions from digital camera
images and classifying chest x-rays with multiple labels, and demonstrated that self-guided learning on
ImageNet followed by additional self-supervised learning on domain-specific unlabeled medical images
significantly improved the accuracy of medical image classifiers.
Chen et al. [13] proposes a novel label co-occurrence learning framework based on graph
convolution networks (GCNs) for explicitly learning dependencies between pathologies for a multi-labeled
chest X-ray (CXR) classification problem, which it calls chest X-ray graph convolution networks
(CheXGCN). Ho and Gwak [14] precise localization and classification of lung anomalies on X-ray images
are important for clinical diagnosis and treatment strategy. However, multi-label classification, in which
medical images are interpreted to identify multiple existing or suspected pathologies, presents practical
limitations. Zheng et al. [15] paper proposes a method for classifying cardiac pathology based on a novel
approach to image-derived feature extraction to characterize the shape and movement of the heart. An
original semi-supervised learning procedure that makes efficient use of a large number of unsegmented
images and a small number of manually segmented images by experts is designed to generate a pixel-by-
pixel visible stream between two 2D+t time points.
In study [16], the study examined the applicability of artificial intelligence (AI) for dental
radiography based on current research. They classified AI applications based on the similarity of the
following purposes: diagnosis of dental caries, periapical pathologies, and periodontal bone loss;
classification of cysts and tumors; cephalometric analysis; osteoporosis screening; tooth recognition and
forensic odontology; recognition of dental implant systems; and improving image quality. The goal of the
study [17] is to evaluate the real-time diagnostic performance of the deep convolutional neural network You
Only Look Once (YOLO) v2, a deep learning algorithm that can simultaneously detect and classify an object,
on panoramic images. radiographs. The images were classified and labeled into four categories:
dentoalveolar cysts, odontogenic keratocysts, ameloblastomas, and no lesions. The results of this study
indicate the usefulness of automatic detection of convolutional networks for detecting certain pathologies and
therefore preventing morbidity in the field of oral and maxillofacial surgery. The work [18] applies various
image preprocessing techniques. The input orthopantomogram (OPG) images are resized, pixels are scaled,
and erroneous data is eliminated. The proposed algorithm is implemented using CNN with Dropout and the
fully connected layer is trained using Genetic algorithms Back propagation error algorithm (GA-BP) hybrid
learning. Using the Dropout regularization technique, overfitting was avoided, which allowed the network to
correctly classify objects. The CNN was implemented with different convolutional layers, and the highest
accuracy of 97.92% was obtained with two convolutional layers.
In a retrospective study [19], the you only look once (YOLO) algorithm was used to obtain hand
images from original radiographs without data loss, and classification was performed by applying transfer
learning with a trained visual geometry group (VGG-16) network. During training, the data augmentation
method was used. Barisoni et al. [20] shows that the field of computer vision can now be effectively applied
to histopathological objects by people who do not have in-depth knowledge of computer vision techniques.
While these new approaches have already advanced disease detection, classification, and prognosis in the
fields of radiology and oncology, renal pathology is just entering the digital age with the creation of consortia
and digital pathology repositories to collect, analyze, and integrate pathology data with other domains.

Int J Elec & Comp Eng ISSN: 2088-8708 
Classification of pathologies on digital chest radiographs using machine learning methods (Murat Aitimov)
1901
2. METHOD
In this work, to solve the classification problem, all input images were initially segmented using the
convolutional network for biomedical image segmentation (U-Net) model [21]–[23], which is a deep neural
network specifically designed for image segmentation tasks in medical applications [24]–[26]. The main idea
of U-Net is to create an efficient architecture for feature extraction and high-quality segmentation while
preserving spatial information. The model has two main parts: an encoder and a decoder. When a model is
trained on a large enough amount of labeled data, it can highlight areas of interest in images and create masks
that can be used for further image analysis and processing. It should be noted that the use of U-Net in
medical problems requires a lot of attention to additional aspects, such as data augmentation, class balancing,
and model adaptation to specific conditions and data types. Using masks generated by the unified neural
network (U-Net) model provides a powerful tool for automatic segmentation and cropping of images as
shown in Figure 1. These masks, created using convolutional neural networks, can highlight objects of
interest in images, such as lungs in X-rays, and then automatically crop the image around those objects.
Figure 1. Architecture for creating masks using a trained model U-Net
To implement pathology classification on digital chest radiographs, machine learning methods such
as the XGBoost algorithm and the ResNet50 deep convolutional neural network were used. The experiment
collected datasets with chest X-rays and associated medical annotations. Augmentation techniques were
applied to increase the size of the dataset and improve model training. Data normalization was also carried
out for data preprocessing. When training the models, the data was divided into training, validation, and test
sets. The XGBoost machine learning method extracts key features from X-ray images using computer vision
techniques or simple statistics. The XGBoost model is trained on training data and tuned on hyperparameters
for better performance. The machine learning method logistic regression is a statistical method used to model
the probability of a binary or multi-class event. In this work, it is used for classification tasks. When
classifying pathologies on chest radiographs, logistic regression was used to model the likelihood of the
presence or absence of a particular pathology.

 ISSN: 2088-8708
1902
For the ResNet50 method, we applied a pre-trained convolutional neural network with 50 layers.
The model was further trained on our data to extract high-level features. The performance of both models
was evaluated using various metrics such as precision, recall, F1-measure, and receiver operating
characteristic (ROC) curve on a test dataset. We conducted a comparative analysis of classification results,
comparing the performance of the XGBoost and ResNet50 methods. Using two different machine learning
methods allowed us to evaluate their performance and select the most appropriate method for the task of
classifying pathologies on chest X-rays. The application of these methods should allow the creation of an
effective system for classifying pathologies on digital chest radiographs.
3. RESULTS AND DISCUSSION
We consider a comparison of two popular machine learning methods, XGBoost and ResNet50, in
the task of automatic diagnosis of pneumonia based on fluorographic images. Both algorithms were trained
on a set of 3,736 pneumonia images and 1,342 normal fluorographic images. The original images had a high
resolution (3000×3000 pixels) and were pre-segmented using a neural network to highlight lung areas. The
images were then scaled to 250×250 pixels for training the models. Preprocessing is a critical step in any
machine learning project, and in this case, it involved two key steps such as segmentation and resizing. To
improve accuracy, a segmentation method performed using a neural network was applied. This stage made it
possible to get rid of areas of the image that were irrelevant for diagnosis, leaving only fragments of the
lungs. After segmentation, the images were scaled to 250×250 pixels to speed up the training process and
reduce computational requirements.
The XGBoost method used was shown to have good performance and accuracy in classifying
pathologies on radiographs. This model showed excellent performance on training data, achieving 100%
accuracy. However, on the validation set the indicators were lower: accuracy was 85.83%, and F1-Score was
85.46% as shown in Figure 2. The performance of the XGBoost method showed good results, which
indicates its applicability in the task of classifying pathologies on radiographs. However, it should be noted
that this method relies on manual feature extraction, which may require more effort in data preprocessing.
Figure 2. XGBoost model training accuracy results
Performance ResNet50 exhibited outstanding performance and handled the classification task with
high accuracy. This model also achieved 100% accuracy on training data. Unlike XGBoost, ResNet50
showed significantly better results on the validation set: accuracy - 96.09%, F1-Score - 96.05% as shown in
Figure 3. Its ability to extract high-level features from images allowed it to achieve high results without the
need to manually define features.
It is important to note that both models showed high values of the ROC area under the curve (AUC)
metric, which indicates their good ability to discriminate between classes. However, ResNet50's higher
F1-Score and accuracy indicate its better performance, possibly due to a better balance between sensitivity
and specificity. Having compared the performance of XGBoost and ResNet50 in the task of diagnosing
pneumonia on fluorographic images, we can conclude that complex models specialized in working with

1903
images can show better results in such tasks. Data preprocessing, including segmentation and scaling, also
has a significant impact on model performance. Despite this, overtraining remains an important aspect that
requires further research and correction.
Figure 3. ResNet50 model training accuracy results
4. CONCLUSION
In this article, a study was conducted on the classification of pathologies in digital chest radiographs
using machine learning methods, in particular, XGBoost and the deep convolutional neural network ResNet
50. The study confirmed that both the machine learning method and deep neural networks have the potential
to successfully classification of pathologies on chest X-rays. Both methods achieved high precision and
recall, but ResNet 50 showed more outstanding performance. Deep neural networks such as ResNet 50 have
shown their ability to extract high-level features from images, which significantly improves classification
accuracy. This makes them a powerful tool in radiology for automated diagnosis.
Future research could focus on improving data preprocessing techniques, as well as collecting larger
datasets to improve model training and generalization. The results of this study may be useful for radiologists
and medical specialists by providing them with tools to automate the process of diagnosing pathologies on
radiographs. Overall, this study highlights the importance of machine learning techniques, especially deep
learning, in the field of medical diagnostics and provides a basis for further research and development of
intelligent systems for medical image analysis.
REFERENCES
[1] A. Shekerbek, S. Serikbayeva, M. Tulenbayev, G. Bakanov, S. Beglerova, and A. Makovetskaya, “Application of mathematical
methods and machine learning algorithms for classification of X-Ray images,” Eastern-European Journal of Enterprise
Technologies, vol. 3, pp. 6–17, Jun. 2022, doi: 10.15587/1729-4061.2022.259710.
[2] G. Abdikerimova et al., “Detection of chest pathologies using autocorrelation functions,” International Journal of Electrical and
Computer Engineering (IJECE), vol. 13, no. 4, pp. 4526–4534, Aug. 2023, doi: 10.11591/ijece.v13i4.pp4526-4534.
[3] A. Naizagarayeva et al., “Detection of heart pathology using deep learning methods,” International Journal of Electrical and
Computer Engineering (IJECE), vol. 13, no. 6, pp. 6673–6680, Dec. 2023, doi: 10.11591/ijece.v13i6.pp6673-6680.
[4] A. Orazayeva, J. Tussupov, W. Wójcik, S. Pavlov, G. Abdikerimova, and L. Savytska, “Methods for detecting and selecting areas
on texture biomedical images of breast cancer,” Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Srodowiska,
vol. 12, no. 2, pp. 69–72, Jun. 2022, doi: 10.35784/iapgos.2951.
[5] D. A. Dias Júnior et al., “Automatic method for classifying COVID-19 patients based on chest X-ray images, using deep features
and PSO-optimized XGBoost,” Expert Systems with Applications, vol. 183, Nov. 2021, doi: 10.1016/j.eswa.2021.115452.
[6] A. Bhattacharyya, D. Bhaik, S. Kumar, P. Thakur, R. Sharma, and R. B. Pachori, “A deep learning based approach for automatic
detection of COVID-19 cases using chest X-ray images,” Biomedical Signal Processing and Control, vol. 71, Jan. 2022, doi:
10.1016/j.bspc.2021.103182.
[7] R. Rajagopal, R. Karthick, P. Meenalochini, and T. Kalaichelvi, “Deep convolutional spiking neural network optimized with
arithmetic optimization algorithm for lung disease detection using chest X-ray images,” Biomedical Signal Processing and
Control, vol. 79, Jan. 2023, doi: 10.1016/j.bspc.2022.104197.
[8] M. Yessenova et al., “Application of informative textural Law’s masks methods for processing space images,” International Journal
of Electrical and Computer Engineering (IJECE), vol. 13, no. 4, pp. 4557–4566, Aug. 2023, doi: 10.11591/ijece.v13i4.pp4557-4566.
[9] M. Hasan, M. Al Mamun, M. C. Das, M. M. Hasan, and A. M. Islam, “The application and comparison of deep learning models
for the prediction of chest cancer prognosis,” 2023 International Conference on Smart Applications, Communications and

 ISSN: 2088-8708
1904
Networking (SmartNets), Istanbul, Turkiye, 2023, pp. 1-5, doi: 10.1109/SmartNets58706.2023.10216201.
[10] A. Elhanashi, S. Saponara, and Q. Zheng, “Classification and localization of multi-type abnormalities on chest X-Rays images,”
IEEE Access, vol. 11, pp. 83264–83277, 2023, doi: 10.1109/ACCESS.2023.3302180.
[11] N. A. Koohbanani, B. Unnikrishnan, S. A. Khurram, P. Krishnaswamy, and N. Rajpoot, “Self-path: self-supervision for
classification of pathology images with limited annotations,” IEEE Transactions on Medical Imaging, vol. 40, no. 10,
pp. 2845–2856, Oct. 2021, doi: 10.1109/TMI.2021.3056023.
[12] S. Azizi et al., “Big self-supervised models advance medical image classification,” in 2021 IEEE/CVF International Conference
on Computer Vision (ICCV), Oct. 2021, pp. 3458–3468, doi: 10.1109/ICCV48922.2021.00346.
[13] B. Chen, J. Li, G. Lu, H. Yu, and D. Zhang, “Label co-occurrence learning with graph convolutional networks for multi-label
chest X-ray image classification,” IEEE Journal of Biomedical and Health Informatics, vol. 24, no. 8, pp. 2292–2302, Aug. 2020,
doi: 10.1109/JBHI.2020.2967084.
[14] T. K. Ho and J. Gwak, “Multiple feature integration for classification of thoracic disease in chest radiography,” Applied Sciences,
vol. 9, no. 19, Oct. 2019, doi: 10.3390/app9194130.
[15] Q. Zheng, H. Delingette, and N. Ayache, “Explainable cardiac pathology classification on cine MRI with motion characterization by
semi-supervised learning of apparent flow,” Medical Image Analysis, vol. 56, pp. 80–95, 2019, doi: 10.1016/j.media.2019.06.001.
[16] R. H. Putra, C. Doi, N. Yoda, E. R. Astuti, and K. Sasaki, “Current applications and development of artificial intelligence for
digital dental radiography,” Dentomaxillofacial Radiology, vol. 51, no. 1, Jan. 2022, doi: 10.1259/DMFR.20210197.
[17] H. Yang et al., “Deep learning for automated detection of cyst and tumors of the jaw in panoramic radiographs,” Journal of
Clinical Medicine, vol. 9, no. 6, pp. 1–14, Jun. 2020, doi: 10.3390/jcm9061839.
[18] A. Laishram and K. Thongam, “Automatic classification of oral pathologies using orthopantomogram radiography images based
on convolutional neural network,” International Journal of Interactive Multimedia and Artificial Intelligence, vol. 7, no. 4,
pp. 69–77, 2022, doi: 10.9781/ijimai.2021.10.009.
[19] K. Üreten and H. H. Maraş, “Automated classification of rheumatoid arthritis, osteoarthritis, and normal hand radiographs with
deep learning methods,” Journal of Digital Imaging, vol. 35, no. 2, pp. 193–199, Jan. 2022, doi: 10.1007/s10278-021-00564-w.
[20] L. Barisoni, K. J. Lafata, S. M. Hewitt, A. Madabhushi, and U. G. J. Balis, “Digital pathology and computational image analysis
in nephropathology,” Nature Reviews Nephrology, vol. 16, no. 11, pp. 669–685, Aug. 2020, doi: 10.1038/s41581-020-0321-6.
[21] J. Zhang, Y. Zhang, Y. Jin, J. Xu, and X. Xu, “MDU-Net: multi-scale densely connected U-Net for biomedical image
segmentation,” Health Information Science and Systems, vol. 11, no. 1, Mar. 2023, doi: 10.1007/s13755-022-00204-9.
[22] H. Song, Y. Wang, S. Zeng, X. Guo, and Z. Li, “OAU-net: outlined attention U-Net for biomedical image segmentation,”
Biomedical Signal Processing and Control, vol. 79, Jan. 2023, doi: 10.1016/j.bspc.2022.104038.
[23] X. Zhou et al., “CUSS-Net: a cascaded unsupervised-based strategy and supervised network for biomedical image diagnosis and
segmentation,” IEEE Journal of Biomedical and Health Informatics, vol. 27, no. 5, pp. 2444–2455, May 2023, doi:
10.1109/JBHI.2023.3238726.
[24] H. Xiao, L. Li, Q. Liu, X. Zhu, and Q. Zhang, “Transformers in medical image segmentation: A review,” Biomedical Signal
Processing and Control, vol. 84, Jul. 2023, doi: 10.1016/j.bspc.2023.104791.
[25] Y. Zhang, T. Zhou, S. Wang, P. Liang, Y. Zhang, and D. Z. Chen, “Input augmentation with SAM: boosting medical image
segmentation with segmentation foundation model,” in Lecture Notes in Computer Science, Springer Nature Switzerland, 2023,
pp. 129–139.
[26] A. Bayegizova et al., “Effectiveness of the use of algorithms and methods of artificial technologies for sign language recognition
for people with disabilities,” Eastern-European Journal of Enterprise Technologies, vol. 4, no. 2–118, pp. 25–31, Aug. 2022, doi:
10.15587/1729-4061.2022.262509.
BIOGRAPHIES OF AUTHORS
Murat Aitimov PhD in philosophy of instrumentation. Dr. Aitimov Murat is an
accomplished scholar with a rich academic background. He currently serves as the Director of
the Kyzylorda Regional Branch at the Academy of Public Administration under the President
of the Republic of Kazakhstan. With an impressive 28 years of experience in both scientific
research and pedagogy, Dr. Murat is a recognized authority in his field. His contributions to
academia are significant, as evidenced by his substantial body of work. He has authored 55
scientific articles, a testament to his dedication to advancing knowledge. Notably, 7 of his
articles have been featured on Scopus, highlighting their impact and relevance in the global
research community. In addition to his articles, Dr. Murat has authored three influential books.
He also holds an innovation patent in the realm of instrumentation and information systems,
showcasing his multidimensional expertise. His expertise and insights continue to drive
progress in the field of instrumentation, making him a valuable resource for researchers and
academics alike. For those seeking to connect with Dr. Aitimov Murat, he can be reached at
email: aitimovmurat07@gmail.com.
Ainur Shekerbek received a bachelor's degree in computer science in 2002 from
Taraz State University named after M.Kh. Dulaty. In 2005, she received a master's degree in
applied mathematics from the South Kazakhstan State University named after M. Auezov,
Kazakhstan, Shymkent. Currently, she is a doctoral student at the Department of Information
Systems of the Eurasian National University. L.N. Gumilev. Her research interests include
image processing, computer vision, radiography, artificial intelligence and machine learning.
She can be contacted at email: shekerbek80@mail.ru.

1905
Igor Pestunov graduated from Novosibirsk State University in 1977. Candidate
of Physical and Mathematical Sciences (1998). He is a leading researcher at the Federal
Research Center for Information and Computational Technologies. His current research
interests include clustering, pattern recognition, and image analysis. He can be contacted at
email: pestunov@ict.nsc.ru.
Galitdin Bakanov graduated from al-Farabi Kazakh National University in 1980.
Doctor of Physical and Mathematical Sciences (2002). He is a professor at the Akhmet
Yassawi International Kazakh-Turkish University. His current research interests include
inverse and ill-posed problems, computed tomography problems. He can be contacted at
email: galitdin.bakanov@ayu.edu.kz.
Aiymkhan Ostayeva candidate of Pedagogical Sciences, currently works as a
senior lecturer of the educational program “Informatics and information and communication
technologies” at Korkyt Ata Kyzylorda University. She is the author of more than 75 scientific
papers in the field of computer science, information technology, computer engineering,
including 4 articles in the Scopus database and 3 articles in Web of Science. She can be
contacted at email: aimak73@mail.ru.
Gulzat Ziyatbekova Ph.D. in computer science, computer engineering and
management, associate professor, a senior researcher at the RSE Institute of Information and
Computational Technologies of the Committee of Science of the Ministry of Science and
Higher Education of the Republic of Kazakhstan, Associate Professor Al-Farabi Kazakh
National University. She has co-authored more than 40 articles in journals and conference
proceedings. Her professional interests are information systems, computer science, scientific
research, mathematical modeling. She can be contacted at email: ziyatbekova@mail.ru.
Saule Mediyeva graduated from the Kazakh state university of International
Relations and World languages in 1994 in Almaty. In 2000, she graduated from the Karaganda
state university named after Buketov. In 2018 she received a master's degree in applied
pedagogical sciencies from the Karaganda state university named after Buketov. More than 25
years has been working at the Karaganda state medical university. Her research and practical
interests include the studying and analysis of problems connected with the student self- work and
controlling. She can be contacted at emai: Medieva@qmu.kz, saule.medieva.1972@mail.ru.
Gulmira Omarova received her Ph.D. in 2023 in Information Systems from L.N.
Gumilyov Eurasian National University, Kazakhstan. Currently, she is a senior lecturer at the
Department of Information Systems at the same university. Her research interests include
image processing, computer vision, medical images, artificial intelligence and machine
learning. She can be contacted at email: ogs12@mail.ru.

Classification of pathologies on digital chest radiographs using machine learning methods

More Related Content

Similar to Classification of pathologies on digital chest radiographs using machine learning methods (20)

More from IJECEIAES (20)

Recently uploaded (20)

Classification of pathologies on digital chest radiographs using machine learning methods