OPTIMIZED DEEP LEARNING APPROACH FOR PNEUMONIA DETECTION USING CHEST X-RAY IMAGES
DOI:
https://doi.org/10.4314/njt.2026.5954SIKeywords:
Pneumonia, Classification, Deep Learning, Convolution Neural Network (CNN), Neural Networks, EfficientNetB3 Architecture, Radiological X-Ray Images, DetectionAbstract
Pneumonia is a virus that ranges from moderate to severe in that there is inflammation of the air sacs located in the lungs, the symptoms of which are fever, cough and difficulty in breathing. Pneumonia may be due to bacteria or viruses. Treatment will generally be by means of antibiotics or antiviral medicines, according to the cause. Preventive medicine would consist of vaccination and good hygiene. Deep learning, a subset of artificial intelligence, has proved to be a mainstay for the development of predictive models. Detection of pneumonia may be made by many methods, such as CT and pulse oximetry, but X-ray tomography is the method mostly employed. The interpretation of chest X-rays (CXR) is an inherently subjective process. In this study we investigate the field of pneumonia lung classification, using images of CXR. The data set consists of 5863 images all of which are labeled as pneumonia positive and normal. This work confines itself to a particular anatomical region as regards the analysis of the disease, so that therefore the performance of pneumonia detection will be studied in detail. The study focuses on the comparative analysis of deep learning models with special emphasis on traditional CNN as compared to techniques of Transfer Learning such as Inception Net, VGG16 and ResNet. The implementation suggested underlines a granular understanding of model effectiveness which was able to jointly reveal that CNN using EfficientNetB3 architecture resulted in higher performance characteristic to pneumonia detection for the dataset selected. We highlight, through intensive experimentation and relative comparisons, the advantages of leveraging base frameworks of CNNs rather than transfer learning paradigms in the specific detection of pneumonia from CXR. The designed CNN architecture has also achieved a commendable accuracy rate of 99.60% compared to other deep learning techniques and existing models on the same dataset.
References
[1] Adil, M., Royaidar, J., Yassa, R., GonzagaLeong-on, M.,Iqbal, F., Hussain, A., Ali, Q. and Rasheed, A. "Epidemiology and Resistance Pattern in Microbial Pneumonia: A Review: Epidemiology and Resistance Pattern in Microbial Pneumonia." Pakistan Journal of Health Sciences, 3(5), pp.27-31, 2022. https://doi.org/10.54393/pjhs.v3i05.229
[2] Pitre, T., Ellen,P., Chaudhuri,D., Saha, R., Rudd, K.,Villar, J., Berry, L. "Corticosteroids for adult patients hospitalised with non-viral community-acquired pneumonia: a systematic review and meta-analysis", Intensive Care Medicine, 51(1), pp.917-929, 2025. https://doi.org/10.1007/s00134-025-07912-2
[3] Sable,N., Rathod, V., Deshmukh,J., and Sonavane, S., "Significance of Blockchain Technology in Industrial Applications with an Emphasis on Security Considerations", In Machine Learning for Environmental Monitoring in Wireless Sensor Networks, Hershey, PA, USA: IGI Global, pp.305-330, 2025. doi: 10.1016/j.rser.2025.116302.
[4] Ahmed, R. "Human Metapneumovirus (HMPV): Epidemiology, Pathophysiology, Clinical Management, and Future Prospects in Treatment and Prevention", Radinka Journal of Health Science, 2(3), pp. 312-327, 2025. https://doi.org/10.56778/rjhs.v2i3.417
[5] Andriani, K., and Debi, S., "Semantic Literature Analysis and Deep Learning Algorithm Comparison for Identifying Paru-Paru Diseases from X-rays", JMATech-Journal of Medical Ai & Technology, 1(1), pp.7-10, 2026.
[6] Alshanketi, F., Abdulrahman, A., Mathew, K., Vahid, M., Shams, T.,Rana, N., and Ali, T., "Pneumonia detection from chest x-ray images using deep learning and transfer learning for imbalanced datasets", Journal of imaging informatics in medicine, 38(4), pp. 2021-2040, 2025. https://doi.org/10.1007/s10278-024-01334-0
[7] Lestari, D., Anggi, M., Aghnia, T., Ryando, R., and Habibah, D., "Deep Learning for Pneumonia Detection in Chest X-Rays using Different Algorithms and Transfer Learning Architectures", Public Research Journal of Engineering, Data Technology and Computer Science, 3(1), pp.10-21, 2025. https://doi.org/10.57152/predatecs.v3i1.1553
[8] Shaikh, M., Qasim, A., and Rabeea,J., "AG-FSNet: A hybrid attention and feature selection network with feature contribution scoring for pneumonia segmentation using CXR images", Biomedical Signal Processing and Control, 114 (1), pp.109248-109285, 2026. https://doi.org/10.1016/j.bspc.2025.109248
[9] Sadiq, B., "Maximizing network capacity, control and management in designing a Telemedicine network: a review and recent challenges", Nigerian Journal of Technology, 43(1), pp. 80-100, 2024.
[10] Rathod, V., Shyamrao, V., and Himadri, M., "The Role of Congestion-Adaptive Routing Protocol in Mobile Ad Hoc Networks." The Scientific World Journal, 1(1), pp. 6296169, 2025. https://doi.org/10.1155/tswj/6296169
[11] Waisi, N., "Comparative Evaluation of Resnet-50 and Efficientnet-B1 for Pneumonia Detection in Chest X-Ray Images Using Transfer Learning", International Journal of Computational & Electronic Aspects in Engineering, 6(2), 2025. https://doi.org/10.26706/ijceae.6.2.20250405
[12] Rathod, V., Gumaste,V., Guttula, R., Zade, S., and Singh, R., "Optimization of energy consumption in mobile ad-hoc networks with a swarm intelligence-based ABC algorithm", Discover Applied Sciences, 8, pp. 805, 2025. https://doi.org/10.1007/s42452-025-07472-6
[13] Khojare, S., and Bhattacharjee, S., "Pneumonia Identification based on Transfer Learning over Chest X-Ray Images", In 2025 International Conference on Emerging Trends in Industry 4.0 Technologies (ICETI4T), IEEE, pp. 1-6, 2025. doi: 10.1109/ICETI4T63625.2025.11132185
[14] Kumar, P.,Verma, S., Shubhanshi, R., Yadav, J., and Sing, S., "Quantitative Analysis of Lung Opacities on Routine Chest X-Ray Radiograph", International Journal of Scientific Research and Technology, 3(1), pp.144-151, 2026. https://doi.org/10.5281/zenodo.18220096
[15] Rabbah,J., Mohammed, R., and Larbi, H., "Improving pneumonia diagnosis with high-accuracy CNN-Based Chest X-ray image classification and integrated gradient", Biomedical Signal Processing and Control, 101(1), pp.107239, 2025. https://doi.org/10.1016/j.bspc.2024.107239
[16] Abdalla, K., Fatemeh, D., Eghbal, H., and Mourad, O., "HyOCNN: Hybrid-optimized convolutional neural network for robust image classification", Biomedical Signal
Processing and Control, 113 (1), pp.109040, 2026. https://doi.org/10.1016/j.bspc.2025.109040
[17] Widhiyanti, E., Guruh, F., Aisyah, N., Talitha, O., and Edi, J., "Comparative Analysis of CNN and Fine-Tuned InceptionV3 with and Without Transposition Layer for Chest X-ray Image Pneumonia Detection", In 2025 International Seminar on Application for Technology of Information and Communication (iSemantic), IEEE, pp. 127-132, 2025. doi: 10.1109/ISemantic67418.2025.11291892
[18] Waisi, N., "Comparative Evaluation of Resnet-50 and Efficientnet-B1 for Pneumonia Detection in Chest X-Ray Images Using Transfer Learning", International Journal of Computational & Electronic Aspects in Engineering (IJCEAE), 6(2), pp.89-97, 2025. https://doi.org/10.26706/ijceae.6.2.20250405
[19] Dataset: P. Mooney, “Chest X-Ray Images (Pneumonia),” Kaggle, 5,863 images, 2 categories, 2018. [Online]. Available: https://www.kaggle.com/datasets/paultimothymooney/chest-xray-pneumonia
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Nigerian Journal of Technology
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
The contents of the articles are the sole opinion of the author(s) and not of NIJOTECH.
NIJOTECH allows open access for distribution of the published articles in any media so long as whole (not part) of articles are distributed.
A copyright and statement of originality documents will need to be filled out clearly and signed prior to publication of an accepted article. The Copyright form can be downloaded from http://nijotech.com/downloads/COPYRIGHT%20FORM.pdf while the Statement of Originality is in http://nijotech.com/downloads/Statement%20of%20Originality.pdf
For articles that were developed from funded research, a clear acknowledgement of such support should be mentioned in the article with relevant references. Authors are expected to provide complete information on the sponsorship and intellectual property rights of the article together with all exceptions.
It is forbidden to publish the same research report in more than one journal.
How to Cite
