Pneumonia detection in chest X-ray images using convolutional neural networks

Abstract

Objectives: Pneumonia ranks among the infections and presents a considerable health threat, especially in certain age groups and developing countries. The accurate diagnosis of the disease and prompt identification are crucial for treatment purposes. This study aimed to develope a convolutional deep neural network model that can detect pneumonia using a sufficient number of chest X-ray images that have been verified with a "definite diagnosis" clinically.

Methods: This study uses a dataset that includes 1000 chest X-ray images from a variety of age groups taken as part of patient care at Koç University Faculty of Medicine Hospital Clinics. The dataset sample includes two sets of pictures called normal and pneumonia infected. Various preprocessing techniques were used on the obtained images, thus enabling the training and testing of our developed prediction model.

Results: We improved the accuracy of the model's decisions by applying image processing techniques, successfully achieving high levels of decision accuracy with our model We have elevated the precision of decision-making in our model to outstanding levels and achieved impressive F1 Score and AUC (Area Under the Curve) values (F1 Score: 0.94 and AUC Score: 0.98).

Conclusions: Our model was trained using X-ray images produced from the same devices of the same hospital and achieved very high prediction results, but using images produced from different countries, different hospitals and different devices, especially training and testing the model with much larger data sets, is a necessary need for this study and the model we developed to become more universal, and in this sense, there is a need to develop and expand the study.

Keywords

• Pneumonia
• convolutional neural network
• chest X-ray
• deep neural network
• machine learning
• deep learning
• artificial intelligence

Ethical Statement

This study was approved by the Koç University Biomedical Research Ethics Committee (Decision no. 2025.300.IRB2.141, date: 30.06.2025).

References

1. 1. Marshall DC, Goodson RJ, Xu Y, et al. Trends in mortality from pneumonia in the Europe union: a temporal analysis of the European detailed mortality database between 2001 and 2014. Respir Res. 2018;19(1):81. doi: 10.1186/s12931-018-0781-4.
2. 2. World Health Organization (WHO). Pneumonia in children. 2022. cited 2025 Feb 20. Available from: https://www.who.int/news-room/fact-sheets/detail/pneumonia
3. 3. Mandell LA, Wunderink RG, Anzueto A, et al; Infectious Diseases Society of America; American Thoracic Society. Infectious Diseases Society of America/American Thoracic Society consensus guidelines on the management of community-acquired pneumonia in adults. Clin Infect Dis. 2007;44 Suppl 2(Suppl 2): S27-72. doi: 10.1086/511159.
4. 4. Bartlett JG. Diagnostic tests for agents of community-acquired pneumonia. Clin Infect Dis. 2011;52 Suppl 4: S296-304. doi: 10.1093/cid/cir045.
5. 5. Mick E, Tsitsiklis A, Kamm J, et al. Integrated host/microbe metagenomics enables accurate lower respiratory tract infection diagnosis in critically ill children. J Clin Invest. 2023 Apr 3;133(7):e165904. doi: 10.1172/JCI165904.
6. 6. Metlay JP, Waterer GW, Long AC, et al. Diagnosis and Treatment of Adults with Community-acquired Pneumonia. An Official Clinical Practice Guideline of the American Thoracic Society and Infectious Diseases Society of America. Am J Respir Crit Care Med. 2019;200(7): e45-e67. doi: 10.1164/rccm.201908-1581ST.
7. 7. File TM Jr, Ramirez JA. Community-Acquired Pneumonia. N Engl J Med. 2023;389(7):632-641. doi: 10.1056/NEJMcp2303286.
8. 8. He K, Zhang X, Ren S, Sun J. Deep Residual Learning for Image Recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, 2016, pp. 770-778, doi: 10.1109/CVPR.2016.90.

9. 9. Huang G, Liu Z, van der Maaten L, Weinberger KQ. Densely connected Convolutional Networks. Densely Connected Convolutional Networks. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA, 2017, pp. 2261-2269, doi: 10.1109/CVPR.2017.243.
10. 10. Chollet F. Xception: Deep Learning with Depthwise Separable Convolutions. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA, 2017, pp. 1800-1807, doi: 10.1109/CVPR.2017.195.
11. 11. Howard AG, Zhu M, Chen B, et al. Mobilenets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv:1704.04861. 2017 Apr 17.
12. 12. Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D. Grad-cam: Visual explanations from deep networks via gradient-based localization. 2017 IEEE International Conference on Computer Vision (ICCV), Venice, Italy, 2017, pp. 618-626, doi: 10.1109/ICCV.2017.74.
13. 13. Hossin M, Sulaiman MN. A review on evaluation metrics for data classification evaluations. International Journal of Data Mining & Knowledge Management Process. 2015;5(2):1-11. doi: 10.5121/ijdkp.2015.5201.
14. 14. Christen P, Hand DJ, Kirielle N. A Review of the F-Measure: Its History, Properties, Criticism, and Alternatives. ACM Comput. Surv. 2024;56(3):73. doi: 10.1145/3606367.