Detecting respiratory diseases in chest x-ray images using convolutional neural network (CNN)

Edgar M. Adina; Mylen L. Aala-Capuno; Hamil Josef P. Roleda; Divina A. Tuason

doi:10.1117/12.3012651

19 December 2023 Detecting respiratory diseases in chest x-ray images using convolutional neural network (CNN)

Edgar M. Adina, Mylen L. Aala-Capuno, Hamil Josef P. Roleda, Divina A. Tuason

Proceedings Volume 12936, International Conference on Mathematical and Statistical Physics, Computational Science, Education and Communication (ICMSCE 2023); 129360A (2023) https://doi.org/10.1117/12.3012651
Event: International Conference on Mathematical and Statistical Physics, Computational Science, Education and Communication (ICMSCE 2023), 2023, Istanbul, Turkey

Abstract

Respiratory diseases such as pneumonia, lung cancer, and tuberculosis are examples of chronic respiratory diseases which are considered as the leading causes of non-communicable disease deaths worldwide. A decision support system for doctors could help make the proper diagnosis, thereby reducing diagnostic errors. One promising field that could help in doing this is by using deep learning methods with image processing, specifically Convolutional Neural Networks (CNN). CNN has been proven effective in classifying chest x-ray images in previous studies. However, most of these mainly focused on classifying two classes or diseases only. This study aimed to use CNN’s to detect multiple respiratory diseases, specifically healthy or normal, lung cancer, tuberculosis, and pneumonia, from chest x-rays. The study was limited to these four classes and only used CNN architecture to detect the aforementioned types. Moreover, image preprocessing methods such as gray level transformation, histogram equalization, and data augmentation were used to improve the dataset's quality and quantity. After applying preprocessing techniques, the images were fed to CNNs through transfer learning, wherein three different models were used, namely, EfficientNetV2, NasNet Large, and ResNet50-V2. Using convolutional neural networks through transfer learning proved effective, with all models obtaining an accuracy of greater than 95%. Convolutional Neural Networks effectively and accurately classified the four types of chest x-ray images. Additionally, the results of this study aid hospitals and doctors in making immediate diagnoses while also being accurate, which is a need, especially at a time when hospitals could be overloaded.

Citation Download Citation

Edgar M. Adina, Mylen L. Aala-Capuno, Hamil Josef P. Roleda, and Divina A. Tuason "Detecting respiratory diseases in chest x-ray images using convolutional neural network (CNN)", Proc. SPIE 12936, International Conference on Mathematical and Statistical Physics, Computational Science, Education and Communication (ICMSCE 2023), 129360A (19 December 2023); https://doi.org/10.1117/12.3012651

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
10 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Chest imaging

Performance modeling

Pulmonary disorders

Lung cancer

Image classification

Convolutional neural networks

Education and training

Show All Keywords

Keywords/Phrases

Search In:

Publication Years