Skip to main content

ORIGINAL RESEARCH article

Front. Med.
Sec. Pathology
Volume 11 - 2024 | doi: 10.3389/fmed.2024.1511389
This article is part of the Research Topic Artificial Intelligence-Assisted Medical Imaging Solutions for Integrating Pathology and Radiology Automated Systems - Volume II View all 11 articles

DLAAD-Deep Learning Algorithms Assisted Diagnosis of Chest Disease Using Radiographic Medical Images

Provisionally accepted
Mosleh Hmoud Al-Adhaileh Mosleh Hmoud Al-Adhaileh 1Bayan M Alsharbi Bayan M Alsharbi 2Theyazn Aldhyani Theyazn Aldhyani 1*Sultan Ahmad Sultan Ahmad 3Mohammed Almaiah Mohammed Almaiah 4Zeyad A T Ahmed Zeyad A T Ahmed 5Saad M Abdelrahman Saad M Abdelrahman 1Elham Alzain Elham Alzain 1Shilpi Singh Shilpi Singh 6
  • 1 King Faisal University, Al-Ahsa, Saudi Arabia
  • 2 Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia
  • 3 Prince Sattam Bin Abdulaziz University, Al-Kharj, Saudi Arabia
  • 4 King Abdullah the II IT School, the University of Jordan, Amman, Jordan
  • 5 Dr. Babasaheb Ambedkar Marathwada University, Aurangabad, Maharashtra, India
  • 6 Amity School of Engineering and Technology, Amity University, Patna, India

The final, formatted version of the article will be published soon.

    Viral infections can cause pneumonia, which is difficult to diagnose using chest X-rays due to its similarities with other respiratory conditions. Current pneumonia diagnosis techniques have limited accuracy. Novelty, of this research is developed a application of deep learning algorithms is essential in enhancing the medical infrastructure used in the diagnosis of chest diseases via the integration of modern technologies into medical devices. This study presents a transfer learning approach, using MobileNetV2, VGG-16, and ResNet50V2 to categorize chest disorders via X-ray images, with the objective of improving the efficiency and accuracy of computer-aided diagnostic systems (CADs). This research project examines the suggested transfer learning methodology using a dataset of 5,863 chest X-ray images classified into two categories: pneumonia and normal. The dataset was restructured to 224x224 pixels, and augmentation techniques were used during the training of deep learning models to mitigate overfitting in the proposed system. The classification head was subjected to regularization to improve performance. Many performance criteria are typically used to evaluate the effectiveness of the suggested strategies. The performance of MobileNetV2, given its regularized classification head, exceeds that of the previous models. The suggested system identifies images as .members of the two categories (pneumonia and normal) with 92% accuracy. The suggested technique exhibits superior accuracy as compared to currently available ones regarding the diagnosis the chest diseases. This system can help enhance the domain of medical imaging and establish a basis for future progress in deep-learning-based diagnostic systems for pulmonary disorders.

    Keywords: deep learning, medical images, diagnosis, Chest diseases, Radiography

    Received: 14 Oct 2024; Accepted: 27 Nov 2024.

    Copyright: © 2024 Al-Adhaileh, Alsharbi, Aldhyani, Ahmad, Almaiah, Ahmed, Abdelrahman, Alzain and Singh. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

    * Correspondence: Theyazn Aldhyani, King Faisal University, Al-Ahsa, Saudi Arabia

    Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.