Diagnosis of Pneumonia from Chest X-Ray Images using Deep Learning | ||||
| Delta University Scientific Journal | ||||
| Volume 7, Issue 3, November 2024, Page 353-362 PDF (1.03 MB) | ||||
| Document Type: Original research papers | ||||
| DOI: 10.21608/dusj.2024.433478 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
Amany M Sarhan  1; Aml Alaa Khattab2; Mahmoud Shaheen3
| ||||
| 1Computer and Control Engineering, Faculty of Engineering, Tanta University, Egypt | ||||
| 2Computer and Control Engineering Department, Faculty of Engineering, Tanta University, Tanta, Egypt | ||||
| 3Faculty of Artificial Intelligence, Delta University for Science and Technology, Gamasa, Egypt | ||||
| Abstract | ||||
| Pneumonia is a disease of the lungs caused by a bacterial infection. Early diagnosis is critical to the outcome of treatment. In general, a trained radiologist may identify the condition using chest X-ray pictures. Diagnoses might be subjective for a variety of reasons, including the appearance of disease, which may be ambiguous in chest X ray pictures or mistaken with other conditions. As a result, computer-aided diagnostic systems are required to advise practitioners. In this study, we employed two well-known convolutional neural network models, Xception and VGG16, and a custom CNN model to diagnose pneumonia. We employed transfer learning and fine-tuning throughout the training step. The test findings indicated that the custom CNN model outperformed VGG16 network and the Xception model with 93% accuracy. | ||||
| Keywords | ||||
| Pneumonia; transfer learning; Xception; VGG16; deep learning | ||||
|
Statistics Article View: 70 PDF Download: 27 |
||||
