To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
Enhancing X-Ray Augmented Images for COVID-19 Disease Diagnosis Using Deep Learning Techniques |
|
Author |
Abdelmoty M. Ahmed, Belgacem Bouallegue, Amr Mohmed Soliman, Mahmoud M. Khattab4and, Hammam M. Abdelaal |
| Citation |
Vol. 26 No. 4 pp. 79-88 |
|
Abstract |
In most cases of diagnosing chest diseases, pneumonia, or even acute respiratory syndrome caused by the coronavirus (COVID-19), we rely on a complex set of data for clinical symptoms or x-rays of the chest or respiratory tract. Due to this complexity and the difficulty of interpreting data and feature extraction for x-rays, the importance of scientific research in this area increases. This importance is due to the urgent need for the effectiveness and accurate prediction and diagnosis of chest diseases and acute respiratory syndrome (COVID-19) in light of the circumstances of this epidemic around the world. Therefore, this paper aims to design and develop a model to deal with a dataset of clinical and diagnostic symptoms of patients with thoracic disease and pneumonia caused by the coronavirus to work on the accurate diagnosis of the disease. In this paper, we implement preprocessing of data and images of the chest and respiratory x-rays to obtain features that help us in the classification and diagnosis process. We conduct several experiments based on a set of algorithms. Firstly, we use the Very Deep Convolutional Networks model (VGG16) to extract and learn the features in addition to classify the data. Secondly, we use the Linear Discriminant Analysis (LDA) for dimensionality reduction. Finally, we use the Support Vector Machine (SVM) and Neural Network for classification; the experimental results indicate that the proposed models have enhanced efficiency compared to other models. |
|
Keywords |
COVID-19, Chest disease, pneumonia, Convolutional Neural Network, VGGNet-16, Machine learning, Deep Learning |
|
URL |
