The ongoing COVID-19 pandemic has emphasized the critical role of early diagnosis using chest X-rays. However, manually distinguishing COVID-19 infections from chest X-ray images presents significant challenges. These challenges arise due to time constraints, the expertise required by radiologists, and the subtle differences between positive and negative X-ray images.
To address these challenges, we introduce an automated COVID-19 screening framework that leverages artificial intelligence techniques, combined with a transfer learning approach, to diagnose COVID-19 from chest X-ray images.
- Transfer Learning: Utilized for efficient feature extraction, reducing the need for extensive training on large datasets.
- Modified Deep Neural Networks: Tailored neural network architectures to better process and interpret X-ray images.
- Grad-CAM Visualization: Employed to visually support and validate the predicted diagnoses, offering insight into areas of interest within the images.
Our experiments on publicly available datasets reveal that our convolutional neural network model, while being straightforward, outperforms other deep learning techniques across various metrics, including accuracy, precision, recall, and F-measure.
Automated COVID-19 Screening Framework via Deep Convolutional Neural Network With Chest X-ray Medical Images | IEEE Conference Publication | IEEE Xplore