| Abstrakt: |
Today's major global pandemic, COVID-19, is caused by coronavirus disease. Every day, a large number of people all over the world become infected with this virus, and many of these individuals die from it. Diagnosing and treating this virus as soon as possible is essential due to its highly infectious nature. This study uses deep learning networks to automatically diagnose COVID-19 from chest X-ray pictures. To enhance the effectiveness of the deep learning network in this study, we suggest combining convolutional networks with type-2 fuzzy activation functions. A generative adversarial network has also been used to increase the amount of data. Based on the classification of three groups of healthy, COVID-19, and pneumonia states, 99.45% accuracy was achieved. Furthermore, the suggested method has demonstrated a sensitivity and specificity of 100 and 99%, respectively, compared to recent studies; this is a remarkable result in terms of specificity, accuracy, and sensitivity. The obtained results (irrespective of the activation functions used) confirm that our method has outperformed state-of-the-art methods used for COVID-19 classification. [ABSTRACT FROM AUTHOR] |
Full text from SpringerLink