| Autor: |
Vinod, Dasari Naga, Kapileswar, N., Simon, Judy, P., Phani Kumar, M., Saraswathy |
| Předmět: |
|
| Zdroj: |
International Journal of Intelligent Engineering & Systems; 2024, Vol. 17 Issue 4, p598-610, 13p |
| Abstrakt: |
The worldwide problem instigated by multiple mutant forms of the prevalent COVID-19 epidemic; the severe clinical diagnosis remains uncertain. Various clinical prognostic imaging approaches have previously been offered to medical practitioners to identify COVID-19 people. In this article, we introduce a novel diagnostic approach leveraging Generative Adversarial Networks (GAN) and machine learning techniques to enhance the precision and efficiency of COVID-19 diagnosis. Our method integrates advanced image processing algorithms with deep learning models to accurately identify patterns indicative of COVID-19 contamination in medical imaging data. By harnessing the power of GANs, we facilitate the creation of synthetic data for training, thus overcoming limitations posed by sparse datasets. Through rigorous experimentation and validation, we demonstrate the efficacy of our approach in achieving superior diagnostic accuracy compared to existing methods. This research represents a significant advancement in the field of medical imaging diagnostics and holds promise for more effective identification and management of COVID-19 cases. The latter technique supplements Computed Tomography as well as chest X-ray imaging. As a result, utilizing the picture database (7050 Ultrasound Images), our innovative approach utilizes gradient mapping and different haralick characteristics. Distinct techniques were employed to evaluate the classification performance of test sets accompanied by 2110 clinical imaging data. Interestingly, the paper demonstrates that the suggested model's multiple classification accuracy reached 98.1% efficiency among the normal, Pneumonia, as well as COVID-19 by ultrasound image dataset. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
