COVID-19 and Pneumonia classification Using Ensembling with Transfer Learning
10th International Conference on Reliability, Infocom Technologies and Optimization ,Trends and Future Directions, ICRITO 2022
; 2022.
Article
in English
| Scopus | ID: covidwho-2191925
ABSTRACT
The world has been rapidly devastated by the Covid-19 virus, which first appeared in the Republic of China. For medical imaging, deep learning-based algorithms show promising results for quick and accurate diagnosis. Various research has been done for the earlier diagnosis of the disease using various deep learning models. Researchers use different medical imaging for the classification of COVID-19. This study explores COVID-19 diagnosis using a chest X-Ray. The Chest X-Ray images were classified with the help of transfer learning using VGG16, DenseNet, and MobileNet. To ensure better results Ensemble Learning is incorporated to provide a strong learner by using the aggregation of weak learners. These models are trained on three different classes of patients COVID-19, Pneumonia, and Normal. The final testing results using ensembling aggregation show an overall accuracy of 95.2%, which is significantly higher than the model performances individually. The result obtained through the proposed model can be used in conjunction with the X-Ray images to classify COVID-19, thus the process can be implemented as an alternative to RT-PCR. © 2022 IEEE.
Full text:
Available
Collection:
Databases of international organizations
Database:
Scopus
Language:
English
Journal:
10th International Conference on Reliability, Infocom Technologies and Optimization ,Trends and Future Directions, ICRITO 2022
Year:
2022
Document Type:
Article
Similar
MEDLINE
...
LILACS
LIS