Classification of COVID-19 using convolutional neural network - Resnet-50
| Autor: | R. M. Suganthy, Fathima. R Tabassum, S Kaviya, E Karpagambal |
|---|---|
| Rok vydání: | 2022 |
| Předmět: | |
| Zdroj: | International journal of health sciences. :4645-4657 |
| ISSN: | 2550-696X 2550-6978 |
| DOI: | 10.53730/ijhs.v6ns1.5950 |
| Popis: | The main significance of employing chest X-ray images is to recognize and determine if it is covid or pneumonia. By using these images, it plays a vital role for doctors to save the lives of patients. This is even more important in nations where laboratory kits for testing are not readily available.Deep learning based recognition of covid-19 using images obtained from chest X-ray is been proposed.For training and testing purposes the ResNet 50 and Xception network is used as convolutional neural network. ResNet 50 contains 48 layers that includes a Max and Average pool.Xception network has 71 layers .Softmax is used as the activation function here to predict multinominal probability distribution.Stochastic Gradient Descent (SGD) is employed for maximising accuracy / SGD is used to enhance accuracy. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|