A Predictive Framework for Detection of Corona Virus based on Transfer Learning Approach
DOI:
https://doi.org/10.47750/pnr.2022.13.S03.098Keywords:
Covid Detection, Pneumonia, X-ray, Convolution Neural Networks (CNN), Transfer Learning.Abstract
Early identification of COVID -19 has a substantial influence on reducing COVID -19 transmission at a faster rate and is the need of the
moment. AI diagnostics utilizing deep learning models trained on X-ray pictures of COVID-infected and uninfected persons is a viable new technique for early prediction and diagnosis of COVID-infected patients. This study presents a technique that can be used to automatically identify the corona virus from machine-made chest X-ray images in less than five minutes. For this we use a collection of chest X-ray images of pneumonia, COVID 19 disease, and healthy infected patients. Transfer Learning is used because it has the advantage of reducing training times for a neural network model. The result shows 99.49% accuracy in predicting Corona virus from an X-ray of a suspect patient using the VGG Transfer Learning framework.
