Important note You can find all codes and jupyter notebook files in each model's folder

Covid-19-detection-with-chest-scan

This repo contains covid-19 detection with three algorithms , CNN , ResNet 50 and VGG 16 . out of which I got best accuracy with vgg16 as 98.8%

• Dataset of chest X-ray is downloaded from Kaggle.com and it is visualised for the proper understanding of the data before processing. Data augmentation is applied on the images so that it can effectively classify the covid-19 cases.
• CNN is proposed for the detection of covid-19 and hence controls the fast spreading covid-19 cases.
• Pre trained deep learning model that is transfer learning algorithm VGG-16 is used to increase the accuracy of the proposed model

Sample Dataset Image

CNN model

CNN ACCURACY CURVE

CNN Confusion Matrics

CNN ROC

ResNet50 ACCURACY CURVE

ResNet50 Confusion Matrics

ResNet50 ROC

VGG16 ACCURACY CURVE

VGG16 Confusion Matrics

VGG16 ROC

Accuracy comarision table

Accuracy Comparision graphs