COVIDXRayNet

This project leverages CNN models (Xception, VGG16, ResNet50) to detect COVID-19 from chest X-ray images. It compares model performance based on accuracy, precision, recall, and F1-score, offering insights into the best architecture for reliable medical image classification.

Accuracy & Loss Over Epochs:

Model Name	Accuracy	Loss
Xception
ResNet50
VGG16

Confusion Matrix:

Model Name	Plot
Xception
ResNet50
VGG16

Model Prediction:

Model Name	COVID	non-COVID
Xception
ResNet50
VGG16

Model Comparison:

Model Name	Accuracy	Loss	Recall (COVID)	Recall (non-COVID)	Precision (COVID)	Precision (non-COVID)	F1-Score (COVID)	F1-Score (non-COVID)	Overall Accuracy	Model Size (MB)
Xception	0.81	0.39	0.72	0.91	0.89	0.78	0.79	0.84	0.82	86.4
ResNet50	0.84	0.41	0.82	0.87	0.86	0.84	0.84	0.85	0.85	214
VGG16	0.81	0.39	0.97	0.96	0.96	0.97	0.96	0.96	0.96	204