Table 6.

Comparison of performance between state of the art and proposed approaches.

State-of-the-art	Methods	Computational approaches	Accuracy (%)
Das et al. [19]	Xception	Pneumonia vs Covid-19 vs Other	97.40
Singh et al. [20]	MADE-based CNN	non-Covid-19 vs Covid-19	94.48
Ucar et al. [21]	Deep Bayes-SqueezeNet	Healthy vs Pneumonia vs Covid-19	98.26
Wang et al. [22]	Covid-net	Healthy vs Pneumonia vs Covid-19	93.30
Afshar et al. [23]	Covid-caps	non-Covid-19 vs Covid-19	95.70
Chowdhury et al. [24]	MobileNetv2, SqueezeNet, ResNet-18, ResNet-101,	Normal vs Covid-19	99.70
	DenseNet-201, CheXNet, Inception-v3 and VGG-19	Normal vs Covid-19 vs Pneumonia Viral	97.90
Khan et al. [25]	CoroNet	Covid-19 vs Pneumonia Bacterial vs Pneumonia Viral vs Normal	89.60
		Covid-19 vs Pneumonia vs Normal	95.00
Sahinbas et al. [26]	ResNet, DenseNet, InceptionV3, VGG-16 and VGG-19	non-Covid-19 vs Covid-19	80.00
Apostolopoulos et al. [27]	MobileNetv2	non-Covid-19 vs Covid-19	99.18
Zulkifley et al. [28]	LightCovidNet	Healthy vs Pneumonia vs Covid-19	96.97
Our proposal	DenseNet-121, DenseNet-161, ResNet-18,	Healthy vs Pneumonia, tested with Covid-19	97.06
	ResNet-34, VGG-16 and VGG-19
Our proposal	DenseNet-121, DenseNet-161, ResNet-18,	Healthy vs Pneumonia/Covid-19	98.39
	ResNet-34, VGG-16 and VGG-19
Our proposal	DenseNet-121, DenseNet-161, ResNet-18,	Healthy/Pneumonia vs Covid-19	97.44
	ResNet-34, VGG-16 and VGG-19
Our proposal	DenseNet-121, DenseNet-161, ResNet-18,	Healthy vs Pneumonia vs Covid-19	97.44
	ResNet-34, VGG-16 and VGG-19