Table 5. Comparison of AI-assisted recent studies for COVID-19 lung infection.

Authors	Modality	Subjects	Method	Performance
Ghoshal et al. [69]	X-Ray	COVID-19 90 and other conditions	CNN	92.9% (Acc.)
Sethy et al. [70]	X-ray	COVID-19 and Normal 25 images	ResNet50 and SVM	95.33%(Acc.)
Ying et al. [71]	CT	COVID-19 777 images and 708 images of Normal	DRE-Net	86% (Acc.)
Hussain et al. [72]	X-Ray	COVID-19 Bacterial & Viral 145 images and 138 Normal	Texture features using Machine learning. Two-class classification i) covid-19 vs normal ii) Covid-19 vs viral pneumonia iii) Covid-19 vs Bacterial pneumonia iv) Four-class (Covid-19, Bacteria, Viral and Normal)	100% accuracy 97.56% Accuracy 97.44% Accuracy 79.52% Accuracy
Pratiwi et al. [73]	CT	Covid—(1251) Non-Covid–(1229)	Two Classes Deep learning VGG-16	88.54% Accuracy
This study	X-Ray	COVID-19 (N = 1525), non-COVID-19 normal (N = 1525), viral pneumonia (N = 1342) and bacterial Pneumonia (N = 2521) After augmentation N = 2521	4-class (Normal, Bacterial Pneumonia, viral Pneumonia and COVID-19) using ESN-MDFS approach	96.18% Accuracy AUC of 0.99