Table 3. Comparing the recognition performance of the proposed model with other models in the recent literature.

	Method	Modality	Accuracy (%)	Senstivity (%)	Specificty (%)	Precision (%)	F1-score (%)	FPR	MCC (%)
4-Classes	Alexnet (Loey, Smarandache & Khalifa, 2020)	X-ray	66.67	66.67	–	64.68	65.66	–	–
	Resnet18 (Loey, Smarandache & Khalifa, 2020)	X-ray	69.46	66.67	–	72.50	69.46	–	–
	ShuffleNet + SVM (Sethy & Behera, 2020)	X-ray	70.66	65.26	–	–	58.79	17.36	–
	Googlenet (Loey, Smarandache & Khalifa, 2020)	X-ray	80.56	80.56	–	84.17	82.32	–	–
	CNN (Zhao et al., 2020)	CT	84.7	76.2	–	97.0	85.3	–	–
	Inception-v3 + SVM (Sethy & Behera, 2020)	X-ray	96.00	90.26	–	–	90.28	4.86	–
	DenseNet201 + SVM (Sethy & Behera, 2020)	X-ray	97.33	93.86	–	–	93.86	3.06	–
	XceptionNet + SVM (Sethy & Behera, 2020)	X-ray	97.33	93.00	–	–	93.00	3.50	–
	VGG-16 + SVM (Sethy & Behera, 2020)	X-ray	97.33	94.20	–	–	94.20	2.90	–
	InceptionResnetV2 + SVM (Sethy & Behera, 2020)	X-ray	97.33	91.13	–	–	91.74	4.43	–
	Ours TL-Incep-V3	X-ray	98.1	98.02	98.03	98.2	98.2	2	–
2-Classes
	DRE-Net (Song et al., 2020)	CT	64	92	96.12	96	94	3.85	88.3
	DenseNet	CT	96.25	96.29	96.21	96.29	96.29	–	–
	VGG-16	CT	96.93	99.20	94.67	94.90	97.0	5.33	93.96
	Resnet-50	CT	97.33	99.20	95.47	95.63	97.38	4.53	94.73
	GoogleNet	CT	97.87	96.93	98.80	98.78	97.85	1.2	95.75
	Ozkaya, Ozturk & Barstugan (2020)	CT	98.27	98.93	97.60	97.63	98.28	2.4	96.54
	MobileNet v2	X-ray	97.40	99.10	97.09	–	–	–	–
	VGG19	X-ray	98.75	92.85	98.75	–	–	–	–
	Ours TL-Incep-V3	X-ray	99.4	99.5	99.1	99.1	99.3	0.9	98.7
	Ours TL-Incep-V3	CT + X-ray	99.5	99.8	98.2	99.2	99.5	0.81	99.0