Table 8. Comparison with other CNN architectures.

Where the best results are highlighted in bold.

VGG19 Pre-trained model	Tuning mode	Performance metrics (%)
		Accuracy	Sensitivity	Specificity
Abbas, Abdelsamea & Gaber (2020)	Shallow (KNN)	93.42	89.71	95.7
	Deep learning (CNN)	94.59	91.64	93.08
Apostolopoulos & Mpesiana (2020)	Deep learning (CNN)	98.75	92.85	98.75
Luz et al. (2020)	Deep learning (CNN)	75.3	77.4	50.0
Proposed framework	Shallow (KNN)	96.93	57.14	99.2
	Deep learning (CNN)	99.04	85.71	99.6