Table 3.

Comparison of classification accuracy of several state-of-the-art methods with DRISTI

Method	Approach	Accuracy
[32]	CNN	44.22%
[25]	CNN	56%
[25]	CNN (two-class)	71%
[23]	CAD system (two-class)	93.75%
[29]	Pretrained model	63.23%
[3]	CNN	74%.
[7]	VGG16 CNN	75.83%
Ours	VGG16+Capsule (five-class)	82.06%
	VGG16+Capsule (two-class)	96.24%