Table 7.

Image-wise comparisons of the sensitivity and specificity metrics with respect to the ICIAR2018 dataset for the maximum voting criterion. The best results are shown in bold.

Class	Model	Sensitivity (Maximum)
		Without Data				Data
		Augmentation				Augmentation
		Non-Carcinoma		Carcinoma		Non-Carcinoma		Carcinoma
Binary	DNet	0.77		0.89		0.72		1.0
	MSI-MFNet	0.79		0.97		0.69		1.0
		Benign	Normal	In Situ	Invasive	Benign	Normal	In Situ	Invasive
Multi	DNet	0.27	1.0	0.78	0.74	0.29	1.0	0.78	0.95
	MSI-MFNet	0.29	0.91	0.88	0.95	0.32	1.0	0.83	0.96
		Specificity (Maximum)
		Without Data				Data
		Augmentation				Augmentation
		Non-Carcinoma		Carcinoma		Non-Carcinoma		Carcinoma
Binary	DNet	0.89		0.77		1.0		0.70
	MSI-MFNet	0.97		0.80		1.0		0.72
		Benign	Normal	In Situ	Invasive	Benign	Normal	In Situ	Invasive
Multi	DNet	0.91	0.78	0.95	0.82	0.98	0.85	0.91	0.84
	MSI-MFNet	0.97	0.90	0.89	0.86	0.98	0.89	0.93	0.85