Table 5.

Evaluation based on two-way class classification.

Model (Trainable/Total Parameters)	Phase	Acc	Sen	PPV	F1
VGG-16 (165,250/14,880,706)	Training	96.61	96.61	96.63	96.62
	Validation	98.01	98.01	98.01	98.01
	Testing	94.28	94.29	94.29	94.29
VGG-19 (263,682/20,288,066)	Training	89.54	89.43	89.54	89.48
	Validation	92.61	92.62	92.61	92.61
	Testing	92.56	92.54	92.55	92.54
Inception-V3 (427,394/54,764,898)	Training	87.33	87.50	87.03	87.26
	Validation	91.76	92.24	91.19	91.72
	Testing	89.60	89.94	89.46	89.70
ResNet-50 (13,141,378/36,729,858)	Training	84.75	84.75	84.88	84.81
	Validation	89.20	89.34	89.12	89.23
	Testing	89.08	89.18	88.78	88.98
Proposed BNCNN (2,786,306/2,787,554)	Training	97.28	97.28	97.23	97.25
	Validation	98.55	98.52	98.53	98.52
	Testing	99.27	99.45	98.83	99.14