TABLE 4.

Comparison of BCNNs and Fast BCNNs.

		Accuracy (%)
		Magnification Factor
Backbone network	Type of BCNNs	40×	100×	200×	400×	Mixed	FPS	NOP (million)
Inceptionv3	BCNNs	95.74 ± 0.21	94.72 ± 0.18	94.78 ± 0.26	94.51 ± 0.23	96.14 ± 0.16	45.5	81.4
	Fast BCNNs	95.99 ± 0.17	95.84 ± 0.16	94.70 ± 0.28	94.51 ± 0.15	95.95 ± 0.19	62.5	59.6
ResNet50	BCNNs	95.74 ± 0.24	95.44 ± 0.32	94.53 ± 0.25	94.88 ± 0.34	95.27 ± 0.26	38.5	84.9
	Fast BCNNs	95.49 ± 0.27	95.78 ± 0.29	94.29 ± 0.26	94.43 ± 0.19	94.96 ± 0.23	58.8	61.3
InceptionResNetV2	BCNNs	95.24 ± 0.13	94.16 ± 0.22	94.03 ± 0.29	93.70 ± 0.23	95.08 ± 0.14	27.8	130.0
	Fast BCNNs	95.99 ± 0.21	93.53 ± 0.18	94.79 ± 0.24	93.78 ± 0.15	95.20 ± 0.17	45.5	75.6

Frames per second (FPS) and Number of Parameters (NOP) were calculated to evaluate the speed. Average accuracies ± standard deviations were calculated from 5-fold cross validation of the BreaKHis dataset.