A Pyramid Architecture-Based Deep Learning Framework for Breast Cancer Detection

Table 1

Comparison results of different methods.

Methods/accuracy	Normal (%)	Macro (%)	Micro (%)	ITCs (%)	Average (tumor, %)
FCN	64.3	61.1	53.6	17.3	44
U-Net	77.8	67.1	55.2	21.2	47.83
Ours	80.1	73.6	57.7	20.6	50.63
	68.7	64.7	54.2	14.1	44.33
	79.6	70.1	58.4	23.1	50.53
	85.3	74.3	60.1	24.6	53