Table 2 Comparison of the experimental results with enhancement using SBE (proposed), Attention U-Net and R2U-Net.

From: Breast lesion detection using an anchor-free network from ultrasound images with segmentation-based enhancement

Dataset

Enhancement method

B-Precision

B-Recall

M-Precision

M-Recall

mAP

BUS

SBE (proposed)

0.710

0.846

0.865

1.000

0.788

Attention U-Net

0.779

0.846

0.644

1.000

0.712

R2U-Net

0.666

0.845

0.756

1.000

0.711

BUSI

SBE (proposed)

0.816

0.932

0.789

0.889

0.802

Attention U-Net

0.796

0.909

0.814

0.833

0.805

R2U-Net

0.762

0.886

0.803

0.889

0.783

BUS+BUSI

SBE (proposed)

0.917

0.980

0.888

0.963

0.902

Attention U-Net

0.951

1.000

0.805

0.926

0.878

R2U-Net

0.934

0.980

0.729

0.963

0.832

  1. Significant values are in bold.