Skip to main content

Table 2 The segmentation performance measured by Dice score and Jaccard index for both datasets in the cross-validation

From: Quantifying the tumour vasculature environment from CD-31 immunohistochemistry images of breast cancer using deep learning based semantic segmentation

 

Behold

Mission

Dice

Jaccard

Dice

Jaccard

Vessel

0.875

0.777

0.856

0.748

Tumour

0.701

0.539

0.815

0.688

Benign

0.476

0.312

0.556

0.385

Adipose

0.898

0.814

0.915

0.843

Lymphocytes

0.698

0.536

0.648

0.480

Muscle

0.714

0.555

0.736

0.582

Leukocytes

0.715

0.557

0.780

0.640

Nerve

0.764

0.618

0.873

0.775

Stroma

0.886

0.796

0.899

0.817

Background

0.753

0.604

0.954

0.913