Table 4 Dice and sensitivity comparison of the proposed model with other methods on BraTS 2018 validation sets.

From: DTASUnet: a local and global dual transformer with the attention supervision U-network for brain tumor segmentation

Method

Dice

Sensitivity

ET

WT

TC

Mean

ET

WT

TC

Mean

Ben et al.38

0.725

0.860

0.744

0.776

0.753

0.842

0.751

0.782

Kong et al.14

0.741

0.871

0.789

0.800

0.756

0.890

0.796

0.814

Zhang et al.39

0.744

0.896

0.792

0.811

0.803

0.895

0.853

0.850

Tong et al.40

0.792

0.887

0.813

0.831

0.851

0.901

0.820

0.857

Liu et al.41

0.767

0.898

0.834

0.833

–

–

–

–

Akbar et al.29

0.777

0.896

0.798

0.824

0.814

0.926

0.805

0.848

Jiang et al.42

0.784

0.899

0.838

0.840

0.780

0.894

0.816

0.830

Ours

0.808

0.905

0.845

0.853

0.820

0.916

0.837

0.858

  1. Bold results denote the two best scores.
Back to article page