Table 9 Model performance comparison on BRATS 2020.

From: Enhanced brain tumor segmentation in medical imaging using multi-modal multi-scale contextual aggregation and attention fusion

Model

Loss

Accuracy

Mean IoU

Precision

Sensitivity

Specificity

Dice Score

Dice Score (Necrotic)

MM-MSCA-AF

0.1065

0.9968

0.8589

0.8589

0.9349

0.9650

0.8158

0.7982

HCA41

0.2301

0.9922

0.845

0.840

0.920

0.960

0.801

0.7828

nnU-Net23

0.015

0.9960

0.850

0.850

0.930

0.963

0.810

0.7950

UNet++26

0.120

0.9957

0.848

0.845

0.935

0.961

0.808

0.7900

SegNet44

0.130

0.9945

0.840

0.860

0.915

0.958

0.800

0.7750

Attention U-Net43

0.118

0.9958

0.847

0.842

0.929

0.962

0.807

0.7880