Skip to main content

Table 5 Comparison of the performances of some state-of-the-art models on the ADAM test set

From: DCAU-Net: dense convolutional attention U-Net for segmentation of intracranial aneurysm images

Model

Dice

Sensitivity

Specificity

Precision

  MIP + 2D CNN

0.6642

0.7201

0.9993

0.5865

  HeadXNet [29]

0.6462

0.7473

0.9994

0.6771

  DeepMedic [32]

0.7421

0.7247

0.9997

0.7521

  GLIA-Net [33]

0.7443

0.7195

0.9997

0.7615

  DAResUNet [30]

0.7376

0.7014

0.9995

0.7553

  3D U-Net [37]

0.6544

0.7181

0.9993

0.6036

  Ours

0.7455

0.7881

0.9996

0.7881