Skip to main content

Table 2 Comparative analysis of CNN-Based and transformer-based models on the KCGMH dataset

From: Automatic segmentation of white matter lesions on multi-parametric MRI: convolutional neural network versus vision transformer

Model

DRLoc

Pre-training

Foreground DSC

Lesion F1-Score

Lesion sensitivity

CNN-based

Without

Without

0.6128

0.4544

0.4586

 

With

Without

0.6021

0.4323

0.4273

Transformer-based

Without

Without

0.6353

0.4193

0.4514

 

With

Without

0.6497

0.4392

0.444

 

With

With

0.6585

0.5105

0.5363

Back to article page

BMC Neurology

ISSN: 1471-2377

Contact us