Table 5 Comparison of methods of stroke diagnosis based on EEG.

From: A novel stroke classification model based on EEG feature fusion

Model

Feature

Method

Accuracy

AUC

Laura21

DAR[i]

Statistics

 

0.83

Tong42

cwPLI + SaEn[ii]

MSE-VGG[iii]

0.9017

 

Ours

ApFu

TPELGBM

0.9689

0.9963

  1. [i] DAR: delta-alpha ratio of EEG.
  2. [ii] cwPLI + SaEn: fusion feature of correlation-weighted Phase Lag Index and sample entropy.
  3. [iii] MSE-VGG: VGG model with multiple squeeze-and-excitation (SE) modules.
Back to article page