Skip to main content

Table 2 Epilepsy detection performance achieved using multiscale feature extraction methods on the SWEC-ETHZ dataset

From: Technological Vanguard: the outstanding performance of the LTY-CNN model for the early prediction of epileptic seizures

Patient id

Acc (%)

Sen (%)

Spe (%)

AUC

Pt 01

100.0

100.0

100.0

1.000

Pt 02

100.0

100.0

100.0

1.000

Pt 03

100.0

100.0

100.0

1.000

Pt 04

100.0

100.0

100.0

1.000

Pt 05

100.0

100.0

100.0

1.000

Pt 06

100.0

100.0

100.0

1.000

Pt 07

100.0

100.0

100.0

1.000

Pt 08

100.0

100.0

100.0

1.000

Pt 10

99.5

99.7

80.0

0.899

Pt 11

100.0

100.0

100.0

1.000

Pt 12

100.0

100.0

100.0

1.000

Pt 13

100.0

100.0

100.0

1.000

Pt 14

100.0

100.0

100.0

1.000

Pt 15

99.2

99.2

100.0

0.925

Pt 16

100.0

100.0

100.0

1.000

Pt 17

100.0

100.0

100.0

1.000

Pt 18

100.0

100.0

100.0

1.000

Average

99.9

99.9

98.8

0.990