Table 5 Comparison to baseline models on our dataset.

From: Automatic snoring detection using a hybrid 1D–2D convolutional neural network

Models

Accuracy (%)

Sensitivity (%)

Specificity (%)

AUC

2D CNN8

84.7 ± 4.5

90.3 ± 7.1

66.9 ± 28.9

0.839 ± 0.091

CNN-LSTM44

86.3 ± 3.2

90.7 ± 2.1

76.3 ± 8.0

0.908 ± 0.020

1D CNN

85.2 ± 7.6

89.5 ± 8.3

70.5 ± 16.0

0.870 ± 0.062

Our work

89.3 ± 1.8

89.7 ± 1.4

88.5 ± 3.5

0.947 ± 0.024