Skip to main content

Table 4 Classification results for tremor prediction applying different feature sets and classifiers. The results were mean ± std for a 10-fold cross validation

From: Parkinson’s disease tremor prediction using EEG data analysis-A preliminary and feasibility study

Feature set

Classifier

Accuracy (%)

Sensitivity

Specificity

AUC

Feature set No.1

KNN

73.67 ± 9.56

70.90 ± 14.89

87.18 ± 10.00

61.52 ± 14.53

90.54 ± 8.92

92.50 ± 6.10

79.4 ± 9.30

0.74 ± 0.09

Multi-class SVM

45.44 ± 7.96

42.31 ± 30.09

67.37 ± 25.56

20.57 ± 23.33

72.19 ± 11.93

65.8 ± 35.3

74.4 ± 16.31

0.45 ± 0.08

Decision tree

63.62 ± 8.32

55.25 ± 17.38

82.22 ± 15.40

61.92 ± 14.46

83.81 ± 12.62

89.74 ± 5.25

72.66 ± 12.27

0.64 ± 0.08

Nave Bayes

61.85 ± 13.25

74.68 ± 30.19

74.37 ± 28.23

63.63 ± 38.91

89.62 ± 8.44

92.76 ± 6.54

72.53 ± 16.50

0.62 ± 0.13

Discriminant analysis

54.78 ± 5.85

59.67 ± 31.05

85.65 ± 15.10

41.78 ± 13.10

70.17 ± 8.11

80.01 ± 8.11

90.23 ± 11.31

0.55 ± 0.06

Feature set No.2

KNN

81.29 ± 9.19

79.15 ± 11.38

84.86 ± 11.53

83.94 ± 20.93

90.37 ± 10.33

95.62 ± 6.29

86.57 ± 4.50

0.81 ± 0.09

Multi-class SVM

51.64 ± 8.97

51.71 ± 30.99

64.55 ± 24.15

23.10 ± 29.89

71.68 ± 12.46

88.17 ± 7.71

78.48 ± 15.71

0.52 ± 0.09

Decision tree

65.96 ± 4.41

66.85 ± 6.62

72.95 ± 13.55

59.93 ± 17.38

86.57 ± 5.05

82.72 ± 4.83

79.62 ± 5.44

0.66 ± 0.04

Nave Bayes

71.49 ± 11.44

85.98 ± 14.09

76.62 ± 22.58

67.84 ± 19.76

82.81 ± 5.98

91.67 ± 8.55

86.74 ± 12.66

0.71 ± 0.11

Discriminant analysis

56.32 ± 3.69

81.67 ± 21.08

71.31 ± 10.16

47.63 ± 6.02

67.51 ± 7.08

87.06 ± 8.16

88.85 ± 10.69

0.56 ± 0.04