Table 1 Comparison of different methods for early PD vs. HC classification.

From: Temporal and spatial variability of dynamic microstate brain network in early Parkinson’s disease

Method

Accuracy (%)

Sensitivity (%)

Specificity (%)

F1 Score

AUC

SN1-SVM

51.05 ± 0.1551

42.30 ± 0.2667

60.21 ± 0.1874

38.63 ± 0.2164

0.50 ± 0.0018

SN2-SVM

49.49 ± 0.1684

37.95 ± 0.2553

61.11 ± 0.1924

36.40 ± 0.2139

0.49 ± 0.0018

SN3-SVM

49.96 ± 0.1435

40.11 ± 0.2222

60.73 ± 0.2054

37.50 ± 0.1810

0.48 ± 0.0018

MN-SVM

69.27 ± 0.1503

41.83 ± 0.2254

92.81 ± 0.1294

50.74 ± 0.2182

0.86 ± 0.0015

MCN-SVM

95.69 ± 0.0636

92.97 ± 0.1136

98.41 ± 0.0515

93.67 ± 0.0954

0.99 ± 0.0002

  1. Significant differences are displayed in bold and italics type. SD, standard deviation. Adjustment for multiple comparisons: FDR. Significant statistical difference with *p < 0.05 and **p < 0.01.
Back to article page