Skip to main content

Table 1 Classification performance for each model

From: Enhancing automated lower limb rehabilitation exercise task recognition through multi-sensor data fusion in tele-rehabilitation

Model

LOSO

Accuracya (%)

Precision (%)

Recall (%)

F1 Score (%)

DC

93.81 ± 7.98

93.85 ± 8.45

93.81 ± 8.48

93.8 ± 7.7

PM

81.43 ± 16.14

81.75 ± 8.56

81.43 ± 8.36

81.45 ± 7.12

DC-PM

95.71 ± 7.51

95.83 ± 6.32

95.71 ± 5.35

95.74 ± 5.19

 

 LMSO

Accuracy (%)

Precision (%)

Recall (%)

F1 Score (%)

DC

90.95 ± 4.49

91.35 ± 8.77

90.95 ± 11.17

90.83 ± 8.28

PM

75.71 ± 5.89

75.48 ± 8.58

75.71 ± 14.49

75.28 ± 10

DC-PM

94.76 ± 1.96

94.84 ± 6.72

94.76 ± 6.34

94.77 ± 5.84

  1. aStandard deviations of accuracies are between-subject, and between-class for other metrics