Skip to main content
. 2022 Nov 11;22(22):8733. doi: 10.3390/s22228733

Table 5.

Mean average value of accuracy and F1 scores of supervised gesture recognition from four to seven gesture labels (based on the hand movements of 19 stroke patients) without and with PCA.

Gesture Classification without PCA
(GL4, GL5, GL6, GL7)
Gesture Classification with PCA *
(GL4, GL5, GL6, GL7)
Non-Paretic Side (NP19) Paretic Side (P19) Non-Paretic Side (NP19) Paretic Side (P19)
Classifier Acc (%) F1
(%)
Acc (%) F1
(%)
Acc
(%)
F1
(%)
Acc (%) F1
(%)
SVM 94.80 89.74 88.71 77.53 92.81 85.63 87.59 74.90
94.19 85.69 88.48 71.49 91.41 78.77 86.07 65.16
94.13 82.67 88.60 66.01 91.36 74.33 85.98 57.82
94.73 81.82 89.75 64.26 91.97 72.00 86.40 52.20
LDA 87.63 75.42 78.16 55.48 94.05 88.13 88.64 77.15
90.92 77.52 77.09 41.42 93.79 84.55 87.96 69.94
92.35 77.39 77.39 30.63 93.85 81.66 87.97 63.98
93.27 76.95 77.43 20.09 94.02 79.25 88.17 58.61
k-NN 92.84 85.82 87.32 74.73 86.30 72.58 82.64 65.36
91.98 80.10 87.38 68.67 87.12 68.03 83.26 58.37
91.70 75.34 86.81 60.72 87.70 63.76 84.24 53.03
91.71 71.12 87.31 55.86 88.19 59.61 85.23 48.49

* PCA had the best specific parameters for each classifier and extremity side; GL indicates the mean scores of all possible gesture combinations, while integer values of 4–7 specify the model complexity (number of gesture labels).