Page 151 - Control Theory in Biomedical Engineering

136 Control theory in biomedical engineering

Table 1 Considered handwriting database (Chihi et al., 2018).
Description of the form Form Description of the form Form
Horizontal line (1) Circle (1)
(left/right/left) (to the right)

Horizontal line (2) Circle (2)
(right/left/right) (to the left)

Vertical line (1) Triangle (1)
(up/down/up) (to the right)

Vertical line (2) Triangle (2)
(down/up/down) (to the left)

Advanced signal processing techniques were applied to extract useful
information from the recorded EMG signals. These techniques filtered
the fluctuation of EMG magnitudes to define new curves called integrated
EMG (IEMG), represented by blue curves (gray curves in print version) in
Fig. 4 (Chihi et al., 2017). We note IEMG1 and IEMG2, the integrated
EMG of channel 1 and channel 2 respectively.
The type of graphic traces, considered in this experimental approach, was
judiciously chosen. Indeed, based on Thomassen’s analysis, which con-
firmed that when a person produces traits or shapes in nonpreferred direc-
tions the performance will be less precise, more unstable and is manifested in

Fig. 3 Experimental approach and positions of the electrodes on the forearm (Chihi
et al., 2015).