Biomedical Engineering Reference
In-Depth Information
followed with a very sharp rising phase as shown in
Figure 8.3
)
, then to set the
output signal to be zero during a preset blocking period (3-5 ms in this study),
which was obtained though observation on the stimulation artifact duration on
a subject before experiment. It was noticed that besides the EMG generated
from voluntary muscle contractions, M-waves evoked by electrical stimulation
also could be captured during tracking tasks assisted with FES (see
Figure 8.3
).
EMG signals used in
Eq. 8.3
were the signals after the sample-and-hold process,
the rectification, and then the moving average. The FES-robot system also could
generate interactive resistive torques as the interactive robot system introduced in
Chapter 5
.
8.3 SYSTEM PERFORMANCE EVALUATION
The developed FES-robot system was first evaluated on its assistive function on
wrist tracking tasks by persons after stroke. In the evaluation, two different wrist
tracking speeds, i.e., 10
◦
/s and 20
◦
/s, were used. For each tracking trial, there
were 5 cycles of wrist flexion and extension from -45
◦
to 60
◦
. In our previous
continuous intention-driven robot-assisted wrist training only 10
◦
/s was selected
(Hu
et al
. 2007; Hu
et al
. 2008; Song
et al
. 2008; Hu
et al
. 2009). It was
an empirical speed that the subjects would not feel too easy or too difficult to
follow during the training process. In the FES-robot evaluation, five subjects with
chronic stroke (at lease 1 year after onset of stroke, male, aged from 36 to 56
with mean
7) were recruited after obtaining approval from the Human
Subjects Ethics Sub-Committee of the Hong Kong Polytechnic University. All
of the subjects were hemiplegia due to single cerebral unilateral ischemic event,
whose spasticity of wrist flexion/extension at the paretic side was measured by
the Modified Ashworth Scale (MAS, equal to 1.8
±
SD=48
±
0.7 at the wrist joint) (Ashworth
1964). The voluntary motor function of the paretic upper limb for the subjects
was scored by the Fugl-Meyer Assessment (FMA, 19.8
±
±
5shoulder/elbow,and
11.2
4.2 wrist/hand) (Fugl-Meyer
et al
. 1975). Each subject was required to
conduct wrist tracking tasks supported by the FES-robot systemwith five different
±
Tab l e 8 . 1
Combination of assistances from the FES and robot parts.
System Assitance
FES Gain
Robot Gain
M5f5
50%
50%
M1f1
100%
100%
M0f1
100%
0%
M0f0
0%
0%
M1f0
0%
100%
The FES ain and robot gain are in accordance with the definition in Eq. 5.2 and in
Eq. 5.3 of
Chapter 5
.
