Myoelectric signals recorded with electrodes on the skin
surface overlying the residual arm muscles have been
used in control of motorized upper-limb prostheses for
several decades [1-23]. A significant improvement over
the traditional EMG control method in myoelectric
prostheses is the use of EMG pattern recognition based
control strategy, which is grounded on the assumption
that the patterns of EMG signals regarding the intended
movements are consistent and repeatable. Most previous
efforts focused on evaluating the capability of EMG
pattern-recognition algorithms in identifying a number
of motion classes in an ideal laboratory setting. Because
of some disparities between laboratory investigation and
practical use of a myoelectric prosthesis, it should be
required to test the control performance in the conditions
of the clinical setting before the myoelectric prosthesis
systems can be clinically viable. Recently, the
influences of some possible issues associated with clinical
applications on the control performance of a multifunctional
myoelectric prosthesis have come to the
attention. To minimize the effect of unintended movements
caused by motion misclassification during the
real-time EMG pattern-recognition control, Simon et al.
reported the use of decision-based velocity ramp that
could attenuate movement speed after a change in classifier
decision [17]. Their post-processing approach
could provide a finer and smooth transition from
current motion class to next identified one. In clinical
use of a myoelectric prosthesis, misalignment inevitably
occurs during prosthesis donning and doffing, resulting
in a change of electrode locations contacted with skin.
Young et al. investigated how the size of the electrode
detection surface and the electrode orientation affected
the robustness of EMG pattern-recognition based prosthesis
control system with electrode shift [18]. While
these reported progresses have been significantly made
towards the clinical applications of EMG patternrecognition
based control, there are still some important
disparities between the laboratory research results and
the clinical performance that remain to be addressed before
the multifunctional myoelectric prostheses are available
for clinical use.
In most reported studies of EMG pattern recognition
systems for multifunctional prosthesis control, subjects
generally took a seated position with a tested arm resting
on a plate surface such as chair arm or table and multichannel
EMG signals were acquired with a number of
surface electrodes placed on either the muscles of forearm
and hand for an able-bodied subject or the residual
muscles for an upper-limb amputee. One portion of the
acquired EMG data was used to train a classifier and
then remaining portion was loaded into the trained classifier
for calculating the offline classification
Myoelectric signals recorded with electrodes on the skin
surface overlying the residual arm muscles have been
used in control of motorized upper-limb prostheses for
several decades [1-23]. A significant improvement over
the traditional EMG control method in myoelectric
prostheses is the use of EMG pattern recognition based
control strategy, which is grounded on the assumption
that the patterns of EMG signals regarding the intended
movements are consistent and repeatable. Most previous
efforts focused on evaluating the capability of EMG
pattern-recognition algorithms in identifying a number
of motion classes in an ideal laboratory setting. Because
of some disparities between laboratory investigation and
practical use of a myoelectric prosthesis, it should be
required to test the control performance in the conditions
of the clinical setting before the myoelectric prosthesis
systems can be clinically viable. Recently, the
influences of some possible issues associated with clinical
applications on the control performance of a multifunctional
myoelectric prosthesis have come to the
attention. To minimize the effect of unintended movements
caused by motion misclassification during the
real-time EMG pattern-recognition control, Simon et al.
reported the use of decision-based velocity ramp that
could attenuate movement speed after a change in classifier
decision [17]. Their post-processing approach
could provide a finer and smooth transition from
current motion class to next identified one. In clinical
use of a myoelectric prosthesis, misalignment inevitably
occurs during prosthesis donning and doffing, resulting
in a change of electrode locations contacted with skin.
Young et al. investigated how the size of the electrode
detection surface and the electrode orientation affected
the robustness of EMG pattern-recognition based prosthesis
control system with electrode shift [18]. While
these reported progresses have been significantly made
towards the clinical applications of EMG patternrecognition
based control, there are still some important
disparities between the laboratory research results and
the clinical performance that remain to be addressed before
the multifunctional myoelectric prostheses are available
for clinical use.
In most reported studies of EMG pattern recognition
systems for multifunctional prosthesis control, subjects
generally took a seated position with a tested arm resting
on a plate surface such as chair arm or table and multichannel
EMG signals were acquired with a number of
surface electrodes placed on either the muscles of forearm
and hand for an able-bodied subject or the residual
muscles for an upper-limb amputee. One portion of the
acquired EMG data was used to train a classifier and
then remaining portion was loaded into the trained classifier
for calculating the offline classification
การแปล กรุณารอสักครู่..
![](//thimg.ilovetranslation.com/pic/loading_3.gif?v=b9814dd30c1d7c59_8619)