Note: This content is accessible to all versions of every browser. However, this browser does not seem to support current Web standards, preventing the display of our site's design details.

  

Toward a wearable forearm EMG system for residual volitional control of FES

Author(s):

Jeroen Buil
Conference/Journal:

Master Thesis, HS12 (10177)
Abstract:

Applying functional electrical stimulation (FES) is a recognised rehabilita- tion strategy for stroke and incomplete SCI patients, however it requires voluntary e ort from the patient to be e ective. Quantifying the amount of voluntary e ort during FES is hindered as electrical currents create large artefacts in the EMG signal. A LabVIEW system has been developed that aides the patient performing hand movements. It makes use of a lter that extracts the voluntary e ort by blanking the stimulation artefacts and elim- inating the muscle responses by predicting the shape using a linear adaptive lter. The lter output is fed to a controller that is able to detect the de- sired hand position according to a nite state model with Schmitt Trigger. The stimulation is adjusted linearly based on the current voluntary input as a percentage of the maximum voluntary contraction (MVC). For this system to work with a wearable FES garment for the forearm, it is mod- i ed to allow simultaneous EMG acquisition. Wireless EMG probes are connected to the garment for FES with a custom PCB. As the garment was not originally designed to allow EMG, it is investigated if the signal is not a ected in a signi cant way. The frequency response analysis shows that the signal is not a ected in the range in which surface EMG operates. An average noise level of 0,62 mV does indicate a high noise sensitivity as the electrode array acts as an antenna. The conductive ability of the continuous layer of gel on the array also introduces crosstalk between the electrodes. Cross-correlation results show that this can be reduced by removing a strip of gel between the channels on the array. The results of this study show that, despite relatively high noise levels, proper calibration and ne tuning of the thresholds is sucient to allow the hand states to be recognised by the system. Further testing with simultaneous electrical stimulation will show if the system is suitable FES control.

Year:

2013
Type of Publication:

(12)Diploma/Master Thesis
Supervisor:

A. Crema

No Files for download available.
% Autogenerated BibTeX entry
@PhdThesis { Xxx:2013:IFA_4380
}
Permanent link