Towards electrodeless EMG linear envelope signal recording for myo-activated prostheses control

  • Nawadita Parajuli

Western Sydney University thesis: Doctoral thesis

Abstract

After amputation, the residual muscles of the limb may function in a normal way, enabling the electromyogram (EMG) signals recorded from them to be used to drive a replacement limb. These replacement limbs are called myoelectric prosthesis. The prostheses that use EMG have always been the first choice for both clinicians and engineers. Unfortunately, due to the many drawbacks of EMG (e.g. skin preparation, electromagnetic interferences, high sample rate, etc.); researchers have aspired to find suitable alternatives. One proposes the dry-contact, low-cost sensor based on a force-sensitive resistor (FSR) as a valid alternative which instead of detecting electrical events, detects mechanical events of muscle. FSR sensor is placed on the skin through a hard, circular base to sense the muscle contraction and to acquire the signal. Similarly, to reduce the output drift (resistance) caused by FSR edges (creep) and to maintain the FSR sensitivity over a wide input force range, signal conditioning (Voltage output proportional to force) is implemented. This FSR signal acquired using FSR sensor can be used directly to replace the EMG linear envelope (an important control signal in prosthetics applications). To find the best FSR position(s) to replace a single EMG lead, the simultaneous recording of EMG and FSR output is performed. Three FSRs are placed directly over the EMG electrodes, in the middle of the targeted muscle and then the individual (FSR1, FSR2 and FSR3) and combination of FSR (e.g. FSR1+FSR2, FSR2-FSR3) is evaluated. The experiment is performed on a small sample of five volunteer subjects. The result shows a high correlation (up to 0.94) between FSR output and EMG linear envelope. Consequently, the usage of the best FSR sensor position shows the ability of electrode less FSR-LE to proportionally control the prosthesis (3-D claw). Furthermore, FSR can be used to develop a universal programmable muscle signal sensor that can be suitable to control the myo-activated prosthesis.
Date of Award2020
Original languageEnglish

Keywords

  • myoelectric prosthesis
  • electromyography
  • biomedical engineering
  • biosensors

Cite this

'