Related Experiment Video
Updated: May 14, 2026
Assessment of Neuromuscular Function Using Percutaneous Electrical Nerve Stimulation
Published on: September 13, 2015
Automated determination of peripheral nerve stimulation parameters to achieve desired effector response - a
Paweł Maciejasz1, Wiesław Marcol, Roman Paśniczek
1Institute of Metrology and Biomedical Engineering, Warsaw University of Technology, Warsaw, Poland. pawel.maciejasz@inria.fr
Background:
The feasibility of selectively stimulating fascicles and fibers within peripheral nerves has been demonstrated by a number of groups. Although various multi-contact electrodes have been developed for this purpose, the lack of procedures for fast determination of stimulation parameters to produce the desired effector activity hampers the clinical application of these techniques.In this paper, we propose an automated search routine that may facilitate the determination of stimulation parameters. To verify the routine's performance, we also developed an another routine that performs systematic stimulus-response mapping (the mapping routine).
Method:
The mapping routine performs systematic mapping of all possible combinations of the allowed stimulation parameters (i.e. combinations of electrode contacts used to provide the stimulus and sets of stimulus parameters values) and the observed displacements. The proposed automated search routine, similarly to the mapping routine, maps stimulation parameters to muscle responses, but it first investigates stimuli of the low charge and during the mapping process it compares the recorded responses with the desired one. Depending on the result of that comparison, it decides whether the use of a particular combination of electrode contacts should be further investigated or skipped.Both approaches were implemented on a custom-made closed-loop FES platform and preliminary experiments were performed on a rat model. The rat's sciatic nerve was stimulated with a 12-contact cuff electrode and the resulting displacement of the rat's paw was determined using a MEMS accelerometer.
Results:
The automated search routine was faster than the mapping routine; however, it failed to find correct stimulation parameters in one out of three searches. This could be due to unexpectedly high variability in the responses to a constant stimulus.
Conclusion:
Our initial tests have proven that the proposed method determines the desired stimulation parameters much more quickly than systematic stimulus-response mapping. However, the factors influencing the variability of responses to constant stimuli should be identified, and their influence diminished; the remaining essential variability can then be identified. Thereafter, the criteria influencing the search process should be investigated and refined.Further improvements to the search routine are also proposed.
