Reliability and normalization of long-term versus substituted electromyography electrodes set-ups

Abstract

Normalization methods are used to minimize the influence of external disturbances in the analysis of myo-electrical signals. Whether these methods actually minimize the error or only relocate it has not been evaluated so far. This verification is only possible in comparison to constant discharge conditions. The aims of this study were (I) to compare the inter-day test-retest reliability of myo-electrical signals from long-term with substituted electrodes and (II) the impact of different normalization methods on test-retest reliability. Myo-electrical activity of the tibialis anterior (TA) and vastus medialis (VA) was recorded from 14 cyclists during 1min cycling with permanent and substituted electrodes and then normalized by selected methods. A good reliability (ICC>0.86) was shown for both muscles with the long-term electrodes whereas poor reliability (ICC=0.32) was found for the substituted electrodes set-up (VM). Only the sprint normalization improved the reliability of the long-term electrode set-up, while the other functional approaches improved at least the reliability from substituted electrodes. Only the sprint method achieved the reliability of permanently attached electrodes. Thus, the results rather suggest the use of permanent electrode set ups. If this is not possible due to the study design, the use of functional normalization procedures is recommended.