Reliability of EMG normalisation methods for upper-limb muscles.

The study investigated different electromyographic (EMG) normalisation methods for upper-limb muscles. This assessment aimed at comparing the EMG amplitude and the reliability of EMG values obtained with each method. Eighteen male tennis players completed isometric maximal voluntary contractions and dynamic strength exercises (push-ups and chin-ups) on three separate test sessions over at least 7 days. Surface EMG activity of nine upper body muscles was recorded. For each muscle, an analysis of variance for repeated measures was used to compare maximal EMG amplitudes between test conditions. The intra-class correlation coefficient, the coefficient of variation and the standard error of measurement were calculated to determine the EMG reliability of each condition. On the basis of a compromise between maximal EMG amplitude and high reliability, the chin-ups appeared to be the optimal normalisation method for M. latissimus dorsi, M. posterior deltoid, M. biceps brachii, M. flexor carpi radialis and M. extensor carpi radialis. The push-ups seemed relevant to normalise M. anterior deltoid and M. triceps brachii activity, while isometric maximal voluntary contraction remained the most appropriate method for M. pectoralis major and M. middle deltoid. Thus, original methods are proposed to normalise EMG signal of upper-limb muscles.