A Comparison of Bioelectric and Biomechanical EMG Normalization Techniques in Healthy Older and Young Adults during Walking Gait

This study compares biomechanical and bioelectric electromyography (EMG) normalization techniques across disparate age cohorts during walking to assess the impact of normalization methods on the functional interpretation of EMG data. The biomechanical method involved scaling EMG to a target absolute torque (EMG_TS) from a joint-specific task and the chosen bioelectric methods were peak and mean normalization taken from the EMG signal during gait, referred to as dynamic mean and dynamic peak normalization (EMG_Mean and EMG_Peak). The effects of normalization on EMG amplitude, activation pattern, and inter-subject variability were compared between disparate cohorts, including OLD (76.6 yrs N = 12) and YOUNG (26.6 yrs N = 12), in five lower-limb muscles. EMG_Peak normalization resulted in differences between YOUNG and OLD cohorts in Biceps Femoris (BF) and Medial Gastrocnemius (MG) that were not observed with EMG_Mean or EMG_TS normalization. EMG_Peak and EMG_Mean normalization also demonstrated interactions between age and the phase of gait in BF that were not seen with EMG_TS. Correlations showed that activation patterns across the gait cycle were similar between all methods for both age groups and the coefficient of variation comparisons found that EMG_TS produced the greatest inter-subject variability. We have shown that the normalization technique can influence the interpretation of findings when comparing disparate populations, highlighting the need to carefully interpret functional differences in EMG between disparate cohorts.