Extended Data Fig. 2: Augmented residual muscle afferents by restored natural muscle dynamics.
From: Continuous neural control of a bionic limb restores biomimetic gait after amputation
a, Assessment of residual muscle neuromechanics. b, Profiles of the recorded EMG, muscle fascicle strains, and estimated muscle afferents for each cohort are reported (n = 7 per cohort; bolded lines, mean; shaded regions, SEM). Imp s-1, impulses per second. c, EMG, muscle fascicle strain, and estimated muscle afferents during the steady state of the phantom ankle movements (bars, mean; error bars, SEM). Larger agonistic shortening and antagonistic lengthening were found in the AMI compared to those of the CTL cohort. The AMI dorsiflexor and plantar flexor produce larger muscle afferents when each muscle works as an antagonist (lengthening) compared to when it works as an agonist (shortening). In contrast, no significant differences are found between dorsiflexor and plantar flexor afferent signaling when each muscle works as an agonist and antagonist in the CTL cohort. For these analyses, two-sided Mann–Whitney U test was used for the GAS EMG comparison during DF, two-sided paired and unpaired t-tests for other within and between group comparisons, respectively (n = 7 per cohort, *P < 0.035, **P < 0.0068, ***P < 4.9\(\times\)10-4). NS, not significant.