Extended Data Fig. 4: Neuroprosthesis restores symmetric lower-extremity kinematics during gait.
From: Continuous neural control of a bionic limb restores biomimetic gait after amputation
a, Range of motion (ROM) analysis of lower-extremity joint kinematics for all gait conditions tested. For level-ground walking, both speed-matched and each cohort’s maximum speed walking comparisons are reported. Two-sided Mann–Whitney U test was used for the contralateral knee ROM comparison of maximum CTL and moderate AMI level-ground walking and ankle ROM comparisons of slope adaptation, two-sided unpaired t-tests for other comparisons (bars, mean; error bars, SEM; n = 7 per cohort, *P < 0.034, **P < 0.0084, ***P < 5.8\(\times\)10-4). NS, not significant. b, Principal component analysis (PCA) applied on ROM and symmetry data of lower-extremity kinematics. The maximum speed AMI walking data were excluded from the analysis to focus on data obtained from the matched conditions. The analysis results show that, without the muscle afferent augmentation, CTL subjects increased bionic limb side knee motions to compensate for the constrained bionic ankle, which prevented them from performing symmetric gait. CTL subjects also increased bionic-limb side-hip motions during terrain adaptation to compensate for the constrained bionic ankle. Two-sided unpaired t-test was used for the PC1 comparison (bars, mean; error bars, SEM; n = 7 per cohort, ***P = 7.0\(\times\)10-5).