Table 1.
List of subjects for whom the walking data is used in this study.
Complete details on amputation can be found in [26].
Fig 1.
a): Placement of tibia markers for leg orientation and velocity calculation, b) A typical shank velocity signal with raw (grey) and filtered (black) data, c) Enlarged view of one gait cycle from the velocity signal. The algorithm starts with the detection of the largest positive peaks in the signal (marked as MS) which define the intervals for gait events. TO is identified as the last negative peak (or minima) just before the MS while HS is defined as the negative peak just after the MS.
Table 2.
Mean error values for heel strike and toe-off events.
Fig 2.
Bland Altman plots illustrating the agreement of four temporal parameters calculated using the algorithm-based method and those derived from the force-plate data.
The average difference is specified by a solid line while limits of agreement (± 1.96SD) are represented by dotted lines. The figure files have also been provided in supplement to this article. RPC: Reproducibility co-efficient. CV: Coefficient of variation.
Table 3.
Absolute error values for gait events and parameters.
Fig 3.
Heel-strike (left) and Toe-off (right) absolute error values separated by subject groups: Black bars for K-level 2 subjects and white bars for K-level 3. P-values are indicated for the Mann-Whitney U test comparing the two subject groups. In general, K3 subjects showed smaller error values and/or standard deviations.
Fig 4.
Speed-wise heel-strike (top), toe-off (middle), and stance/swing time (bottom) absolute error values, further separated by subject groups (left: K-level 2 subjects and right: K-level 3 subjects) and leg side (blue: sound, red: prosthetic).
Table 4.
A comparison with the error magnitudes found in this study and the available relevant literature, ME: Mean Error, MAE: Mean Absolute Error.