Accuracy of three methods in gait event detection during overground running

Inertial measurement units (IMUs) have been extensively used to detect gait events. Various methods have been proposed for detecting initial contact (IC) and toe-off (TO) using IMUs affixed at various anatomical locations. However, the accuracy of such methods has yet to be compared. This study evaluated the accuracy of three common methods used for detecting gait events during jogging and running: (1) S-method, in which IC is identified as the instant of peak foot-resultant acceleration and TO is identified when the acceleration exceeds a threshold of 2g in the region of interest; (2) M-method, in which IC and TO are defined as the minimum before the positive peak shank vertical acceleration and the minimum in the region of interest, respectively; and (3) L-method, in which IC is indicated by the instant of peak pelvis anteroposterior acceleration and TO is identified by the maximum in the region of interest. The performance of the IMU-based methods in detecting IC and TO and estimating stance time (ST) were tested on 11 participants at jogging and running speeds against a reference provided by a force-platform method. The S-method was the most accurate for IC detection (overall mean absolute difference (MAD): 4.7 +/- 4.1 ms). The M-method was the most accurate for TO detection (overall MAD: 7.0 +/- 3.5 ms). A combination of M-and S-methods, called the MS-method, was the most accurate for ST estimation (overall MAD: 9.0 +/- 3.9 ms). Thus, the MS-method is recommended for ST estimation; however, this method requires four IMUs for bilateral estimation.