A validation test of a simple method of stride length measurement only with inertial sensors and a preliminary test in FES-assisted hemiplegic gait

The purpose of this study was to test stride length measurement only with inertial sensors to for the simple gait evaluation system developed in our previous study to support rehabilitation, healthcare and so on. In this paper, a calculation method of the stride length was developed, in which acceleration and angular velocity signals of the foot were used. The integration period of the acceleration for calculation of each stride length was automatically detected by the acceleration signal. First, the measurement method was evaluated in 10 m walking of 6 neurologically intact subjects. The average values of error and absolute error of measured stride lengths were 0.38±4.77% and 3.36±3.04%, respectively. Correlation coefficient of the measured length with the reference data was 0.910 and the slope of the regression equation was 0.978. Then, stride lengths of a hemiplegic subject were measured in 10 m walking with and without FES-assisted foot drop correction. Although there was no difference in stride length between with and without the FES-assist, the calculated data from inertial sensor signals supported that the time for 10 m walking measured by therapists decreased when the subject walked with the FES-assist. It was expected that the stride length measurement only with inertial sensor would be practical, and it can be implemented easily.