Stride-to-stride gait variability in daily life measured using accelerometers attached to the wrist

A stride-to-stride gait variability is an important predictor for various neurological and age-related diseases [1], [2]. Stride-to-stride variation is generally estimated from the gait cycle time measured at each gait cycle. Foot-worn sensors provide an accurate estimation of gait cycle time, however, the sensors may be removed multiple times a day. An ideal sensor location for activity monitoring is the wrist. Although today’s wrist fitness trackers are used for step counting they are not able to provide gait variability. In this study we propose a new method to detect locomotion and accurately estimate stride-to-stride variability using a wrist sensor.