Gait events detection using inertial sensors, Apple Watch, and the G-WALK reference system

The main challenge in developing gait analysis systems is to ensure accuracy in the detection of gait eve nts. To achieve this, the development of algorithms that detect these events is critical. Besides, knowledge of the places where the events appear on the sensor signals is important. In this study, a method to detect two gait cycle events, in healthy subjects walking at their natural speed, was developed. Angular velocity signals acquired from three devices were used to evaluate the method. The devices were the G-WALK (Reference device), two Inertial Measurement Unit (IMU), and two Apple Watch. Experimental results demonstrated the ability of the method to detect initial contact or heel-strike (HS) and pre-balance or toe-off (TO) events, regardless of the location of the sensor.