On-the-Edge Gait Analysis Using a Smart Earable Inertial Measurement Unit

Wearable devices are becoming more and more common, especially for what concerns continuous health monitoring. Among them, earables refer to earbuds or headphones, which can be exploited for health monitoring applications when embedded with sensors such as Inertial Measurement Units (IMUs). This paper explores the feasibility of implementing a real-time gait analysis algorithm on-the-edge, thanks to a 6-axis smart IMU developed by STMicroelectronics, the LSM6DSO16IS, featuring an Intelligent Sensor Processing Unit (ISPU). In particular, an earable device embedded with one single IMU was used in order to acquire the data necessary for the study. Starting from the detection of Heel Strike (HS) and Toe Off (TO), an implemented algorithm able to extract various gait parameters (number of steps, step and stride duration, stance and swing phases, single and double support phases, step and stride length, gait speed) led to an average absolute error below 1.5% with respect to the reference in the number of counted steps. The other results were compared to those given by a Python algorithm, taken as gold standard. The algorithm was then embedded in the ISPU of the LSM6DSO16IS and was able to compute all previously named gait parameters in real-time.