Low power activity monitoring using accelerometer data from an inertial measurement unit (IMU)

Abstract

A custom-made device equipped with a low-power IMU sensor and BLE functionality was developed for comprehensive activity monitoring of elderly individuals and geriatric patients. The device automatically recognizes various types of movements and orientations, storing data on a microcontroller for up to a week before transferring it to a Raspberry Pi and then to a database for visualization and evaluation. Unlike similar approaches, this method doesn't rely on machine learning, resulting in lower computing power requirements, decreased energy consumption, longer battery life, and improved user acceptance due to a smaller device size and less frequent need for recharging.