Continuous functional activity monitoring based on wearable tri-axial accelerometer and gyroscope

Given the growing number of elderly people and patients diagnosed with Parkinson´s disease, monitoring functional activities using wearable wireless sensors can be used to promote the Quality of Life and healthier life styles. We propose a novel and practical solution using three small wearable wireless Functional Activity Monitor (FAM) sensors and a smartphone to store, transmit, analyze and update data. Three sensors, each composed of a tri-axial accelerometer and a tri-axial gyroscope, are attached to the chest and both thighs. A computationally efficient signal processing algorithm is designed to accurately measure tilting angles. A continuous activity recognition algorithm is developed using a decision tree based on time series data and spectrum analysis; this algorithm can identify activities of daily life in three general categories: (1) postures such as standing, sitting, and lying; (2) locomotion such as walking; and (3) transitions such as sit-to-stand and stand-to-sit. The results show an accurate angle measurement compared to the motion capture system Optotrak 3020 and a reliable detection of all activities with sensitivity at least 96.2% compared to video recordings.