Skip to Main Content
The development of functional activity monitors (FAMs) will allow rehabilitation researchers and clinicians to evaluate treatment efficacy, to monitor compliance to exercise instructions, and to provide real time feedback in the treatment of movement disorders during the performance of daily activities. The purpose of the present study was to develop and test a small sized wearable FAM system comprised of three sensors positioned on the sternum and both thighs, wireless Bluetooth transmission capability to a smartphone, and computationally efficient activity detection algorithms for the accurate detection of functional activities. Each sensor was composed of a tri-axial accelerometer and a tri-axial gyroscope. Computationally efficient activity recognition algorithms were developed, using a sliding window of 1 second, the variability of the tilt angle time series and power spectral analysis. In addition, it includes a decision tree that identifies postures such as sitting, standing and lying, walking at comfortable, slow and fast speeds, transitions between these functional activities (e.g, sit-to-stand and stand-to-sit), activity duration and step frequency. In a research lab setting the output of the FAM system, video recordings and a 3D motion analysis system were compared in 10 healthy young adults. The results show that the agreement between the FAM system and the video recordings ranged from 98.10% to 100% for all postures, transfers and walking periods. There were no significant differences in activity durations and step frequency between measurement instruments.