UDC 612.766.1
METRIC DEVELOPMENT FOR EXERCISE RECOGNITION IN REMOTE REHABILITATION SYSTEMS
A. V. Alpatov, Ph.D. (Tech.), associate professor, RSREU, Ryazan; This email address is being protected from spambots. You need JavaScript enabled to view it.
M. S. Ashapkina, assistant, RSREU, Ryazan; This email address is being protected from spambots. You need JavaScript enabled to view it.
The problem of developing quantitative metrics for automated recognition of the types of exercises performed within the system of remote control over the process of physical rehabilitation is studied. The aim is to develop metrics for assessing the correctness of physical exercises, allowing to obtain quantitative characteristics of the correctness of following the protocols of rehabilitation exercises in order to increase objectivity and reliability of remote control. The goal of controlling the correctness of rehabilitation exercises implementation based on the assessment of similarity measure of real trajectories with model representation is found. Comparative analysis of metrics characterizing the distance between two implementations is carried out, using the example of model trajectories. It is concluded that it is necessary to use the metric being invariant to the deformations of trajectory form and time. It is proposed to use dynamic time warping algorithm with additional metrics that localize the properties of the function to minimum strain path.
Key words: rehabilitation, knee, distance metrics, dynamic time warping algorithm, trajectory of movement, waveform, inertial sensors, mobile app.