Mobile smart systems to detect balance motion in rehabilitation

Abstract

In the present paper, a mobile smart system is introduced to assess the individual’s balance in remote rehabilitation. Gyroscope sensor is used in order to analyze the individual’s balance during the remote rehabilitation for five movement activities. The data acquired from the sensor are transmitted via edge layer and SDN controllers to the server database leading to reduction of costs, control of network traffic, and mitigation of delay. The transmitted raw data are analyzed using unsupervised K Means algorithm. The respective algorithm performed the best separation with K values equal to 5 and 8. In fact, accuracy of this method for the 5 movement activities is equal to 0.7 and 0.8 for k = 5 and k = 8, respectively.