Abstract—Tele-rehabilitation is becoming extremely popular in the health scenario. Although this new medical approach has several advantages over traditional therapy, it is important to identify its limitations and risks in the recovery process. This study focuses on the modeling and implementation of a module to assess in real time the quality of movement and the emotional state of patients when they execute the rehabilitation exercises provided by a web-based platform. An algorithm based on dynamic time warping is proposed and tested to assess the movement kinematic. Furthermore, pain is discriminated through a support vector machine classifier. Both methods allow a high identification of gestures and emotions, respectively. These results are discussed in terms of the most suitable solutions to develop an efficient telerehabilitation system.
Index Terms—Dynamic time warping, machine learning, motion assessment, eHealth. affective computing
Yves Rybarczyk is with the University of Skövde, Skövde 54128, Sweden (e-mail: y.rybarczyk@fct.unl.pt). Louis Leconte is with the Ecole Normale Supérieure, Paris-Saclay 94235, France (e-mail: louis.leconte@ens-paris-saclay.fr). Jorge Luis Pérez Medina, Karina Jimenes Vargas, Patricia Acosta- Vargas, and Danilo Esparza are with the Universidad de Las Américas, Quito, Ecuador (e-mail: jorge.perez.medina@udla.edu.ec, karina.jimenes@udla.edu.ec; patricia.acosta@udla.edu.ec; wilmer.esparza@udla.edu.ec).
[PDF]
Cite: Yves Rybarczyk, Louis Leconte, Jorge Luis Pérez Medina, Karina Jimenes Vargas, Patricia Acosta- Vargas, and Danilo Esparza, "Design of a Kinematic and Emotional Assessment Module for the Tele-rehabilitation Platform," International Journal of Modeling and Optimization vol. 9, no. 2, pp. 页, 2019.