Design and Development of ReMoVES Platform for Motion and Cognitive Rehabilitation

Exergames have recently gained popularity and scientific reliability in the field of assistive computing technology for human well-being. The ReMoVES platform, developed by the author, provides motor and cognitive exergames to be performed by elderly or disabled people, in conjunction with traditional rehabilitation. Data acquisition during the exercise takes place through Microsoft Kinect, Leap Motion and touchscreen monitor. The therapist is provided with feedback on patients' activity over time in order to assess their weakness and correct inaccurate movement attitudes. This work describes the technical characteristics of the ReMoVES platform, designed to be used by multiple locations such as rehabilitation centers or the patient's home, while providing a centralized data collection server. The system includes 15 exergames, developed from scratch by the author, with the aim of promoting motor and cognitive activity through patient entertainment. The ReMoVES platform differs from similar solutions for the automatic data processing features in support of the therapist. Three methods are presented: based on classic data analysis, on Support Vector Machine classification, and finally on Recurrent Neural Networks. The results describe how it is possible to discern patient gaming sessions with adequate performance from those with incorrect movements with an accuracy of up to 92%. The system has been used with real patients and a data database is made available to the scientific community. The aim is to encourage the dissemination of such data to lay the foundations for a comparison between similar studies.