The last decade has seen a growing interest in the use of brain computer interfaces (BCI) in gaming for rehabilitation. There have been cases in which use game dynamics that seek to enrich the flow and entertainment BCI games, however the complexity and randomness of the electrical signals of the brain extracted often hinder the design of video games by limiting the possibilities of interaction and the commands used. In this paper we propose a hBCI that articulates the Kinect sensor as additional sensor input commands of the game, providing motion capture in real time to the interpretation of gestures and movements and the neuroheadset Emotiv EPOC for generating SSVEP by visual stimuli, in this case, a particular animation. We implement an interactive room in a medical center that uses the two interfaces together into a video game designed for rehabilitation. Preliminary results were obtained from a patient with hemiparesis, the accuracy of the EPOC software for classifying SSVEPs within the game dynamics was 64 % , lower than the levels attained in a stage of user training prior to the interaction taht was 86 %. The goniometric analysis performed to the user by software that we developed for the analysis of motion capture data from the Kinect sensor produced that in the sagittal plane the arc of movement of the shoulder on the affected side is Fl/Ex 72/12/0, taken while the user was playing, which is short on elevation. Finally, it highlights some features such as portability, fluidity of interaction, the quality of the user experience and the low cost of the mounted system, which becomes an ideal tool for rehabilitation of patients with neuromotor diseases.