Diseño y desarrollo de una aplicación para controlar un teléfono móvil mediante sistemas Brain Computer Interface (BCI) orientada a personas con grave discapacidad

Abstract

El objetivo de este TFM es desarrollar y evaluar una aplicación BCI que permita el control de un Smartphone y sus principales funcionalidades. Se ha realizado una revisión del estado del arte para elegir las mejores herramientas y métodos disponibles. Posteriormente, se comenzó la etapa de desarrollo de la aplicación, diseñada en Android. La aplicación consta de varias subaplicaciones que permiten controlar llamadas y SMS, un navegador web, alarmas, temporizador, cronómetro, twitter, y la agenda de contactos. Una vez desarrollada la aplicación, se diseñó un protocolo de evaluación con cinco sujetos de control. Se realizaron diversas pruebas repartidas en dos sesiones, cuya precisión media fue del 94,9%.

Since the development of electroencephalogram (EEG) by Hans Berger in 1929, electroencephalography has grown up greatly. Thanks to this discovery, the brain waves began to be studied for the diagnosis of diseases such as epilepsy or sleep disorders. It was in the second half of the twentieth century when the scientific community began to suggest the application of brain signals for communication without the mediation of peripheral muscles or nerves. This could improve people’s lives with severe motor disability. This first system was developed by Dr. Jacques Vidal in 1977, naming it BCI (Brain Computer Interface). These systems allow to control machines, translating the user's intentions into commands by means of extracting and translating the EEG signal features. This paper aims to develop and evaluate a BCI application that allows the users to control a mobile phone and its main features. After doing a review of the state of the art, it is concluded that the best way to accomplish this goal is by developing a BCI application based on the oddball paradigm. This BCI paradigm selects commands represented in the cells of a matrix by detecting P300 potentials in the EEG, provoked by random visual stimuli and generated by the user, endogenously. The designed system is distributed between a computer and the mobile device which the user has to control. Due to its greater capacity, oddball paradigm and the signal processing are performed on the computer that translates the user’s intentions into commands. Afterwards, they are sent to the mobile device using Bluetooth technology. As soon as the command is received, the mobile device will run it. The tasks available in the application allow the user to make calls, send SMS, add, view or delete contacts, take pictures and display them, or add alarms and events to the calendar. The application was developed and evaluated by five healthy subjects, who had to perform various tasks to control the Smartphone with increasing difficulty, spread over two sessions held on different days. The results obtained were satisfactory, with an average accuracy of 94.38% performed under controlled laboratory conditions. After these tasks, the subjects completed a satisfaction questionnaire that allowed to know their personal opinion of the system, and fed back the project with their suggestions.