Fundación General CSIC


Notebooks of the Fundación General CSIC <span>Digital Edition</span>

Go back to Articles


Ageing-related R&D: FGCSIC Proyectos Cero


Grupo de Ingeniería Biomédica (GIB). Universidad de Valladolid

Brain Computer Interface (BCI) systems applied to cognitive training and home automation control to offset the effects of ageing

The Biomedical Engineering Group at the University of Valladolid aims to use BCI (Brain-Computer Interface) systems to translate users’ intentions into commands and as a cognitive training tool to help offset the effects of ageing. They also plan to develop an assistive BCI application that can be used to control home automation systems and other household electronic devices.

Share |
Brain-Computer Interface (BCI)
Ever since Hans Berger demonstrated the electroencephalography technique for the first time in 1929, it has been one of the basic tools doctors and scientists have used to investigate how the brain works. There has also been speculation about the possibility of using electroencephalograms (EEGs) to decode people’s HIGHLIGHTSProfile: Biomedical Engineering Group
intentions, such that people could use their brain activity to control devices directly. This is the definition of a brain-computer interface (BCI), as a communication system that monitors brain activity and translates certain characteristics, corresponding to users’ intentions, into commands that operate a device. Defined this way, BCI systems could be very useful for people who are dependent on others, as a result of either advanced age or a severe disability, as it would open up a new channel of communication for them.

There are several different methods that can be used to record brain activity: EEG, electro-corticography (ECoG), magnetoencephalography (MEG), positron emission tomography (PET), or functional magnetic resonance imaging (fMRI). ECoG requires an operation to fit the electrodes onto the surface of the cortex, making it an invasive technique. MEG, PET and fMRI require very expensive equipment and facilities. Therefore, the most widely used method for recording brain activity in BCI systems is EEG, as it is a technique that is simple, non-invasive, port­able and low cost.

BCI systems can be classed as endogenous or exogenous depending on the nature of the input signal:

Endogenous BCI systems depend on the user’s ability to control their electrophysiological activity, such as the EEG amplitude in a specific frequency band on a specific area of the cerebral cortex. Endogenous BCI systems include those based on motor images (sensorimotor rhythms) or slow cortical potentials (SCPs). Both require a period of intensive training. These systems work as follows:

  • BCI based on slow cortical potentials (SCPs). SCPs are slow voltage changes generated on the cerebral cortex, with a duration varying between 0.5 and 10 seconds. Negative SCPs are typically associated with movement and other functions that imply cortex activity. It has been demonstrated that people are able to learn to control these potentials.
  • BCI based on motor images or sensorimotor rhythms. This is based on a paradigm of two or more classes of motor images (moving the right or left hand, the feet or tongue, etc.) or other mental tasks (rotating a cube, doing mental arithmetic, etc.) Mental tasks of this type produce changes in the amplitude of the µ (8-12 Hz) and ß (16-24 Hz) sensorimotor rhythms, registered on the somatosensorial and motor zone of the cerebral cortex. These rhythms change when either making an actual movement or imagining or preparing for a movement.

Exogenous BCI systems depend on the electrophysical activity evoked by external stimuli and do not require intensive training. The main characteristics of exogenous systems based on P300 evoked potentials or steady state visual evoked potentials (SSVEP) are described below.

  • BCI based on P300 evoked potentials. The P300 potential is an amplitude peak that appears on the EEG approximately 300 ms after an infrequent auditory or visual stimulus. The user is normally presented with a series of stimuli of which only a few are related to the user’s intention. This means that, as the stimuli of interest are infrequent and mixed with other much more common stimuli, they cause a P300 potential to appear in the user’s brain activity. This potential is observed mainly in the central and parietal regions of the cerebral cortex.
  • BCI based on steady state evoked visual potentials. Evoked visual potentials are detected on the EEG recording on the visual region of the cerebral cortex after application of a visual stimulus to the user. These potentials stabil­ise if the rate at which the visual stimulus is presented is over 6 Hz (6 repetitions a second). When the user turns his or her gaze onto an image that is flickering at a particular rate, it is possible to detect this frequency by analysing the spectrum of the EEG signal, as the amplitude of the SSVEP is increased by the frequency of the second and third harmonic of the flickering image.

Signal processing in BCI systems is usually divided into four stages. It starts with an initial pre-processing stage in which the EEG signals are filtered and some of the possible artefacts superimposed on the signal of interest (flickering, eye movements, electrocardiogram, muscle movements, etc.) are eliminated. This is followed by a second stage in which certain specific characteristics of the EEG signal are extracted. Then, feature selection methods are used to choose the most significant, i.e. those which codify the user’s intention, from extracted set. Finally, classification algorithms translate the selected set of characteristics into a specific command, related to the user’s intention.

Applications of BCI systems
Ageing and dependence are two terms that are increasingly interrelated. Increased life expectancy in Western countries is leading to a gradual increase in the number of people dependent on others. An ageing society calls for new solutions to assist elderly people who find their ability to perform everyday tasks limited and need help to carry them out. BCI systems could turn out to be extremely useful in this respect, as they offer a new way of interacting with the various different devices present in the everyday envir­onment. This would allow
certain basic communication, comfort, leisure and mobility needs to be met. These systems could therefore contribute to improving the ability of people who are dependent on others for their care to live more autonomously, improving their quality of life and social integration.

Set of typical household electronic devices that could be controlled using the proposed BCI application to increase the functional independence of dependent elderly persons: television, DVD player, lamps, stereo, multimedia disk drive, heater, fan, and telephone.

The most frequent applications are aimed at enabling communication, controlling a wheelchair or prosthesis, or environmental control. Similarly, applications to control a computer or browse the Internet have also been developed.

BCI-based communication applications have also been developed. These present the user with the letters of the alphabet in the form of a matrix or a computer keyboard. The user forms words and phrases by selecting letters from the matrix. The technique of selecting letters and characters has been implemented with various types of BCI: based on P300, on slow cortical potentials or on motor images. At present, considerable interest and effort is being devoted to the development of systems which combine the selection of letters with Internet browsing. Along these lines are BCI applications to publish messages on Twitter and BCI browsers to explore web pages.

Various types of BCI systems (based on P300 or on motor images) have been used to control movement of a wheelchair. These systems allow the wheelchair to be started, stopped, turned or instructed to travel to a nearby user-selected location. The wheelchair is also equipped with sensors and other devices allowing it to detect possible obstacles and help avoid them to always ensure user safety. Ageing and dependence are two terms that are increasingly interrelatedA similar application to control routes, movements and turns, that has been implemented with BCI systems, is the remote operation of robots, i.e. controlling robots that may be located thousands of kilometres away.

Lastly, research is being undertaken into the application of BCI systems oriented towards performing mental tasks to boost users’ cognitive capacities. The first applications developed have been promising, although from the point of view of cognitive training and delaying the effects of ageing, there is still a long way to go.

The aim of the “Brain-Computer Interface for cognitive training and domotic assistance against the effects of ageing” project, funded by the Fundación General del CSIC’s Proyectos Cero program, is to develop new assistive applications using BCI systems based on motor images and P300 potentials. These applications will allow various different cognitive processes to be trained and a multitude of heating and air-conditioning devices, lights, entertainment (TV, DVD, musical equipment, etc.) and communication devices (telephone) to be controlled.
Fundación General CSIC Proyecto Cero: “Brain-Computer Interface for cognitive training and domotic assistance against the effects of ageing.”

According to the World Health Organisation (WHO), ageing is the progressive and generalised deterioration of functions, producing a loss of adaptive response to stress and increased risk of suffering ageing-related illnesses. Over time, ageing can lead to situ­ations of disability and depend­ence. In 2008 it was estimated that there were over 3.7 million disabled people in Spain. Almost three million have some form of impediment to carrying out ordinary daily activities, and of this group, over 1.7 million were aged over 64 years. A BCI home-automation control application would allow many of these people to interact with their home environment and so bolster their personal independence. Moreover, the mental tasks used in some types of BCI system could be useful for people in the early stages of ageing, so as to prevent or slow their cognitive decline.

The aim of this Proyecto Cero project is to develop new tools to make life easier for elderly people. And it sets out to use BCI to do so. The goal is therefore, firstly, to prevent or slow the cognitive decline associated with ageing, and, secondly, to facilitate dependent persons’ access to devices present in their normal environment.
The initial aim is to develop a cognitive training application using a BCI system based on motor images. This type of BCI system requires an intensive training stage. During this stage users have to perform a variety of mental tasks (visualising hand, foot or tongue movements, the rotation of a cube, mental arithmetic, etc.). The objective consists of adapting these mental tasks so that they are valid cognitive training tasks for elderly persons. Performing several training sessions with the application developed by the project will help slow the process of cognitive decline, and may even enhance users cognitive capacities.

User performing tests on a BCI system The photograph shows the electrodes that have to be worn to record EEG activity and the compact biomedical signal amplifier used.

The cognitive training BCI application will use µ (8-12 Hz) and ß (16-24 Hz). sensori­motor rhythms. These rhythms exhibit variations in the motor region of the cerebral cortex when the user either makes The objective is to make it easier for elderly people to interact with household electronic devicesa movement or thinks about it. To identify the user’s intention properly, an in-depth review of state of the art in EEG signal processing methods to control BCI systems based on motor images will be carried out. Fourier-transform spectral analysis methods will be used to study wavelets or auto­regressive models; spatial filter­ing methods, such as the common spatial patterns (CSP) and non-linear methods will also be used. In order to select and classify the characteristics, an evolutionary approach with pattern-recognition methods such as genetic algorithms, will be used, along with linear discrimination classification, Bayes classification, vectorial support machines and neuronal networks.

The cognitive tasks implemented in the BCI training application will be designed in collaboration with therapists and elderly users at the National Reference Centre on Disability and Dependence (Centro de Referencia Estatal (CRE) de Discapacidad y Dependencia) in San Andrés del Rabanedo (León). The proposed tasks will be divided into different groups and levels of difficulty so as to evaluate the progress of users’ cognitive skills. The intention is for the application to train and reinforce users’ cognitive skills, so they are able to prevent or slow the effects of ageing.

Secondly, a home-automation control application will be developed using a BCI system based on P300 evoked potentials. The objective is to facilitate the interaction between dependent elderly people and the electronic devices present in the home, so they can lead more comfortable and independent lives. Given that P300-based BCI systems do not require a training stage, this application is aimed at people at advanced stages of ageing and in situations of dependence, with a view to enhancing their independence in the home and thus improve their quality of life. This application will take the final users’ main needs into account: entertainment, comfort, communication, etc. It will therefore allow them to control home automation and electronic devices commonly found in the home and use their main functions:

  • Television: turn on/off, turn the volume up/down, mute, change channels, access the set-up menu, etc.
  • DVD player: turn on/off, explore the DVD contents, play, pause or stop a film, etc.
  • Stereo: turn on/off, turn the volume up/down, mute, change the radio or CD function, select a track or station, etc.
  • Multimedia disk drive: turn on/off, explore the hard drive’s contents, play, pause or stop a film, etc.
  • Telephone: hang up, pick up the phone, dial a number, access the address book, consult the list of missed calls, calls made or received, etc.
  • Lights: turn on/off, change colour, increase or decrease the light intensity.
  • Heating: turn on/off, raise or lower the temperature, switch the timer on/off, etc.
  • Fan: turn on/off, raise or lower the power, switch the timer on/off, etc.

The home automation control application will use P300 potentials produced when the user is shown an infrequent stimulus mixed with multiple frequent stimuli. To do so, the user will be shown a series of images representing the different devices and commands to execute, which will be illumin­ated randomly. The BCI system will identify which of the options displayed on the screen caused the user’s P300 potential when it was illuminated. This will enable the user’s intention to be identified and execute the relevant command. To identify the user’s intention properly, an in-depth review of state of the art in EEG signal processing methods to control BCI systems based on P300 will be carried out. Methods of artefact minimisation and peak detection in the time domain will be studied. For the selection and classification of the characteristics, methods such as Fischer linear discrimination, stepwise linear discrim­inant analysis (SWLDA), Pearson correlation and SVM networks will be studied.

As in the case of the cognitive training application, input from therapists and users at the Reference Centre on Disability and Dependence (CRE de Discapacidad y Dependencia) will be taken into account. Again, the home automation application will be evaluated by elderly users who will conduct various sessions with the application, controlling devices in their environment.

Finally, both applications will be subjected to an overall assessment. In particular, it will be examined whether having performing cognitive training tasks with the first application leads to improved cognitive skills and, therefore, a better control over the home automation application.

Profile: Biomedical Engineering Group

The Biomedical Engineering Group at the University of Valladolid is a multi-disciplinary research team whose lines of research focus mainly on medical image and signal processing, and the development of BCI (brain-computer interface) systems aimed at disabled and elderly people. It currently comprises four lecturers from the School of Telecommunications Engineering, six contract researchers, and seven doctors from various fields of medicine, who are working together on the various research lines. Since its creation, the group has grown considerably in terms of size, involvement in research projects at national and international level, papers published in scientific journals, and prizes awarded. For more information visit the group’s website at:

Published in No. 08

  • ® Fundación General CSIC.
    All rights reserved.
  • Lychnos. ISSN: 2171-6463 (Spanish print edition),
    2172-0207 (English print edition), 2174-5102 (online edition)
  • Privacy and legal notices
  • Contact

Like what we do? Keep up to date
with our latest
news and activities
on Facebook, Twitter or YouTube

Search options