A Review on Brain Imaging Techniques for BCI Applications

Introduction

The idea of communicating with computers just by thought has intrigued mankind for a long time and the advent in cognitive science and a large number of technologies for mapping the brain activity has helped it to become a reality. The mapping of neural signals or imaging of activities taking place in different regions of the human brain has been made possible by using different sensors which have the capability of detecting any physical change in those regions. At the beginning, scientists thought that, this system will be very helpful for reinstating people with disabilities to their former state and improve the quality of life (Lebedev, M. A., 2006). For that purpose, people are trained to manipulate their thought so as to control the external device (e.g. prosthetic arm, wheelchair and likewise). This technology is particularly useful for providing rehabilitation to people suffering from various neural diseases like amyotrophic lateral sclerosis (Nijboer, F.et al., 2008), paralysis (Birbaumer, N.et al.., 2008), cerebral palsy and also to people who have lost one or more of their limb(s) (Neuper, C.et al.., 2003). Apart from rehabilitation, brain computer interface (BCI) has found application in various other domains like communication (Wolpaw, J. R.et al.., 2002), computer gaming (Krepki, R.et al.., 2007), robot control (Galán, F.et al.., 2008) and virtual reality.

A brain computer interface (BCI) basically translates thethoughtsto the machine action (computer) by decoding the brain activities. The signals generated by the nerves in the brain are acquired first and then a number of steps are undergone to find out whether these signals form a pattern or not. The main objective of any BCI system is to detect any such pattern from the neural activity of a person's brain and then decode the actual intent of that person from it to drive or communicate with machines. This system then converts the acquired signals into machine understandable signals so that those can be used for controlling any external device. There are mainly three types of BCI systems available, namely invasive, partially-invasive and non-invasive BCI (Vourvopoulos, A.et al., 2012). In case of non-invasive BCIs, the signals are acquired superficially using a number of technologies. The invasive BCIs are used mostly for surgical purposes and the signals are obtained by placing the electrodes directly within grey matter of the brain. Whereas partially-invasive BCIs are connects the electrodes inside the brain, but the rest of the components are placed outside.