Motor Imagery for Mouse Automation and Control by ijcsiseditor


More Info
									                                                         (IJCSIS) International Journal of Computer Science and Information Security,
                                                         Vol. 9, No. 10, October 2011

      Motor Imagery for Mouse Automation and Control
                     Bedi Rajneesh Kaur                                                        Kad Reshma Hanumant
                 Dept. of computer engineering,                                              Dept. of computer engineering,
                           MIT COE,                                                          MIT COE,Pune, India, 411038
                      Pune, India, 411038                                           
                                                                                               Katariya Payal Jawahar
                     Bhor Rohan Tatyaba                                                      Dept. of computer engineering,
                 Dept. of computer engineering,                                                        MIT COE,
                           MIT COE,                                                               Pune, India, 411038
                      Pune, India, 411038                                           
                                                                                             Gove Nitinkumar Rajendra
                                                                                             Dept. of computer engineering,
                                                                                                       MIT COE,
                                                                                                  Pune, India, 411038

Abstract—A brain-computer interface (BCI) basically transforms            phenomenon is referred as Event-related Desynchronization
the brain’s electrical activity into commands that can be used to         (ERD)[6].
control devices such as robotic arms, pianos and other devices.                The Graz-BCI developed at Graz university of technology
With this, BCI provides a non-muscular communication channel,             by the pfurtscheller's group during nineties was the firt online
which can be used to help people with highly compromised motor            BCI sytem that used ERD classification in signle EEG trials to
abilities or functions. Mental imagery is the mental rehearsal of
actions without overt execution. A study of motor imagery can
                                                                          differentiate between various types of motor execution and
help us to develop better neuroprosthetic systems. In this paper,         motor imagery. After these basic studies, ERD during motor
we describe general concepts about motor imagery and other                imagery has been investigated for its usability for device
aspects associated with it. Recent researches in this field, has          control by various scientists.
employed motor imagery in normal and brain-damaged subjects
to understand the content and structure of covert processes that                       II.    PHYSIOLOGICAL ASPECTS RELATED
occur before execution of action. Finally, we propose a new                                          TO MOTOR IMAGERY
system “μMAC”, which will automate and control basic mouse
operations using motor imagery.                                                Simulating a particular activity mentally leads to
                                                                          activation of motor pathways. An increase is seen in muscular
Keywords- Mu waves, Motor imagery, EEG, Neuroprosthesis, BCI,             activity during the motor imagery [7]. During this scenario,
Mouse Control.                                                            electromyography is limited to specifically those muscles
                                                                          which participate in simulated action [8]. Motor imagery is
            I.      INTRODUCTION                                          independent of ability to execute the movement and is
                                                                          dependent on central processing mechanism.
     Motor imagery is a one of the most studied and
                                                                               It has been demonstrated by using various brain imaging
researched topic in the field of cognitive neuroscience.
                                                                          methods that different distinct regions of cortex are activated
Roughly stated, motor imagery is a mental state wherein a
                                                                          during motor imagery i.e. MI [9]. It has been revealed in
subject imagines something. To be more specific, motor
                                                                          neural studies that imagined and actual actions share the same
imagery is a dynamic state during which the subject mentally
                                                                          subtrates or brain areas. Various brain areas that get activated
simulates a given action.
                                                                          during motor imagery are supplementary motor area, primary
     According to Jeannerod, motor imagery is a result of
                                                                          motor cortex, the inferior parietal cortex, basal ganglia and the
conscious access to the contents of intent of movement [1][2].
Motor imagery is a cognitive state which can be experienced
                                                                               Fig 1 shows pattern of cortical activation during mental
virtually by anyone without more training. It is similar to
                                                                          motor imagery in normal subjects. The main Brodmann areas
many real time situations that are experienced in life like
                                                                          activated during motor imagery have been outlined on
watching others performing action with intention to imitate it,
                                                                          schematic views of a left hemisphere [7]. As shown in figure,
making moves, imagining oneself performing action and many
                                                                          there is consistent involvement of pre-motor area 6, without
more [3][4]. While preparing and imagining a particular
                                                                          involvement of primary motor cortex (M1). The AC-PC line
movement, the mu and central beta rhythm are desynchronized
                                                                          defines the horizontal reference line in magnetic resonance
over the contralateral primary sensorimotor area [5]. This
                                                                          imaging (MRI) scan. The vertical line passing though the AC

                                                                                                      ISSN 1947-5500
                                                          (IJCSIS) International Journal of Computer Science and Information Security,
                                                          Vol. 9, No. 10, October 2011

(VAC) defines a verticofrontal plane. VPC is the vertical line                    III.  MENTAL REHEARSAL STRATEGIES FOR
passing through the PC [10].                                                                           MOTOR IMAGERY
     The two rhythms that are strongly related with motor                       Basically, there are two different strategies that a subject
imagery are mu and central beta rhythms. The main                          may take or opt when asked to rehearse mentally a motor task
characteristic that defines the mu rhythm is that it attenuates in            These are -
one cerebral hemisphere during preparation of contralateral                             1. Visual Imagery
extremity movement [5], the thought of the contralateral                                2. Kinetic Imagery
movement or tactile electrical simulation of a contralateral                  1. Visual Imagery:
limb. As these rhythms are associated with cortical areas                           In this strategy, the subject produces a visual
having most direct connection with the brain's normal motor                   representation of their moving limb(s). The subject views
output channels, they are quite promising for BCI research.                   himself from third person perspective (e.g. seeing one
     Other thing which should be considered is that, the                      running from an external point of reference).
frequencies that are easy to be performed during ME may be                         This type of imagery is also referred to as external
too fast to imagine for a subject who is not used to motor                    imagery as for a person to view movements must have a
imagery training. Due to this, most of the researchers use                    third person perspective. VI activates regions primarily
motor imagery with half of the velocity (0.5Hz) that are used                 concerned with visual processing and does not obey Fitt’s
for movement execution in simple movements [12].                              law nor is it correlated with excitability of the cortico-
                                                                              spinal path as assessed by transcranial magnetic
                                                                              stimulation [11].
                                                                              2. Kinetic Imagery:
                                                                                    In this strategy, the subject rehearses or practices the
                                                                              particular movements using the kinesthetic feeling of the
                                                                              movement. Here, the subject sees himself from first
                                                                              person perspective. This type of imagery is also referred
                                                                              to as internal imagery. Each type of motor imagery has
                                                                              different properties with respect to both psychophysical
                                                                              and physiological perspectives. The motor and sensory
                                                                              regions that are activated during KI are same as those
                                                                              activated during overt movement [11].
                                                                                   Motor or kinesthetic imagery has to be differentiated
                                                                              from visual imagery because it shows different qualities:
                                                                              not the virtual environment is imagined in a third person’s
                                                                              view but introspective kinesthetic feelings of moving the
                                                                              limb in the first person’s view [10].

                                                                                           IV. TRAINING MOTOR SKILL
                                                                               A subject doing mental practice/task with MI is required
                                                                               to have all the declarative knowledge about the various
                                                                               component of that specific activity/task before practicing
                                                                               it [13]. So, a proper training should be given to subjects
                                                                               about the various components of an activity/task that they
                                                                               are going to rehearse or practice.
                                                                                    The non-conscious processes involved in mental task
                                                                               training are best activated by the internally driven images
                                                                               which promote the kinesthetic feeling of movement [13].
                                                                               Mental training and execution training are two
                                                                               complementary techniques.
                                                                                    According to Gandevia, motor imagery improves the
    Fig.1 Pattern of cortical activation during mental motor                   dynamics of motor performance, for instance the
imagery in normal subjects [7].                                                movement trajectories [14]. The lower effect of MI
                                                                               training compared to ME training may be caused by
                                                                               lacking sensorimotor feedback which results in decreased
                                                                               progress in motor training in lesion patients [15].
                                                                               Sufficient level of complexity of imagined motor
                                                                               task/activity ensures occurrence of lateralizing effect of
                                                                               brain activation during MI [16]. An everyday activity can

                                                                                                      ISSN 1947-5500
                                                   (IJCSIS) International Journal of Computer Science and Information Security,
                                                   Vol. 9, No. 10, October 2011

also be used for study of brain activations during MI in             Signal Acquisition Unit:
training.                                                                    The proposed system works on multi-channel EEG
     This has two potential advantages [17]:                         signals that are generated for each motor imagery activity.
     1. Easy modulation in their complexity.                         This unit receives the EEG signals from the sensors that
                                                                     are attached to the scalp of the subject’s head. The signals
    2.   Familiarity of task to subject helps him to                 captured by the signal acquisition unit are then passed to
                                                                     the spike sorting unit for further processing.
         generate vivid mental representation without any
         prior practice.                                             Spike Sorting Unit:
                                                                            The signal captured by signal acquisition system
     Motor imagery is widely used by athletes and
                                                                     contains noise and other unwanted spikes. These are then
musicians of improving their performance. It can be used
                                                                     processed by the spike sorting unit. The signal here is
for automation and control of mouse operations on
                                                                     processed in three phases:
system. Various studies have elaborated and demonstrated
applications of motor imagery for controlling mouse
                                                                          a) Preprocessing:
operations [21-24].
                                                                                   This phase is responsible for artifact
            V. THE PROPOSED SYSTEM                                             removal from the acquired EEG signals.
         The systems that are proposed in these studies try
to implement 1-D or 2-D control of mouse operations.                      b) Feature Extraction:
        Here, we propose a system that will try to automate
all the operations of mouse by using motor imagery. This                           This phase extracts differed desired features
includes mouse movement, left click, right click and                           from the processed signal.
double click. Following figure fig.2 shows a block
diagram of the proposed system. Different parts of system                 c)   Detection and classification:
are explained below:
                                                                                     This phase is responsible for actual spike
                                                                               detection and its clustering into different classes.

                                                                       Signal Decoding Module:
                                                                             This module actually decodes/detects a particular
                                                                     motor imagery signal of system’s concern which is further
                                                                     used by control module to automate the mouse operation.

                                                                       Control Module:
                                                                              This module on receiving the decoded signal
                                                                     from signal decoding module actually replicates the
                                                                     desired mouse operation on the monitor.

                                                                              This is an actual display on which mouse
                                                                     operation is replicated.
                                                                         Finally, the user receives the video feedback in the
                                                                     form of the mouse operation. This helps in monitoring the
                                                                     performance of the system.


                                                                          This paper explains the basics of motor imagery, its
                                                                          Applications and other factors related to it. It also
                                                                          proposes a system for automation and control of
                                                                          mouse operation using brain mu and beta rhythms
                                                                          that are fired during this activity. This system will
                                                                          eventually make the existing systems more
                                                                          interactive and usable for physically challenged
                                                                          people. Apart from this, the system is quite sensitive
         Fig 2 Block Diagram of proposed system

                                                                                              ISSN 1947-5500
                                                                      (IJCSIS) International Journal of Computer Science and Information Security,
                                                                      Vol. 9, No. 10, October 2011

         to the level of excellence with which the respective                            control by combining Mu/Beta rhythm and P300 potential” , 2010, IEEE
                                                                                         trans., On Biomedical Engineering
         subject rehearses the desired movement or action.
     In the future work, we plan to implement this system to                                                  AUTHORS PROFILE
make the proposed system usage easier and interactive for
physically challenged people.
                                                                                          1.   Prof. Rajneesh kaur Bedi
                            ACKNOWLEDGMENT                                                     Head of Department
                                                                                               Department of computer engineering
  We are thankful to the department of computer engineering,                                   MIT COE, Pune, India
MIT COE, for their kind support.                                                               411038

                               REFERENCES                                                 2.   Mr. Bhor Rohan Tatyaba
[1] Jeannerod M. “The representing brain. Neural correlates of motor intention                 Currently pursuing bachelor’s degree in computer
     and imagery”, Behav Sci 1994, 17:187-245                                                  engineering at,        Department of computer
[2] Jeannerod M. “Mental imagery in the motor context”, Neuropsychologia
     1995, 33:1419-1432
                                                                                               engineering MIT COE, Pune, India, 411038
[3] Annet J. “Motor imagery perception or action”, Neuropshycologiy”, 1995,
     33:1395-1417,                                                                        3.   Miss. Kad Reshma Hanumant
[4] Williams JD, Rippon G., Stone BM, Annett J                   “Psychological                Currently pursuing bachelors degree in computer
     correlates of dynamic imagery”. Br J Psychol, 1995, 86:283-300.
[5] Pfurtscheller G., and Neuper ch., Flotzinger D. ,Pregenzer “EEG based
                                                                                               engineering at, Department of computer engineering
     discrimination between imagery right and left movement” Electroenceph                     MIT COE, Pune, India, 411038
     Clin Neurophysiol 1997.
[6] Schoegl A.,Pfurtscheller G “Subject specific EEG patterns during motor                4.   Miss. Katariya Payal Jawahar
     imagery”,19th international conference IEEE 1997
[7] Marc Jeannerod,Jean Decety “Mental motor imagery: a window into the
                                                                                               Currently pursuing bachelors degree in computer
     representational statges of action” Neurobiology 1995, 5:727-732                          engineering at, Department of computer engineering
[8] Wehner T, Bogi S , Stadler M “Task specific EMG characteristics during                     MIT COE, Pune, India, 411038
     mental training” Psychol Res 1984 , 46:389-401
[9] A. Solodkin, P... Huichin, E chen “Small cerebral cortes.” Cerebral cortex
     14(2004) 1246-1255
                                                                                          5.   Mr. Gove Nitinkumar Rajendra
 [10] Decety J, Perani D, Jeannerod M, Bettinardi V,Tadary B, Woods R,                         Currently pursuing bachelors degree in computer
     Mazziotta JC, Fazio F “Mapping motor representation with positron                         engineering at, Department of computer engineering
     emission tomography” Natur 1994, 371:600-602                                              MIT COE, Pune, India, 411038
[11] John Milton, Steven L. Small, Ana Solodkin “Imaging motor imagery:
     Methodological issues related to expertise” 2008
[12] Martin Lotze,Ulrike Halsband “Motor Imagery” 2006
[13] Jackson,P/L., Lafleur M. F. ,Maloin F.,Richards C., Doyon J., “Potential
     role of mental practice using motor imagery in neurologic rehabilitation” ,
     2001, Arch Phys Med.82:1133-114
[14] Gandevia S.C. “Mind, muscles and motor neurons” J.Sci. Med. Sport
     2,167-180 1999
 [15] Feltz D.L., Landers D.M. “The effect of mental practice on motor skill
     learning and performance – A meta analysis” J. Spot psycho 5.25-27
[16] A.J. Szameitat, S. Shen A. Starr, European journal of neuroscience 26
     (2005) 3303-3308
[17] A.J. Szameitat, S. Shen , A. Starr , Neuroimage 34 (2007) 702-713
[18] Sakamoto T., Porter, Asanuma H. “Long lasting potentiation of synaptic
     potentials in the motor cortex produeced by simulation of the sensory
     cortex in the cat: basis of motor learning” 1987 Brain Res. 413, 360-364
[19] Cumming J., Hall C., “Deliberate imagery practice: the development of
     imagery skills in competitive athletes” 2002, J. Sports Sci. 20, 137-145
[20] Bangert M., Haeusler u. , Altenmuller E. , “On practice: how the brain
     connects piono keys and piono sounds” , 2001, Ann. N Y Acad. Sci 930,
[21] M. Cheng, W.Y. Jia, X.R., Gao, S.K. and F.S.Yang “mu rhythm based
     mouse control: an offline analysis”, 2004, Clinical Neurophysiology, vol
     115 pp, 745-751
[22] G.E. Fabianim D.J. McFarland, J.R. Wolpaw and G. Pfurtscheller,”
     conversion of EEG activity into cursor movement by a brain computer
     interface (BCI)” IEEE trans. Neural systems and rehabilitation
     engineering . vol 12 (3) pp ,331-338 ,2004
[23] L.j. Trejo, r. Rosipal, B. Matthews “Brain computer interfaces for 1-D
     and 2-D cursor control: designs using volitical control of the EEg
     spectrum or steady state visual evoked potentials” IEEE trans. Neural
     systems and rehabilitation engineering, Vol 14 no 4, 331- 354, 2006
[24] Yuanqing Li , Jinyi long, Tianyou Yu, Zhuliang Yu, Chuanchu Wang,
     Haihong Zhang, Cuntai Guan “An EEG based BCI system for 2-D cursor

                                                                                                                   ISSN 1947-5500

To top