Abstract:The Electroencephalographic (EEG) signals are those which recordsthe electrical status of the human brain.The signal pattern varies according tothe chemical reaction in the human brain. Therecorded waveforms reflect the cortical electrical activity. EEG activity isquite small, measured in microvolts (mV).
Acquistition of the EEG signal can be used in many researchfields. Depending on the frequency of obtained signal it can be classified intofive types they are: delta, theta, alpha and beta. Introduction: The brain is the most complex part in the human body. Itcontrols the overall activity of our body, it consist of billions of neuronswhich communicates with each other in achieving it. Inside the human brainthere will be a lot of chemical reaction taking place in it.The evolution ofthe EEG signals took place in 1929.
Hans Berger was the first scientist to foundabout the EEG signals and on the later stages it found a drastic improvement inthe research about the various fields.The different types of brain waves havespecific frequency such as beta(14-30hz), alpha(8-13hz), theta(4-7hz) and delta(<3.5hz). These waves have certain characteristics, by analysing those wavepattern we can do various research in different fields such as BCI (Braincomputer interface), Phycological factors, neuro-imaging, etc Neuromarketing:Acquistition of EEG signals led to the new field of sciencecalled neuromarketing.
It is a new way to get the feedback that you couldmeasure with a consumer device to discover which types of advertising areeffective and useful, and which types are embrassing. Neuromarketingresearchers believe that consumers’decisions are made in a split second, those decisions are made subconsciously. Theystrongly believe that decision of consumers are not factual and they aretotally taken in a matter of seconds by simple attraction that the companyadvertises.The function of neuromarketing is to analyse how the customersemotions are triggered depending on the advertisement they see, how their subconcious mind react to it. The data it generates is extremely useful for thecompanies to develop an advertisement which attracts the customers they target.The data is gathered by monitoring certain biometrics, including: Eye tracking Facial coding Galvanic skin response and electrothermal activity Electroencephalography (EEG)Some neuromarketing research is conducted using fMRI, whichmeasures brain activity by detecting changes in blood flow in response tostimuli. It yields accurate data, but it is challenging for the following reasons: It requires subjects to lie completely still in a large MRI chamber, which can be a total discomfort to the subjects.
Stimuli cannot be encountered in the same way the test subject would usually be exposed to it—you can’t take an MRI chamber into a retail store. It takes a lot of time and its also expensive stratergy.EEG technique, on the other hand, allow neuromarketing researchto be conducted efficiently fromanywhere. This methodology helps the researchers to measure consumer responseto an testing environment , such as a movie theatre, bar, mall. Small biosensors can be placed at distinctplaces on the head, allowing for accurate measurement of brain activity whilegiving the test subject full range of motion and ensuring their comfort.
Changes in state of the brain can be interpreted using thesuitable technique and the current status of the individual such as sleepiness,focused state ,laziness etc can be found. In a concentrated state, a 30 second commercial advertisement isenough to hold them to watch carefully.The EEG reading taken from that consumerin test reveals that ,he/she will be fully attentive for the first 10 secondsand lost their attention for next 10 seconds finally they pay attention at thefinal 10 seconds.Thus improving the middle content of the video based on thefeeedback could help us to create a more creative commercial.Neuromarketing helps firms to create more effective and creativeadvertisements. This not only benefits the venture, but also the consumers who areexposed to hundreds of ads per day, so creating more informative, emotionallyrewarding, and useful ads can enhance a customer’s experience with a product orbrand long before they consider buying.Brain computer interface: Introduction Themajor field where the EEG signals can be effectively utilised is the BrainComputer interface (BCI).
“A brain–computer interface is a communication systemthat does not depend on the brain’s normal output pathways of peripheral nervesand muscles.” It reflects the principal reason for the interest in BCIdevelopment—the possibilities it offers for providing new augmentativecommunication technology to those who are paralyzed or have other severemovement deficits. All other augmentative communication technologies requiresome form of muscle control, and thus may not be useful for those with the mostsevere motor disabilities, such as late-stage amyotrophic lateral sclerosis,brainstem stroke, or severe cerebral palsy.Therefore by using this technique wecan make a lot paralyzed patients to act on their own without depending onanyone. Essentialsfeatures of BCIBCI operationdepends on the communication which takes place between the two adaptivecontrollers, the user’s brain, which produces the activity (EEG signals) measuredby the BCI system, and the system itself, which translates that activity intospecific commands to perform the tasks. Completing the BCI operation is a newskill,it does not control our muscular organs but it controls our EEG signalsas a single unit.
Each BCI uses certain algorithm to translate the obtainedinput into required output control signals of our requirements. This algorithmmight include linear or nonlinear equations, a neural network, or othermethods, and might incorporate continual adaptation of important parameters tokey aspects of the input provided by the user. BCI outputs can be cursormovement, letter or icon selection, or another form of device control, andprovides the feedback that the user and the BCI can use to adapt so as tooptimize communication. Adding to its input, translation algorithm, and output,each BCI has several other distinctive characteristics which should bemonitored. These include its On/Off mechanism (e.g.
, EEG signals orconventional control); response time, speed and accuracy and their combinationinto information transfer rate, appropriate user population, applications andconstraints imposed on concurrent conventional sensory input and motor MatchingBCI and the Input to user.The inputfeatures of proposed BCI system should be properso that it can be broadlyapplied to the communication needs of users with different disabilities. MostBCI systems use EEG or single-unit features that originate mainly insomatosensory or motor areas of cortex. These areas may be severely damaged inpeople with stroke or neurogenerative disease. Use of features from other CNS regionsmay prove necessary.
In EEG-based BCI system, effective multielectroderecording which are performed initiallyand then periodically, can detect the changes in the user’s performance and,and can thereby guide selection of optimal recording locations and EEG features.Some areas of the brain may not be effectively used for the interaction becauseof slow potentials and rhythms. BCI system should be designed such that itworks on the wide variety of EEGsignals.A system works on the slow potentials,rhythms,P300 potentials, etc areunder the research.Signal analysisand translation algorithms:Signal analysisis done in the BCI system in order to enhance the signal-to-noise ratio (SNR)of the EEG or single-unit features that carry the user’s messages and commands.
To accomplish this, consideration of the major sources of noise is essential.Noise has two types of sources bothnonneural sources (e.g., eye movements, EMG, 60-Hz line noise) and neuralsources (e.g., EEG features other than those used for communication). Noisedetection and eliminating those noises will be very difficult if the frequency,amplitude and other parameters of noises are similar to the required system..
While they can enhance the signal-to-noise ratio, they cannot directly addressthe impact of changes in the signal itself. Factors such as motivation,intention, frustration, fatigue, and learning affect the input features thatthe user provides. From that we can state that proper interaction between theuser and the system and a effective signal processing methods helps in the BCIdevelopment. A translation algorithm is a series of computations thattransforms the BCI input features derived by the signal processing stage intoactual device control commands. Stated in a different way, a translationalgorithm takes abstract feature vectors that encodes the message that the user wants to communicate andtransforms those vectors into application-dependent device commands.
DifferentBCI’s use different translation algorithms . Each algorithm can be classifiedin terms of three key features: transfer function, adaptive capacity, andoutput. The transfer function can be linear (e.g., linear discriminantanalysis, linear equations) or nonlinear (e.g.
, neural networks). The algorithmcan be adaptive or nonadaptive. Adaptive algorithms can use simple handcraftedrules or more sophisticated machine-learning algorithms. The output of thealgorithm may be discrete (e.g., letter selection) or continuous (e.g., cursormovement).
The diversity in translation algorithms among research groups is duein part to diversity in their intended real-world applications. Nevertheless,in all cases the goal is to maximize performance and practicability for thechosen application.BCI application:Figure shows thevarious fields in which the BCI’s are used.
Application ofEEG signals in epilepsy diagnosis:Epilepsy:Epilepsy is a disorder that causes different types of seizures.A seizure is a sudden surge in the electrical conditions in the brain . It consistof two main types. Generalized seizures which affects the whole brain.
Focal,or partial seizures, which affect just one part of the brain . Impacts ofseizures vary according the type which the person is affected. There areseveral symptoms for epilepsy fever, stroke, etc It is a common disorder whichaffects millions of people around the world.Types of seizures are Focal(partial) seizures, A simplepartial seizure and Complexpartial seizures . Generalized seizuresGeneralized seizuresaffects the whole brain. The different types are they are Absence seizures,Tonic seizures , Atonic seizures , Clonic Myoclonicseizures and TonicseizuresEEG Analysis:An EEG test only decribes about the electrical activity of thebrain at the time of the test bring conducted. If a person is affected by theseizure he/she has unusal brain activity. At other time brain activity is normal.
So, if your EEG testresults are normal, it usually meansthat there is no epileptic activity in your brain at the time the test is beingdone. People affected by epilepsy have unusual electrical activity in theirbrain all the time, even when they are not having a seizure. From the testresults a doctor can recognise the pattern of waves and he/she can diagnose it.Some people may have unusal brain patterns but they wont have epilepsy . Thesecould be caused by other medical conditions, problems with their vision, orbrain damage.
So it may found that this technique may not be correct everytime.Itcan also show up some types of seizure. But it might not show up some focal (partial)seizures .The EEG signals only gives the brain activity and not the location ofaffected areas.
There’s a very small risk that you could have a seizure duringan EEG test. This could be caused by looking at a flashing light or breathingdeeply. These activities are usually part of the test.Your doctor may ask youto reduce your epilepsy medicine or have less sleep than usual before you havesome types of EEG tests. This would also increase the risk that you would havea seizure around the time of having the test.If you hold a driving licence,having a seizure could mean that youshould not drive until you have been seizure free for 12months.
There are several ways an EEG test can be done.Standard EEG tests. You may be asked to breathe deeply for some minutes and alsoto look at a flashing light. These activities can change the electricalactivity in your brain, and this will show on the computer.You will be asked tokeep as still as possible during the test. Any movement can change theelectrical activity in your brain, which can affect the results.Routine EEGrecordings usually take 20 to 40 minutes.Sleep EEG testsEEG test is taken when you are asleep.
Before the test, you maybe given some medicine to make you go to sleep. The test lasts for one to twohours.It is useful when epilepsy is suspected in children under 5. This isbecause there are some types of epilepsy which are common in young children,where seizures mainly happen in sleep.
Sleep-deprivedEEG testsThese tests are done when you have had less sleepthan usual. At that time, there is more chance of unusual electrical activity inthe brain.It can show up subtle seizures,. Before you have a sleep-deprived EEG test, your doctor may askyou not to go to sleep at all the night before.The patient sleeping timingsshould be altered. You may then fall asleep while recording the activity inyour brain.
It extends upto few hours.AmbulatoryEEG testsThese tests are conducted when the patient is walking. It is designedto record the activity in your brain over a few hours, days or weeks. Thismeans there is more chance that it will pick up unusual electrical activity inyour brain, than during a standard or sleep EEG test.
However, the electrodesthat are attached to your head are plugged in to a small machine that recordsthe results. You can wear the machine on a belt, so you are able to go aboutyour daily business. You don’t usually stay in hospital .The person should keeptrack of activities which they do.Diagnosis: Identifying what kind of seizure is the patient is beingaffected is the most critical step in diagnosis of epilepsy.Different seizuresare to be treated in the unique way ,if it is not clearly identified false treatment may affect the patient in agreater level.
The diagnosis includes meditation, medicines, etc