Analysis and simulation of EEG Brain Signal Data using MATLAB

Transcription

1 Chapter 4 Analysis and simulation of EEG Brain Signal Data using MATLAB 4.1 INTRODUCTION Electroencephalogram (EEG) remains a brain signal processing technique that let gaining the appreciative of the difficult internal machines of the brain and irregular brain waves ensures to be connected through articular brain disorders. The analysis of brain waves shows a significant part in the diagnosis of dissimilar brain syndromes. MATLAB delivers a cooperative graphical user interface (GUI) letting users to openly and interactively route their high density EEG dataset then additional brain signal information dissimilar methods like independent component analysis (ICA) and time/frequency analysis (TFA). In addition to fixed averaging methods. The research work resolve display dissimilar brain signals through associating, analysing then simulating datasets which is before encumbered in the MATLAB software to practice the EEG signals. The human brain is one of the greatest composite structures in the creation. Currently many technologies are to record brain waves then electroencephalography (EEG) remains one of them. This remains one of the brain signals processing technique that permits attainment the thoughtful of the difficult internal mechanisms of the brain and abnormal brain waves have shown to be associated with particular brain disorders. 4.2 LITERATURE REVIEW Creusere et al (2012), Assessment of subjective brain wave form quality from EEG brain replies via time space frequency analysis, page Theories give details herein and research work is the problem of quantifying changes in the perceived quality of signals by directly measuring the brain wave responses of human subjects using EEG technique. Ideas taken on from this research work are that has preferred an approach constructed on time space frequency analysis of EEG wave form set for detecting different brain disorders. Jutgla et al (2012) Diagnosis of Alzheimer s disease from EEG by means of synchrony measures in optimized frequency bands, page Theories give 38

2 details herein research work is the EEG is considered as a promising diagnostic tool for analysing brain disorders symptoms because of its non-invasive safe and easy to use properties. EEG has the potential to complement or replace some of the current tradition diagnostic techniques. Ideas taken from this research work are EEG datasets of the patients with different brain disorders symptoms have been collected to diagnosis the seizures symptoms related to the patients. Sosa et al (2011) reported in theories give details herein research work is the operational procedures of EEGLAB and efficiency of EEG signal processing for students and professionals to perform and analysis of the EEG signals. Its use as a starting point for the comparison of different brain signal processing algorithms. Ideas taken from this research work are Capabilities of EEGLAB for diagnosis purpose and basic explanation of the working procedure of that tool for signal processing such as loading the dataset, plotting techniques to get the proper result, etc. Bhattacharya et al (2011) theories give details herein research work Presented the information about EEGLAB software for Brain-computer interface (BCI) is an emerging technology which aims to convey people's intentions to the outside world directly from their thoughts. Ideas taken from this research work are the Feature learning of EEG to the classification among frequencies in tribunals and within recording locations. Methods to allow users to remove data channels, artifacts by accepting or rejecting visually. Michalopolous et al (2011) reported that the Characterization of evoked and induced activity in EEG and assessment of intertrail variability, page Theories give details herein research work is the brain reply to an internal or external experience is poised through the superposition of suggested and persuaded brain activity which reproduces dissimilar brain mechanisms involved. Caminiti (2010) reported that the identification of different brain activities through EEG assessment procedure. Ideas taken from this research work are identifying brain activities for diagnostic purposes and provide useful tools for brain computer interfaces through insight on the activation of different brain channels Ye Yuan (2010) theories give details herein research work; EEG dataset is collected after analysing the entire length of the EEG recording the patient frequently 39

3 for long time to detect traces of different human brain activities. Ideas taken from this research work are change of the structure of different brain activities during seizures is observed by the change of embedding dimension of EEG signals if the human brain is considered as a nonlinear dynamic system. Duque Grajales J.E., Múnera Perafán A., Trujillo Cano D., Urrego Higuita D.A., Hernández Valdivieso A.M.(2009), System for Processing and Simulation of Brain Signals, Page Theories give details herein research work has presented the methodology used to develop a system useful in the simulation of brain signals. It has been described in detail the procedure in the modelling of EEG signals and insight brain signals recorded during surgical procedures. Ideas taken from this research work are processing and simulation of brain signals from different signal processing models which allows going deep into the study of brain function during sleeping and pathological situations and facilitated the assessment of the effect of different drugs in different brain disorders. 4.3 BRAIN SIGNAL PROCESSING Signal processing is the supporting technology for the generation, conversion, also understanding of data. On dissimilar phases of period, human brain responds contrarily. All these brain signals castoff for several purposes so that it is conceivable to train the functionalities of brain suitably by creating, converting and interpreting the collected signal. This progression is acknowledged as brain signal processing Brain Waves and EEG The study of brain waves shows a significant part in the analysis of dissimilar brain disorders. Brain is fabricated of billions of brain cells named neurons, which use electricity to interconnect with each other. Wallace et al (2012) reported that the permutation of millions of neurons distributing the signals simultaneously to create an massive volume of electrical movement in the brain, can be perceived by consuming complex medical tools such as an EEG which processes electrical levels over areas of the scalp. Michalopolous et al (2011) reported that the electroencephalogram (EEG) recording is a suitable tool for learning the functional state of the brain and for 40

4 analysing certain syndromes. The mixture of electrical movement of the brain is usually called a brainwave pattern because of its wave-like nature Overview of EEG Signals EEG signals contain more relevant information about brain disorders and different types of artifacts. Signals in the form of dataset are already loaded to the tool so that it will be using those signals to plot the data and visualization of the timefrequency domain plots which can be displayed all together. Basically the work will be monitoring the EEG signals according to the placement of electrodes which are called montages. After that the research work will observe the EEG signals to recognize and eliminate different disease related artifacts. Then unwanted signal will be subtracted by differential amplifier. Finally the work will proceed for the signal filtering based on the different types of brainwave frequencies to diagnosis and simulate variety of brain disorders by using MATLAB. Figure 4.1 Overview of the EEG Signal Processing Technique 4.4 ANALYSIS AND DESIGN Study of Existing EEG Hardware Techniques Steps involved in the existing techniques: the electrodes are placed on the brain by wires and electrical activities of the brain are recorded in a computer. It will display the movement as a sequence of wavy lines drained as an image on the computer screen. Patients need to lie down and close their eyes during the recording. The recording might be motionless since there should be time to permit the patient for widening and repositioning. Different things will be done by the patient during the test to record the brain activity at that time. Such as taking breathe deeply and rapidly for few minutes and looking at a bright, flashing light for checking the stimulation. After recording the brain activities like the above mentioned process brain disorder symptoms will be detected. 41

5 Requirement Analysis Functional Requirements This research technique will provide the solution for the patients who will be able to see their brainwaves while using this software called MATLAB; and then 2. User will be able to see the data related to the processed brainwave signal. There are different research techniques and features are defined as follows: EEG montages: Montage means the location of the electrodes. EEG can also be examined with a bipolar montage or a referential one. Bipolar means to use two electrodes on the scalp on all the sides and reference electrode for one side of the brain. The referential montage means only having a common reference electrode in both the side of the brain. Channels: The electrical activity of the brain is conducted by wires from the scalp and electrodes are placed by using EEG machine. The inputs to the hardware EEG machine are then used to combine a montage, which is an exact organization or array of electrodes that show the EEG signal. Sensitivity: Amplitude is the magnitude of the EEG activity which is measured in microvolts (µv). It is determined by measuring the brainwave deflection in millimetres (mm) at specified machine sensitivity (µv /mm). Filtering: Low- pass filtering is used for smoothening the brainwaves and high-pass filtering is used for sharpening the brainwaves in order to make the signals more clearly to the viewer. Frequency Sweeping: Sweeping basically reduces the complexity towards analysing the brainwaves for EEG signal processing. It is possible for the new user and can also use these techniques and features with the data through MATLAB for brain signal processing and other purposes Non-Functional Requirements There are two non-functional requirements used in this research work first one is the performance analysis for the MATLAB software tool which is efficient in analysing and processing the signals in a proper way so it will be easier for the user to observe the signals properly. Then the second one is the reliability checking tool used for analysing the EEG signals, removing and recognizing the artifacts to process the signal datasets. 42

6 Background Design Requirements EEG signal processing in built plug-ins under MATLAB environment, while looking at all the possible technologies, libraries, platforms that will use in this research work; it has seen that the most convenient programming language for this work is inbuilt plug-ins which work under MATLAB environment. 4.5 SYSTEM DESCRIPTION In this research work it has encompassed several features. Such by means of EEG montages; Montage capitals the location of the electrodes. The EEG can also be observed with a bipolar montage or a referential one. Bipolar means that just to use two electrodes on the scalp on all the sides and for reference electrode one side of the brain. The referential montage means only having a common reference electrode in both the sides of the brain. In this part, we will be presenting how brainwaves will differ according to the placement of electrodes. i) Right Montages: Patient information has composed conferring to electrodes that are located in the right side of the brain, so it will show the waves related to the right side of the brain based on time-frequency analysis. ii) Left Montages: Patient information has composed conferring to electrodes that are located in the left side of the brain, so it will show the waves related to the right side of the brain based on time-frequency analysis. iii) Both Side Montages: Patient information has composed conferring to electrodes that are located in the left and right side of the brain, so it will show the waves related to the right side of the brain based on time-frequency analysis EEG Channels The electrical activity of the brain is conducted by wires from the scalp and electrodes are placed by using EEG machine. The inputs to the hardware EEG machine are then castoff to comprise a montage, which is an exact preparation or array of electrodes that show the EEG signal. In this research the work is dealing with basically 20 channels of the brain because EEG hardware machine deals with only 20 channels of the brain. Each channel basically compares input data taken based on 43

7 placement of the two electrodes. Upward deflection of the wave is defined as negative and occurs when the first input data is negative with respect to the second input data or second input data is positive with respect to the first input data. A descending refraction of the brainwave is clear as positive then follows once first input data is positive with deference to second input data or then second input data is negative with respect to first input data Sensitivity Amplitude is the magnitude of the EEG activity which is measured in microvolts (μv). It is determined by measuring the brainwave deflection in millimetres (mm) at specified machine sensitivity (μv /mm). The work have analysed the brainwaves according to the collected sensitivity values of the patients, EEG procedures are performed at a sensitivity rate of 7 μv /mm, such that a 10 mm deflection of waves signifies amplitude of 70 μv. This work ensure the dignified sensitivity values as 10 μv /mm, 15 μv /mm, 20 μv /mm, 30 μv /mm, 50 μv /mm because it is easier to determine the brainwave patters with these values Filtering Low- pass filtering is used for smoothening the brainwaves and high-pass filtering is used for sharpening the brainwaves in order to make the signals more clearly to the viewer. According to the patients EEG hardware data collected, this research work have shown two types of filtering technique options such as Low pass frequency filters and High- pass frequency filters For the low pass frequency filters generally setting the maximum range is till 1Hz and for high pass frequency filters setting the maximum range is till 70Hz because this the standard limit of filters Frequency Sweeping Sweeping basically reduces the complexity towards analysing the brainwaves for EEG signal processing. At last this work will sweep the signal to reduce the complexity for the visualization of the brainwaves. Nolan et al (2009) reported that in 44

8 this work will resolve using linear frequency extensive and it requires a stable rate of frequency per interval. Essentially it is dignified as Hz/sec. 4.6 ARCHITECTURAL DESIGN FOR EEG ANALYSIS Figure 4.2 System architectural designs 4.7 METHODOLOGY Procedures i) Dissimilar EEG signals are composed as a form of dataset in the MATLAB; ii) Load the data into the software for brain signal processing and then practice the datasets; iii) Remove and select the particular features for different EEG datasets; iv) Classify the datasets conferring to the product features such as- montages, channels, sensitivity, filtering and sweeping v) Check the difference of dissimilar brain waves based on their characteristics, vi) Select the specific montages such as left side montage, right side montage or both side montages to check change in different brainwaves. 45

9 vii) Then select the channels out of 20 channels for viewing more detailed waveforms. ix) By fixing the sensitivity values of the collected EEG data it will be setting filtering range of the signals for high frequency (50-70 Hz) or for low frequency (0.1-1 Hz); then it will change the values of sweeping also according to the data. x) At last it will get the final EEG signal in waveform Flow chart SELECT RIGHT/ LEFT / BOTH MONTAGES 46

10 4.8 IMPLEMENTATION OF BRAIN SIGNAL DATA ANALYSIS AND RESULTS The figure 4.3 shows the options for checking the exiting patient s records by selecting the first option and for diagnosis of different brainwaves forms selecting the second option. Figure 4.3 Sample results and other records page Figure 4.4 Brainwaves pattern The figure 4.4 displays the result obtained by the changes made in montage module just by selecting right montage of the brain and different frequencies of brainwaves can be easily determined. Figure 4.5 Signals and frequency variation The above figure 4.5 displays the report of the test generated according to various data which are collected from patients and they have been imported in to the program so that it will be easier to undergo many changes according to different modules. 47

11 4.8.1 Testing Table 4.1 different patient s data of montages e e e e e e e e e e-001 Table 4.2 different patient s data of brain channels e e e e e e e e-001 Table 4.3 different patient s data of brain sensitivity e e e e e e e e-001 Table 4.4 different patient s data of signal filtering e e e e e e e e-001 Table 4.5 different patient s data of signal sweeping e e e e e e e e-001 The overall research work consists of five modules which were completed in their respective time frame. The implementation and functionalities procedures seemed a daunting task, but were successfully completed to achieve the desired objective. After the successful implementation of test results, this research work can be applicable for monitoring alertness, coma and brain death; testing drug effects; investigating sleep disorders; Investigating mental disorders; Locating areas of damage following head injury, stroke and tumour and Monitoring the brain development. 4.9 SUMMARY The above mentioned research work has clearly demonstrated the concepts about open source plug-ins, running under the platform MATLAB environment and its ability to process biophysical data by different ways such as using simplicity of its command line language or using the many MATLAB functions and the methodology related to the analysis of the brain signal processing through MATLAB software toolbox. It has been described in detail, the procedure in the modelling of EEG signals and insight brain signals recorded during surgical procedure. 48