EEG based Brain Computer Interface (BCI) establishes a new channel between human brain and the
surrounding environment in order to disseminate instructions to the outside world. It is based on the
recording of temporary EEG changes during different types of motor imagery such as imagination of
different hand movements. The spatial pattern of activated cortical areas during motor imagery is similar
to that of real time executed movement. Time domain features and frequency domain features are extracted
with emphasis on recognizing discriminative features representing EEG trials recorded during imagination
of different hand movements. Then, classification into different hand movements is carried out.
Correlation Analysis of Electromyogram SignalsIJMTST Journal
An inability to adapt myoelectric interfaces to a user’s unique style of hand motion. The system also adapts
the motion style of an opposite limb. These are the important factors inhibiting the practical application of
myoelectric interfaces. This is mainly attributed to the individual differences in the exhibited electromyogram
(EMG) signals generated by the muscles of different limbs. In this project myoelectric interface easily adapts
the signal from the users and maintains good movement recognition performance. At the initial stage the
myoelectric signal is extracted from the user by using the data acquisition system. A new set of features
describing the movements of user’s is extracted and the user’s features are classifed using SVM
classification. The given signal is then compared with the database signal with the accuracy of 90.910 %
across all the EMG signals.
he main idea of the current work is to use a wireless Electroencephalography (EEG) headset as a remote control for the mouse cursor of a personal computer. The proposed system uses EEG signals as a communication link between brains and computers. Signal records obtained from the PhysioNet EEG dataset were analyzed using the Coif lets wavelets and many features were extracted using different amplitude estimators for the wavelet coefficients. The extracted features were inputted into machine learning algorithms to generate the decision rules required for our application. The suggested real time implementation of the system was tested and very good performance was achieved. This system could be helpful for disabled people as they can control computer applications via the imagination of fists and feet movements in addition to closing eyes for a short period of time
Analysis of EEG data Using ICA and Algorithm Development for Energy Comparisonijsrd.com
This Electroencephalogram (EEG) signal analysis very useful in clinical research and brain computer interface application. EEG signal (brain wave) recordings are highly susceptible from artifacts which are originated from the non-cerebral origin of the brain. EEG detection and rejection of artifacts are necessary for acquiring correct information from EEG signal. Emotiv, Epoc headset can record 16 channels from the scalp of the electrode. EEGLAB allows analysis of EEG signal through Event related potential (ERP) analysis, Independent component analysis (ICA), and time/frequency analysis. Independent component analysis (ICA) may be suitable method for detecting artifacts. We analyzed EEG data which are recorded using emotiv epoc in a different situation for a single person. EEG data are preprocessed by EEGLAB and decomposes the data by the ICA. Using statistical method, analyzed the all the dataset and finding the relationship among the dataset. T- Test shows that EEG pattern is unique in a person. EEG data is divided into different frequency band to find the relationship between the dataset. Also develop the algorithm for calculating energy of dataset for each channel. Comparing the energy for each dataset and each channel to find the maximum and minimum value of energy. In higher frequency range (13-100 Hz) dataset D (meditation) contains maximum value of energy for most channels among all datasets.
METHODS OF COMMAND RECOGNITION USING SINGLE-CHANNEL EEGSijistjournal
This work proposes to recognize a user's commands by analysing his/her brainwaves captured with single channel electroencephalogram (EEG). Whenever a user intends to issue one of the pre-defined commands, the proposed system prompts him/her all the candidate commands in turn. Then, the user is
asked to be concentrated as possible as he/she can, when the desired command is shown. It is assumed
that the concentration will present a certain pattern of “Yes” in the captured EEG, as opposed to a
certain pattern of “No” when the user is relaxed. Accordingly, the task is to determine that the captured EEG is “Yes” or not. This work compares three recognition methods, respectively, based on Gaussian mixture models, hidden Markov models and recurrent neural network, and conducts experiments using
2400 test EEG samples recorded from 10 subjects.
Classification of EEG Signals for Brain-Computer InterfaceAzoft
This e-book gives you a sneak peak into how the classification of right hand movements via EEG could contribute to the development of a brain-computer interface. The Azoft R&D department, along with Sergey Alyamkin and Expasoft provide detailed data from research done for the "Grasp-and-Lift EEG Detection" competition organized by Kaggle. You’ll learn why the deep learning algorithms can be effective in various types of signal classifications and how to apply convolutional neural networks for a specific task such as identifying hand motions from EEG recordings.
See more details on our website: http://rnd.azoft.com/classification-eeg-signals-brain-computer-interface/
This document will examine issues pertaining to feature extraction, classification and prediction. It will
consider the application of these techniques to unlabelled Electroencephalogram (E.E.G.) data in an
attempt to discriminate between left and right hand imagery movements
Correlation Analysis of Electromyogram SignalsIJMTST Journal
An inability to adapt myoelectric interfaces to a user’s unique style of hand motion. The system also adapts
the motion style of an opposite limb. These are the important factors inhibiting the practical application of
myoelectric interfaces. This is mainly attributed to the individual differences in the exhibited electromyogram
(EMG) signals generated by the muscles of different limbs. In this project myoelectric interface easily adapts
the signal from the users and maintains good movement recognition performance. At the initial stage the
myoelectric signal is extracted from the user by using the data acquisition system. A new set of features
describing the movements of user’s is extracted and the user’s features are classifed using SVM
classification. The given signal is then compared with the database signal with the accuracy of 90.910 %
across all the EMG signals.
he main idea of the current work is to use a wireless Electroencephalography (EEG) headset as a remote control for the mouse cursor of a personal computer. The proposed system uses EEG signals as a communication link between brains and computers. Signal records obtained from the PhysioNet EEG dataset were analyzed using the Coif lets wavelets and many features were extracted using different amplitude estimators for the wavelet coefficients. The extracted features were inputted into machine learning algorithms to generate the decision rules required for our application. The suggested real time implementation of the system was tested and very good performance was achieved. This system could be helpful for disabled people as they can control computer applications via the imagination of fists and feet movements in addition to closing eyes for a short period of time
Analysis of EEG data Using ICA and Algorithm Development for Energy Comparisonijsrd.com
This Electroencephalogram (EEG) signal analysis very useful in clinical research and brain computer interface application. EEG signal (brain wave) recordings are highly susceptible from artifacts which are originated from the non-cerebral origin of the brain. EEG detection and rejection of artifacts are necessary for acquiring correct information from EEG signal. Emotiv, Epoc headset can record 16 channels from the scalp of the electrode. EEGLAB allows analysis of EEG signal through Event related potential (ERP) analysis, Independent component analysis (ICA), and time/frequency analysis. Independent component analysis (ICA) may be suitable method for detecting artifacts. We analyzed EEG data which are recorded using emotiv epoc in a different situation for a single person. EEG data are preprocessed by EEGLAB and decomposes the data by the ICA. Using statistical method, analyzed the all the dataset and finding the relationship among the dataset. T- Test shows that EEG pattern is unique in a person. EEG data is divided into different frequency band to find the relationship between the dataset. Also develop the algorithm for calculating energy of dataset for each channel. Comparing the energy for each dataset and each channel to find the maximum and minimum value of energy. In higher frequency range (13-100 Hz) dataset D (meditation) contains maximum value of energy for most channels among all datasets.
METHODS OF COMMAND RECOGNITION USING SINGLE-CHANNEL EEGSijistjournal
This work proposes to recognize a user's commands by analysing his/her brainwaves captured with single channel electroencephalogram (EEG). Whenever a user intends to issue one of the pre-defined commands, the proposed system prompts him/her all the candidate commands in turn. Then, the user is
asked to be concentrated as possible as he/she can, when the desired command is shown. It is assumed
that the concentration will present a certain pattern of “Yes” in the captured EEG, as opposed to a
certain pattern of “No” when the user is relaxed. Accordingly, the task is to determine that the captured EEG is “Yes” or not. This work compares three recognition methods, respectively, based on Gaussian mixture models, hidden Markov models and recurrent neural network, and conducts experiments using
2400 test EEG samples recorded from 10 subjects.
Classification of EEG Signals for Brain-Computer InterfaceAzoft
This e-book gives you a sneak peak into how the classification of right hand movements via EEG could contribute to the development of a brain-computer interface. The Azoft R&D department, along with Sergey Alyamkin and Expasoft provide detailed data from research done for the "Grasp-and-Lift EEG Detection" competition organized by Kaggle. You’ll learn why the deep learning algorithms can be effective in various types of signal classifications and how to apply convolutional neural networks for a specific task such as identifying hand motions from EEG recordings.
See more details on our website: http://rnd.azoft.com/classification-eeg-signals-brain-computer-interface/
This document will examine issues pertaining to feature extraction, classification and prediction. It will
consider the application of these techniques to unlabelled Electroencephalogram (E.E.G.) data in an
attempt to discriminate between left and right hand imagery movements
ENERGY COMPUTATION FOR BCI USING DCT AND MOVING AVERAGE WINDOW FOR NOISE SMOO...IJCSEA Journal
Brain computer interface (BCI) is a fast evolving field of research enabling computers and machines to be directly controlled by the human neural system. This enables people with muscular disability to directly control machines using their thought process. The brain signals are recorded using Electroencephalography (EEG) and patterns extracted so that the BCI system should be able to classify various patterns of brain signal accurately to perform different tasks. The raw EEG signal contains different kinds of interference waveforms (artifacts) and noise. Thus raw signals cannot be directly used for classification, the EEG signals has to undergo preprocessing, to remove artifacts and to extract the right attributes for classification. In this paper it is proposed to extract the energies in the EEG signal and classify the signal using Naïve Bayes and Instance based learners. The proposed method performs well for the two class problem in the multiple datasets used..
SUITABLE MOTHER WAVELET SELECTION FOR EEG SIGNALS ANALYSIS: FREQUENCY BANDS D...sipij
Wavelet transform (WT) is a powerful modern tool for time-frequency analysis of non-stationary signals such as electroencephalogram (EEG). The aim of this study is to choose the best and suitable mother wavelet function (MWT) for analyzing normal, seizure-free and seizured EEG signals. Several MWTs can be used, but the best MWT is the one that conserves the quasi-totality of information of the original signal on wavelet coefficients and gather more EEG rhythms in terms of frequency. In this study, Daubechies, Symlets and Coiflets orthogonal families were used as bsis mother wavelet functions. The percentage rootmeans square difference (PRD), the signal to noise ratio (SNR) and the simulated frequencies as the selection metrics. Simulation results indicate Daubechies wavelet at level 4 (Db4) as the most suitable MWT for EEG frequency bands decomposition.Furthermore, due to the redundancy of the extracted features, linear discriminant analysis (LDA) is applied for feature selection. Scatter plot showed that the selected feature vector represents the amount of changes in frequency distribution and carries most of the discriminative and representative information about their classes. Then, this study can provide a reference for the selection of a suitable MWT and discriminativefeatures.
Robot Motion Control Using the Emotiv EPOC EEG SystemjournalBEEI
Brain-computer interfaces have been explored for years with the intent of using human thoughts to control mechanical system. By capturing the transmission of signals directly from the human brain or electroencephalogram (EEG), human thoughts can be made as motion commands to the robot. This paper presents a prototype for an electroencephalogram (EEG) based brain-actuated robot control system using mental commands. In this study, Linear Discriminant Analysis (LDA) and Support Vector Machine (SVM) method were combined to establish the best model. Dataset containing features of EEG signals were obtained from the subject non-invasively using Emotiv EPOC headset. The best model was then used by Brain-Computer Interface (BCI) to classify the EEG signals into robot motion commands to control the robot directly. The result of the classification gave the average accuracy of 69.06%.
International Journal of Computational Engineering Research(IJCER)ijceronline
International Journal of Computational Engineering Research (IJCER) is dedicated to protecting personal information and will make every reasonable effort to handle collected information appropriately. All information collected, as well as related requests, will be handled as carefully and efficiently as possible in accordance with IJCER standards for integrity and objectivity.
Brain-computer interface of focus and motor imagery using wavelet and recurre...TELKOMNIKA JOURNAL
Brain-computer interface is a technology that allows operating a device without involving muscles and sound, but directly from the brain through the processed electrical signals. The technology works by capturing electrical or magnetic signals from the brain, which are then processed to obtain information contained therein. Usually, BCI uses information from electroencephalogram (EEG) signals based on various variables reviewed. This study proposed BCI to move external devices such as a drone simulator based on EEG signal information. From the EEG signal was extracted to get motor imagery (MI) and focus variable using wavelet. Then, they were classified by recurrent neural networks (RNN). In overcoming the problem of vanishing memory from RNN, was used long short-term memory (LSTM). The results showed that BCI used wavelet, and RNN can drive external devices of non-training data with an accuracy of 79.6%. The experiment gave AdaDelta model is better than the Adam model in terms of accuracy and value losses. Whereas in computational learning time, Adam's model is faster than AdaDelta's model.
IJERA (International journal of Engineering Research and Applications) is International online, ... peer reviewed journal. For more detail or submit your article, please visit www.ijera.com
Overview of Machine Learning and Deep Learning Methods in Brain Computer Inte...IJCSES Journal
Research under the field of Brain Computer Interfaces is adapting various Machine Learning and Deep
Learning techniques in recent times. With the advent of modern BCI, the data generated by various devices
is now capable of detecting brain signals more accurately. This paper gives an overview of all the steps
involved in the process of applying Machine Learning as well as Deep Learning methods from Data
Acquisition to application of algorithms. It aims to study techniques currently employed to extract data,
features from brain data, different algorithms employed to draw insights from the extracted features, and
how it can be used in various BCI applications. By this study, I aim to put forward current Machine
Learning and Deep Learning Trends in the field of BCI.
IJRET : International Journal of Research in Engineering and Technology is an international peer reviewed, online journal published by eSAT Publishing House for the enhancement of research in various disciplines of Engineering and Technology. The aim and scope of the journal is to provide an academic medium and an important reference for the advancement and dissemination of research results that support high-level learning, teaching and research in the fields of Engineering and Technology. We bring together Scientists, Academician, Field Engineers, Scholars and Students of related fields of Engineering and Technology.
Current motorized limb prostheses provide rudimentary functionality for the application in everyday life. Together with a
poor cosmetic appearance, this is the reason why a large percentage of amputees do not use their prosthetic device regularly. This
paper seeks to present an overview of current state of the art research on neural interfaces. The focus lies on non-invasive
recording with EMG and especially High-Density EMG sensors. Additionally, direct machine learning and pattern recognition
algorithms for the decoding of the recorded signals are discussed. Finally, promising research directions for advanced prosthesis
control will be discussed. The bionic arm uses EMG signals to control each action of the hand. In order to control them, we need to
record the EMG signal for different actions. And compare it with real-time values to move the hand in a different manner. There
are separate servo motors to control the actions of each finger separately. So these are programmed by using microcontrollers.
Recognition of new gestures using myo armband for myoelectric prosthetic appl...IJECEIAES
Myoelectric prostheses are a viable solution for people with amputations. The chal- lenge in implementing a usable myoelectric prosthesis lies in accurately recognizing different hand gestures. The current myoelectric devices usually implement very few hand gestures. In order to approximate a real hand functionality, a myoelectric prosthesis should implement a large number of hand and finger gestures. However, increasing number of gestures can lead to a decrease in recognition accuracy. In this work a Myo armband device is used to recognize fourteen gestures (five build in gestures of Myo armband in addition to nine new gestures). The data in this research is collected from three body-able subjects for a period of 7 seconds per gesture. The proposed method uses a pattern recognition technique based on Multi-Layer Perceptron Neural Network (MLPNN). The results show an average accuracy of 90.5% in recognizing the proposed fourteen gestures.
Motor Imagery Recognition of EEG Signal using Cuckoo Search Masking Empirical...ijtsrd
Brain Computer Interface BCI aims at providing an alternate means of communication and control to people with severe cognitive or sensory motor disabilities. Brain Computer Interface in electroencephalogram EEG is of great important but it is challenging to manage the non stationary EEG. EEG signals are more vulnerable to contamination due to noise and artifacts. In our proposed work, we used Cuckoo Search Masking Empirical Mode decomposition to ignore such vulnerable things. Initially, the features of EEG signals are taken such as Energy, AR Coefficients, Morphological features and Fuzzy Approximate Entropy. Then, for Feature extraction method, Masking Empirical Mode Decomposition MEMD is applied to deal with motor imagery MI recognition tasks. The EEG signal is decomposed by MEMD and hybrid features are then extracted from the first two intrinsic mode functions IMFs . After the extracted features, Cuckoo Search algorithm is used to select the significant features. Different weights for the relevance and redundancy in the fitness function of the proposed algorithm are used to further improve their performance in terms of the number of features and the classification accuracy and finally they are fed into Linear Discriminant Analysis for classification. This analysis produces models whose accuracy is as good as more complex method. The results show that our proposed method can achieve the highest accuracy, maximal MI, recall as well as precision for Motor Imagery Recognition tasks. Our proposed method is comparable or superior than existing method. Jaipriya D ""Motor Imagery Recognition of EEG Signal using Cuckoo-Search Masking Empirical Mode Decomposition"" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-4 | Issue-2 , February 2020,
URL: https://www.ijtsrd.com/papers/ijtsrd30020.pdf
Paper Url : https://www.ijtsrd.com/engineering/electronics-and-communication-engineering/30020/motor-imagery-recognition-of-eeg-signal-using-cuckoo-search-masking-empirical-mode-decomposition/jaipriya-d
Distributed Approach for Clock Synchronization in Wireless Sensor NetworkEditor IJMTER
Time synchronization is an important service in WSNs. existing time synchronization algorithms
provide on average good synchronization between arbitrary nodes, however, as we show in this paper, close-by
nodes in a network may be synchronized poorly. We propose the Distributed Time Synchronization Algorithm
(DTSA) which is designed to provide accurately synchronized clocks between nearest-neighbours. DTSA works
in a completely decentralized fashion: Every node periodically broadcasts its time information. Synchronization
messages received from direct neighbours are used to calibrate the logical clock. The algorithm requires neither a
tree topology nor a reference node, which makes it robust against link and node failures.
Finding difference between west and eastijbesjournal
Yellow Fever is a fatal disease that causes yellowness and high fever. It is known to be highly contagious
and it is mostly prevalent in Africa, where it is know n to be originated from. However, the derivations of
this disease have reached other continents across the sea and they seem to be in different order. We mostly
focused on the derivations within Africa and discovered that the geographical derivations of this disease
also have different frequencies. We proved this phenomenon by analysing the sequences through apriori
and decision tree.
A look at the rise of mobile, the Facebook mobile unbundling strategy, the advantages of developing a mobile constellation, and the long term fit into the FB vision. From www.jamesmreed.com
A UTOMATIC S EGMENTATION IN B REAST C ANCER U SING W ATERSHED A LGORITHMijbesjournal
Accurate and reproducible delineation of breast les
ions can be challenging, as the lesions may have
complicated topological structures and heterogeneou
s intensity distributions. Diagnosis using magnetic
resonance imaging (MRI) with an appropriate automat
ic segmentation algorithm can be a better imaging
technique for the early detection of malignant brea
st tumours. The main objective of this system is to
develop a method for automatic segmentation and the
early detection of breast cancer based on the
application of the watershed transform to MRI image
s. The algorithm was separated into three major
sections: pre-processing, watershed and post-proces
sing. After computing different segments, the final
image was cleared of all noise and superimposed on
the original MRI image to generate the final modifi
ed image. The algorithm successfully resulted in the a
utomatic segmentation of the MRI images, and this c
an be a beneficial tool for the early detection of bre
ast cancer. This study showed that the automatic re
sults correctly agree with manual detection.
ENERGY COMPUTATION FOR BCI USING DCT AND MOVING AVERAGE WINDOW FOR NOISE SMOO...IJCSEA Journal
Brain computer interface (BCI) is a fast evolving field of research enabling computers and machines to be directly controlled by the human neural system. This enables people with muscular disability to directly control machines using their thought process. The brain signals are recorded using Electroencephalography (EEG) and patterns extracted so that the BCI system should be able to classify various patterns of brain signal accurately to perform different tasks. The raw EEG signal contains different kinds of interference waveforms (artifacts) and noise. Thus raw signals cannot be directly used for classification, the EEG signals has to undergo preprocessing, to remove artifacts and to extract the right attributes for classification. In this paper it is proposed to extract the energies in the EEG signal and classify the signal using Naïve Bayes and Instance based learners. The proposed method performs well for the two class problem in the multiple datasets used..
SUITABLE MOTHER WAVELET SELECTION FOR EEG SIGNALS ANALYSIS: FREQUENCY BANDS D...sipij
Wavelet transform (WT) is a powerful modern tool for time-frequency analysis of non-stationary signals such as electroencephalogram (EEG). The aim of this study is to choose the best and suitable mother wavelet function (MWT) for analyzing normal, seizure-free and seizured EEG signals. Several MWTs can be used, but the best MWT is the one that conserves the quasi-totality of information of the original signal on wavelet coefficients and gather more EEG rhythms in terms of frequency. In this study, Daubechies, Symlets and Coiflets orthogonal families were used as bsis mother wavelet functions. The percentage rootmeans square difference (PRD), the signal to noise ratio (SNR) and the simulated frequencies as the selection metrics. Simulation results indicate Daubechies wavelet at level 4 (Db4) as the most suitable MWT for EEG frequency bands decomposition.Furthermore, due to the redundancy of the extracted features, linear discriminant analysis (LDA) is applied for feature selection. Scatter plot showed that the selected feature vector represents the amount of changes in frequency distribution and carries most of the discriminative and representative information about their classes. Then, this study can provide a reference for the selection of a suitable MWT and discriminativefeatures.
Robot Motion Control Using the Emotiv EPOC EEG SystemjournalBEEI
Brain-computer interfaces have been explored for years with the intent of using human thoughts to control mechanical system. By capturing the transmission of signals directly from the human brain or electroencephalogram (EEG), human thoughts can be made as motion commands to the robot. This paper presents a prototype for an electroencephalogram (EEG) based brain-actuated robot control system using mental commands. In this study, Linear Discriminant Analysis (LDA) and Support Vector Machine (SVM) method were combined to establish the best model. Dataset containing features of EEG signals were obtained from the subject non-invasively using Emotiv EPOC headset. The best model was then used by Brain-Computer Interface (BCI) to classify the EEG signals into robot motion commands to control the robot directly. The result of the classification gave the average accuracy of 69.06%.
International Journal of Computational Engineering Research(IJCER)ijceronline
International Journal of Computational Engineering Research (IJCER) is dedicated to protecting personal information and will make every reasonable effort to handle collected information appropriately. All information collected, as well as related requests, will be handled as carefully and efficiently as possible in accordance with IJCER standards for integrity and objectivity.
Brain-computer interface of focus and motor imagery using wavelet and recurre...TELKOMNIKA JOURNAL
Brain-computer interface is a technology that allows operating a device without involving muscles and sound, but directly from the brain through the processed electrical signals. The technology works by capturing electrical or magnetic signals from the brain, which are then processed to obtain information contained therein. Usually, BCI uses information from electroencephalogram (EEG) signals based on various variables reviewed. This study proposed BCI to move external devices such as a drone simulator based on EEG signal information. From the EEG signal was extracted to get motor imagery (MI) and focus variable using wavelet. Then, they were classified by recurrent neural networks (RNN). In overcoming the problem of vanishing memory from RNN, was used long short-term memory (LSTM). The results showed that BCI used wavelet, and RNN can drive external devices of non-training data with an accuracy of 79.6%. The experiment gave AdaDelta model is better than the Adam model in terms of accuracy and value losses. Whereas in computational learning time, Adam's model is faster than AdaDelta's model.
IJERA (International journal of Engineering Research and Applications) is International online, ... peer reviewed journal. For more detail or submit your article, please visit www.ijera.com
Overview of Machine Learning and Deep Learning Methods in Brain Computer Inte...IJCSES Journal
Research under the field of Brain Computer Interfaces is adapting various Machine Learning and Deep
Learning techniques in recent times. With the advent of modern BCI, the data generated by various devices
is now capable of detecting brain signals more accurately. This paper gives an overview of all the steps
involved in the process of applying Machine Learning as well as Deep Learning methods from Data
Acquisition to application of algorithms. It aims to study techniques currently employed to extract data,
features from brain data, different algorithms employed to draw insights from the extracted features, and
how it can be used in various BCI applications. By this study, I aim to put forward current Machine
Learning and Deep Learning Trends in the field of BCI.
IJRET : International Journal of Research in Engineering and Technology is an international peer reviewed, online journal published by eSAT Publishing House for the enhancement of research in various disciplines of Engineering and Technology. The aim and scope of the journal is to provide an academic medium and an important reference for the advancement and dissemination of research results that support high-level learning, teaching and research in the fields of Engineering and Technology. We bring together Scientists, Academician, Field Engineers, Scholars and Students of related fields of Engineering and Technology.
Current motorized limb prostheses provide rudimentary functionality for the application in everyday life. Together with a
poor cosmetic appearance, this is the reason why a large percentage of amputees do not use their prosthetic device regularly. This
paper seeks to present an overview of current state of the art research on neural interfaces. The focus lies on non-invasive
recording with EMG and especially High-Density EMG sensors. Additionally, direct machine learning and pattern recognition
algorithms for the decoding of the recorded signals are discussed. Finally, promising research directions for advanced prosthesis
control will be discussed. The bionic arm uses EMG signals to control each action of the hand. In order to control them, we need to
record the EMG signal for different actions. And compare it with real-time values to move the hand in a different manner. There
are separate servo motors to control the actions of each finger separately. So these are programmed by using microcontrollers.
Recognition of new gestures using myo armband for myoelectric prosthetic appl...IJECEIAES
Myoelectric prostheses are a viable solution for people with amputations. The chal- lenge in implementing a usable myoelectric prosthesis lies in accurately recognizing different hand gestures. The current myoelectric devices usually implement very few hand gestures. In order to approximate a real hand functionality, a myoelectric prosthesis should implement a large number of hand and finger gestures. However, increasing number of gestures can lead to a decrease in recognition accuracy. In this work a Myo armband device is used to recognize fourteen gestures (five build in gestures of Myo armband in addition to nine new gestures). The data in this research is collected from three body-able subjects for a period of 7 seconds per gesture. The proposed method uses a pattern recognition technique based on Multi-Layer Perceptron Neural Network (MLPNN). The results show an average accuracy of 90.5% in recognizing the proposed fourteen gestures.
Motor Imagery Recognition of EEG Signal using Cuckoo Search Masking Empirical...ijtsrd
Brain Computer Interface BCI aims at providing an alternate means of communication and control to people with severe cognitive or sensory motor disabilities. Brain Computer Interface in electroencephalogram EEG is of great important but it is challenging to manage the non stationary EEG. EEG signals are more vulnerable to contamination due to noise and artifacts. In our proposed work, we used Cuckoo Search Masking Empirical Mode decomposition to ignore such vulnerable things. Initially, the features of EEG signals are taken such as Energy, AR Coefficients, Morphological features and Fuzzy Approximate Entropy. Then, for Feature extraction method, Masking Empirical Mode Decomposition MEMD is applied to deal with motor imagery MI recognition tasks. The EEG signal is decomposed by MEMD and hybrid features are then extracted from the first two intrinsic mode functions IMFs . After the extracted features, Cuckoo Search algorithm is used to select the significant features. Different weights for the relevance and redundancy in the fitness function of the proposed algorithm are used to further improve their performance in terms of the number of features and the classification accuracy and finally they are fed into Linear Discriminant Analysis for classification. This analysis produces models whose accuracy is as good as more complex method. The results show that our proposed method can achieve the highest accuracy, maximal MI, recall as well as precision for Motor Imagery Recognition tasks. Our proposed method is comparable or superior than existing method. Jaipriya D ""Motor Imagery Recognition of EEG Signal using Cuckoo-Search Masking Empirical Mode Decomposition"" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-4 | Issue-2 , February 2020,
URL: https://www.ijtsrd.com/papers/ijtsrd30020.pdf
Paper Url : https://www.ijtsrd.com/engineering/electronics-and-communication-engineering/30020/motor-imagery-recognition-of-eeg-signal-using-cuckoo-search-masking-empirical-mode-decomposition/jaipriya-d
Distributed Approach for Clock Synchronization in Wireless Sensor NetworkEditor IJMTER
Time synchronization is an important service in WSNs. existing time synchronization algorithms
provide on average good synchronization between arbitrary nodes, however, as we show in this paper, close-by
nodes in a network may be synchronized poorly. We propose the Distributed Time Synchronization Algorithm
(DTSA) which is designed to provide accurately synchronized clocks between nearest-neighbours. DTSA works
in a completely decentralized fashion: Every node periodically broadcasts its time information. Synchronization
messages received from direct neighbours are used to calibrate the logical clock. The algorithm requires neither a
tree topology nor a reference node, which makes it robust against link and node failures.
Finding difference between west and eastijbesjournal
Yellow Fever is a fatal disease that causes yellowness and high fever. It is known to be highly contagious
and it is mostly prevalent in Africa, where it is know n to be originated from. However, the derivations of
this disease have reached other continents across the sea and they seem to be in different order. We mostly
focused on the derivations within Africa and discovered that the geographical derivations of this disease
also have different frequencies. We proved this phenomenon by analysing the sequences through apriori
and decision tree.
A look at the rise of mobile, the Facebook mobile unbundling strategy, the advantages of developing a mobile constellation, and the long term fit into the FB vision. From www.jamesmreed.com
A UTOMATIC S EGMENTATION IN B REAST C ANCER U SING W ATERSHED A LGORITHMijbesjournal
Accurate and reproducible delineation of breast les
ions can be challenging, as the lesions may have
complicated topological structures and heterogeneou
s intensity distributions. Diagnosis using magnetic
resonance imaging (MRI) with an appropriate automat
ic segmentation algorithm can be a better imaging
technique for the early detection of malignant brea
st tumours. The main objective of this system is to
develop a method for automatic segmentation and the
early detection of breast cancer based on the
application of the watershed transform to MRI image
s. The algorithm was separated into three major
sections: pre-processing, watershed and post-proces
sing. After computing different segments, the final
image was cleared of all noise and superimposed on
the original MRI image to generate the final modifi
ed image. The algorithm successfully resulted in the a
utomatic segmentation of the MRI images, and this c
an be a beneficial tool for the early detection of bre
ast cancer. This study showed that the automatic re
sults correctly agree with manual detection.
Characterization of effective mechanical strength of chitosan porous tissue s...ijbesjournal
Tissue engineering can be understand as the development of functional substitute to replace missing or malfunctioning human tissue and organs by using biodegradable or non-biodegradable biomaterials such
as scaffolds to direct specific cell types to organize into three dimensional structures and perform
differentiated function of targeted tissue. The important factors to be considered in designing of
microstructure and there structure material were type of bio-material porosity, pore size, and pore
structure with respect to nutrient supply for transplanted and regenerated cells. Performance of various
functions of the tissue structure depends on porous scaffold microstructures with dimensions of specific
porosity, pore size, characteristics that influence the behaviorand development of the incorporated cells.
Finite element Methods (FEM) and Computer Aided Design (CAD) combines with manufacturing
technologies such as Solid Freeform Fabrication (SFF) helpful to allow virtual design and fabrication,
characterization and production of porous scaffold optimized for tissue replacement with appropriate pore
size and proper dimension. In this paper we found that with the increase in the porosity of tissue
scaffolds(PCL, HAP, PGAL & Chitosan) their effective mechanical strength decreases by performing the
modeling & simulation with CAD & FEM package (Pro/E & ANSYS respectively) and validating the results with in vitro fabrication of Chitosan scaffold by performing in vivo mechanical testing.
METHODS OF COMMAND RECOGNITION USING SINGLE-CHANNEL EEGSijistjournal
This work proposes to recognize a user's commands by analysing his/her brainwaves captured with single channel electroencephalogram (EEG). Whenever a user intends to issue one of the pre-defined commands, the proposed system prompts him/her all the candidate commands in turn. Then, the user is asked to be concentrated as possible as he/she can, when the desired command is shown. It is assumed that the concentration will present a certain pattern of “Yes” in the captured EEG, as opposed to a certain pattern of “No” when the user is relaxed. Accordingly, the task is to determine that the captured EEG is “Yes” or not. This work compares three recognition methods, respectively, based on Gaussian mixture models, hidden Markov models and recurrent neural network, and conducts experiments using 2400 test EEG samples recorded from 10 subjects.
METHODS OF COMMAND RECOGNITION USING SINGLE-CHANNEL EEGSijistjournal
This work proposes to recognize a user's commands by analysing his/her brainwaves captured with single channel electroencephalogram (EEG). Whenever a user intends to issue one of the pre-defined commands, the proposed system prompts him/her all the candidate commands in turn. Then, the user is asked to be concentrated as possible as he/she can, when the desired command is shown. It is assumed that the concentration will present a certain pattern of “Yes” in the captured EEG, as opposed to a certain pattern of “No” when the user is relaxed. Accordingly, the task is to determine that the captured EEG is “Yes” or not. This work compares three recognition methods, respectively, based on Gaussian mixture models, hidden Markov models and recurrent neural network, and conducts experiments using 2400 test EEG samples recorded from 10 subjects.
METHODS OF COMMAND RECOGNITION USING SINGLE-CHANNEL EEGSijistjournal
This work proposes to recognize a user's commands by analysing his/her brainwaves captured with single
channel electroencephalogram (EEG). Whenever a user intends to issue one of the pre-defined
commands, the proposed system prompts him/her all the candidate commands in turn. Then, the user is
asked to be concentrated as possible as he/she can, when the desired command is shown. It is assumed
that the concentration will present a certain pattern of “Yes” in the captured EEG, as opposed to a
certain pattern of “No” when the user is relaxed. Accordingly, the task is to determine that the captured
EEG is “Yes” or not. This work compares three recognition methods, respectively, based on Gaussian
mixture models, hidden Markov models and recurrent neural network, and conducts experiments using
2400 test EEG samples recorded from 10 subjects.
EEG SIGNAL CLASSIFICATION USING LDA AND MLP CLASSIFIERhiij
Electroencephalography (EEG) is the recording of electrical activities along the scalp. EEG measures
voltage fluctuations resulting from ionic current flows within the neurons of the brain. Diagnostic
applications generally focus on the spectral content of EEG, which is the type of neural oscillations that
can be observed in EEG signal. EEG is most often used to diagnose epilepsy, which causes obvious
abnormalities in EEG readings. This powerful property confirms the rich potential for EEG analysis and
motivates the need for advanced signal processing techniques to aid clinicians in their interpretations.
This paper describes the application of Wavelet Transform (WT) for the processing of
Electroencephalogram (EEG) signals. Furthermore, the linear discriminant analysis (LDA) is applied for
feature selection and dimensionality reduction where the informative and discriminative two-dimension
features are used as a benchmark for classification purposes through a Multi-Layers Perceptron (MLP)
neural network. For five classification problems, the proposed model achieves a high sensitivity,
specificity and accuracy of 100%.Finally, the comparison of the results obtained with the proposed
methods and those obtained with previous literature methods shows the superiority of our approach for
EEG signals classification and automated diagnosis
MHEALTH APPLICATIONS DEVELOPED BY THE MINISTRY OF HEALTH FOR PUBLIC USERS INK...hiij
mHealth applications have shown promise in supporting the delivery of health services in peoples’ daily life. Recently, the Ministry of Health in the Kingdom of Saudi Arabia (MOH) has launched several mHealth applications to develop work mechanisms. Our study aimed to identify and understand the design of mHealth apps by classifying their persuasive features using the Persuasive Systems Design (PSD) model and expert evaluation method. This paper presents the distinct persuasive features applied in recent applications launched by MOH for public users called “Sehha & Mawid” Apps. The results revealed the extensive use of persuasive features; particularly features related to credibility support, dialogue support and primary task support respectively. The implementation and design of social support features were found to be poor; this could be due to the nature of the apps or lack of knowledge from the developers’
perspectives. The findings suggest some features that may improve the persuasion for the evaluated apps.
5. detection and separation of eeg artifacts using wavelet transform nov 11, ...IAESIJEECS
Bio-medical signal processing is one of the most important techniques of multichannel sensor network and it has a substantial concentration in medical application. However, the real-time and recorded signals in multisensory instruments contains different and huge amount of noise, and great work has been completed in developing most favorable structures for estimating the signal source from the noisy signal in multichannel observations. Methods have been developed to obtain the optimal linear estimation of the output signal through the Wide-Sense-Stationary (WSS) process with the help of time-invariant filters. In this process, the input signal and the noise signal are assumed to achieve the linear output signal. During the process, the non-stationary signals arise in the bio-medical signal processing in addition to it there is no effective structure to deal with them. Wavelets transform has been proved to be the efficient tool for handling the non-stationary signals, but wavelet provide any possible way to approach multichannel signal processing. Based on the basic structure of linear estimation of non-stationary multichannel data and statistical models of spatial signal coherence acquire through the wavelet transform in multichannel estimation. The above methods can be used for Electroencephalography (EEG) signal denoising through the original signal and then implement the noise reduction technique to evaluate their performance such as SNR, MSE and computation time.
A New Approach to Denoising EEG Signals - Merger of Translation Invariant Wav...CSCJournals
In this paper we present a new algorithm using a merger of Independent Component Analysis and Translation Invariant Wavelet Transform. The efficacy of this algorithm is evaluated by applying contaminated EEG signals. Its performance was compared to three fixed-point ICA algorithms (FastICA, EFICA and Pearson-ICA) using Mean Square Error (MSE), Peak Signal to Noise Ratio (PSNR), Signal to Distortion Ratio (SDR), and Amari Performance Index. Experiments reveal that our new technique is the most accurate separation method.
Performance Comparison of Known ICA Algorithms to a Wavelet-ICA MergerCSCJournals
Electroencephalogram (EEG) is useful for biological research and clinical diagnosis. These signals are however contaminated with artifacts which must be removed to have pure EEG signals. These artifacts can be removed by using Independent Component Analysis (ICA). In this paper we studied the performance of three ICA algorithms (FastICA, JADE, and Radical) as well as our newly developed ICA technique which utilizes wavelet transform. Comparing these ICA algorithms, it is observed that our new technique performs as well as these algorithms at denoising EEG signals.
METHODS OF COMMAND RECOGNITION USING SINGLE-CHANNEL EEGSijistjournal
This work proposes to recognize a user's commands by analysing his/her brainwaves captured with single
channel electroencephalogram (EEG). Whenever a user intends to issue one of the pre-defined
commands, the proposed system prompts him/her all the candidate commands in turn. Then, the user is
asked to be concentrated as possible as he/she can, when the desired command is shown. It is assumed
that the concentration will present a certain pattern of “Yes” in the captured EEG, as opposed to a
certain pattern of “No” when the user is relaxed. Accordingly, the task is to determine that the captured
EEG is “Yes” or not. This work compares three recognition methods, respectively, based on Gaussian
mixture models, hidden Markov models and recurrent neural network, and conducts experiments using
2400 test EEG samples recorded from 10 subjects.
HUMAN EMOTION ESTIMATION FROM EEG AND FACE USING STATISTICAL FEATURES AND SVMcsandit
An approach is presented in this paper for automated estimation of human emotions from
combination of multimodal data: electroencephalogram and facial images. The used EEG
features are the Hjorth parameters calculated for theta, alpha, beta and gamma bands taken
from pre-defined channels. For face emotion estimation PCA feature are selected. Classification
is performed with support vector machines. Since the human emotions are modelled as
combinations from physiological elements such as arousal, valence, dominance, liking, etc.,
these quantities are the classifier’s outputs. The best achieved correct classification
performance for EEG is about 76%. Classifier combination is used to return the final score for
the particular subject.
IJERA (International journal of Engineering Research and Applications) is International online, ... peer reviewed journal. For more detail or submit your article, please visit www.ijera.com
We have compiled the most important slides from each speaker's presentation. This year’s compilation, available for free, captures the key insights and contributions shared during the DfMAy 2024 conference.
Literature Review Basics and Understanding Reference Management.pptxDr Ramhari Poudyal
Three-day training on academic research focuses on analytical tools at United Technical College, supported by the University Grant Commission, Nepal. 24-26 May 2024
KuberTENes Birthday Bash Guadalajara - K8sGPT first impressionsVictor Morales
K8sGPT is a tool that analyzes and diagnoses Kubernetes clusters. This presentation was used to share the requirements and dependencies to deploy K8sGPT in a local environment.
HEAP SORT ILLUSTRATED WITH HEAPIFY, BUILD HEAP FOR DYNAMIC ARRAYS.
Heap sort is a comparison-based sorting technique based on Binary Heap data structure. It is similar to the selection sort where we first find the minimum element and place the minimum element at the beginning. Repeat the same process for the remaining elements.
Using recycled concrete aggregates (RCA) for pavements is crucial to achieving sustainability. Implementing RCA for new pavement can minimize carbon footprint, conserve natural resources, reduce harmful emissions, and lower life cycle costs. Compared to natural aggregate (NA), RCA pavement has fewer comprehensive studies and sustainability assessments.
Hierarchical Digital Twin of a Naval Power SystemKerry Sado
A hierarchical digital twin of a Naval DC power system has been developed and experimentally verified. Similar to other state-of-the-art digital twins, this technology creates a digital replica of the physical system executed in real-time or faster, which can modify hardware controls. However, its advantage stems from distributing computational efforts by utilizing a hierarchical structure composed of lower-level digital twin blocks and a higher-level system digital twin. Each digital twin block is associated with a physical subsystem of the hardware and communicates with a singular system digital twin, which creates a system-level response. By extracting information from each level of the hierarchy, power system controls of the hardware were reconfigured autonomously. This hierarchical digital twin development offers several advantages over other digital twins, particularly in the field of naval power systems. The hierarchical structure allows for greater computational efficiency and scalability while the ability to autonomously reconfigure hardware controls offers increased flexibility and responsiveness. The hierarchical decomposition and models utilized were well aligned with the physical twin, as indicated by the maximum deviations between the developed digital twin hierarchy and the hardware.
TOP 10 B TECH COLLEGES IN JAIPUR 2024.pptxnikitacareer3
Looking for the best engineering colleges in Jaipur for 2024?
Check out our list of the top 10 B.Tech colleges to help you make the right choice for your future career!
1) MNIT
2) MANIPAL UNIV
3) LNMIIT
4) NIMS UNIV
5) JECRC
6) VIVEKANANDA GLOBAL UNIV
7) BIT JAIPUR
8) APEX UNIV
9) AMITY UNIV.
10) JNU
TO KNOW MORE ABOUT COLLEGES, FEES AND PLACEMENT, WATCH THE FULL VIDEO GIVEN BELOW ON "TOP 10 B TECH COLLEGES IN JAIPUR"
https://www.youtube.com/watch?v=vSNje0MBh7g
VISIT CAREER MANTRA PORTAL TO KNOW MORE ABOUT COLLEGES/UNIVERSITITES in Jaipur:
https://careermantra.net/colleges/3378/Jaipur/b-tech
Get all the information you need to plan your next steps in your medical career with Career Mantra!
https://careermantra.net/
NUMERICAL SIMULATIONS OF HEAT AND MASS TRANSFER IN CONDENSING HEAT EXCHANGERS...ssuser7dcef0
Power plants release a large amount of water vapor into the
atmosphere through the stack. The flue gas can be a potential
source for obtaining much needed cooling water for a power
plant. If a power plant could recover and reuse a portion of this
moisture, it could reduce its total cooling water intake
requirement. One of the most practical way to recover water
from flue gas is to use a condensing heat exchanger. The power
plant could also recover latent heat due to condensation as well
as sensible heat due to lowering the flue gas exit temperature.
Additionally, harmful acids released from the stack can be
reduced in a condensing heat exchanger by acid condensation. reduced in a condensing heat exchanger by acid condensation.
Condensation of vapors in flue gas is a complicated
phenomenon since heat and mass transfer of water vapor and
various acids simultaneously occur in the presence of noncondensable
gases such as nitrogen and oxygen. Design of a
condenser depends on the knowledge and understanding of the
heat and mass transfer processes. A computer program for
numerical simulations of water (H2O) and sulfuric acid (H2SO4)
condensation in a flue gas condensing heat exchanger was
developed using MATLAB. Governing equations based on
mass and energy balances for the system were derived to
predict variables such as flue gas exit temperature, cooling
water outlet temperature, mole fraction and condensation rates
of water and sulfuric acid vapors. The equations were solved
using an iterative solution technique with calculations of heat
and mass transfer coefficients and physical properties.
PROJECT FORMAT FOR EVS AMITY UNIVERSITY GWALIOR.ppt
Analysis of eeg for motor imagery
1. International Journal of Biomedical Engineering and Science (IJBES), Vol. 2, No. 3, July 2015
11
ANALYSIS OF EEG FOR MOTOR IMAGERY
BASED CLASSIFICATION OF HAND
ACTIVITIES
A.Sivakami1
and S.Shenbaga Devi2
1
PG Student, Department of ECE College of Engineering Guindy, Anna University
Chennai-600025, India
2
Professor, Center for Medical Electronics, Department of ECE College of Engineering
Guindy, Anna University Chennai-600025, India
ABSTRACT
EEG based Brain Computer Interface (BCI) establishes a new channel between human brain and the
surrounding environment in order to disseminate instructions to the outside world. It is based on the
recording of temporary EEG changes during different types of motor imagery such as imagination of
different hand movements. The spatial pattern of activated cortical areas during motor imagery is similar
to that of real time executed movement. Time domain features and frequency domain features are extracted
with emphasis on recognizing discriminative features representing EEG trials recorded during imagination
of different hand movements. Then, classification into different hand movements is carried out.
KEYWORDS
EEG motor imagery, Brain Computer Interface, event-related desynchronization (ERD), wavelet transform
1. INTRODUCTION
Patients in a late stage of amyotrophic lateral sclerosis (ALS), severe neuromuscular disorders
and those paralyzed from higher level spinal cord injury are not able to produce any voluntary
muscle movements. Sensory and cognitive functions are only minimally affected by such
diseases. Communication based on EEG signals does not require neuromuscular control and the
individuals who have no more control over any of their conventional communication abilities
may still be able to communicate through a direct Brain Computer Interface (BCI). EEG
recordings during hand motor imagery can be used as control signals for BCI applications like
cursor control, selection of letters or words, control of prosthesis, navigation of wheelchair, etc.
Such systems will increase the disables’ independence, leading to an improvement in quality of
life and reduced social costs. A lot of other techniques can monitor brain activity like functional
Magnetic Resonance Imaging (fMRI), Magnetoencephalography (MEG), Positron Emission
Tomography (PET) and Single Photon Emission Computer Tomography (SPECT). Although
fMRI, PET and SPECT are more accurate and have better spatial resolution than EEG, they are
not candidates for BCI applications because, due to their large size, heavy weight, they cannot act
as portable devices. EEG, on the other hand has better temporal resolution, is portable and cost
effective
2. International Journal of Biomedical Engineering and Science (IJBES), Vol. 2, No. 3, July 2015
12
Motor imagery is defined as imagining a motor action without any efferent information to
neuromuscular system. Thoughts and actions are intimately linked. A confirmation of this
prediction is found in the spatial patterning of activated cortical areas seen with functional brain
imaging techniques such as PET and fMRI.
Studies have shown that when the subject performs or even imagines limb movement, specific
frequency components of EEG such as mu and central beta rhythms are (de)synchronized over
contralateral (ipsilateral) sensorimotor area.EEG recordings during left and right motor imagery
can be used as control signals for a BCI.
Athena Akrami et al (2005) extracted quantitative changes in EEG due to movement imagination.
The features extracted are logarithmic power of different frequency bands of EEG which are
extracted from various combinations of channels [1].
Xiu Zhang and Xingyu Wang (2008) used Canonical Variate Analysis (CVA) for classification of
mental imagery. Temporal features are extracted as squared band pass filtered EEG and
frequency features are extracted as energy in specific rhythms. Features in time and frequency
domains are projected into canonical discriminant spatial feature space and classification was
done with Support Vector Machine [2].
Pawel Herman et al (2008) conducted a comparative study of spectral approaches and quantified
relevant spectral content. Features were extracted from a variety of techniques like atomic
decomposition, quadratic energy distribution, wavelet packets, Discrete Wavelet Transform and
AR model. They concluded that the effective spectral method is subject specific. Few cases of
techniques achieve good performance in one subject and poor performance in another one [3].
Kavitha P.Thomas et al (2009) proposed a new discriminative filter bank common spatial
algorithm to extract subject specific frequency bands using Fisher ratio of filtered EEG signal
from channels C3 or C4. Classification was done using Support Vector Machine [4].
In our work, multi-dimension feature extraction is done. Features are extracted using Discrete
Wavelet Transform, FFT Power Spectrum, Event Related Desynchronization (ERD) and
classification is done using Back Propagation Neural Network.
2. METHODOLOGY
Forty healthy volunteers mostly aged between 20 and 25, participated in the study. The subjects
are seated in a relaxing chair with armrests, approximately 100 cm from a computer screen. Prior
to the experiment, each subject is given the opportunity to practice and perform actual movements
of the hand. Electrodes are positioned according to the international 10-20 system. The motor
imagery EEG signal is predominant in sensorimotor cortex, corresponding to electrode positions
C3 and C4. The block diagram is shown in Figure 1.
The recordings are made with 32 channel RMS EEG machine. The EEG signal is sampled at 256
Hz. Higher frequency can be seen as noise caused by muscle activity, blink of eyes and other
noises. It is filtered between 1 and 35Hz with Notch filter ON. It is possible to set the sensitivity,
filter cut-off frequencies for each channel individually.
3. International Journal of Biomedical Engineering and Science (IJBES), Vol. 2, No. 3, July 2015
13
Figure 1. Block Diagram
2.1. Experimental paradigm
The experimental paradigm decides the environmental conditions in order to prompt the user to
generate the controlling EEG activity.
Two different experimental paradigms are followed. One is using auditory cue and the other is
using visual cue. The task is to perform left hand or right hand imagery according to the cue.
Subjects are explicitly instructed to imagine the kinesthetic sensation of movement and not to
imagine mere visualization of movement. The order of cues is random. The experiment consists
of several runs with 12 trials each after each. Each trial lasts 20 seconds.
In the case of auditory cue, the subject is instructed to keep eyes closed. In this type of cue, there
is no problem of EOG artifact. The EEG signal is elicited by having the subject executing
different mental tasks such as imagination of left hand or right hand movement in response to the
pre-recorded instructions, while remaining in a totally passive state. During execution, subjects
are asked to perform only mild movement and not overt movement. The timeline of the trials is
illustrated in Figure 2. The EEG activity of the rest period from 1-4 seconds is used as a baseline
for subsequent analysis of the mental tasks.
Figure 2. Auditory Cue
In the case of visual cue, arrows are used to indicate the imagination tasks to be performed (left
vs. right arm movement imaginations). The timeline of the trials is illustrated in Figure 3. After
trial begins, the first 3s are quiet, at t=3s an acoustic stimulus indicated the beginning of the trial,
and a fixation cross “+” is displayed, which remained on screen for the rest of the trial period.
The subjects are instructed to focus their gaze and attention at the centre of fixation cross. An
arrow indicating the direction of the imagination task appears 3 seconds after the cross is made
visible. The arrow remains on screen for 4 seconds, i.e. for the duration of the imagination period
which starts at the appearance of the arrow. When, the arrow disappears, the subject instructed to
perform actual movement. Blinking and swallowing are permitted only during rest time.
4. International Journal of Biomedical Engineering and Science (IJBES), Vol. 2, No. 3, July 2015
14
Figure 3. Visual Cue
2.2. EEG Referencing and Preprocessing
Three types of EEG referencing methods are adopted in this study, namely unipolar signals
(signals with respect to ear reference), bipolar signals and Surface Laplacian (SL) filtered signals.
The advantage of using unipolar is that, it localizes event of interest and that of using bipolar is
that, it reduces shared artifact like ECG.
All the trials are visually checked for EEG artifacts during the movement imagery period. Hand
movements during imagination period or EOG activity are omitted from further analysis. Eye
blink is said to have occurred if change in magnitude of greater than 100 µV occurred within 10
ms period. EMG artifact may occur as a result of muscles that lift eyebrows, close the jaw. All the
signals are filtered using a digital butterworth FIR band pass filter in the range 8-30 Hz because
the mu band and central beta band fall within this range.
2.3. SL Estimation
Raw EEG scalp signals are known to have poor spatial resolution due to volume conduction
effect. Only half the contribution to each scalp electrode comes from sources within 3 cm radius.
This is in particular a problem if the signal of interest is weak like the sensorimotor rhythms.
Surface laplacian (SL) method is superior to the ear reference method, because it is a high pass
spatial filter, which uses linear combination of simultaneous input samples. It favours signals
originating from hand areas over signals that originate from other areas. It takes the difference
between target electrode and several electrodes that surround it i.e. between C3 electrode and its
neighbouring electrodes and between C4 and its neighbouring electrodes. A low resolution SL
computation needs a total of 9 electrodes whereas a high resolution SL computation needs 26
electrodes. In this work, data is collected from F3, F4, T3, T4, C3, C4, CZ, P3, and P4 of the
international 10-20 system for SL computation. However there is an acceptable correlation
between the two and hence a low resolution SL transform is applied to the acquired signal [5].
3. FEATURE EXTRACTION
The signals are processed to extract distinct features. For this purpose, Multidimensional feature
extraction is carried out. i.e. Features are extracted from both time domain and frequency domain
for all three types of recorded signals namely raw, bipolar, SL filtered signals. The techniques
used are ERD quantification, FFT Power Spectrum and Discrete Wavelet Transform.
3.1. ERD Quantification
Pfurtscheller and Aranibar first quantified event-related desynchronization (ERD) in 1977. Event
Related Desynchronization (ERD) is a reduction of a specific frequency component and is related
to an increase in neural activity. Event Related Synchronization (ERS) is an increase in a specific
5. International Journal of Biomedical Engineering and Science (IJBES), Vol. 2, No. 3, July 2015
15
frequency component and is related to neural suppression. This technique is based on the
assumption that if an assembly of neurons engaged in the same process works synchronously in
frequency, the power at this frequency increases [6]. When EEG signal is recorded from
sensorimotor area, alpha rhythm is called as Mu rhythm.
Movement or preparation for movement is typically accompanied by a decrease in mu activity
over, specifically contralateral to the movement causing ERD. It’s opposite, rhythm increase or
"event-related synchronization" (ERS) occurs in the post-movement period and with relaxation.
Thus, ERD and ERS can occur independent of activity in the brain's normal output channels of
peripheral nerves and muscles, and can thus, be used as the basis for a BCI.
Steps to compute the time course of ERD are as follows:
1.Bandpass filtering of all event-related trials.
2.Squaring of the amplitude samples to obtain power samples.
3.Averaging of power samples across all trials
4.Averaging over time samples to smooth the data and reduce variability
ERD is quantified as percentage EEG power decrease or increase within specific frequency
bands.
where A is the power within the frequency band of interest i.e. mu band and R is the power of
reference interval before the warning beep.
Figure 4. shows ERD% wherein during right hand imagery, at the contra-lateral (left)
sensorimotor cortex there is a reduction in power when compared with baseline period. During
right hand movement imagery, in the imagery period from 5-9s there is an attenuation in the C3
component when compared to C4.Features are obtained from this 5 to 9 second imagery period. A
feature vector is constructed by taking the average ERD% for every second. For each channel,
there are 4 ERD features. Thus, a total of 8 ERD features are obtained.
Figure 4. ERD%
3.2. Power Spectrum
The time domain signals obtained from EEG amplifier reflects only one side of EEG. For precise
analysis, we need to look at the signal in the frequency domain. In the frequency domain, ERD is
6. International Journal of Biomedical Engineering and Science (IJBES), Vol. 2, No. 3, July 2015
16
characterised by a reduction in the power spectrum corresponding to mu band in the contra-lateral
cortex.Different segments of EEG like baseline, imagery and actual movement are analysed.
It is observed that the pattern of mu band spectrum of movement imagery is similar to that of
actual movement. There is a difference only in terms of magnitude i.e. peaks of mu rhythm are
higher in imagery than movement. This is illustrated in Figure 5. The actual movement spectrum
depicted in green colour has reduced amplitude than the imagery spectrum.
Figure 5. Power spectrum of movement vs imagery
Planning for movement leads to a short-lasting circumscribed attenuation in the mu rhythm.
Hence, during right hand imagery, in the channel C3 imagery power spectrum, the 8-12 Hz
component is attenuated when compared to the baseline. This is evident in Figure 6.
Figure 6. Power spectrum of baseline vs imagery
In right hand motor imagery, peak value of PSD is lesser in C3 than C4. This is depicted in Figure
7.In the case of left hand imagery, peak value of PSD is more in C3 when compared to C4.
The figures plotted are for the same subject. The location of the mu peak tends to vary from
person to person. Generally it varies between 8 and 12 Hz. For the subject shown here it is
observed that the characteristic peak is around 9 Hz.
Figure 7. Power spectrum for right hand imagery
7. International Journal of Biomedical Engineering and Science (IJBES), Vol. 2, No. 3, July 2015
17
Motor imagery period of 4 sec length i.e. from 5th second to 9th second of the trial is considered
for processing. The power values in the spectrum are averaged together in groups defined by
frequency ranges to yield power in the mu and central beta bands since motor imagery
information is available only in these bands. The features extracted are alpha and central beta
mean values of C3 and C4 signals and also the relative alpha power percentage and beta power
percentage are considered.
For each type of signal i.e. unipolar, bipolar and SL signal, the above mentioned features are
considered. While recording EEG itself, the subjects are asked to perform actual movements.
From the classification point of view, only imagery and baseline data are needed.
3.3. Discrete Wavelet Transform
Wavelet, which is called a mathematical microscope for analyzing signals, has the ability to
analyze signal which is localized in time domain or frequency domain. EEG signal's non-
stationary and transient characteristics make it difficult to extract the exact characteristics of EEG
through the ordinary spectrum analysis methods. The wavelet transform decomposes a signal into
a set of functions obtained by shifting and dilating one single function called mother wavelet.
In wavelet analysis there are the low frequencies, high scale component of the signal-
approximation coefficient A and the high frequency, low frequency component - detail coefficient
D. The hierarchically organized wavelet decomposition scheme is illustrated in Figure 8.
Figure 8. Wavelet decomposition
3.4. DWT of EEG
Daubechies6 wavelet is chosen as the wavelet base. Db6 is a 6 band orthogonal wavelet. The
sampling frequency of the EEG data acquisition unit is 256 Hz. The number of levels of
decomposition is chosen based on the dominant frequency components of the signal. We chose
the level as 5. As a result, the EEG signal is decomposed into detail coefficients D1-D5 and
approximation coefficient A5. D5 coefficients correspond to alpha band and D4 coefficients
correspond to central beta band. The 1280 points of raw EEG data are considered in wavelet
transform.
3.5. EEG Feature Selection
The wavelet decomposed sub-bands are illustrated in Figure 9.It shows the decomposition of both
C3and C4 signal for right hand motor imagery. The above analysis shows that there is an obvious
8. International Journal of Biomedical Engineering and Science (IJBES), Vol. 2, No. 3, July 2015
18
difference between the two signals at the level of D4 and D5.The C3 component has reduced
magnitude when compared to the C4 component.
The Statistical features are extracted from the sub bands corresponding to alpha and central beta
waves. The features considered are the mean of the absolute value of wavelet coefficients and
standard deviation of the relevant sub band coefficients– D4 and D5. These features represent the
frequency distribution and the amount of changes in the frequency distribution. Thus, for each
channel, 4 features namely mean and standard deviations corresponding to alpha and beta sub-
bands are extracted as features.
0 100 200 300 400 500 600 700
-5
0
5
cD1
C3
C4
0 50 100 150 200 250 300 350
-50
0
50
cD2
0 20 40 60 80 100 120 140 160 180
-200
0
200
cD3
0 10 20 30 40 50 60 70 80 90
-500
0
500
cD4
C3
C4
0 5 10 15 20 25 30 35 40 45 50
-200
0
200
cD5
0 5 10 15 20 25 30 35 40 45 50
-1000
0
1000
cA5
Figure 9. Wavelet decomposition of C3 and C4 for right hand imagery
The feature values computed for one subject during right hand imagery is shown in Table 1.
Table 1. Feature values during right hand imagery
9. International Journal of Biomedical Engineering and Science (IJBES), Vol. 2, No. 3, July 2015
19
4. CLASSIFICATION
The very aim of BCI is to translate brain activity into a command for a computer. The
classification stage involves the identification of the feature patterns to facilitate the
categorization of the user’s intents. The output of the classification stage is the controlling input
of the device.
The various classification algorithms used to design BCI systems are: linear classifiers (Linear
Discriminant Analysis- LDA, Support Vector Machine- SVM), neural networks and non-linear
Bayesian classifiers. Out of these, the Linear Discriminant Analysis and Neural Networks are
widely used in BCI systems. The main drawback of LDA is its linearity that can provide poor
results on complex non-linear EEG data.
Neural network is an assembly of several artificial neurons which enables to produce non-linear
decision boundaries. A back propagation network with 3 layers as shown in Figure 10 is used for
classification.
Figure 10. 3 layer Back Propagation Network
Traingdx learning algorithm is used to train the network. Traingdx uses gradient descend with
momentum factor set as 0.86 to avoid a shallow local minimum and a variable learning rate to
make the learning as fast as possible while maintaining stability. The initial learning rate is set as
0.05.Batch learning is used to update the network weights after all training data is presented.
Maximum number of epochs for training is given as 1000. Two-thirds of the data is utilized for
training and the remaining one-third is kept for testing. i.e. 23 samples are used for training and
12 samples are used for testing.
In this work, different combinations of signal (i.e. unipolar, bipolar, SL signals) and feature
extraction methods are tried. This is shown in Table 2. Out of the 3 types of signals considered, it
is found that bipolar and SL signals possess more discriminatory properties for right vs. left
classification. 6 BPN’s with different types of input features are tried in order to identify the
processing method which yields best results. After many trials it is found that a BPN with 12
features namely ERD 1st
second, ERD 2nd
second, ERD 3rd
second, ERD 4th
second values in
channel C3 and C4, Bipolar channel alpha mean and central beta mean in channels C3 and C4
provided the best results. This can be seen in row no. 1 in Table 2. The 4 seconds of ERD
correspond to the period of motor imagery.
10. International Journal of Biomedical Engineering and Science (IJBES), Vol. 2, No. 3, July 2015
20
Table 2. BPN Classification Accuracies
5. RESULTS AND CONCLUSION
The EEG motor imagery data has been collected from 45 healthy volunteers in the age group of
20-25. Out of the forty five subjects, for 5 subjects, the imagery trials lacked notable variations.
So, they are not considered for classification. Features associated with topographical variations of
EEG activity offer significant information regarding the origins of the dominant neural
community that contributes most to the recording.
First the EEG signal is filtered in the alpha range and the power spectrum is obtained for the 4
second imagery period. The mean alpha band power values are calculated for all the 19 electrode
positions. The values at other locations are interpolated by 4 nearest neighbour interpolation
method. Figure 11(a), shows the EEG map for left motor imagery and Figure 11(b), shows the
EEG map for right motor imagery. The colour scale range is depicted below it.
(a) EEG map for left hand imagery (b) EEG map for right hand imagery
Figure 11. EEG mean alpha power spatial distribution
From the colour map, it is evident that during left motor imagery, there is a significant decrease in
the mean alpha power seen in the right hemisphere (visible as blue shade). Also, during right
motor imagery, there is a reduction in mean alpha power seen in the left hemisphere (visible as
yellow shade).
11. International Journal of Biomedical Engineering and Science (IJBES), Vol. 2, No. 3, July 2015
21
A cluster plot is obtained for the features C3 and C4 alpha mean values vs. beta mean values are
plotted for right hand imagery. It is shown in Figure 12(a). As seen in the scatter plot, there is
variation between C3 and C4 values during right hand imagery. Figure 12(b) shows C3 and C4
alpha mean values vs. beta mean values for left hand imagery.
Figure 12. C3 vs. C4 features for motor imagery
The plot shown in Figure 13 depicts the C3 channel alpha and beta mean values during right hand
and left hand imagery. Since the two classes are not linearly separable, we go for neural network
classifier.
Figure 13. Right vs. Left imagery features from channel C
A BPN classifier with all 24 features taken at the same time (ERD features (8), Wavelet features
(8), Bipolar mean values (4), Bipolar Percentage values (4)) yielded a classification accuracy of
69% for right hand imagery and 73% for left hand imagery, which is almost the same for
classification of both the hands. The individual accuracies of the BPN taking few features at a
time are discussed earlier in Table 2.It is identified that a bipolar EEG signal with alpha and beta
mean as features yielded a better accuracy of 70% for right hand imagery and 87% for left hand
imagery.
In future, this work can be expanded by using more electrode locations and feature sets to
improve the classification accuracy.
ACKNOWLEDGEMENTS
The authors thank Life Sciences Research Board, Defense Research and Development
Organization for funding this project.
12. International Journal of Biomedical Engineering and Science (IJBES), Vol. 2, No. 3, July 2015
22
REFERENCES
[1] Athena Akrami, Soroosh Solhjoo, Ali Motie-Nasrabadi and Mohammed Reza Hashemi-Golpayegani,
“EEG-Based Mental Task Classification: Linear and Nonlinear Classification of Movement
Imagery”, Proceedings of the 27th Annual Conference of the IEEE Engineering in Medicine and
Biology Society, Shanghai, China, September 1-4, 2005.
[2] Xiu Zhang and Xingyu Wang, “Temporal and Frequency Feature Extraction with Canonical Variate
Analysis for Multi-Class Imagery Task”, Proceedings of the Control and Decision Conference, 2008.
[3] Pawel Herman, Girijesh Prasad, Thomas Martin McGinnity and Damien Coyle, “Comparative
Analysis of Spectral Approaches to Feature Extraction for EEG-Based Motor Imagery Classification”,
IEEE Transactions on Neural Systems and Rehabilitation Engineering, Vol.16, No.4, August 2008.
[4] Kavitha P. Thomas, Cuntai Guan, Chiew Tong Lau, A.P.Vinod and Kai Keng Ang, “ A New
Discriminative Common Spatial Pattern Method for Motor Imagery Brain-Computer Interfaces”,
IEEE Transactions on Biomedical Engineering ,Vol.56,No.11,November 2009.
[5] Febo Cincotti, Luigi Bianchi and Jose del R.Millan “Brain Computer Interface: the use of low
resolution Surface Laplacian and Linear Classifiers for the Recognition of Imagined Hand
Movements”, Proceedings of the 23rd Annual EMBS International Conference, 2001.
[6] G. Pfurstcheller, C. Neuper, “Motor imagery and direct Brain-Computer Communication,” Proc.
IEEE, vol. 89, pp. 1123-1134, 2001.
[7] A.Sivakami, S.Shenbaga Devi, “Analysis of EEG Signal During Motor Imagery of Hand Movement”,
Proceedings of the International Conference on Recent Trends in Computer Science and Engineering,
ICRTCSE, 2012
Authors Bio
1.A.Sivakami
A.Sivakami is a PG student in Medical Electronics, Anna University Chennai. Her areas of interest are
Biomedical Instrumentation, Artificial Intelligence, Bio signal Processing
2.S.Shenbaga Devi
Dr. S.Shenbaga Devi is Professor and Director of Center for Medical Electronics in Anna University,
Chennai. She has over 30 years of teaching experience. Her professional interests include Bio Signal
Processing, Medical Image Processing and BCI applications