D. N. Tibarewala

Performance analysis of LDA, QDA and KNN algorithms in left-right limb movement classification from EEG data

Brain Computer Interface (BCI) improve the lifestyle of the normal people by enhancing their performance levels. It also provides a way of communication for the disabled people with their surrounding who are otherwise unable to physically communicate. BCI can be used to control computers, robots, prosthetic devices and other assistive technologies for rehabilitation. The dataset used for this study has been obtained from the BCI competition II 2003 databank provided by the University of Technology, Graz. After pre-processing of the signals from their electrodes (C3 & C4), the wavelet coefficients, Power Spectral Density of the alpha and the central beta band and the average power of the respective bands have been employed as features for classification. In one of the approaches we fed all the extracted features individually and in the other approach we considered all features together and submitted them to LDA, QDA and KNN algorithms distinctly to classify left and right limb movement. The aim of this study is to analyze the performance of linear discriminant analysis (LDA), quadratic discriminant analysis (QDA) and K-nearest neighbor (KNN) algorithms in differentiating the raw EEG data obtained, into their associative movement, namely, left-right movement. Also the importance of the feature vectors selected is highlighted in this study. The total set to feature vector comprising all the features (i.e., wavelet coefficients, PSD and average band power estimate) performed better with the classifiers without much deviation in the classification accuracy, i.e., 80%, 80% and 75.71% with LDA, QDA and KNN respectively. Wavelet coefficients performed best with QDA classifier with an accuracy of 80%. PSD vector resulted in superior performance of 81.43% with both QDA and KNN. Average band power estimate vector showed highest accuracy of 84.29% with KNN algorithm. Our approach presented in this paper is quite simple, easy to execute and is validated robustly with a large dataset.

Performance analysis of left/right hand movement classification from EEG signal by intelligent algorithms

Brain Computer interfaces (BCI) has immense potentials to improve human lifestyle including that of the disabled. BCI has possible applications in the next generation human-computer, human-robot and prosthetic/assistive devices for rehabilitation. The dataset used for this study has been obtained from the BCI competition-II 2003 databank provided by the University of Technology, Graz. After pre-processing of the signals from their electrodes (C3 & C4), the wavelet coefficients, Power Spectral Density of the alpha and the central beta band and the average power of the respective bands have been employed as features for classification. This paper presents a comparative study of different classification methods including linear discriminant analysis (LDA), Quadratic discriminant analysis (QDA), k-nearest neighbor (KNN) algorithm, linear support vector machine (SVM), radial basis function (RBF) SVM and naive Bayesian classifiers algorithms in differentiating the raw EEG data obtained, into their associative left/right hand movements. Performance of left/right hand classification is studied using both original features and reduced features. The feature reduction here has been performed using Principal component Analysis (PCA). It is as observed that RBF kernelised SVM classifier indicates the highest performance accuracy of 82.14% with both original and reduced feature set. However, experimental results further envisage that all the other classification techniques provide better classification accuracy for reduced data set in comparison to the original data. It is also noted that the KNN classifier improves the classification accuracy by 5% when reduced features are used instead of the original.

Automatic feature selection of motor imagery EEG signals using differential evolution and learning automata

Brain–computer interfacing (BCI) has been the most researched technology in neuroprosthesis in the last two decades. Feature extractors and classifiers play an important role in BCI research for the generation of suitable control signals to drive an assistive device. Due to the high dimensionality of feature vectors in practical BCI systems, implantation of efficient feature selection algorithms has been an integral area of research in the past decade. This article proposes an efficient feature selection technique, realized by means of an evolutionary algorithm, which attempts to overcome some of the shortcomings of several state-of-the-art approaches in this field. The outlined scheme produces a subset of salient features which improves the classification accuracy while maintaining a trade-off with the computational speed of the complete scheme. For this purpose, an efficient memetic algorithm has also been proposed for the optimization purpose. Extensive experimental validations have been conducted on two real-world datasets to establish the efficacy of our approach. We have compared our approach to existing algorithms and have established the superiority of our algorithm to the rest.

Object-shape recognition from tactile images using a feed-forward neural network

The sense of touch is an extremely important sensory system in the human body which helps to understand object shape, texture, hardness in the world around us. Incorporating artificial haptic sensory systems in rehabilitative aids and in various other human computer interfaces is a thrust area of research presently. This paper presents a novel approach of shape recognition and classification from the tactile pressure images by touching the surface of various real life objects. Here four objects (viz. a planar surface, object with one edge, a cuboid i.e. object with two edges and a cylindrical object) are used for shape recognition. The obtained tactile pressure images of the object surfaces are subjected to segmentation, edge detection and a mapping procedure to finally reconstruct the particular object shapes. The reconstructed images are used as features. The processed tactile pressure images are classified with feed- forward neural network (FFNN) using extracted features. The classifier performance is tested with different signal-to-noise (SNR) ratios. Is is observed that classifier accuracy decreases with decrease in SNR, but at SNR value 6 i.e. when the noise power is one sixth of the signal power, the mean classification accuracy of the classifier is 88%. This shows the robustness of feed-forward neural network in the classification purpose. The performance of FFNN is compared with four classifiers (Linear Discriminant Analysis, Linear Support vector machine, Radial Basis Function SVM, k-Nearest Neighbor). FFNN performed best acquiring first rank with a average classification accuracy of 94.0%.

Single channel electrooculogram(EOG) based interface for mobility aid

Human computer interfacing (HCI) technology has emerged as a new pathway towards the improvement of different rehabilitative aids. In this paper, new approach to control the motorized human computer interface using electrooculogram (EOG) is proposed. A mobility interface controlled by eye movements has been developed to help the disabled individuals with motor impairment who cannot even speak. Electrooculogram(EOG) is the potential generated in due the movement of the eyeballs and can be acquired from the surrounding region of eye socket. The signal is easy to acquire noninvasively and has a simple pattern. A low cost data acquisition system for EOG is designed. Horizontal electrooculographic signal is recorded by placing electrodes at the outer region of the orbit of eyes, and a reference electrode at neck. Using different combinations of eye movements in right and left direction a simple control strategy has been developed to drive motors. Control signals have been first generated using 8051 microcontroller. To meet the problems occurred while using 8051, ATMEGA microcontroller has been adapted. Directional movements of a small prototype of mobility aid (a toy car) with DC motors in right, left and forward is controlled and start and stop of movement is also implemented with ATMEGA. These control signals can be further used to command rehabilitative assistive device with eye movement sequences.

A Two-fold classification for composite decision about localized arm movement from EEG by SVM and QDA techniques

Disabled people now expect better quality of life with the development of brain computer interfaces (BCIs) and neuroprosthetics. EEG (electroencephalograph) based BCI research for robot arm control mainly concentrates on distinguishing the left/right arm movement. But for controlling artificial arm in real life scenario with greater degrees of freedom, it is essential to classify the left/right arm movement further into different joint movements. In this paper we have classified the raw EEG signal for left and right hand movement, followed by further classification of each hand movement into elbow, finger and shoulder movements. From the two electrodes of interest, namely, C3 and C4, wavelet coefficients, power spectral density (PSD) estimates for the alpha and beta bands and their corresponding powers were selected as the features for this study. These features are further fed into the quadratic discriminant analysis (QDA), linear support vector machine (LSVM) and radial basis function kernelized support vector machine (RSVM) to classify into the intended classes. For left-right hand movement, the maximum classification accuracy of 87.50% is obtained using wavelet coefficient for RSVM classifier. For the multi-class classification, i.e., Finger-Elbow-Shoulder classification the maximum classification accuracy of 80.11% for elbow, 93.26% for finger and 81.12% for shoulder is obtained using the features obtained from power spectral density for RSVM classifier. The results presented in this paper indicates that elbow-finger-shoulder movement can be successfully classified using the given set of features.

A differential evolution based energy trajectory planner for artificial limb control using motor imagery EEG signal

Abstract Current research on neuro-prosthetics is aimed at designing several computational models and techniques to trigger the neuro-motor rehabilitative aids. Researchers are taking keen interest to accurately classify the stimulated electroencephalography (EEG) signals to interpret motor imagery tasks. In this paper we aim to classify the finger-, elbow- and shoulder-classification along with left- and right-hand classification to move a simulated robot arm in 3D space towards a target of known location. The contribution of the paper lies in the design of an energy optimal trajectory planner, based on differential evolution, which would decide the optimal path for the robot arm to move towards the target based on the classifier output. Each different set of movements consists of a trajectory planner which is activated by the classifier output. The energy distribution of wavelet coefficients of the incoming EEG signals is used as features to be used as inputs in a naive Bayesian classifier to discriminate among the different mental tasks. The average training classification accuracy obtained is 76.88% and the success rate of the simulated robot arm reaching the target is 85%.

Artificial Bee Colony Based Feature Selection for Motor Imagery EEG Data

Brain-computer Interface (BCI) has widespread use in Neuro-rehabilitation engineering. Electroencephalograph (EEG) based BCI research aims to decode the various movement related data generated from the motor areas of the brain. One of the issues in BCI research is the presence of redundant data in the features of a given dataset, which not only increases the dimensions but also reduces the accuracy of the classifiers. In this paper, we aim to reduce the redundant features of a dataset to improve the accuracy of classification. For this, we have employed Artificial Bee Colony (ABC) cluster algorithm to reduce the features and have acquired their corresponding accuracy. It is seen that for a reduced features of 200, the highest accuracy of 64.29 %. The results in this paper validate our claim.

Object shape recognition from tactile images using regional descriptors

This paper presents a novel approach of shape recognition from the tactile images by touching the surface of various real life objects. Here four geometric shaped objects (viz. a planar surface, object with one edge, a cubical object i.e. object with two edges and a cylindrical object) are used for shape recognition. The high pressure regions denoting surface edges have been segmented out via multilevel thresholding. These high pressure regions hereby obtained were unique to different object classes. Some regional descriptors have been used to uniquely describe the high pressure regions. These regional descriptors have been employed as the features needed for the classification purpose. Linear Support Vector Machine (LSVM) classifier is used for object shape classification. In noise free environment the classifier gives an average accuracy of 92.6%. Some statistical tests have been performed to prove the efficacy of the classification process. The classifier performance is also tested in noisy environment with different signal-to-noise (SNR) ratios.

An interval type-2 fuzzy approach for real-time EEG-based control of wrist and finger movement

Abstract Feature extraction and automatic classification of mental states is an interesting and open area of research in the field of brain–computer interfacing (BCI). A well-trained classifier would allow the BCI system to control an external assistive device in real world problems. Sometimes, standard existing classifiers fail to generalize the components of a non-stationary signal, like Electroencephalography (EEG) which may pose one or more problems during real-time usage of the BCI system. In this paper, we aim to tackle this issue by designing an interval type-2 fuzzy classifier which deals with the uncertainties of the EEG signal over various sessions. Our designed classifier is used to decode various movements concerning the wrist (extension and flexion) and finger (opening and closing of a fist). For this purpose, we have employed extreme energy ratio (EER) to construct the feature vector. The average classification accuracy achieved during offline training and online testing over eight subjects are 86.45% and 78.44%, respectively. On comparison with other related works, it is shown that our designed IT2FS classifier presents a better performance.