Temel Kayikcioglu

A polynomial fitting and k-NN based approach for improving classification of motor imagery BCI data

Speed and accuracy in classification of electroencephalographic (EEG) signals are key issues in brain computer interface (BCI) technology. In this paper, we propose a fast and accurate classification method for cursor movement imagery EEG data. A two-dimensional feature vector is obtained from coefficients of the second order polynomial applied to signals of only one channel. Then, the features are classified by using the k-nearest neighbor (k-NN) algorithm. We obtained significant improvement for the speed and accuracy of the classification for data set Ia, which is a typical representative of one kind of BCI competition 2003 data. Compared with the Multiple Layer Perceptron (MLP) and the Support Vector Machine (SVM) algorithms, the k-NN algorithm not only provides better classification accuracy but also needs less training and testing times.

An improved motion-compensated restoration method for damaged color motion picture films

In this paper, we propose an improved motion-compensated restoration method for color motion picture films deteriorated due to flashing blotches. The method consists of an improved multiresolution block matching with log-D search, a rank ordered differences-based blotch detection and 3D vector median filtering for interpolation of missing data, and utilizes five consecutive frames. Performance of the method is tested on artificially corrupted image sequences and real motion picture films, and is compared to that of the three-frame-based method which involves similar algorithms except improved motion estimation and blotch detection. The results show that the method efficiently works even in severely blotched and motion regions in image sequences.

Decision tree structure based classification of EEG signals recorded during two dimensional cursor movement imagery

BACKGROUND Input signals of an EEG based brain computer interface (BCI) system are naturally non-stationary, have poor signal to noise ratio, depend on physical or mental tasks and are contaminated with various artifacts such as external electromagnetic waves, electromyogram and electrooculogram. All these disadvantages have motivated researchers to substantially improve speed and accuracy of all components of the communication system between brain and a BCI output device. NEW METHOD In this study, a fast and accurate decision tree structure based classification method was proposed for classifying EEG data to up/down/right/left computer cursor movement imagery EEG data. The data sets were acquired from three healthy human subjects in age group of between 24 and 29 years old in two sessions on different days. RESULTS The proposed decision tree structure based method was successfully applied to the present data sets and achieved 55.92%, 57.90% and 82.24% classification accuracy rate on the test data of three subjects. COMPARISON WITH EXISTING METHOD(S) The results indicated that the proposed method provided 12.25% improvement over the best results of the most closely related studies although the EEG signals were collected on two different sessions with about 1 week interval. CONCLUSIONS The proposed method required only a training set of the subject and automatically generated specific DTS for each new subject by determining the most appropriate feature set and classifier for each node. Additionally, with further developments of feature extraction and/or classification algorithms, any existing node can be easily replaced with new one without breaking the whole DTS. This attribute makes the proposed method flexible.

Fast and accurate PLS-based classification of EEG sleep using single channel data

Fast classification of sleep and wake stages using a single EEG channel is proposed.The dataset was provided by Physionet.Speed and accuracy of PLS were compared with those of k-NN, Bayes and LDC classifiers.Results indicated that the Pz-Cz channel had better accuracy than the Fpz-Cz channel.We achieved 91% classification accuracy by selecting PLS as the classifier. Since speed of classification is important to real-time applications, this study proposed fast classification of sleep and wake stages using a single electroencephalograph (EEG) channel. Changes in the sleep and wake stages are accompanied by changes in the frequency spectrum of the EEG signals; so, the features extracted from the 5-s epoch of the EEG using auto-regressive (AR) coefficients were used to represent EEG signals of different sleep and wake stages. The proposed fast classification method was based on partial least squares regression (PLS), which was used to classify these features by finding an optimum beta using K-fold cross validation. The Physionet database was used to confirm accuracy and speed of the proposed classification system. This system could be used in real-time implementations because of its high classification rate, speed and capability to be implemented on hardware owing to be very comfortable. Finally, results of the PLS were compared with those of other classifiers such as k-nearest neighborhood (k-NN), linear discriminant classifier (LDC) and Bayes. We achieved 91% classification accuracy by selecting PLS as the classifier. These comparisons revealed that the proposed algorithm could recognize an emergency situation in less than 1s with high accuracy.

A surface-based method for detection of coronary vessel boundaries in poor quality X-ray angiogram images

In this paper, we propose a surface-based method for simultaneous detection of left and right coronary borders that is suitable for analysis of poor quality X-ray angiogram images. Coronary artery is modelled with a 3D generalized cylinder (GC) with elliptic cross-sections. Based on this model, we developed a 2D surface function for the projection intensity distribution of a vessel part. The parameters associated with vessel edges are estimated from this model. The model takes into account local background intensity, noise and blurring. In simulation and real experiments over a range of imaging conditions, the proposed method consistently produced lower estimation error and variability in detecting edges than our previously reported 1D profile-based method. The improvement is most significant especially for noisy and low-contrast angiograms.

Reconstructing coronary arterial segments from three projection boundaries

Abstract This paper describes a method for the three-dimensional (3D) reconstruction of coronary arterial segments from three projection boundaries. The method is based on the 3D generalized cylinder (GC) model with elliptic cross-sections and mainly consists of a model-based estimation of the vessel boundaries, precise computation of the ellipse parameters of arterial cross-sections from the estimated boundaries, and obtaining the 3D structures by linking consecutive cross-sections. Synthetic and real experiments showed that the method is robust to degradation sources and consequently superior to the two-view method because it uses boundary instead of gray-level information.

Reconstructing ellipsoids from three projection contours

Abstract In this paper, a method for reconstructing ellipsoids from three projection contours is described. Line integral projection of an ellipsoid with uniform density is developed to obtain the equation of the projection contour for any projection view. The reconstruction method starts from the estimation of parameters of a projection contour that has elliptic shape. Then an error function utilizing three sets of ellipse parameters estimated in the previous step is minimized to estimate ellipsoid parameters. Simulation and real experiments demonstrate the validity, usefulness and accuracy of the proposed method.

Classification of EEG signal during gaze on the different rotating vanes

In this paper a novel brain-computer interface based on the gaze on rotating vane using five channels of EEG signal is proposed. Classification of EEG signal is done in three sessions: 1-when vane rotates fast and slow in an anti-clockwise manner, 2-when vane rotates slow in a clockwise and rotates fast in an anti-clockwise manner, 3-when vane rotates slow in a clockwise and rotates slow in an anti-clockwise manner. The signals were obtained from seven healthy human subjects in age groups between 25 and 32 years old. Discrete Wavelet transform (DWT) were used to extracted feature vectors. The features classified by two classifiers, k-nearest neighbor (k-NN) algorithm and Linear Discriminant Classifier (LDC). Our results demonstrated that LDC was more accurate compared to k-NN. Analysis was carried out using MATLAB Software.

Performance evaluation of five classification algorithms in low-dimensional feature vectors extracted from EEG signals

There are lots of classification and feature extraction algorithms in the field of brain computer interface. It is significant to use optimal classification algorithm and fewer features to implement a fast and accurate brain computer interface system. In this paper, we evaluate the performances of five classical classifiers in different aspects including classification accuracy, sensitivity, specificity, Kappa and computational time in low-dimensional feature vectors extracted from EEG signals. The experiments show that naive Bayes is the most appropriate classifier for low dimensional feature vectors compared to k-nearest neighbor, support vector machine, linear discriminant analysis and decision tree classifiers.

Classification of various facial movement artifacts in EEG signals

In this paper, a novel approach to classify various facial movement artifacts in EEG signals is presented. EEG signals were obtained in EEG Laboratory from three healthy human subjects in age groups between 28 and 30 years old and on different days. Extracted feature vectors based on root mean square, polynomial fitting and Hjorth descriptors were classified by k-nearest neighbor algorithm. The proposed method was successfully applied to the data sets and achieved an average classification rate of 94% on the test data.