Jasmin Kevric

Comparison of signal decomposition methods in classification of EEG signals for motor-imagery BCI system

Abstract In this study, three popular signal processing techniques (Empirical Mode Decomposition, Discrete Wavelet Transform, and Wavelet Packet Decomposition) were investigated for the decomposition of Electroencephalography (EEG) Signals in Brain Computer Interface (BCI) system for a classification task. Publicly available BCI competition III dataset IVa, a multichannel 2-class motor-imagery dataset, was used for this purpose. Multiscale Principal Component Analysis method was applied for the purpose of noise removal. In addition, different sets of features were formed to examine the effect of a particular group of features. The parameter selection process for signal decomposition methods was thoroughly explained as well. Our results show that the combination of Multiscale Principal Component Analysis de-noising and higher order statistics features extracted from wavelet packet decomposition sub-bands resulted in highest average classification accuracy of 92.8%. Our study is one among very few that provides a comprehensive comparison between signal decomposition methods in combination with higher order statistics in classification of BCI signals. In addition, we stressed the importance of higher frequency ranges in improving the classification task for EEG signals in Brain Computer Interface Systems. Obtained results indicate that the proposed model has the potential to obtain a reliable classification of motor imagery EEG signals, and can thus be used as a practical system for controlling a wheelchair. It can also further enhance the current rehabilitation therapies where appropriate feedback is delivered once the individual executes the correct movement. In that way, motor rehabilitation outcomes may improve over time.

Performance evaluation of empirical mode decomposition, discrete wavelet transform, and wavelet packed decomposition for automated epileptic seizure detection and prediction

Abstract This study proposes a new model which is fully specified for automated seizure onset detection and seizure onset prediction based on electroencephalography (EEG) measurements. We processed two archetypal EEG databases, Freiburg (intracranial EEG) and CHB-MIT (scalp EEG), to find if our model could outperform the state-of-the art models. Four key components define our model: (1) multiscale principal component analysis for EEG de-noising, (2) EEG signal decomposition using either empirical mode decomposition, discrete wavelet transform or wavelet packet decomposition, (3) statistical measures to extract relevant features, (4) machine learning algorithms. Our model achieved overall accuracy of 100% in ictal vs. inter-ictal EEG for both databases. In seizure onset prediction, it could discriminate between inter-ictal, pre-ictal, and ictal EEG with the accuracy of 99.77%, and between inter-ictal and pre-ictal EEG states with the accuracy of 99.70%. The proposed model is general and should prove applicable to other classification tasks including detection and prediction regarding bio-signals such as EMG and ECG.

Epileptic seizure detection using hybrid machine learning methods

The aim of this study is to establish a hybrid model for epileptic seizure detection with genetic algorithm (GA) and particle swarm optimization (PSO) to determine the optimum parameters of support vector machines (SVMs) for classification of EEG data. SVMs are one of the robust machine learning techniques and have been extensively used in many application areas. The kernel parameter’s setting for SVMs in training process effects the classification accuracy. We used GA- and PSO-based approach to optimize the SVM parameters. Compared to the GA algorithm, the PSO-based approach significantly improves the classification accuracy. It is shown that the proposed Hybrid SVM can reach a classification accuracy of up to 99.38% for the EEG datasets. Hence, the proposed Hybrid SVM is an efficient tool for neuroscientists to detect epileptic seizure in EEG.

Cloud computing-based parallel genetic algorithm for gene selection in cancer classification

Cancer classification is one of the main steps during patient healing process. This fact enforces modern clinical researchers to use advanced bioinformatics methods for cancer classification. Cancer classification is usually performed using gene expression data gained in microarray experiment and advanced machine learning methods. Microarray experiment generates huge amount of data, and its processing via machine learning methods represents a big challenge. In this study, two-step classification paradigm which merges genetic algorithm feature selection and machine learning classifiers is utilized. Genetic algorithm is built in MapReduce programming spirit which makes this algorithm highly scalable for Hadoop cluster. In order to improve the performance of the proposed algorithm, it is extended into a parallel algorithm which process on microarray data in distributed manner using the Hadoop MapReduce framework. In this paper, the algorithm was tested on eleven GEMS data sets (9 tumors, 11 tumors, 14 tumors, brain tumor 1, lung cancer, brain tumor 2, leukemia 1, DLBCL, leukemia 2, SRBCT, and prostate tumor) and its accuracy reached 100% for less than 25 selected features. The proposed cloud computing-based MapReduce parallel genetic algorithm performed well on gene expression data. In addition, the scalability of the suggested algorithm is unlimited because of underlying Hadoop MapReduce platform. The presented results indicate that the proposed method can be effectively implemented for real-world microarray data in the cloud environment. In addition, the Hadoop MapReduce framework demonstrates substantial decrease in the computation time.

Ovary Cancer Detection using Decision Tree Classifiers based on Historical Dataof Ovary Cancer Patients

This research implements decision tree classifiers and artificial neural network to predict whether the patient will live with ovary cancer or not. Dataset was obtained from Danish Cancer Register and contains five Input parameters. Dataset contains some missing values and a noticeable improvement in accuracy was detected after removing them. Three features of the original dataset were shown to be the most significant: Mobility of the cancer, Surface of the cancer, and the Consistency of the cancer. The addition of the other two features (Size of the cancer and age of the patient) did not improve the results significantly. It was noticed that the patients with a cystic, but fixed and even cancer have always died from the ovary cancer. In contrast, the patients with uneven, but fixed and solid cancer have always survived the cancer. It is recommended to include more information about either the cancer or the patient to increase the chance of predicting the output of such patients.

Diagnosis of Chronic Kidney Disease by Using Random Forest

Chronic kidney disease (CKD) is a global public health problem, affecting approximately 10% of the population worldwide. Yet, there is little direct evidence on how CKD can be diagnosed in a system ...

Cloud-based mobile platform for EEG signal analysis

It is estimated that there are millions of people with epilepsy around the world. Seizure detection and prediction systems are built to improve lifestyle of patients. Closed-loop systems are designed to predict and detect seizures and inform patient and caretakers. Ideally, wireless technologies are used in order not to interfere with patients life. We build a prototype for closed-loop systems consisting of Mind Wave EEG capturing device and Android application communicating via Bluetooth. The application can store signals locally or send them to cloud and then process them for different applications such as BCI, Neurofeedback, epileptic seizure prediction, etc.

Wavelet and Teager Energy Operator (TEO) for Heart Sound Processing and Identification

The heart sound signal (heartbeat) recorded from normal subjects usually contains two separate tones, S1 and S2. In addition, an auscultation technique used to provide physicians with accurate and objective interpretation of heart sounds can be used to detect four sounds, namely, S1, S2, S3, and S4, during the heart cycle. In this project, we propose a technique to detect these four heartbeats effectively using the combination of multi-scale wavelet transform and Teager Energy Operator to increase the precision of the detection process. The purpose of combining TEO with Wavelets is to observe how different details obtained from the Wavelet Transform influence the Teager Operator success in detecting S1, S2, S3, and S4 heart sounds. The effectiveness of the proposed approach is evaluated in experiments related to different cardiac conditions, achieving 88 % accuracy for localization of S1 and S2, and 86 % accuracy for S3 and/or S4.

An extended model of a level and flow control system

FESTO Compact Workstation is a well known didactic tool in process control. This paper aims at providing an improved transfer function model of this systems level and flow control loops. This higher order model is compared to existing first order system approximations of the level control loop in various input-output scenarios to verify its applicability and superiority. Results are obtained using MATLAB System Identification Toolbox after data acquisition in LabVIEW. MATLAB Simulink is used for cascade PI and single loop PI experiments to show the improvement cascade control on the new model brings. Together with the practical value the results have, the procedure conducted here can serve as a primer and a tutorial for system identification class using this or similar apparatus.

Brainiac’s Arm—Robotic Arm Controlled by Human Brain

This paper shows electroencephalograph (EEG) controlled robotic arm based on Brain–computer interfaces (BCI). BCIs are systems that enable bypassing conventional methods of communication (i.e., muscles and thoughts) and provide direct communication and control between the human brain and physical devices using the power of the human brain. The main goal of the project work is to develop a robotic arm that can assist the disabled people in their daily life and by it make their work independent on others.