Classifying Various EMG and EOG Artifacts in EEG Signals

EEG is the most popular potential non-invasive interface, mainly due to its fine temporal resolution, ease of use, portability and low set-up cost. However, it has some limitations. The main limitation is that EEG is frequently contaminated by various artifacts. In this paper, a novel approach to classify various electromyography and electrooculography artifacts in EEG signals is presented. EEG signals were acquired at the Department of Electrical and Electronics Engineering Karadeniz Technical University from three healthy human subjects in age groups between 28 and 30 years old and on two 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. Streszczenie. W artykule przedstawiono nową metodę analizy sygnałów w technice EEG pod względem klasyfikacji błędów zakłóceniowych w wynikach badań elektromiografii i elektrookulografii. Badanie przeprowadzone zostało na podstawie rzeczywistych wyników EEG. (Klasyfikacja zakłóceń sygnałów w technice EEG w badaniach EMG i EOG)