چكيده به لاتين
Blind Source Separation (BSS) techniques, such as Second-Order Blind Identification (SOBI) and Independent Component Analysis (ICA), are based on a spread spectrum of unsupervised learning algorithms. They have potential applications in many areas of applied sciences such as biomedical engineering, image processing, speech enhancement, geophysical data processing, and wireless communication. The BSS and ICA refer to the problem of recovering statistically independent signals from a linear mixture. The term “blind” refers to the fact that there is no information about the mixing process or about the source signals.
A Brain Computer Interface (BCI) system can be defined as a “communication and control channel which does not depend on the brain’s normal output channels of peripheral nerves and muscles”. In BCI systems, Electroencephalography (EEG) signals help to restore sensory and motor functions in patients who have severe motor disabilities. Therefore, there is an ever-increasing need for developing automatic classification methods to evaluate and diagnose neurological disorders. A big challenge is for BCI systems to correctly and efficiently identify different EEG signals of different Motor Imagery (MI). The BCIs use appropriate classification algorithms to assist motor disabled patients in communication. The EEG signals are highly susceptible to artifacts. Contamination of the EEG signals by artifacts affects the interest signal and makes the analysis difficult. Thus, automatic artifact removal from the EEG signals is important to ensure a correct classification.
In this research, the BSS methods have been used to remove artifacts in a Motor Imagery-based Brain Computer Interface (MI-BCI) system. The results indicate that the artifact removal causes a decrease of 6% of the average classification performance from 63% to 57%.
Keywords: Blind Source Separation, Artifact Removal, Brain Computer Interface, EEG, Motor Imagery.
