Thesis (Ph.D.) - Cairo University - Faculty of Computers and Information - Department of Computer Science

Brain Computer Interface (BCI) has become extremely popular in recent decades. It gained its significance from the intention of helping paralyzed people communicate with the external environment. The Motor Imagery (MI) based BCI system is one of the most popular systems in the BCI area. Motor imagination does not involve motor output from the human. This can be used to help disabledpeople to accomplish primitive tasks by themselves and be able to communicate with the surroundingenvironment.One of the major challenges facing MI-BCI systems is obtaining reliable classification accuracy of motor imagery (MI) mental tasks. Thus, this study aims to improve the classification accuracy of the MI-BCI systems and overcome some drawbacks in existing studies.Therefore, a new bio-inspired algorithm for feature selection and classifier optimization is proposed.To reach the stated objective, this thesisproposesanoptimumselection of time interval for each subject. The features are extracted from the EEG signal using the Common Spatial Pattern (CSP). Binary CSP is extended to a multi-class problem by utilizing the One-vs-One strategy. The new hybrid feature selection approach is obtained by combining the Attractor Metagene (AM) algorithm with the Bat optimization Algorithm (BA) to select the most discriminant CSP features and optimize the SVM parameters automatically

Issued also as CD