Term of Award

Spring 2016

Degree Name

Master of Science in Applied Engineering (M.S.A.E.)

Document Type and Release Option

Thesis (open access)


Department of Electrical Engineering

Committee Chair

Rami Haddad

Committee Member 1

Mohammad Ahad

Committee Member 2

Rocio Alba


Using Electroencephalography (EEG) to detect imaginary motions from brain waves, to interface human and computer is a very nascent and challenging field that started developing rapidly in the past few decades. The technique involves some electrodes attached on the scalp of the patient and the signals generated by the brain while the thought process of the patient is captured and recorded in a computer. This technique of human and computer interfacing is termed as Brain Computer Interface (BCI). Disability is a serious problem of our nation and hence BCI is extremely important in case of people who are incapable of communicating due to spinal cord injury. This technique uses the brain signals to make decisions, control objects and communicate with the world using brain integration with peripheral devices and systems. This requires some intelligence to classify these motions. Neural network have been used as a mean to classify motions, however, the accuracy of classification for certain motion was limited. The novelty of the proposed approach is in using a majority vote system for a network of artificial neural networks (ANNs) that is used to optimally classify imaginary motions performed by multiple subjects. Three kinds of imaginary motionswere classified which are imaginary left hand movement, imaginary right hand movement, and imagination of words starting with the same letter. Using an optimized set of electrodes, classification accuracywas optimized for the three users as a group and also individually. The optimization procedure was conducted based on the rank of the electrodes 2 according to their individual classification accuracy, and the eliminating electrodes with the lowest accuracies. The group optimization of 3 subjects altogether resulted in an electrode structure consisting of 15 electrodes with a relatively high classification accuracy of almost 80%. The individual optimization for each subject resulted in an electrode structure of 20 for subject 1 and subject 3 with classification accuracies of 63:63% and 84:33% respectively and single electrode structure for subject 2 with an accuracy of 94:01%. The overall average classification accuracy of all the users with the individual optimization of electrodes was as high as 82:32%.

Cover letter.docx (11 kB)
Changes made in revised version