Date of Award

Spring 2019

Document Type

Thesis

Degree Name

Master of Science (MS) in Computer Science

Department

Computer Science

Abstract

The electroencephalogram (EEG) has proven to be useful in a wide variety of applications, including: diagnosis of mental disorders, psychological research, neurofeedback, and brain-computer interfacing. Most such applications of the EEG benefit from an ability to automatically detect when the subject is in a relaxed state. Recently, inexpensive and relatively easy to use EEG systems, with multiple electrodes, have become available at prices comparable to cellular phones or game machines. This project’s purpose is to investigate the feasibility of real-time classification of a subject's relaxation state using one such consumer-grade EEG system, the Emotiv Epoc. The subject's state is classified as relaxed or non-relaxed by monitoring the EEG signals over the occipital brain region and monitoring alpha wave activity. Said activity is characterized using an adaptive subject-specific threshold algorithm. Different variations of the threshold algorithm were investigated and their performance was compared using receiver operating characteristic graphs.

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Share

COinS