Detecting Changes In Neurological Status Using Electroencephalography Signals

Amirsalar Mansouri, University of Nebraska-Lincoln

Document Type Article

A DISSERTATION Presented to the Faculty of The Graduate College at the University of Nebraska In Partial Fulfillment of Requirements For the Degree of Doctor of Philosophy, Major: Electrical Engineering, Under the Supervision of Professor Khalid Sayood. Lincoln, Nebraska: December, 2021

Copyright © 2021 Amirsalar Mansouri

Abstract

Advisor: Khalid Sayood

Different neuroimaging techniques can monitor the brain activity. Electroencephalogram (EEG) is the most accessible neuroimaging technique; it is not only cost-efficient, but also its initial setup does not need extensive training or expertise. Employing EEG systems can facilitate monitoring the progress of neurophysiological disorders or detecting their onsets. In this dissertation, we introduce a comprehensive method to cover a variety of brain disorders such as detecting seizures, which affects more than 70 million people around the world, and reporting potential concussive injuries; about 3 million people experience one concussive incident every year in the US. Besides detecting and diagnosing neurological disorders, EEG signals can also be used in developing applications to increase the quality of life of individuals whose ability to control their movements or communicate is hindered, as observed in paralyzed or amyotrophic lateral sclerosis (ALS) patients. Brain–computer interface (BCI) systems convey commands from the brain to an external machine instead of the muscle, which is not connected to the brain due to neurodegenerative diseases that target the muscle neurons.

EEG applications are growing rapidly; to increase the feasibility of the EEG applications, a universal approach is needed to aggregate different applications. One of the challenges of previously automated diagnostic systems is enhancing the system performance, where most of the algorithms are patient (subject) -specific and need a prior sample of abnormal EEG signals for further detection or analysis of the same category of disorders.

It is therefore essential to develop a system for monitoring the healthy brain, like systems used for a regular check-up of the body. These systems need to be cost-efficient, with minimum calibration setup, and accessible for the majority of the population. Here, we introduce a comprehensive non-patient-specific EEG technique which can be used in multiple applications. In particular we use it to report concussive injuries and seizure attacks to the brain. We also show how this approach can be used to perform as a BCI system to help people increase their quality of life.

Advisor: Khalid Sayood