Toronto Metropolitan University
Browse
0647ea1f870fa1d987e639b21c634b34.pdf (2.77 MB)

Classifying Severity of Depression and Anxiety by Analyzing Electroencephalography (EEG) Signals for Neurophysiological Biomarkers

Download (2.77 MB)
thesis
posted on 2023-12-18, 16:11 authored by Laura Minkowski

Biomarkers detected in neurophysiological signals can be analyzed to determine indicators of disorders. Electroencephalography (EEG) detects neural activity in the brain and the signals can be analyzed to diagnose stress and mood disorders. The objective is to analyze EEG signals to identify and delineate the severity of depression and/or anxiety validated by the results of psychological test scores. Signals were analyzed from a public database of 119 participants aged 18 to 24 with 45 individuals having moderate to severe anxiety and/or depression and the remaining 74 people having minimal or none. Using extracted signal features, individual variations were compared during a testing protocol for both groups, affected and unaffected. Similarities, and asymmetry, were numerically and visually examined between the left and right brain hemispheres as well as the specific channels. In addition, machine learning classification was performed to predict the class based on the input data. The results demonstrate indications of physiological differences between participants, indicating a likely presence or absence of a mood disorder. Understanding the complexities of how mood and anxiety disorders, including its comorbidities, are physiologically manifested is critical for accurate and objective diagnosis.

History

Language

eng

Degree

  • Master of Applied Science

Program

  • Biomedical Engineering

Granting Institution

Ryerson University

LAC Thesis Type

  • Thesis

Thesis Advisor

Kristiina Mai

Year

2021

Usage metrics

    Biomedical Engineering (Theses)

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC