Date of Award

2022

Document Type

Masters Thesis

Degree Name

M.S.

Organizational Unit

Daniel Felix Ritchie School of Engineering and Computer Science, Electrical and Computer Engineering

First Advisor

Mohammad H. Mahoor

Second Advisor

Daniel Paredes

Third Advisor

Yun-bo Yi

Fourth Advisor

Mohammad Matin

Keywords

Capillary electrophoresis, Machine learning, Parkinson's disease

Abstract

Parkinson’s disease (PD) is a neurodegenerative movement disorder that progresses gradually over time. The onset of symptoms in people who are suffering from PD can vary from case to case, and it depends on the progression of the disease in each patient. The PD symptoms gradually develop and exacerbate the patient’s movements throughout time. An early diagnosis of PD could improve the outcomes of treatments and could potentially delay the progression of this disorder and that makes discovering a new diagnostic method valuable. In this study, I investigate the feasibility of using a machine learning (ML) approach to classify PD patients from a healthy group. A set of plasma samples were collected from both PD patients and healthy people. Then the data were processed in a custom-designed capillary zone electrophoresis (CZE) system. CZE allows us to study metabolomics, which is the chemical processes and comprehensive analysis of small molecules in regard to metabolism within an organism such as a cell or body fluids like plasma. Metabolic profiling can demonstrate changes in the composition and therefore it can potentially reflect an underlying condition or may provide valuable information for disease diagnosis. After preprocessing the generated electropherograms or output data from the CZE system, I developed and applied various machine learning algorithms to distinguish the PD samples from the healthy samples based on the biomarkers extracted using the CZE system. Our experimental results demonstrate that there are clearly different features in two groups of samples. Therefore, it was possible to reach the classification accuracy of 94% in a very small set of samples.

Publication Statement

Copyright is held by the author. User is responsible for all copyright compliance.

Rights Holder

Soroush Dehghan

Provenance

Received from ProQuest

File Format

application/pdf

Language

en

File Size

59 pgs

Discipline

Computer engineering



Share

COinS