Back to search

University of Ioannina

Ανάλυση ηλεκτροεγκεφαλογραφήματος σε συνδυασμό με δεδομένα από φορέσιμες συσκευές για τη μελέτη νευρολογικών διαταραχών και γνωσιακών καταστάσεων

Abstract

dc:description

Brain activity is monitored through the electroencephalogram (EEG), which provides important information of the brain and remains an indispensable tool for evaluating the abnormal brain activity. The EEG is the diagnostic tool for epilepsy and is used in the examination of dementia. Since the early 1970’s scientists have developed a variety of methodologies and signal processing techniques to detect the epileptic activity and study dementia. In the literature, different methodologies focus on the analysis of EEG characteristics to support the clinical practice and shed light on the mechanisms of several brain disorders, especially on epilepsy and Alzheimer’s Disease. This dissertation focuses on the development of innovative methods for EEG analysis, incorporating data mining and machine learning algorithms to extract knowledge for epilepsy and Alzheimer's disease (AD). In this context, a systematic review of signal processing techniques for seizure detection and AD analysis is presented. AD analysis incorporates both the detection of the disease and the correlation of cognitive decline in AD with quantitative EEG characteristics. The systematic review of the literature is performed according to a well-known and established technique (PRISMA methodology) for the collection, the evaluation and the analysis of the experimental studies. In detail, for AD analysis two methodologies are developed and presented that are applied in original clinical EEG recordings, collected from 24 subjects with mild and moderate AD. The classification methodology studies a feature vector of linear and non-linear EEG characteristics that are used to train a Random Forest classification model. The proposed classification methodology shows comparable and for some classification problems better results compared to other EEG-based AD studies. Also, a Mulitple Regression Analysis methodology is presented in order to correlate a cognitive metric that assesses the severity of AD with quantitative EEG characteristics. On the other hand, an innovative, robust methodology for automatic seizure detection is presented, which is applied to two public available EEG databases. According to the methodology spectral and statistical EEG features are extracted and are given as an input to a Random Forests classifier. Results of the methodology show better classification performance compared to other experimental studies that have applied their methodology on these two epileptic EEG databases.

Degree

thesis:*
Grantor dc:publisher
University of Ioannina
Year dc:date
2020

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Tzimourta, Aikaterini
  • Τζημούρτα, Αικατερίνη

Subjects

dc:subject × 21

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/48915

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Tzimourta, Aikaterini; Τζημούρτα, Αικατερίνη. Ανάλυση ηλεκτροεγκεφαλογραφήματος σε συνδυασμό με δεδομένα από φορέσιμες συσκευές για τη μελέτη νευρολογικών διαταραχών και γνωσιακών καταστάσεων. University of Ioannina, 2020. http://hdl.handle.net/10442/hedi/48915