A master thesis from Aalborg University
Dependency Analysis of Electroencephalography Signals: A Theoretical and Data Driven Approach to Quantifying Dependencies in Multivariate Signals
[Afhængighedsanalyse af Electroencephalografisignaler: En Teoretisk og Datadrevet Tilgang til Kvantificering af Afhængigheder i Multivariate Signaler]
Term
4. semester
Education
Publication year
2023
Submitted on
2023-06-02
Pages
145 pages
Abstract
I dette projekt præsenteres informationsteori, grafteori samt omegakompleksitet med henblik på at udføre en analyse af afhængigheder i EEG signaler. Der udføres en analyse af omegakompleksiteten, og der introduceres en generaliseret udgave af omegakompleksitet for at forbedre nogle af de præsenterede mangler. Metoderne testes på koblede Rösslersystemer og multivariate autoregressive processer, da disse har vist sig i nogen grad at have EEG-lignende opførsel. Først analyseres et EEG-datasæt, der er indsamlet fra en person udsat for et høj- og lav-SNR-miljø, dog observeres der ingen signifikante ændringer mellem de to. Herefter anvendes de introducerede metoder på et iEEG-datasæt fra en person med epilepsi, hvilket viser signifikante ændringer i afhængigheder i EEG-signalerne før og under et anfald. Dermed er metoderne i nogen grad i stand til at fange ændringer i afhængigheder i EEG-signaler.
In this project, information theory, graph theory, as well as the omega complexity is presented in order to analyse dependencies in EEG signals. An analysis of the omega complexity is performed, and a generalised omega complexity is introduced to combat some of the presented deficiencies. The methods are tested on coupled Rössler systems and multivariate autoregressive processes as these have proven to be comparable with EEG signals in their behaviour. Initially an EEG data set obtained from a subject exposed to a high and low SNR environment is analysed, although no significant changes between the two are found. Next, the presented methods are applied to an iEEG data set on a subject with epilepsy, resulting in significant changes between dependencies in the EEG signals prior to a seizure and during a seizure. Hence the methods introduced are to some degree able to capture changes in dependencies in EEG signals.
Keywords
Documents
Colophon: This page is part of the AAU Student Projects portal, which is run by Aalborg University. Here, you can find and download publicly available bachelor's theses and master's projects from across the university dating from 2008 onwards. Student projects from before 2008 are available in printed form at Aalborg University Library.
If you have any questions about AAU Student Projects or the research registration, dissemination and analysis at Aalborg University, please feel free to contact the VBN team. You can also find more information in the AAU Student Projects FAQs.