National Technical University of Athens (NTUA)
Aνάπτυξη τεχνικών επεξεργασίας δεδομένων λειτουργικού μαγνητικού συντονισμού για τον προσδιορισμό δικτύων λειτουργικής συνδεσιμότητας σε ασθενή και υγιή πληθυσμό
Abstract
dc:descriptionThe main scope of the present PhD thesis is to investigate functional changes in brain areas and functional networks that are associated with cognitive and motor functions in patients with Small Cell Lung Cancer (SCLC) after chemotherapy and before treatment with Prophylactic Cranial Irradiation (PCA) aiming to identify the underlying functional brain alterations following lung cancer chemotherapy treatment. In this context, two experiments were conducted using functional Magnetic Resonance (fMRI) technique: (a) a finger-tapping task and (b) resting-state fMRI / rs-fMRI recordings.Advanced preprocessing and statistical analysis techniques were applied in the task-based fMRI data. Specifically, data were preprocessed both in time and space in order to remove artefacts that may affect the following statistical analysis. After preprocessing, the timeseries of all voxels of the brain were extracted and modeled using the General Linear Model (GLM) in order to create the activation maps. The activation maps represent the activation and the hemodynamic response following a particular task. Results did not demonstrate any functional changes in the motor cortex of the brain of the SCLC patients compared to controls, demonstrating that this particular brain area is not affected by the toxic effects of chemotherapy.As in the first experiment, resting-state fMRI data were initially preprocessed. Three different methodological approaches were applied in order to extract the functional brain networks of interest and to explore brain functional connectivity alterations in SCLC patients. Particularly, Independent Component Analysis (ICA) was implemented in order to extract a set of statistically independent spatial maps (brain networks) along with their timeseries. Additionally, Seed-based Correlation Analysis and ROI-to-ROI Analysis were applied to demonstrate functional connectivity maps that express the temporal correlation of the brain regions that consistute the brain networks of interest. Results from both analyses revealed brain functional connectivity alterations in SCLC patients in brain regions associated with cognitive and motor functions.Additionally, a graph theory based algorithm was developed and applied in the resting-state fMRI data aiming at determining the small world topological properties that govern the relationship between the brain networks. Resting-state fMRI data were transformed into a graphical structure were each node represents a brain network and the connectivity between nodes represents the similarity in the activity of this networks. Each graph can then be summarized by its connectivity properties. In the present study, a variety of graph-theory connectivity metrics such as the average path length, the graph density, the clustering coefficient and the small-worldness were explored. Alongside, an alternative approach of calculating the smallwroldness is proposed, which characterizes more accurately the small world brain networks compared to the established approach. After applying the algorithm to the resting state fMRI data of the present study, no statistically significant alterations were observed between the SCLC group and the controls. Both groups exhibited small world characteristics with the controls displaying network properties that are closer to that of a lattice network (i.e. higher clustering coeefficient), while patients exhibited properties that are closer to that of a random network (i.e. shorter average path length).
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2017
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Bromis, Konstantinos
- Μπρόμης, Κωνσταντίνος
Subjects
dc:subject × 24- Λειτουργική απεικόνιση μαγνητικού συντονισμού
- Πείραμα με βάση ερέθισμα
- Πείραμα σε κατάσταση ηρεμίας
- Λειτουργική συνδεσιμότητα
- Δίκτυα εγκεφάλου
- Μικροκοσμικότητα
- Θεωρία γράφων
- Ανάλυση ανεξάρτητων συνιστωσών
- Χάρτες ενεργοποίησης
- Μικροκυτταρικός καρκίνος πνεύμονα
- Functional magnetic resonance imaging (FMRI)
- Task-based fMRΙ
- Resting state fMRI
- Functional connectivity
- Brain networks
- Smallworldness
- Graph theory
- Independent component analysis
- Activation maps
- Small cell lung cancer
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Engineering and Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/42566