National Technical University of Athens (NTUA)
Ανάλυση και επεξεργασία ιατρικών δεδομένων
Abstract
dc:descriptionThe scope of the present thesis was the study of the effect of chemotherapy in the brain of patients suffering from small cell lung cancer (SCLC) by employing the latest techniques of Magnetic Resonance Imaging (MRI, T1, T2) as well as of Diffusion tensor imaging (DTI) and functional magnetic resonance imaging (fΜRI). For this purpose, Τ1, Τ2, DTI and fMRI data were collected from 24 SCLC patients after chemotherapy as well as from 14 healthy volunteers. The first goal of the data analysis process was to investigate whether there was disruption of the fiber tracts architecture of the white matter of the patients’ brains. The latter was achieved by the realization of two different processing methodologies of the DTI data: the deterministic analysis and the voxel-to-voxel analysis. Furthermore, we studied the potential occurrence of structural destruction of the gray matter by the utilization of titration on the T1 images while the analysis of the fMRI data at the functional connectivity level enabled the study of potential existing functional deficiencies. The overall results of the abovementioned analysis showed that the patients’ group had discrepancies in functional connectivity as well as deteriorations in their gray and white matter. The following analysis phase focused on connecting/associating DTI with Graph Theory in order to proceed to the study of the topological organization of the SCLC patients’ white matter’s structural networks as well as the structural connectivity of these networks. Moreover, a variety of graph-theory connectivity metrics such as the average path length, the clustering coefficient, the local and global efficiency, the hubs and the small world properties were explored/investigated. The findings of the latter analysis demonstrated that patients had clearly a disruption of the structural networks of the brain while presenting differentiated structural connectivity when compared to the healthy individuals. Finally, within the framework of this thesis Machine Learning methods were developed that succeeded in separating the SCLC patients from the healthy individuals with high accuracy by using the networks of structural connectivity. The overall findings of the current thesis offer a better understanding of the effect of chemotherapy on the human brain.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2022
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Mentzelopoulos, Anastasios
- Μεντζελόπουλος, Αναστάσιος
Subjects
dc:subject × 27- Απεικόνισης του τανυστή διάχυσης
- Δομική απεικόνιση μαγνητικού συντονισμού
- Λειτουργική συνδεσιμότητα
- Μέθοδοι μηχανικής εκμάθησης
- Θεωρία των γράφων
- Δομική συνδεσιμότητα
- Μικροκυταρικός καρκίνος του πνεύμονα
- Χημειοθεραπεία
- Diffusion tensor imaging
- Structural Magnetic Resonance Imaging
- Functional connectivity
- Graph theory
- Brain structural connectivity
- Machine learning
- Small-cell lung cancer
- Chemotherapy
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Φυσικές Επιστήμες
- Επιστήμη ηλεκτρονικών υπολογιστών και Πληροφορική, άλλοι τομείς
- Βιοϊατρική Μηχανική
- Επιστήμες Μηχανικού και Τεχνολογία
- Computer and Information Sciences
- Natural Sciences
- Computer and Information sciences, miscellaneous
- Medical Engineering
- Engineering and Technology
- Biomedical Engineering
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/52064