Back to search

Aristotle University Of Thessaloniki (AUTH)

Μεθοδολογίες συστημάτων ταξινόμησης για την αντιμετώπιση του προβλήματος ιστολογικού χαρακτηρισμού αγγειακού τοιχώματος από εικόνες ενδοστεφανιαίου υπερηχογραφήματος

Abstract

dc:description

Vulnerable plaque identification in the vessel wall is of primary importance concerning a potential vessel rupture that would cause an acute coronary syndrome. The purpose of the present doctoral dissertation is to develop efficient methodologies to encounter the problem of vessel wall tissue characterization and, consequently, vulnerable plaque detection, using intravascular ultrasonic images (IVUS). To successfully address the problem, innovative feature extraction and feature selection techniques are suggested, as well as fuzzy classification models. The development of the above methods is based on the principles of fuzzy logic and machine learning. A new way of fuzzy coloring is suggested for the vessel wall, based on the membership degree to which the classifier identifies each tissue type. Post processes aid the revelation of fibro-calcific tissue, which comprises an intermediate stage between calcium and fibrous tissue. Moreover, the grayscale IVUS image undergoes a transformation to polar coordinates through a novel procedure, aiming to retrieve the initial radiofrequency signal of the image, enhancing the quality of the final result. Furthermore, decision fusion algorithms are applied for the first time on the tissue characterization feat, in order to achieve maximum exploitation of the extracted features, while a new method on this field is implemented using an SVM classifier to fuse the various decisions. Finally, fuzzy rule based genetic systems are properly adapted to aid the difficult task of elaborate vessel wall component detection analysis.

Degree

thesis:*
Grantor dc:publisher
Aristotle University Of Thessaloniki (AUTH)
Year dc:date
2013

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Giannoglou, Vasileios
  • Γιαννόγλου, Βασίλειος

Subjects

dc:subject × 20

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/37429

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

Giannoglou, Vasileios; Γιαννόγλου, Βασίλειος. Μεθοδολογίες συστημάτων ταξινόμησης για την αντιμετώπιση του προβλήματος ιστολογικού χαρακτηρισμού αγγειακού τοιχώματος από εικόνες ενδοστεφανιαίου υπερηχογραφήματος. Aristotle University Of Thessaloniki (AUTH), 2013. http://hdl.handle.net/10442/hedi/37429