Συντάχθηκε 10-12-2013 14:53
από Eleni Stamataki
Email συντάκτη: estamataki<στο>tuc.gr
Ενημερώθηκε:
-
Ιδιότητα: σύνταξη/αποχώρηση υπάλληλος.
Σχολή Ηλεκτρονικών Μηχανικών και Μηχανικών Υπολογιστών
Πρόγραμμα Μεταπτυχιακών Σπουδών
ΠΑΡΟΥΣΙΑΣΗ ΜΕΤΑΠΤΥΧΙΑΚΗΣ ΔΙΑΤΡΙΒΗΣ
ΚΑΒΒΑΔΙΑ ΒΑΣΙΛΕΙΟΥ
με θέμα
Simultaneous Multi-Spectral Imaging System: Application in Real-Time, Unsupervised Spectral Classification in Endometrial Endoscopy
Παρασκευή 13 Δεκεμβρίου 2013, 11πμ
Αίθουσα Συνεδριάσεων, Κτίριο Επιστημών, Πολυτεχνειούπολη
Εξεταστική Επιτροπή
Αν. Καθηγητής Μπάλας Κωνσταντίνος (επιβλέπων)
Καθηγητής Ζερβάκης Μιχάλης
Επ. Καθηγητής Μπούχερ Ματτίας
Abstract
We report the first real-time spectral mapper based on the combination of snap-shot spectral imaging and spectral classification algorithms. High accuracy is accompanied with much faster, by 3 orders of magnitude, spectral mapping, as compared with scanning spectral systems. This new technology is intended to enable spectral mapping at nearly video rates in all kinds of dynamic bio-optical effects as well as in applications where the target-probe relative position is randomly and fast changing. With this system adapted to a thin hysteroscope for imaging of the endometrial tissue, we also report, for the first time, spectral analysis of the endometrium and unsupervised/objective clustering of the spectra. We have implemented a method combining the k-means algorithm with the silhouette criterion for estimating the number of the distinguishable spectral classes that may correspond to different medical conditions of the tissue. It was found that there are five-well defined clusters of spectra, while preliminary clinical data seem to correlate well with the tissue pathology.
