Back to search

National and Kapodistrian University of Athens

Μελέτη αλλοιώσεων του ενδομητρίου με τεχνικές ανάλυσης εικόνας

Abstract

dc:description

The incidence of endometrial cancer has increased in recent years, so that in western world it has become the most frequent gynaecological cancer. Efforts for the early detection of the disease have led to the use of endometrial devices which can be used as outpatient diagnostic methods. Such devices can collect endometrial sample and provide a reliable cytοlogical diagnosis. The recent developments in computers and the use of image analysis techniques in Cytology, makes it feasible to use systems which can collect and analyse morphometric measurments of cells. According to these morphometric characteristics, a classification in diagnostic categories can be achieved In the present study, 106 postmenopausal women were studied. Endometrial samples were collected from all of them with the use of Gynoscann endometrial device. An experimental system of image analysis was used. The system achieved a right chassification of atrophic endometrial cells in 90,21% of the cases and in 89% of the cases of proliferative and secretory endometrial cells. The method achieved the correct diagnosis in 50% of the cases of cystic and adenomatous hyperplasias and only in 25% of the cases of adenomatous hyperplasias and adenomatous hyperplasias with atypias. This reflects the difficulty of the system to distinguish between cells of cystic and adenomatous hyperplasia and between cells of adenomatous and atypical adenomatous hyperplasia. Yet, when the Fox classification was used, the system could classificate correctly the 97% of the cases with atrophy, the 99,95 of the cases with hyperplasia and the 96,12% of the cases with atypical hyperplasia. Finally, when we divided cur material in cases without atypias (secretory endometrium, proliferative endometrium, atrophy, polyps, hyperplasias without atypias) and cases with atypias (atypical hyperplasias, and endometrial cancer), the diagnostic accuracy of the method was 93,4% and a correct classification of cases with and without atypias was achieved in 95,1% and 91,12% respectively. Since nowdays it is well established that malignancy potential have only atypical cells, the system achieves a correct classification in most cases. It seems that a correct classification is not feasible in 100% of the cases, since progression of cells without malignancy potential to cells with malignancy potential has not a distinguish line. This could also explain the diagnostic difficulties of cytopathologists in clinical praxis.

Degree

thesis:*
Grantor dc:publisher
National and Kapodistrian University of Athens
Year dc:date
1996

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Giannacopoulos, Constantinos
  • Γιαννακόπουλος, Κωνσταντίνος

Subjects

dc:subject × 16

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/6101

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

Giannacopoulos, Constantinos; Γιαννακόπουλος, Κωνσταντίνος. Μελέτη αλλοιώσεων του ενδομητρίου με τεχνικές ανάλυσης εικόνας. National and Kapodistrian University of Athens, 1996. http://hdl.handle.net/10442/hedi/6101