University of Patras
Ανάλυση χαρακτηριστικών περιεμμηνοπαυσιακού και μετεμμηνοπαυσιακού ενδομητρίου στη δισδιάστατη υπερηχοτομογραφία με χρήση τεχνικών ανάλυσης εικόνας
Abstract
dc:descriptionIntroduction: Cancer of the corpus uteri represents the fourth commonest neoplasm among European women and the tenth most common cause of death attributed to cancer. Irrespective whether the use of transvaginal ultrasonography (TVS) as a screening tool for detecting endometrial cancer in asymptomatic postmenopausal women is warranted, TVS dominates most diagnostic algorithms in assessing metrorrhagias to exclude this cancer. Alongside the potential benefits stemming from the integration of Saline Infusion Sonography) and Doppler modalities in endometrial imaging, gray scale TVS showed remarkable advances in the previous decades, largely attributed to the evolution in computer sciences. Following the introduction of multifrequency transvaginal probes and harmonic imaging, modern scanners are equipped with software options that enhance the resolution or the contrast between different structures, fine tune while assessing different types of tissue, implement different depth of focusing, transmit the ultrasonic beam in oblique directions to the acoustic window; all these features, in addition to de-speckle filters optimize the endometrial depiction, facilitating its assessment, even in the hands of moderately skilled operators. Double stripe endometrial thickness has illustrated a remarkable robustness over time as a quantitative indicator of endometrial cancer, especially in the presence of postmenopausal bleeding. The combined consideration of endometrial morphology and thickness has proven particularly beneficial, especially in the assessment of the 4-10 mm endometrial thickness “grey zone”, although the findings of the “morphologic” studies haven’t always been consistent. Given the importance of morphology in assessing endometrial tissue, and aiming to overcome the inherent subjectivity of the qualitative consideration of ultrasonic images, implementation of automated techniques assessing objective morphologic features such as “computerized texture analysis” would be beneficial. In digital images, texture reflects tonal (intensities of image pixels) and structural (spatial distribution of pixel intensities) properties. Texture analysis refers to algorithms that quantify texture content that may, or may not, be visually perceived. Since medical images capture various properties of biological structures, texture analysis of medical images can provide quantitative metrics relevant to structure, morphology and status of biological tissues. Texture based classification schemes have been successfully implemented in a variety of ultrasound applications. Computerized TVS assessment of endometrial morphology, has been applied mainly in assisted reproduction techniques; however, computerized texture analysis has not been implemented for diagnosing endometrial malignancies in grey scale TVS. The aim of this study is to investigate the feasibility of computerized texture analysis in characterizing endometrial tissue as depicted in 2D grey scale TVS images. Furthermore, we assess the effect of a wavelet-based image processing technique in the segmentation and subsequent characterization tasks of endometrial tissue. […]
Degree
thesis:*- Grantor dc:publisher
- University of Patras
- Year dc:date
- 2007
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Μιχαήλ, Γεώργιος
- Michail, Georgios
Subjects
dc:subject × 20- Διακολπική υπερηχογραφία
- Ενδομητρικός καρκίνος
- Περιεμμηνόπαυση
- Μετεμμηνόπαυση
- Μητρορραγία
- Αυτοματοποιημένη διάγνωση
- Ανάλυση υφής
- Ποσοτικά χαρακτηριστικά
- Transvaginal ultranography (TVS)
- Endometrial cancer
- Perimenopause
- Postmenopause
- Vaginal bleeding
- Automated diagnosis
- Texture analysis
- Quantitative features
- Ιατρική και Επιστήμες Υγείας
- Κλινική Ιατρική
- Medical and Health Sciences
- Clinical Medicine
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/26541