University of Ioannina
Τεχνικές ψηφιακής επεξεργασίας εικόνας για την ανάλυση μικροσκοπικής απεικόνισης ιστοπαθολογίας
Abstract
dc:descriptionThe biopsy is the examination that gives the full picture of the pathological state of the patient’s tissues and is a key factor in determining the subsequent treatment of the disease. This has led to the emergence of a relatively new field of research, with significant growth in recent years, focusing on the evaluation of histological findings via digital microscopic scanners. Its primary objective is the processing of biopsy sample images for their accurate quantitative analysis and the possibility of making valid diagnostic decisions. Consequently, doctors decide whether the condition is pharmacologically treatable or whether an organ needs to be removed or replaced with a transplant. The aim of this PhD is to apply modern image analysis techniques for the accurate quantitative evaluation of the spread of pathological findings in microscopic liver, pancreatic and kidney biopsy samples and to reduce diagnostic errors resulting from clinicians’ semi-quantitative visual estimates. This direction has led to the implementation of novel computer vision methodologies for the quantitative measurement of fat accumulation in digitized histological biopsy samples from patients with non-alcoholic fatty liver disease and non-alcoholic fatty pancreas disease. The findings of the automated tools show little diagnostic deviation from the semi-quantitative assessments of expert histopathologists, emphasizing their reliability and the necessity of adopting them in modern clinical practice. Both the above conditions, can lead to irreversible complications and increase the risk of cancer development and metastasis to other organs. A typical example of malignant infiltration is clear cell renal cell carcinoma (ccRCC), the most common cancer of the urinary system. In the context of the doctoral research, a methodology was developed for the detection of cellular abnormalities in kidney biopsy images that contribute to the progression of ccRCC according to the Fuhrman scoring system. Based on this research, in a subsequent postdoctoral phase, the diagnostic system will be further used to perform the grading of detected structures for the Fuhrman staging of ccRCC.
Degree
thesis:*- Grantor dc:publisher
- University of Ioannina
- Year dc:date
- 2023
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Bantaloukas-Arjmand, Alexandros
- Μπανταλούκας-Αρτζμάντ, Αλέξανδρος
Subjects
dc:subject × 23- Ψηφιακή Παθολογία
- Ιατρική απεικόνιση
- Ψηφιακή επεξεργασία εικόνας
- Κατάτμηση εικόνας
- Μηχανική μάθηση
- Βαθιά μάθηση
- Digital Pathology
- Medical imaging
- Digital image processing
- Image segmentation
- Machine learning
- Deep learning
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Φυσικές Επιστήμες
- Τεχνητή νοημοσύνη
- Βιοϊατρική Μηχανική
- Επιστήμες Μηχανικού και Τεχνολογία
- Computer and Information Sciences
- Natural Sciences
- Artificial Intelligence
- Medical Engineering
- Engineering and Technology
- Biomedical Engineering
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/54380