Institutional Repository of Vilnius University
Miego arterijų ir aterosklerozinės plokštelės segmentacija kompiuterinės tomografijos vaizduose naudojant giliojo mokymosi metodus /
Abstract
dc:descriptionCarotid artery stenosis is most commonly caused by atherosclerotic plaques, which, by obstructing blood flow, can lead to vision impairments, cognitive disorders, and in the most severe cases, stroke. Diagnosis consists of evaluation of computed tomography images, but this task can be challenging even for experienced specialists, as the boundary between blood vessel and plaque can sometimes be indistinct. This work aims to segment arteries and plaques in CT images. Literature related to the application of deep learning in medical diagnosis is reviewed and most common models for semantic segmentation are discussed. Created dataset was tested with two models, using different loss functions and modifying other hyperparameters. Two different metrics were used for evaluation.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Abramavičiūtė, Iveta,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD210641553&prefLang=en_US