Thesis Details
Segmentace zubních objemových dat
The main goal of this work was to use neural networks for volumetric segmentation of dental CBCT data. As a byproducts, both new dataset including sparse and dense annotations and automatic preprocessing pipeline were produced. Additionally, the possibility of applying transfer learning and multi-phase training in order to improve segmentation results was tested. From the various tests that were carried out, conclusion can be drawn that both multi-phase training and transfer learning showed substantial improvement in dice score for both sparse and dense annotations compared to the baseline method.
Image processing, segmentation, volumetric segmentation, U-Net, CT scans, CBCT scans, medical data, deep learning, convolutional neural networks, sparse annotations, dense annotations, transfer learning, multi-phase training, image restoration
Bartík Vladimír, Ing., Ph.D. (DIFS FIT BUT), člen
Bařina David, Ing., Ph.D. (DCGM FIT BUT), člen
Kočí Radek, Ing., Ph.D. (DITS FIT BUT), člen
Vašíček Zdeněk, doc. Ing., Ph.D. (DCSY FIT BUT), člen
@bachelorsthesis{FITBT22559,
author = "Matej Berezn\'{y}",
type = "Bachelor's thesis",
title = "Segmentace zubn\'{i}ch objemov\'{y}ch dat",
school = "Brno University of Technology, Faculty of Information Technology",
year = 2021,
location = "Brno, CZ",
language = "czech",
url = "https://www.fit.vut.cz/study/thesis/22559/"
}