prikaz prve stranice dokumenta Klasifikacija histoloških snimaka crijeva
No public access
master's thesis
Klasifikacija histoloških snimaka crijeva
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2024. urn:nbn:hr:168:679913
Hrabar, Fran
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Hrabar, F. (2024). Klasifikacija histoloških snimaka crijeva (Master's thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:679913

Hrabar, Fran. "Klasifikacija histoloških snimaka crijeva." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2024. https://urn.nsk.hr/urn:nbn:hr:168:679913

Hrabar, Fran. "Klasifikacija histoloških snimaka crijeva." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2024. https://urn.nsk.hr/urn:nbn:hr:168:679913

Hrabar, F. (2024). 'Klasifikacija histoloških snimaka crijeva', Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 01 June 2024, https://urn.nsk.hr/urn:nbn:hr:168:679913

Hrabar F. Klasifikacija histoloških snimaka crijeva [Master's thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2024 [cited 2024 June 01] Available at: https://urn.nsk.hr/urn:nbn:hr:168:679913

F. Hrabar, "Klasifikacija histoloških snimaka crijeva", Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2024. Available at: https://urn.nsk.hr/urn:nbn:hr:168:679913

Metadata
TitleKlasifikacija histoloških snimaka crijeva
Title (english)Classification of Histological Images of the Intestine
AuthorFran Hrabar
MentorMarko Subašić (mentor)
Committee memberMarko Subašić (predsjednik povjerenstva)
Committee memberZvonko Kostanjčar (član povjerenstva)
Committee memberTomislav Petković (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2024-02-16, Croatia
Abstract
Tema ovog rada je klasifikacija Crohnove bolesti ili ulceroznog kolitisa na temelju histoloških snimki crijeva pomoću dubokih neuronskih mreža. Rad objašnjava postupak predobrade malog, visokodimenzionalnog i neuravnoteženog skupa podataka kao što su segmentacija i razlamanja snimki na zakrpe radi smanjenja dimenzija ulaza, očuvanja detalja, eliminacije šuma te povećanja primjeraka u skupu podataka te uravnotežavanje skupa podataka i podjelu skupa podataka na skup za treniranje i testiranje. Opisuju se arhitektura modela, proces treniranja i evaluacije modela na temelju odabranih metrika. Metrike pokrivaju razne zahtjeve koje bi liječnik mogao imati pri evaluaciji modela te dopušta stručnjaku da odabere najuspješniji model.
Abstract (english)
The topic of this paper is the classification of Crohn's disease or ulcerative colitis based on histological images of the intestine using deep neural networks. The paper explains the preprocessing procedure of a small, high-dimensional and imbalanced dataset, such as segmentation and splitting of recordings into patches in order to reduce input dimensions, preserve details, eliminate noise and increase the number of samples in the data set, as well as balancing the data set and dividing the data set into training and testing sets. The architecture of the model, the process of training and evaluation of the model based on selected metrics are described. The metrics cover the various requirements a physician might have when evaluating a model and allow the practitioner to select the most successful model.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:679913
Study programmeTitle: Computing
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2024-03-28 07:46:38