prikaz prve stranice dokumenta Usporedba različitih modela strojnog učenja za klasifikaciju tkiva rana
  Download
PDF 1.73 MB
master's thesis
Usporedba različitih modela strojnog učenja za klasifikaciju tkiva rana
2024. urn:nbn:hr:200:799137
Antunović, Antonio
Josip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
Department of Computer Engineering and Automation
Chair of Automation and Robotics

Cite this document

Antunović, A. (2024). Usporedba različitih modela strojnog učenja za klasifikaciju tkiva rana (Master's thesis). Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek. Retrieved from https://urn.nsk.hr/urn:nbn:hr:200:799137

Antunović, Antonio. "Usporedba različitih modela strojnog učenja za klasifikaciju tkiva rana." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2024. https://urn.nsk.hr/urn:nbn:hr:200:799137

Antunović, Antonio. "Usporedba različitih modela strojnog učenja za klasifikaciju tkiva rana." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2024. https://urn.nsk.hr/urn:nbn:hr:200:799137

Antunović, A. (2024). 'Usporedba različitih modela strojnog učenja za klasifikaciju tkiva rana', Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, accessed 17 July 2024, https://urn.nsk.hr/urn:nbn:hr:200:799137

Antunović A. Usporedba različitih modela strojnog učenja za klasifikaciju tkiva rana [Master's thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek; 2024 [cited 2024 July 17] Available at: https://urn.nsk.hr/urn:nbn:hr:200:799137

A. Antunović, "Usporedba različitih modela strojnog učenja za klasifikaciju tkiva rana", Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Osijek, 2024. Available at: https://urn.nsk.hr/urn:nbn:hr:200:799137

Please login to the repository to save this object to your list.
Metadata
TitleUsporedba različitih modela strojnog učenja za klasifikaciju tkiva rana
Title (english)Comparison of machine learning models for wound tissue classification
AuthorAntonio Antunović
MentorEmmanuel Karlo Nyarko (mentor)
MentorDamir Filko (sumentor)
Committee memberEmmanuel Karlo Nyarko (predsjednik povjerenstva)
Committee memberPetra Pejić (član povjerenstva)
Committee memberRobert Cupec (član povjerenstva)
GranterJosip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
(Department of Computer Engineering and Automation)
(Chair of Automation and Robotics)
Osijek
Defense date and country2024-07-10, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Artificial Intelligence
Abstract
Proces zacjeljivanja kroničnih rana često je iz nekog razloga usporen, te one zahtijevaju poseban medicinski tretman. Shodno tome, praćenje procesa zacjeljivanja predstavlja značajan korak pri određivanju optimalne terapije za tretiranje kroničnih rana. Napredak u području strojnog učenja i računalnog vida ostvario je mogućnosti razvoja metoda za analizu rana koje bi mogle točno odrediti vrste prisutnih tkiva i pratiti zacjeljivanje, čime se ovaj rad bavi. Ovo istraživanje bilo je usmjereno na razvoj i evaluaciju modela strojnog učenja za zadatak semantičke segmentacije tkiva kroničnih rana. Odabrano je osam različitih arhitektura neuronskih mreža, pri čemu je svaka uparena s jednom od četiri korištene funkcije gubitka. Proces učenja proveden je za svaku arhitekturu, nakon čega su modeli evaluirani na testnom skupu podataka. Nakon evaluacije provedena je statistička analiza rezultata kako bi se utvrdilo postoje li i statistički značajne razlike među modelima.
Abstract (english)
The healing process of chronic wounds is often disrupted for a variety of reasons, and they require special medical treatment. Therefore, monitoring the healing process is a significant step in recognizing and addressing potential problems and creating an appropriate treatment plan. Improvement of machine learning and computer vision enabled the development of methods for wound analysis that would be able to precisely detect types of tissues within the wound and monitor the healing process, which is what this paper deals with. This research focused on the development and evaluation of machine learning models for the semantic segmentation of chronic wound tissue. Eight different neural network architectures were selected, each paired with one of the four loss functions. A training process was carried out for each architecture and then evaluated on a test set. Following that, statistical analysis was conducted to determine whether there were significant differences between the models.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:200:799137
Study programmeTitle: Graduate University Study Programme in Computer Engineering
Study programme type: university
Study level: graduate
Academic / professional title: sveučilišni/a magistar/magistra inženjer/inženjerka računarstva (sveučilišni/a magistar/magistra inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2024-07-16 11:49:00