Klasifikacija ortopantomograma pomoću dubokih neuronskih mreža

master's thesis

• Title (english): Classification of Orthopantomograms Using Deep Neural Networks
• Author: Miljenko Šuflaj
• Mentor: Marko Subašić (mentor)
• Committee member: Marko Subašić (predsjednik povjerenstva)
• Committee member: Sven Lončarić (član povjerenstva)
• Committee member: Zvonko Kostanjčar (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-07-12, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Ponovno posjećujemo problem detekcije i klasifikacije zuba, no ovog puta s modernijim metodama. Koristeći Swin transformer i Mask R-CNN detektor, finetuneamo predtrenirani detektor nad našim podatkovnim skupom. Navodimo kako bi u optimalnom slučaju izgledao podatkovni skup, model, hiperparametri. Prikazujemo naše rezultate te ih komentiramo. Uspoređujemo s bogatom kolekcijom postojećih radova na temu. Prezentiramo ciljeve za buduće istraživanje na temu.

Abstract (english)

We visit the problem of detection and classification of teeth once more, only this time with modern methods. Using a Swin transformer and a Mask R-CNN detector we finetune the pretrained detector on our dataset. We list how an optimal scenario regarding our dataset, model and its hyperparameters would look like. We show our results and comment on them. We compare them to the rich collection of existing work on the topic. Finally, we present goals for future work on the topic.

• Keywords: duboko učenje; ortopantomogram; zubi; transformer
• Keywords (english): deep learning; orthopantomogram; teeth; transformer
• Language: croatian
• URN:NBN: urn:nbn:hr:168:796435
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 16:20:00