Segmentacija medicinskih slika primjenom dubokog učenja

master's thesis

• Title (english): Segmentation of medical images using deep learning
• Author: Filip Martinović
• Mentor: Tomislav Hrkać (mentor)
• Committee member: Tomislav Hrkać (predsjednik povjerenstva)
• Committee member: Siniša Šegvić (član povjerenstva)
• Committee member: Domagoj Jakobović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2024-09-20, Croatia

Abstract

U ovom radu raspravlja se o problemu segmentacije medicinskih slika na skupu podataka DRIVE. Skup sadrži slike mrežnice oka i segmentacijske maske koje sadrže žile. U radu se opisuju tradicionalne metode obrade slike i suvremene metode dubokog učenja. Na skupu podataka se uči segmentacijski model UNet sa ResNet34 koderom. Uspoređuju se rezultati na originalnom skupu, augmentiranom skupu i skupu na koji je primijenjena CLAHE transformacija.

Abstract (english)

In this paper, the problem of segmentation of medical images is discussed. The dataset used for experiments is DRIVE. The dataset contains retinal images and segmentation masks containing vessels. The paper describes traditional image processing methods and modern deep learning methods. The segmentation model UNet with the ResNet34 encoder is trained on the dataset. The results on the original dataset, the augmented dataset and the dataset with CLAHE transformation applied are compared.

• Keywords: segmentacija slike; obrada medicinskih slika; duboko učenje
• Keywords (english): image segmentation; medical images processing; deep learning
• Language: croatian
• URN:NBN: urn:nbn:hr:168:123728
• 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: 2024-10-12 21:29:10