Publication result detail

Segmentation of optic disc and cup in retinal images using of deep learning approaches

NOHEL, M.; KOLÁŘ, R.

Original Title

Segmentation of optic disc and cup in retinal images using of deep learning approaches

English Title

Segmentation of optic disc and cup in retinal images using of deep learning approaches

Type

Paper in proceedings outside WoS and Scopus

Original Abstract

This paper presents a comparative analysis of optic disc and cup segmentation in retinal fundus images using two deep learning models: the classical U-net and its modified version, nnU-Net. The models were trained and tested on publicly available databases consisting of 1295 images for training and 555 images for testing. The results indicate that while nnU-Net demonstrated only slight improvements in disc segmentation on the test database, it significantly outperformed the U-net model in optical cup segmentation.

English abstract

This paper presents a comparative analysis of optic disc and cup segmentation in retinal fundus images using two deep learning models: the classical U-net and its modified version, nnU-Net. The models were trained and tested on publicly available databases consisting of 1295 images for training and 555 images for testing. The results indicate that while nnU-Net demonstrated only slight improvements in disc segmentation on the test database, it significantly outperformed the U-net model in optical cup segmentation.

Keywords

deep learning, convolutional neural networks, vertebrae segmentation, segmentation, spine, vertebra, CT, computed tomography

Key words in English

deep learning, convolutional neural networks, vertebrae segmentation, segmentation, spine, vertebra, CT, computed tomography

Authors

NOHEL, M.; KOLÁŘ, R.

RIV year

2024

Released

25.04.2023

Publisher

Brno University of Technology, Faculty of Electrical Engineering and Communication

Location

Brno, Czech Republic

ISBN

978-80-214-6153-6

Book

Proceedings I of the 29th Conference STUDENT EEICT 2023

Edition

1st edition

Pages from

265

Pages to

269

Pages count

5

URL

https://www.eeict.cz/eeict_download/archiv/sborniky/EEICT_2023_sbornik_1.pdf

BibTex

@inproceedings{BUT184277,
  author="Michal {Nohel} and Radim {Kolář}",
  title="Segmentation of optic disc and cup in retinal images using of deep learning approaches",
  booktitle="Proceedings I of the 29th Conference STUDENT EEICT 2023",
  year="2023",
  series="1st edition",
  pages="265--269",
  publisher="Brno University of Technology, Faculty of Electrical Engineering and Communication",
  address="Brno, Czech Republic",
  isbn="978-80-214-6153-6",
  url="https://www.eeict.cz/eeict_download/archiv/sborniky/EEICT_2023_sbornik_1.pdf"
}