Publication detail

Synthetic Retinal Images from Unconditional GANs

BISWAS, S. ROHDIN, J. DRAHANSKÝ, M.

Original Title

Synthetic Retinal Images from Unconditional GANs

English Title

Synthetic Retinal Images from Unconditional GANs

Type

conference paper

Language

en

Original Abstract

Synthesized retinal images are highly demanded in the development of automated eye applications since they can make machine learning algorithms more robust by increasing the size and heterogeneity of the training database. Recently, conditional Generative Adversarial Networks (cGANs) based synthesizers have been shown to be promising for generating retinal images. However, cGANs based synthesizers require segmented blood vessels (BV) along with RGB retinal images during training. The amount of such data (i.e., retinal images and their corresponding BV) available in public databases is very small. Therefore, for training cGANs, an extra system is necessary either for synthesizing BV or for segmenting BV from retinal images. In this paper, we show that by using unconditional GANs (uGANs) we can generate synthesized retinal images without using BV images.

English abstract

Synthesized retinal images are highly demanded in the development of automated eye applications since they can make machine learning algorithms more robust by increasing the size and heterogeneity of the training database. Recently, conditional Generative Adversarial Networks (cGANs) based synthesizers have been shown to be promising for generating retinal images. However, cGANs based synthesizers require segmented blood vessels (BV) along with RGB retinal images during training. The amount of such data (i.e., retinal images and their corresponding BV) available in public databases is very small. Therefore, for training cGANs, an extra system is necessary either for synthesizing BV or for segmenting BV from retinal images. In this paper, we show that by using unconditional GANs (uGANs) we can generate synthesized retinal images without using BV images.

Keywords

eye retina, blood vessels, GAN, synthetic image

Released

23.07.2019

Publisher

IEEE Computer Society

Location

Berlin

ISBN

978-1-5386-1311-5

Book

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society

Edition

NEUVEDEN

Edition number

NEUVEDEN

Pages from

2736

Pages to

2739

Pages count

4

URL

Documents

BibTex


@inproceedings{BUT161844,
  author="Sangeeta {Biswas} and Johan Andréas {Rohdin} and Martin {Drahanský}",
  title="Synthetic Retinal Images from Unconditional GANs",
  annote="Synthesized retinal images are highly demanded in the development of automated
eye applications since they can make machine learning algorithms more robust by
increasing the size and heterogeneity of the training database. Recently,
conditional Generative Adversarial Networks (cGANs) based synthesizers have been
shown to be promising for generating retinal images. However, cGANs based
synthesizers require segmented blood vessels (BV) along with RGB retinal images
during training. The amount of such data (i.e., retinal images and their
corresponding BV) available in public databases is very small. Therefore, for
training cGANs, an extra system is necessary either for synthesizing BV or for
segmenting BV from retinal images. In this paper, we show that by using
unconditional GANs (uGANs) we can generate synthesized
retinal images without using BV images.",
  address="IEEE Computer Society",
  booktitle="Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society",
  chapter="161844",
  doi="10.1109/EMBC.2019.8857857",
  edition="NEUVEDEN",
  howpublished="print",
  institution="IEEE Computer Society",
  year="2019",
  month="july",
  pages="2736--2739",
  publisher="IEEE Computer Society",
  type="conference paper"
}