Share Email Print

Proceedings Paper

FunSyn-Net: enhanced residual variational auto-encoder and image-to-image translation network for fundus image synthesis
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Medical imaging datasets typically do not contain many training images and are usually not sufficient for training deep learning networks. We propose a deep residual variational auto-encoder and a generative adversarial network based approach that can generate a synthetic retinal fundus image dataset with corresponding blood vessel annotations. In terms of structural statistics comparison of real and artificial our model performed better than existing methods. The generated blood vessel structures achieved a structural similarity value of 0.74 and the artificial dataset achieved a sensitivity of 0.84 and specific city of 0.97 for the blood vessel segmentation task. The successful application of generative models for the generation of synthetic medical data will not only help to mitigate the small dataset problem but will also address the privacy concerns associated with such medical datasets.

Paper Details

Date Published: 10 March 2020
PDF: 7 pages
Proc. SPIE 11313, Medical Imaging 2020: Image Processing, 113132M (10 March 2020); doi: 10.1117/12.2549869
Show Author Affiliations
Sourya Sengupta, Univ. of Waterloo (Canada)
Akshay Athwale, ITT Dhanbad (India)
Tanmay Gulati, Manipal Institute of Technology (India)
John Zelek, Univ. of Waterloo (Canada)
Vasudevan Lakshminarayanan, Univ. of Waterloo (Canada)

Published in SPIE Proceedings Vol. 11313:
Medical Imaging 2020: Image Processing
Ivana Išgum; Bennett A. Landman, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?