Share Email Print

Proceedings Paper

Semantic denoising autoencoders for retinal optical coherence tomography
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Noise in speckle-prone optical coherence tomography tends to obfuscate important details necessary for medical diagnosis. In this paper, a denoising approach that preserves disease characteristics on retinal optical coherence tomography images in ophthalmology is presented. We propose semantic denoising autoencoders, which combine a convolutional denoising autoencoder with a priorly trained ResNet image classifier as regularizer during training. This promotes the perceptibility of delicate details in the denoised images that are important for diagnosis and filters out only informationless background noise. With our approach, higher peak signal-to-noise ratios with PSNR = 31.0 dB and higher classification performance of F1 = 0.92 can be achieved for denoised images compared to state-of-the-art denoising. It is shown that semantically regularized autoencoders are capable of denoising retinal OCT images without blurring details of diseases.

Paper Details

Date Published: 19 July 2019
PDF: 4 pages
Proc. SPIE 11078, Optical Coherence Imaging Techniques and Imaging in Scattering Media III, 1107818 (19 July 2019); doi: 10.1117/12.2526936
Show Author Affiliations
Max-Heinrich Laves, Leibniz Univ. Hannover (Germany)
Sontje Ihler, Leibniz Univ. Hannover (Germany)
Lüder Alexander Kahrs, Leibniz Univ. Hannover (Germany)
Tobias Ortmaier, Leibniz Univ. Hannover (Germany)

Published in SPIE Proceedings Vol. 11078:
Optical Coherence Imaging Techniques and Imaging in Scattering Media III
Maciej Wojtkowski; Stephen A. Boppart; Wang-Yuhl Oh, 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?