Anatomical Conditioning for Contrastive Unpaired Image-to-Image Translation of Optical Coherence Tomography Images

TL;DR

This paper introduces an enhanced unpaired image-to-image translation method for OCT images that improves semantic consistency and segmentation accuracy, aiding disease monitoring across different OCT devices.

Contribution

It proposes a style decoder supported by a segmentation decoder to restore semantic consistency in unpaired OCT image translation, improving biomarker segmentation in unsupervised domain adaptation.

Findings

Increased similarity between style-translated and target images.

Improved segmentation of biomarkers in OCT images.

Potential for disease monitoring across OCT devices.

Abstract

For a unified analysis of medical images from different modalities, data harmonization using image-to-image (I2I) translation is desired. We study this problem employing an optical coherence tomography (OCT) data set of Spectralis-OCT and Home-OCT images. I2I translation is challenging because the images are unpaired, and a bijective mapping does not exist due to the information discrepancy between both domains. This problem has been addressed by the Contrastive Learning for Unpaired I2I Translation (CUT) approach, but it reduces semantic consistency. To restore the semantic consistency, we support the style decoder using an additional segmentation decoder. Our approach increases the similarity between the style-translated images and the target distribution. Importantly, we improve the segmentation of biomarkers in Home-OCT images in an unsupervised domain adaptation scenario. Our data harmonization approach provides potential for the monitoring of diseases, e.g., age related macular disease, using different OCT devices.