Physiology-based simulation of the retinal vasculature enables annotation-free segmentation of OCT angiographs

TL;DR

This paper introduces a physiology-based simulation and physics-based augmentations to generate realistic OCTA images with ground truth labels, enabling annotation-free training of retinal vessel segmentation algorithms.

Contribution

It presents a novel simulation pipeline that produces realistic OCTA images with matching labels, reducing reliance on manual annotations for training segmentation models.

Findings

Synthetic data enables effective training of segmentation algorithms.

The method achieves competitive quantitative performance.

Qualitative results show realistic vessel representations.

Abstract

Optical coherence tomography angiography (OCTA) can non-invasively image the eye's circulatory system. In order to reliably characterize the retinal vasculature, there is a need to automatically extract quantitative metrics from these images. The calculation of such biomarkers requires a precise semantic segmentation of the blood vessels. However, deep-learning-based methods for segmentation mostly rely on supervised training with voxel-level annotations, which are costly to obtain. In this work, we present a pipeline to synthesize large amounts of realistic OCTA images with intrinsically matching ground truth labels; thereby obviating the need for manual annotation of training data. Our proposed method is based on two novel components: 1) a physiology-based simulation that models the various retinal vascular plexuses and 2) a suite of physics-based image augmentations that emulate the OCTA image acquisition process including typical artifacts. In extensive benchmarking experiments, we demonstrate the utility of our synthetic data by successfully training retinal vessel segmentation algorithms. Encouraged by our method's competitive quantitative and superior qualitative performance, we believe that it constitutes a versatile tool to advance the quantitative analysis of OCTA images.