De-speckling of Optical Coherence Tomography Images Using Anscombe Transform and a Noisier2noise Model

TL;DR

This paper introduces a novel OCT image denoising method that uses the Anscombe transform and a Noisier2noise model, enabling effective training with only single noisy observations without needing clean images or multiple noisy samples.

Contribution

It presents a new approach combining Anscombe transform with a Noisier2noise model for OCT denoising, allowing training with single noisy images and outperforming existing methods.

Findings

Outperforms several existing denoising methods on OCT images.

Effectively trains with only a single noisy observation per image.

Validated on a publicly available OCT dataset.

Abstract

Optical Coherence Tomography (OCT) image denoising is a fundamental problem as OCT images suffer from multiplicative speckle noise, resulting in poor visibility of retinal layers. The traditional denoising methods consider specific statistical properties of the noise, which are not always known. Furthermore, recent deep learning-based denoising methods require paired noisy and clean images, which are often difficult to obtain, especially medical images. Noise2Noise family architectures are generally proposed to overcome this issue by learning without noisy-clean image pairs. However, for that, multiple noisy observations from a single image are typically needed. Also, sometimes the experiments are demonstrated by simulating noises on clean synthetic images, which is not a realistic scenario. This work shows how a single real-world noisy observation of each image can be used to train a denoising network. Along with a theoretical understanding, our algorithm is experimentally validated using a publicly available OCT image dataset. Our approach incorporates Anscombe transform to convert the multiplicative noise model to additive Gaussian noise to make it suitable for OCT images. The quantitative results show that this method can outperform several other methods where a single noisy observation of an image is needed for denoising. The code and implementation of this paper will be available publicly upon acceptance of this paper.