Adaptive Extensions of Unbiased Risk Estimators for Unsupervised Magnetic Resonance Image Denoising

TL;DR

This paper introduces and evaluates unsupervised risk estimation methods, including ePURE, for MRI denoising, demonstrating their robustness in medical imaging scenarios with complex noise types where ground truth is unavailable.

Contribution

The paper develops and benchmarks the ePURE method, extending unbiased risk estimators for effective unsupervised MRI denoising in noisy, real-world conditions.

Findings

ePURE outperforms traditional denoising methods in MRI data

Unsupervised estimators are effective without ground truth images

Methods handle Gaussian and Poisson noise effectively

Abstract

The application of Deep Neural Networks (DNNs) to image denoising has notably challenged traditional denoising methods, particularly within complex noise scenarios prevalent in medical imaging. Despite the effectiveness of traditional and some DNN-based methods, their reliance on high-quality, noiseless ground truth images limits their practical utility. In response to this, our work introduces and benchmarks innovative unsupervised learning strategies, notably Stein's Unbiased Risk Estimator (SURE), its extension (eSURE), and our novel implementation, the Extended Poisson Unbiased Risk Estimator (ePURE), within medical imaging frameworks. This paper presents a comprehensive evaluation of these methods on MRI data afflicted with Gaussian and Poisson noise types, a scenario typical in medical imaging but challenging for most denoising algorithms. Our main contribution lies in the effective adaptation and implementation of the SURE, eSURE, and particularly the ePURE frameworks for medical images, showcasing their robustness and efficacy in environments where traditional noiseless ground truth cannot be obtained.