Self-supervised Physics-based Denoising for Computed Tomography

TL;DR

This paper presents a self-supervised CT denoising method that leverages the physics of noise and projection data, enabling effective noise reduction without high-dose ground truth images, thus improving low-dose CT imaging safety.

Contribution

The introduced Noise2NoiseTD-ANM method uniquely combines CT noise modeling with projection relationships for interpretable, no-reference denoising without requiring paired training data.

Findings

Achieves comparable or superior results to supervised models.

Generalizes well across different noise levels.

Outperforms existing self-supervised denoising algorithms.

Abstract

Computed Tomography (CT) imposes risk on the patients due to its inherent X-ray radiation, stimulating the development of low-dose CT (LDCT) imaging methods. Lowering the radiation dose reduces the health risks but leads to noisier measurements, which decreases the tissue contrast and causes artifacts in CT images. Ultimately, these issues could affect the perception of medical personnel and could cause misdiagnosis. Modern deep learning noise suppression methods alleviate the challenge but require low-noise-high-noise CT image pairs for training, rarely collected in regular clinical workflows. In this work, we introduce a new self-supervised approach for CT denoising Noise2NoiseTD-ANM that can be trained without the high-dose CT projection ground truth images. Unlike previously proposed self-supervised techniques, the introduced method exploits the connections between the adjacent projections and the actual model of CT noise distribution. Such a combination allows for interpretable no-reference denoising using nothing but the original noisy LDCT projections. Our experiments with LDCT data demonstrate that the proposed method reaches the level of the fully supervised models, sometimes superseding them, easily generalizes to various noise levels, and outperforms state-of-the-art self-supervised denoising algorithms.