Rethinking Cross-Dose PET Denoising: Mitigating Averaging Effects via Residual Noise Learning

TL;DR

This paper introduces a residual noise learning framework for cross-dose PET denoising, addressing the averaging effect of traditional models and improving generalization across different dose levels.

Contribution

The paper proposes a novel residual noise learning approach that estimates noise directly, outperforming existing models in multi-dose PET denoising tasks.

Findings

Outperforms

one-size-for-all

models and dose-specific U-Net models

Abstract

Cross-dose denoising for low-dose positron emission tomography (LDPET) has been proposed to address the limited generalization of models trained at a single noise level. However, neural networks trained on a specific dose level often fail to generalize to other dose conditions due to variations in noise magnitude and statistical properties. Conventional "one-size-for-all" models attempt to mitigate this variability but tend to learn averaged representations across noise levels, resulting in degraded performance. In this work, we analyze this limitation and show that standard training formulations implicitly optimize an expectation over heterogeneous noise distributions, causing the network to learn an averaged denoising mapping that cannot accurately model dose-specific noise characteristics. We propose a unified residual noise learning framework that estimates noise directly from low-dose PET images rather than predicting full-dose images. Experiments on large-scale multi-dose PET datasets from two medical centers demonstrate that the proposed method outperforms the "one-size-for-all" model, individual dose-specific U-Net models, and dose-conditioned approaches, achieving improved denoising performance. These results indicate that residual noise learning effectively mitigates the averaging effect and enhances generalization for cross-dose PET denoising.