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

2026-04-18

AI Summary

Existing cross-dose PET denoising methods suffer from limited generalization due to varying noise levels, and "one-size-fits-all" models often learn averaged representations that degrade performance. This work proposes a unified residual noise learning framework that directly estimates the noise component from low-dose PET images rather than reconstructing full-dose images, thereby mitigating the averaging effect caused by implicitly optimizing the expectation over heterogeneous noise distributions in conventional approaches. Built upon deep neural networks, the proposed architecture is trained and validated on large-scale, multi-center, multi-dose PET data. It significantly outperforms one-size-fits-all models, dose-specific U-Nets, and dose-conditioned methods on datasets from two medical centers, achieving superior generalization across diverse dose levels.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

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. In practice, 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 handle 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. To this end, 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.

Problem

• cross-dose PET denoising
• low-dose PET
• noise generalization
• averaging effect
• residual noise

Innovation

• residual noise learning
• cross-dose denoising
• PET image reconstruction
• noise distribution heterogeneity
• deep learning