MDAA-Diff: CT-Guided Multi-Dose Adaptive Attention Diffusion Model for PET Denoising

2025-05-08

AI Summary

To address three key challenges in low-dose PET denoising—radiation exposure control, large inter-patient variability in dose response, and absence of CT anatomical priors—this paper proposes a CT-guided multi-dose adaptive denoising framework. Methodologically, we introduce a novel attention module that jointly integrates wavelet-based high-frequency features and CT-derived anatomical priors, coupled with a dose-conditioned channel-spatial joint attention mechanism to enable personalized dose-response modeling and anatomical fidelity preservation. Furthermore, we integrate diffusion modeling with adaptive multimodal feature weighting. Evaluated on dual tracers (¹⁸F-FDG and ⁶⁸Ga-FAPI), our method significantly outperforms state-of-the-art approaches, achieving superior lesion contrast enhancement and edge detail recovery. It enables clinically safe operation at less than 20% of the original injected activity, supporting ultra-low-dose PET imaging without compromising diagnostic quality.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Multi-instance/Multi-view Learning
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Security and Privacy: Privacy-enhancing technologies

Abstract

Acquiring high-quality Positron Emission Tomography (PET) images requires administering high-dose radiotracers, which increases radiation exposure risks. Generating standard-dose PET (SPET) from low-dose PET (LPET) has become a potential solution. However, previous studies have primarily focused on single low-dose PET denoising, neglecting two critical factors: discrepancies in dose response caused by inter-patient variability, and complementary anatomical constraints derived from CT images. In this work, we propose a novel CT-Guided Multi-dose Adaptive Attention Denoising Diffusion Model (MDAA-Diff) for multi-dose PET denoising. Our approach integrates anatomical guidance and dose-level adaptation to achieve superior denoising performance under low-dose conditions. Specifically, this approach incorporates a CT-Guided High-frequency Wavelet Attention (HWA) module, which uses wavelet transforms to separate high-frequency anatomical boundary features from CT images. These extracted features are then incorporated into PET imaging through an adaptive weighted fusion mechanism to enhance edge details. Additionally, we propose the Dose-Adaptive Attention (DAA) module, a dose-conditioned enhancement mechanism that dynamically integrates dose levels into channel-spatial attention weight calculation. Extensive experiments on 18F-FDG and 68Ga-FAPI datasets demonstrate that MDAA-Diff outperforms state-of-the-art approaches in preserving diagnostic quality under reduced-dose conditions. Our code is publicly available.

Problem

• Reducing radiation exposure risks in PET imaging by denoising low-dose PET.
• Addressing inter-patient dose response variability and leveraging CT anatomical constraints.
• Enhancing PET image quality with adaptive attention to dose levels.

Innovation

• CT-Guided High-frequency Wavelet Attention module
• Dose-Adaptive Attention mechanism
• Multi-dose PET denoising with anatomical guidance