Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Adaptive Whole-Body PET Image Denoising Using 3D Diffusion Models with ControlNet

Boxiao Yu, Kuang Gong

TL;DR

Experimental results based on clinical PET datasets show that the proposed framework outperformed other state-of-the-art PET image denoising methods both in visual quality and quantitative metrics.

Abstract

Positron Emission Tomography (PET) is a vital imaging modality widely used in clinical diagnosis and preclinical research but faces limitations in image resolution and signal-to-noise ratio due to inherent physical degradation factors. Current deep learning-based denoising methods face challenges in adapting to the variability of clinical settings, influenced by factors such as scanner types, tracer choices, dose levels, and acquisition times. In this work, we proposed a novel 3D ControlNet-based denoising method for whole-body PET imaging. We first pre-trained a 3D Denoising Diffusion Probabilistic Model (DDPM) using a large dataset of high-quality normal-dose PET images. Following this, we fine-tuned the model on a smaller set of paired low- and normal-dose PET images, integrating low-dose inputs through a 3D ControlNet architecture, thereby making the model adaptable to denoising tasks in diverse clinical settings. Experimental results based on clinical PET datasets show that the proposed framework outperformed other state-of-the-art PET image denoising methods both in visual quality and quantitative metrics. This plug-and-play approach allows large diffusion models to be fine-tuned and adapted to PET images from diverse acquisition protocols.

Adaptive Whole-Body PET Image Denoising Using 3D Diffusion Models with ControlNet

TL;DR

Experimental results based on clinical PET datasets show that the proposed framework outperformed other state-of-the-art PET image denoising methods both in visual quality and quantitative metrics.

Abstract

Positron Emission Tomography (PET) is a vital imaging modality widely used in clinical diagnosis and preclinical research but faces limitations in image resolution and signal-to-noise ratio due to inherent physical degradation factors. Current deep learning-based denoising methods face challenges in adapting to the variability of clinical settings, influenced by factors such as scanner types, tracer choices, dose levels, and acquisition times. In this work, we proposed a novel 3D ControlNet-based denoising method for whole-body PET imaging. We first pre-trained a 3D Denoising Diffusion Probabilistic Model (DDPM) using a large dataset of high-quality normal-dose PET images. Following this, we fine-tuned the model on a smaller set of paired low- and normal-dose PET images, integrating low-dose inputs through a 3D ControlNet architecture, thereby making the model adaptable to denoising tasks in diverse clinical settings. Experimental results based on clinical PET datasets show that the proposed framework outperformed other state-of-the-art PET image denoising methods both in visual quality and quantitative metrics. This plug-and-play approach allows large diffusion models to be fine-tuned and adapted to PET images from diverse acquisition protocols.

Paper Structure

This paper contains 10 sections, 7 equations, 3 figures.

Table of Contents

  1. Introduction
  2. Methodology
  3. Pre-training with 3D DDPM
  4. Fine-tuning with 3D ControlNet
  5. Experiments and Results
  6. Dataset and Implementation Details
  7. Results
  8. Conclusion
  9. Compliance with ethical standards
  10. Acknowledgments

Figures (3)

  • Figure 1: Overview of the proposed 3D ControlNet-based denoising model.
  • Figure 2: Sagittal and coronal view of $1/20$ low-dose PET images and the corresponding denoised results using the proposed 3D ControlNet and other reference methods, along with the normal-dose PET.
  • Figure 3: The PSNR and SSIM values calculated based on 60 $1/20$ low-dose test datasets. *** located at the top of the bar plot represents p-value < $0.001$.