English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Denoising low-dose computed tomography (CT) images is a critical task inmedical image computing. Supervised deep learning-based approaches have madesignificant advancements in this area in recent years. However, these methodstypically require pairs of low-dose and normal-dose CT images for training,which are challenging to obtain in clinical settings. Existing unsuperviseddeep learning-based methods often require training with a large number oflow-dose CT images or rely on specially designed data acquisition processes toobtain training data. To address these limitations, we propose a novelunsupervised method that only utilizes normal-dose CT images during training,enabling zero-shot denoising of low-dose CT images. Our method leverages thediffusion model, a powerful generative model. We begin by training a cascadedunconditional diffusion model capable of generating high-quality normal-dose CTimages from low-resolution to high-resolution. The cascaded architecture makesthe training of high-resolution diffusion models more feasible. Subsequently,we introduce low-dose CT images into the reverse process of the diffusion modelas likelihood, combined with the priors provided by the diffusion model anditeratively solve multiple maximum a posteriori (MAP) problems to achievedenoising. Additionally, we propose methods to adaptively adjust thecoefficients that balance the likelihood and prior in MAP estimations, allowingfor adaptation to different noise levels in low-dose CT images. We test ourmethod on low-dose CT datasets of different regions with varying dose levels.The results demonstrate that our method outperforms the state-of-the-artunsupervised method and surpasses several supervised deep learning-basedmethods. Codes are available in https://github.com/DeepXuan/Dn-Dp.

Diffusion Probabilistic Priors for Zero-Shot Low-Dose CT Image Denoising