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

The noise in diffusion-weighted images (DWIs) decreases the accuracy andprecision of diffusion tensor magnetic resonance imaging (DTI) derivedmicrostructural parameters and leads to prolonged acquisition time forachieving improved signal-to-noise ratio (SNR). Deep learning-based imagedenoising using convolutional neural networks (CNNs) has superior performancebut often requires additional high-SNR data for supervising the training ofCNNs, which reduces the practical feasibility. We develop a self-superviseddeep learning-based method entitled "SDnDTI" for denoising DTI data, which doesnot require additional high-SNR data for training. Specifically, SDnDTI dividesmulti-directional DTI data into many subsets, each consisting of six DWIvolumes along optimally chosen diffusion-encoding directions that are robust tonoise for the tensor fitting, and then synthesizes DWI volumes along allacquired directions from the diffusion tensors fitted using each subset of thedata as the input data of CNNs. On the other hand, SDnDTI synthesizes DWIvolumes along acquired diffusion-encoding directions with higher SNR from thediffusion tensors fitted using all acquired data as the training target. SDnDTIremoves noise from each subset of synthesized DWI volumes using a deep3-dimensional CNN to match the quality of the cleaner target DWI volumes andachieves even higher SNR by averaging all subsets of denoised data. Thedenoising efficacy of SDnDTI is demonstrated on two datasets provided by theHuman Connectome Project (HCP) and the Lifespan HCP in Aging. The SDnDTIresults preserve image sharpness and textural details and substantially improveupon those from the raw data. The results of SDnDTI are comparable to thosefrom supervised learning-based denoising and outperform those fromstate-of-the-art conventional denoising algorithms including BM4D, AONLM andMPPCA.

SDnDTI: Self-supervised deep learning-based denoising for diffusion tensor MRI