Deep learning‐based reconstruction on intensity‐inhomogeneous diffusion magnetic resonance imaging

Abstract Background Ultra high field diffusion magnetic resonance imaging (dMRI) provides diffusion‐weighted (DW) images with a high signal‐to‐noise ratio, but increases inhomogeneity, which affects the accuracy of dMRI metric reconstruction. Current methods for correcting inhomogeneity rarely consi...

Full description

Saved in:
Bibliographic Details
Main Authors: Zaimin Zhu, He Wang, Yong Liu, Fangrong Zong
Format: Article
Language:English
Published: Wiley 2024-12-01
Series:iRADIOLOGY
Subjects:
Online Access:https://doi.org/10.1002/ird3.100
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850101739326275584
author Zaimin Zhu
He Wang
Yong Liu
Fangrong Zong
author_facet Zaimin Zhu
He Wang
Yong Liu
Fangrong Zong
author_sort Zaimin Zhu
collection DOAJ
description Abstract Background Ultra high field diffusion magnetic resonance imaging (dMRI) provides diffusion‐weighted (DW) images with a high signal‐to‐noise ratio, but increases inhomogeneity, which affects the accuracy of dMRI metric reconstruction. Current methods for correcting inhomogeneity rarely consider the accuracy of the reconstructed dMRI metrics. Deep learning models for reconstructing metrics from dMRI signals typically assume that DW images have a homogeneous intensity. To address these challenges, we propose a deep learning model capable of directly reconstructing high‐accuracy dMRI metric maps from inhomogeneous DW images. Methods An attention‐based q‐space inhomogeneity‐resistant reconstruction network (qIRR‐Net) is proposed for the voxel‐wise reconstruction of diffusion tensor imaging and diffusion kurtosis imaging metrics. A training procedure based on data augmentation and consistency loss is introduced to ensure that the reconstruction results of qIRR‐Net are not affected by signal inhomogeneity. The 3T and 7T dMRI data from the Human Connectome Project are used for model training, testing, and evaluation. Results On the 3T dMRI data with simulated inhomogeneity, qIRR‐Net improves the peak signal‐to‐noise ratio by 5.39 and the structural similarity index measure by 0.18 compared with weighted linear least‐squares fitting. On the 7T dMRI data, the metric maps reconstructed by qIRR‐Net not only exhibit clearer tissue structures but also demonstrate greater stability compared with the weighted linear least‐squares results. Conclusions The proposed qIRR‐Net enables the accurate reconstruction of dMRI metrics from inhomogeneous DW images. This approach could potentially be expanded to obtain multiple artifact‐free metric maps from ultrahigh field dMRI for neuroscience research and neurology applications.
format Article
id doaj-art-bdb3f3bca76a461e8451b6e7b5bd0985
institution DOAJ
issn 2834-2860
2834-2879
language English
publishDate 2024-12-01
publisher Wiley
record_format Article
series iRADIOLOGY
spelling doaj-art-bdb3f3bca76a461e8451b6e7b5bd09852025-08-20T02:39:55ZengWileyiRADIOLOGY2834-28602834-28792024-12-012657158310.1002/ird3.100Deep learning‐based reconstruction on intensity‐inhomogeneous diffusion magnetic resonance imagingZaimin Zhu0He Wang1Yong Liu2Fangrong Zong3School of Artificial Intelligence Beijing University of Posts and Telecommunications Beijing ChinaInstitute of Science and Technology for Brain‐Inspired Intelligence Fudan University Shanghai ChinaSchool of Artificial Intelligence Beijing University of Posts and Telecommunications Beijing ChinaSchool of Artificial Intelligence Beijing University of Posts and Telecommunications Beijing ChinaAbstract Background Ultra high field diffusion magnetic resonance imaging (dMRI) provides diffusion‐weighted (DW) images with a high signal‐to‐noise ratio, but increases inhomogeneity, which affects the accuracy of dMRI metric reconstruction. Current methods for correcting inhomogeneity rarely consider the accuracy of the reconstructed dMRI metrics. Deep learning models for reconstructing metrics from dMRI signals typically assume that DW images have a homogeneous intensity. To address these challenges, we propose a deep learning model capable of directly reconstructing high‐accuracy dMRI metric maps from inhomogeneous DW images. Methods An attention‐based q‐space inhomogeneity‐resistant reconstruction network (qIRR‐Net) is proposed for the voxel‐wise reconstruction of diffusion tensor imaging and diffusion kurtosis imaging metrics. A training procedure based on data augmentation and consistency loss is introduced to ensure that the reconstruction results of qIRR‐Net are not affected by signal inhomogeneity. The 3T and 7T dMRI data from the Human Connectome Project are used for model training, testing, and evaluation. Results On the 3T dMRI data with simulated inhomogeneity, qIRR‐Net improves the peak signal‐to‐noise ratio by 5.39 and the structural similarity index measure by 0.18 compared with weighted linear least‐squares fitting. On the 7T dMRI data, the metric maps reconstructed by qIRR‐Net not only exhibit clearer tissue structures but also demonstrate greater stability compared with the weighted linear least‐squares results. Conclusions The proposed qIRR‐Net enables the accurate reconstruction of dMRI metrics from inhomogeneous DW images. This approach could potentially be expanded to obtain multiple artifact‐free metric maps from ultrahigh field dMRI for neuroscience research and neurology applications.https://doi.org/10.1002/ird3.100data augmentationdeep learningdiffusion MRI reconstructioninhomogeneity correctionultrahigh field
spellingShingle Zaimin Zhu
He Wang
Yong Liu
Fangrong Zong
Deep learning‐based reconstruction on intensity‐inhomogeneous diffusion magnetic resonance imaging
iRADIOLOGY
data augmentation
deep learning
diffusion MRI reconstruction
inhomogeneity correction
ultrahigh field
title Deep learning‐based reconstruction on intensity‐inhomogeneous diffusion magnetic resonance imaging
title_full Deep learning‐based reconstruction on intensity‐inhomogeneous diffusion magnetic resonance imaging
title_fullStr Deep learning‐based reconstruction on intensity‐inhomogeneous diffusion magnetic resonance imaging
title_full_unstemmed Deep learning‐based reconstruction on intensity‐inhomogeneous diffusion magnetic resonance imaging
title_short Deep learning‐based reconstruction on intensity‐inhomogeneous diffusion magnetic resonance imaging
title_sort deep learning based reconstruction on intensity inhomogeneous diffusion magnetic resonance imaging
topic data augmentation
deep learning
diffusion MRI reconstruction
inhomogeneity correction
ultrahigh field
url https://doi.org/10.1002/ird3.100
work_keys_str_mv AT zaiminzhu deeplearningbasedreconstructiononintensityinhomogeneousdiffusionmagneticresonanceimaging
AT hewang deeplearningbasedreconstructiononintensityinhomogeneousdiffusionmagneticresonanceimaging
AT yongliu deeplearningbasedreconstructiononintensityinhomogeneousdiffusionmagneticresonanceimaging
AT fangrongzong deeplearningbasedreconstructiononintensityinhomogeneousdiffusionmagneticresonanceimaging