Synthesizing T1 weighted MPRAGE image from multi echo GRE images via deep neural network

Kanghyun Ryu, Na Young Shin, Donghyun Kim, Yoonho Nam

Research output: Contribution to journalArticle

Abstract

For quantitative neuroimaging studies using multi-echo gradient echo (mGRE) images, additional T 1 -weighted magnetization prepared rapid gradient echo (MPRAGE) images are often acquired to supplement the insufficient morphometric information of mGRE for tissue segmentation which require lengthened scan time and additional processing such as image registration. This study investigated the feasibility of generating synthetic MPRAGE images from mGRE images using a deep convolutional neural network. Tissue segmentation results derived from the synthetic MPRAGE showed good agreement with those from actual MPRAGE (DSC = 0.882 ± 0.017). There was no statistically significant difference between the mean susceptibility values obtained with the regions of interest from synthetic and actual MPRAGEs and high correlation between the two measurements.

Original languageEnglish
JournalMagnetic Resonance Imaging
DOIs
Publication statusPublished - 2019 Jan 1

Fingerprint

Magnetization
Feasibility Studies
Neuroimaging
Tissue
Image registration
Neural networks
Deep neural networks
Processing

All Science Journal Classification (ASJC) codes

  • Biophysics
  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Cite this

@article{2db6b26fd5904303a95c30865ad8ef60,
title = "Synthesizing T1 weighted MPRAGE image from multi echo GRE images via deep neural network",
abstract = "For quantitative neuroimaging studies using multi-echo gradient echo (mGRE) images, additional T 1 -weighted magnetization prepared rapid gradient echo (MPRAGE) images are often acquired to supplement the insufficient morphometric information of mGRE for tissue segmentation which require lengthened scan time and additional processing such as image registration. This study investigated the feasibility of generating synthetic MPRAGE images from mGRE images using a deep convolutional neural network. Tissue segmentation results derived from the synthetic MPRAGE showed good agreement with those from actual MPRAGE (DSC = 0.882 ± 0.017). There was no statistically significant difference between the mean susceptibility values obtained with the regions of interest from synthetic and actual MPRAGEs and high correlation between the two measurements.",
author = "Kanghyun Ryu and Shin, {Na Young} and Donghyun Kim and Yoonho Nam",
year = "2019",
month = "1",
day = "1",
doi = "10.1016/j.mri.2019.04.002",
language = "English",
journal = "Magnetic Resonance Imaging",
issn = "0730-725X",
publisher = "Elsevier Inc.",

}

Synthesizing T1 weighted MPRAGE image from multi echo GRE images via deep neural network. / Ryu, Kanghyun; Shin, Na Young; Kim, Donghyun; Nam, Yoonho.

In: Magnetic Resonance Imaging, 01.01.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Synthesizing T1 weighted MPRAGE image from multi echo GRE images via deep neural network

AU - Ryu, Kanghyun

AU - Shin, Na Young

AU - Kim, Donghyun

AU - Nam, Yoonho

PY - 2019/1/1

Y1 - 2019/1/1

N2 - For quantitative neuroimaging studies using multi-echo gradient echo (mGRE) images, additional T 1 -weighted magnetization prepared rapid gradient echo (MPRAGE) images are often acquired to supplement the insufficient morphometric information of mGRE for tissue segmentation which require lengthened scan time and additional processing such as image registration. This study investigated the feasibility of generating synthetic MPRAGE images from mGRE images using a deep convolutional neural network. Tissue segmentation results derived from the synthetic MPRAGE showed good agreement with those from actual MPRAGE (DSC = 0.882 ± 0.017). There was no statistically significant difference between the mean susceptibility values obtained with the regions of interest from synthetic and actual MPRAGEs and high correlation between the two measurements.

AB - For quantitative neuroimaging studies using multi-echo gradient echo (mGRE) images, additional T 1 -weighted magnetization prepared rapid gradient echo (MPRAGE) images are often acquired to supplement the insufficient morphometric information of mGRE for tissue segmentation which require lengthened scan time and additional processing such as image registration. This study investigated the feasibility of generating synthetic MPRAGE images from mGRE images using a deep convolutional neural network. Tissue segmentation results derived from the synthetic MPRAGE showed good agreement with those from actual MPRAGE (DSC = 0.882 ± 0.017). There was no statistically significant difference between the mean susceptibility values obtained with the regions of interest from synthetic and actual MPRAGEs and high correlation between the two measurements.

UR - http://www.scopus.com/inward/record.url?scp=85063980909&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063980909&partnerID=8YFLogxK

U2 - 10.1016/j.mri.2019.04.002

DO - 10.1016/j.mri.2019.04.002

M3 - Article

AN - SCOPUS:85063980909

JO - Magnetic Resonance Imaging

JF - Magnetic Resonance Imaging

SN - 0730-725X

ER -