Data-driven synthetic MRI FLAIR artifact correction via deep neural network

Kanghyun Ryu, Yoonho Nam, Sung Min Gho, Jinhee Jang, Ho Joon Lee, Jihoon Cha, Hye Jin Baek, Jiyong Park, Dong Hyun Kim

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Background: FLAIR (fluid attenuated inversion recovery) imaging via synthetic MRI methods leads to artifacts in the brain, which can cause diagnostic limitations. The main sources of the artifacts are attributed to the partial volume effect and flow, which are difficult to correct by analytical modeling. In this study, a deep learning (DL)-based synthetic FLAIR method was developed, which does not require analytical modeling of the signal. Purpose: To correct artifacts in synthetic FLAIR using a DL method. Study Type: Retrospective. Subjects: A total of 80 subjects with clinical indications (60.6 ± 16.7 years, 38 males, 42 females) were divided into three groups: a training set (56 subjects, 62.1 ± 14.8 years, 25 males, 31 females), a validation set (1 subject, 62 years, male), and the testing set (23 subjects, 57.3 ± 20.4 years, 13 males, 10 females). Field Strength/Sequence: 3 T MRI using a multiple-dynamic multiple-echo acquisition (MDME) sequence for synthetic MRI and a conventional FLAIR sequence. Assessment: Normalized root mean square (NRMSE) and structural similarity (SSIM) were computed for uncorrected synthetic FLAIR and DL-corrected FLAIR. In addition, three neuroradiologists scored the three FLAIR datasets blindly, evaluating image quality and artifacts for sulci/periventricular and intraventricular/cistern space regions. Statistical Tests: Pairwise Student's t-tests and a Wilcoxon test were performed. Results: For quantitative assessment, NRMSE improved from 4.2% to 2.9% (P < 0.0001) and SSIM improved from 0.85 to 0.93 (P < 0.0001). Additionally, NRMSE values significantly improved from 1.58% to 1.26% (P < 0.001), 3.1% to 1.5% (P < 0.0001), and 2.7% to 1.4% (P < 0.0001) in white matter, gray matter, and cerebral spinal fluid (CSF) regions, respectively, when using DL-corrected FLAIR. For qualitative assessment, DL correction achieved improved overall quality, fewer artifacts in sulci and periventricular regions, and in intraventricular and cistern space regions. Data Conclusion: The DL approach provides a promising method to correct artifacts in synthetic FLAIR. Level of Evidence: 4. Technical Efficacy: Stage 1. J. Magn. Reson. Imaging 2019;50:1413–1423.

Original languageEnglish
Pages (from-to)1413-1423
Number of pages11
JournalJournal of Magnetic Resonance Imaging
Volume50
Issue number5
DOIs
Publication statusPublished - 2019 Nov 1

Fingerprint

Artifacts
Learning
Sequence Inversion
Retrospective Studies
Students
Brain

All Science Journal Classification (ASJC) codes

  • Radiology Nuclear Medicine and imaging

Cite this

Ryu, Kanghyun ; Nam, Yoonho ; Gho, Sung Min ; Jang, Jinhee ; Lee, Ho Joon ; Cha, Jihoon ; Baek, Hye Jin ; Park, Jiyong ; Kim, Dong Hyun. / Data-driven synthetic MRI FLAIR artifact correction via deep neural network. In: Journal of Magnetic Resonance Imaging. 2019 ; Vol. 50, No. 5. pp. 1413-1423.
@article{930923693c3a458fac16475527ad73dd,
title = "Data-driven synthetic MRI FLAIR artifact correction via deep neural network",
abstract = "Background: FLAIR (fluid attenuated inversion recovery) imaging via synthetic MRI methods leads to artifacts in the brain, which can cause diagnostic limitations. The main sources of the artifacts are attributed to the partial volume effect and flow, which are difficult to correct by analytical modeling. In this study, a deep learning (DL)-based synthetic FLAIR method was developed, which does not require analytical modeling of the signal. Purpose: To correct artifacts in synthetic FLAIR using a DL method. Study Type: Retrospective. Subjects: A total of 80 subjects with clinical indications (60.6 ± 16.7 years, 38 males, 42 females) were divided into three groups: a training set (56 subjects, 62.1 ± 14.8 years, 25 males, 31 females), a validation set (1 subject, 62 years, male), and the testing set (23 subjects, 57.3 ± 20.4 years, 13 males, 10 females). Field Strength/Sequence: 3 T MRI using a multiple-dynamic multiple-echo acquisition (MDME) sequence for synthetic MRI and a conventional FLAIR sequence. Assessment: Normalized root mean square (NRMSE) and structural similarity (SSIM) were computed for uncorrected synthetic FLAIR and DL-corrected FLAIR. In addition, three neuroradiologists scored the three FLAIR datasets blindly, evaluating image quality and artifacts for sulci/periventricular and intraventricular/cistern space regions. Statistical Tests: Pairwise Student's t-tests and a Wilcoxon test were performed. Results: For quantitative assessment, NRMSE improved from 4.2{\%} to 2.9{\%} (P < 0.0001) and SSIM improved from 0.85 to 0.93 (P < 0.0001). Additionally, NRMSE values significantly improved from 1.58{\%} to 1.26{\%} (P < 0.001), 3.1{\%} to 1.5{\%} (P < 0.0001), and 2.7{\%} to 1.4{\%} (P < 0.0001) in white matter, gray matter, and cerebral spinal fluid (CSF) regions, respectively, when using DL-corrected FLAIR. For qualitative assessment, DL correction achieved improved overall quality, fewer artifacts in sulci and periventricular regions, and in intraventricular and cistern space regions. Data Conclusion: The DL approach provides a promising method to correct artifacts in synthetic FLAIR. Level of Evidence: 4. Technical Efficacy: Stage 1. J. Magn. Reson. Imaging 2019;50:1413–1423.",
author = "Kanghyun Ryu and Yoonho Nam and Gho, {Sung Min} and Jinhee Jang and Lee, {Ho Joon} and Jihoon Cha and Baek, {Hye Jin} and Jiyong Park and Kim, {Dong Hyun}",
year = "2019",
month = "11",
day = "1",
doi = "10.1002/jmri.26712",
language = "English",
volume = "50",
pages = "1413--1423",
journal = "Journal of Magnetic Resonance Imaging",
issn = "1053-1807",
publisher = "John Wiley and Sons Inc.",
number = "5",

}

Ryu, K, Nam, Y, Gho, SM, Jang, J, Lee, HJ, Cha, J, Baek, HJ, Park, J & Kim, DH 2019, 'Data-driven synthetic MRI FLAIR artifact correction via deep neural network', Journal of Magnetic Resonance Imaging, vol. 50, no. 5, pp. 1413-1423. https://doi.org/10.1002/jmri.26712

Data-driven synthetic MRI FLAIR artifact correction via deep neural network. / Ryu, Kanghyun; Nam, Yoonho; Gho, Sung Min; Jang, Jinhee; Lee, Ho Joon; Cha, Jihoon; Baek, Hye Jin; Park, Jiyong; Kim, Dong Hyun.

In: Journal of Magnetic Resonance Imaging, Vol. 50, No. 5, 01.11.2019, p. 1413-1423.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Data-driven synthetic MRI FLAIR artifact correction via deep neural network

AU - Ryu, Kanghyun

AU - Nam, Yoonho

AU - Gho, Sung Min

AU - Jang, Jinhee

AU - Lee, Ho Joon

AU - Cha, Jihoon

AU - Baek, Hye Jin

AU - Park, Jiyong

AU - Kim, Dong Hyun

PY - 2019/11/1

Y1 - 2019/11/1

N2 - Background: FLAIR (fluid attenuated inversion recovery) imaging via synthetic MRI methods leads to artifacts in the brain, which can cause diagnostic limitations. The main sources of the artifacts are attributed to the partial volume effect and flow, which are difficult to correct by analytical modeling. In this study, a deep learning (DL)-based synthetic FLAIR method was developed, which does not require analytical modeling of the signal. Purpose: To correct artifacts in synthetic FLAIR using a DL method. Study Type: Retrospective. Subjects: A total of 80 subjects with clinical indications (60.6 ± 16.7 years, 38 males, 42 females) were divided into three groups: a training set (56 subjects, 62.1 ± 14.8 years, 25 males, 31 females), a validation set (1 subject, 62 years, male), and the testing set (23 subjects, 57.3 ± 20.4 years, 13 males, 10 females). Field Strength/Sequence: 3 T MRI using a multiple-dynamic multiple-echo acquisition (MDME) sequence for synthetic MRI and a conventional FLAIR sequence. Assessment: Normalized root mean square (NRMSE) and structural similarity (SSIM) were computed for uncorrected synthetic FLAIR and DL-corrected FLAIR. In addition, three neuroradiologists scored the three FLAIR datasets blindly, evaluating image quality and artifacts for sulci/periventricular and intraventricular/cistern space regions. Statistical Tests: Pairwise Student's t-tests and a Wilcoxon test were performed. Results: For quantitative assessment, NRMSE improved from 4.2% to 2.9% (P < 0.0001) and SSIM improved from 0.85 to 0.93 (P < 0.0001). Additionally, NRMSE values significantly improved from 1.58% to 1.26% (P < 0.001), 3.1% to 1.5% (P < 0.0001), and 2.7% to 1.4% (P < 0.0001) in white matter, gray matter, and cerebral spinal fluid (CSF) regions, respectively, when using DL-corrected FLAIR. For qualitative assessment, DL correction achieved improved overall quality, fewer artifacts in sulci and periventricular regions, and in intraventricular and cistern space regions. Data Conclusion: The DL approach provides a promising method to correct artifacts in synthetic FLAIR. Level of Evidence: 4. Technical Efficacy: Stage 1. J. Magn. Reson. Imaging 2019;50:1413–1423.

AB - Background: FLAIR (fluid attenuated inversion recovery) imaging via synthetic MRI methods leads to artifacts in the brain, which can cause diagnostic limitations. The main sources of the artifacts are attributed to the partial volume effect and flow, which are difficult to correct by analytical modeling. In this study, a deep learning (DL)-based synthetic FLAIR method was developed, which does not require analytical modeling of the signal. Purpose: To correct artifacts in synthetic FLAIR using a DL method. Study Type: Retrospective. Subjects: A total of 80 subjects with clinical indications (60.6 ± 16.7 years, 38 males, 42 females) were divided into three groups: a training set (56 subjects, 62.1 ± 14.8 years, 25 males, 31 females), a validation set (1 subject, 62 years, male), and the testing set (23 subjects, 57.3 ± 20.4 years, 13 males, 10 females). Field Strength/Sequence: 3 T MRI using a multiple-dynamic multiple-echo acquisition (MDME) sequence for synthetic MRI and a conventional FLAIR sequence. Assessment: Normalized root mean square (NRMSE) and structural similarity (SSIM) were computed for uncorrected synthetic FLAIR and DL-corrected FLAIR. In addition, three neuroradiologists scored the three FLAIR datasets blindly, evaluating image quality and artifacts for sulci/periventricular and intraventricular/cistern space regions. Statistical Tests: Pairwise Student's t-tests and a Wilcoxon test were performed. Results: For quantitative assessment, NRMSE improved from 4.2% to 2.9% (P < 0.0001) and SSIM improved from 0.85 to 0.93 (P < 0.0001). Additionally, NRMSE values significantly improved from 1.58% to 1.26% (P < 0.001), 3.1% to 1.5% (P < 0.0001), and 2.7% to 1.4% (P < 0.0001) in white matter, gray matter, and cerebral spinal fluid (CSF) regions, respectively, when using DL-corrected FLAIR. For qualitative assessment, DL correction achieved improved overall quality, fewer artifacts in sulci and periventricular regions, and in intraventricular and cistern space regions. Data Conclusion: The DL approach provides a promising method to correct artifacts in synthetic FLAIR. Level of Evidence: 4. Technical Efficacy: Stage 1. J. Magn. Reson. Imaging 2019;50:1413–1423.

UR - http://www.scopus.com/inward/record.url?scp=85073314131&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073314131&partnerID=8YFLogxK

U2 - 10.1002/jmri.26712

DO - 10.1002/jmri.26712

M3 - Article

C2 - 30884007

AN - SCOPUS:85073314131

VL - 50

SP - 1413

EP - 1423

JO - Journal of Magnetic Resonance Imaging

JF - Journal of Magnetic Resonance Imaging

SN - 1053-1807

IS - 5

ER -