3D MRI image super-resolution for brain combining rigid and large diffeomorphic registration

Zifei Liang, Xiaohai He, Qizhi Teng, Dan Wu, Lingbo Qing

Research output: Contribution to journalArticle

Abstract

Most of the recent leading multiple magnetic resonance imaging (MRI) super-resolution techniques for brain are limited to rigid motion. In this study, the authors aim to develop a super-resolution technique with diffeomorphism mainly for longitudinal brain MRI data. For the images from different time slots, unpredicted deformation may occur. In previous studies, sole rigid registration or traditional non-rigid registration has been frequently used to achieve multi-plane super-resolution. However, non-rigid motion of two brains from different time slots is difficult to model, since brain contains a wealth of complex structure such as the cerebral cortex. In order to address such problem, rigid and large diffeomorphic registration has been embedded into their super-resolution framework. In addition, many previous researchers use L2 norm to achieve super-resolution framework. In this work, L1 norm minimisation and regularisation based on a bilateral prior are adopted. These operations ensure its robustness to the assumed model of data and noise. Their approach is evaluated using Alzheimer datasets from seven different resolutions. Results show that their reconstructions have advantages over rigid and conventional non-rigid registration-based super-resolution, in terms of the root-mean-square error and structure similarity. Furthermore, their reconstruction results improve the precision of brain automatic segmentation.

Original languageEnglish (US)
Pages (from-to)1291-1301
Number of pages11
JournalIET Image Processing
Volume11
Issue number12
DOIs
StatePublished - Dec 1 2017

Fingerprint

Magnetic resonance
Brain
Imaging techniques
Brain models
Mean square error

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering

Cite this

3D MRI image super-resolution for brain combining rigid and large diffeomorphic registration. / Liang, Zifei; He, Xiaohai; Teng, Qizhi; Wu, Dan; Qing, Lingbo.

In: IET Image Processing, Vol. 11, No. 12, 01.12.2017, p. 1291-1301.

Research output: Contribution to journalArticle

Liang, Zifei ; He, Xiaohai ; Teng, Qizhi ; Wu, Dan ; Qing, Lingbo. / 3D MRI image super-resolution for brain combining rigid and large diffeomorphic registration. In: IET Image Processing. 2017 ; Vol. 11, No. 12. pp. 1291-1301.
@article{d70d5eed58db4948b4d8e986e19d0f5f,
title = "3D MRI image super-resolution for brain combining rigid and large diffeomorphic registration",
abstract = "Most of the recent leading multiple magnetic resonance imaging (MRI) super-resolution techniques for brain are limited to rigid motion. In this study, the authors aim to develop a super-resolution technique with diffeomorphism mainly for longitudinal brain MRI data. For the images from different time slots, unpredicted deformation may occur. In previous studies, sole rigid registration or traditional non-rigid registration has been frequently used to achieve multi-plane super-resolution. However, non-rigid motion of two brains from different time slots is difficult to model, since brain contains a wealth of complex structure such as the cerebral cortex. In order to address such problem, rigid and large diffeomorphic registration has been embedded into their super-resolution framework. In addition, many previous researchers use L2 norm to achieve super-resolution framework. In this work, L1 norm minimisation and regularisation based on a bilateral prior are adopted. These operations ensure its robustness to the assumed model of data and noise. Their approach is evaluated using Alzheimer datasets from seven different resolutions. Results show that their reconstructions have advantages over rigid and conventional non-rigid registration-based super-resolution, in terms of the root-mean-square error and structure similarity. Furthermore, their reconstruction results improve the precision of brain automatic segmentation.",
author = "Zifei Liang and Xiaohai He and Qizhi Teng and Dan Wu and Lingbo Qing",
year = "2017",
month = "12",
day = "1",
doi = "10.1049/iet-ipr.2017.0517",
language = "English (US)",
volume = "11",
pages = "1291--1301",
journal = "IET Image Processing",
issn = "1751-9659",
publisher = "Institution of Engineering and Technology",
number = "12",

}

TY - JOUR

T1 - 3D MRI image super-resolution for brain combining rigid and large diffeomorphic registration

AU - Liang, Zifei

AU - He, Xiaohai

AU - Teng, Qizhi

AU - Wu, Dan

AU - Qing, Lingbo

PY - 2017/12/1

Y1 - 2017/12/1

N2 - Most of the recent leading multiple magnetic resonance imaging (MRI) super-resolution techniques for brain are limited to rigid motion. In this study, the authors aim to develop a super-resolution technique with diffeomorphism mainly for longitudinal brain MRI data. For the images from different time slots, unpredicted deformation may occur. In previous studies, sole rigid registration or traditional non-rigid registration has been frequently used to achieve multi-plane super-resolution. However, non-rigid motion of two brains from different time slots is difficult to model, since brain contains a wealth of complex structure such as the cerebral cortex. In order to address such problem, rigid and large diffeomorphic registration has been embedded into their super-resolution framework. In addition, many previous researchers use L2 norm to achieve super-resolution framework. In this work, L1 norm minimisation and regularisation based on a bilateral prior are adopted. These operations ensure its robustness to the assumed model of data and noise. Their approach is evaluated using Alzheimer datasets from seven different resolutions. Results show that their reconstructions have advantages over rigid and conventional non-rigid registration-based super-resolution, in terms of the root-mean-square error and structure similarity. Furthermore, their reconstruction results improve the precision of brain automatic segmentation.

AB - Most of the recent leading multiple magnetic resonance imaging (MRI) super-resolution techniques for brain are limited to rigid motion. In this study, the authors aim to develop a super-resolution technique with diffeomorphism mainly for longitudinal brain MRI data. For the images from different time slots, unpredicted deformation may occur. In previous studies, sole rigid registration or traditional non-rigid registration has been frequently used to achieve multi-plane super-resolution. However, non-rigid motion of two brains from different time slots is difficult to model, since brain contains a wealth of complex structure such as the cerebral cortex. In order to address such problem, rigid and large diffeomorphic registration has been embedded into their super-resolution framework. In addition, many previous researchers use L2 norm to achieve super-resolution framework. In this work, L1 norm minimisation and regularisation based on a bilateral prior are adopted. These operations ensure its robustness to the assumed model of data and noise. Their approach is evaluated using Alzheimer datasets from seven different resolutions. Results show that their reconstructions have advantages over rigid and conventional non-rigid registration-based super-resolution, in terms of the root-mean-square error and structure similarity. Furthermore, their reconstruction results improve the precision of brain automatic segmentation.

UR - http://www.scopus.com/inward/record.url?scp=85039060983&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85039060983&partnerID=8YFLogxK

U2 - 10.1049/iet-ipr.2017.0517

DO - 10.1049/iet-ipr.2017.0517

M3 - Article

VL - 11

SP - 1291

EP - 1301

JO - IET Image Processing

JF - IET Image Processing

SN - 1751-9659

IS - 12

ER -