An example-based brain MRI simulation framework

Qing He, Snehashis Roy, Amod Jog, Dzung L. Pham

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The simulation of magnetic resonance (MR) images plays an important role in the validation of image analysis algorithms such as image segmentation, due to lack of sufficient ground truth in real MR images. Previous work on MRI simulation has focused on explicitly modeling the MR image formation process. However, because of the overwhelming complexity of MR acquisition these simulations must involve simplifications and approximations that can result in visually unrealistic simulated images. In this work, we describe an example-based simulation framework, which uses an "atlas"? consisting of an MR image and its anatomical models derived from the hard segmentation. The relationships between the MR image intensities and its anatomical models are learned using a patch-based regression that implicitly models the physics of the MR image formation. Given the anatomical models of a new brain, a new MR image can be simulated using the learned regression. This approach has been extended to also simulate intensity inhomogeneity artifacts based on the statistical model of training data. Results show that the example based MRI simulation method is capable of simulating different image contrasts and is robust to different choices of atlas. The simulated images resemble real MR images more than simulations produced by a physics-based model.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2015: Physics of Medical Imaging
PublisherSPIE
Volume9412
ISBN (Print)9781628415025
DOIs
StatePublished - 2015
Externally publishedYes
EventMedical Imaging 2015: Physics of Medical Imaging - Orlando, United States
Duration: Feb 22 2015Feb 25 2015

Other

OtherMedical Imaging 2015: Physics of Medical Imaging
CountryUnited States
CityOrlando
Period2/22/152/25/15

Fingerprint

Magnetic resonance
Magnetic resonance imaging
brain
Brain
Magnetic Resonance Spectroscopy
magnetic resonance
Anatomic Models
simulation
Atlases
Physics
Image processing
regression analysis
Statistical Models
physics
Image segmentation
ground truth
Artifacts
Image analysis
image contrast
simplification

Keywords

  • Brain MRI simulation
  • Example based method
  • Inhomogeneity field
  • Regression ensemble

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics
  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Radiology Nuclear Medicine and imaging

Cite this

He, Q., Roy, S., Jog, A., & Pham, D. L. (2015). An example-based brain MRI simulation framework. In Medical Imaging 2015: Physics of Medical Imaging (Vol. 9412). [94120P] SPIE. https://doi.org/10.1117/12.2075687

An example-based brain MRI simulation framework. / He, Qing; Roy, Snehashis; Jog, Amod; Pham, Dzung L.

Medical Imaging 2015: Physics of Medical Imaging. Vol. 9412 SPIE, 2015. 94120P.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

He, Q, Roy, S, Jog, A & Pham, DL 2015, An example-based brain MRI simulation framework. in Medical Imaging 2015: Physics of Medical Imaging. vol. 9412, 94120P, SPIE, Medical Imaging 2015: Physics of Medical Imaging, Orlando, United States, 2/22/15. https://doi.org/10.1117/12.2075687
He Q, Roy S, Jog A, Pham DL. An example-based brain MRI simulation framework. In Medical Imaging 2015: Physics of Medical Imaging. Vol. 9412. SPIE. 2015. 94120P https://doi.org/10.1117/12.2075687
He, Qing ; Roy, Snehashis ; Jog, Amod ; Pham, Dzung L. / An example-based brain MRI simulation framework. Medical Imaging 2015: Physics of Medical Imaging. Vol. 9412 SPIE, 2015.
@inproceedings{3d9f3219e45643749823f1ad900610dc,
title = "An example-based brain MRI simulation framework",
abstract = "The simulation of magnetic resonance (MR) images plays an important role in the validation of image analysis algorithms such as image segmentation, due to lack of sufficient ground truth in real MR images. Previous work on MRI simulation has focused on explicitly modeling the MR image formation process. However, because of the overwhelming complexity of MR acquisition these simulations must involve simplifications and approximations that can result in visually unrealistic simulated images. In this work, we describe an example-based simulation framework, which uses an {"}atlas{"}? consisting of an MR image and its anatomical models derived from the hard segmentation. The relationships between the MR image intensities and its anatomical models are learned using a patch-based regression that implicitly models the physics of the MR image formation. Given the anatomical models of a new brain, a new MR image can be simulated using the learned regression. This approach has been extended to also simulate intensity inhomogeneity artifacts based on the statistical model of training data. Results show that the example based MRI simulation method is capable of simulating different image contrasts and is robust to different choices of atlas. The simulated images resemble real MR images more than simulations produced by a physics-based model.",
keywords = "Brain MRI simulation, Example based method, Inhomogeneity field, Regression ensemble",
author = "Qing He and Snehashis Roy and Amod Jog and Pham, {Dzung L.}",
year = "2015",
doi = "10.1117/12.2075687",
language = "English (US)",
isbn = "9781628415025",
volume = "9412",
booktitle = "Medical Imaging 2015: Physics of Medical Imaging",
publisher = "SPIE",

}

TY - GEN

T1 - An example-based brain MRI simulation framework

AU - He, Qing

AU - Roy, Snehashis

AU - Jog, Amod

AU - Pham, Dzung L.

PY - 2015

Y1 - 2015

N2 - The simulation of magnetic resonance (MR) images plays an important role in the validation of image analysis algorithms such as image segmentation, due to lack of sufficient ground truth in real MR images. Previous work on MRI simulation has focused on explicitly modeling the MR image formation process. However, because of the overwhelming complexity of MR acquisition these simulations must involve simplifications and approximations that can result in visually unrealistic simulated images. In this work, we describe an example-based simulation framework, which uses an "atlas"? consisting of an MR image and its anatomical models derived from the hard segmentation. The relationships between the MR image intensities and its anatomical models are learned using a patch-based regression that implicitly models the physics of the MR image formation. Given the anatomical models of a new brain, a new MR image can be simulated using the learned regression. This approach has been extended to also simulate intensity inhomogeneity artifacts based on the statistical model of training data. Results show that the example based MRI simulation method is capable of simulating different image contrasts and is robust to different choices of atlas. The simulated images resemble real MR images more than simulations produced by a physics-based model.

AB - The simulation of magnetic resonance (MR) images plays an important role in the validation of image analysis algorithms such as image segmentation, due to lack of sufficient ground truth in real MR images. Previous work on MRI simulation has focused on explicitly modeling the MR image formation process. However, because of the overwhelming complexity of MR acquisition these simulations must involve simplifications and approximations that can result in visually unrealistic simulated images. In this work, we describe an example-based simulation framework, which uses an "atlas"? consisting of an MR image and its anatomical models derived from the hard segmentation. The relationships between the MR image intensities and its anatomical models are learned using a patch-based regression that implicitly models the physics of the MR image formation. Given the anatomical models of a new brain, a new MR image can be simulated using the learned regression. This approach has been extended to also simulate intensity inhomogeneity artifacts based on the statistical model of training data. Results show that the example based MRI simulation method is capable of simulating different image contrasts and is robust to different choices of atlas. The simulated images resemble real MR images more than simulations produced by a physics-based model.

KW - Brain MRI simulation

KW - Example based method

KW - Inhomogeneity field

KW - Regression ensemble

UR - http://www.scopus.com/inward/record.url?scp=84943338809&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84943338809&partnerID=8YFLogxK

U2 - 10.1117/12.2075687

DO - 10.1117/12.2075687

M3 - Conference contribution

SN - 9781628415025

VL - 9412

BT - Medical Imaging 2015: Physics of Medical Imaging

PB - SPIE

ER -