Artificial neural networks for stiffness estimation in magnetic resonance elastography

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Purpose: To investigate the feasibility of using artificial neural networks to estimate stiffness from MR elastography (MRE) data. Methods: Artificial neural networks were fit using model-based training patterns to estimate stiffness from images of displacement using a patch size of ∼1cm in each dimension. These neural network inversions (NNIs) were then evaluated in a set of simulation experiments designed to investigate the effects of wave interference and noise on NNI accuracy. NNI was also tested in vivo, comparing NNI results against currently used methods. Results: In 4 simulation experiments, NNI performed as well or better than direct inversion (DI) for predicting the known stiffness of the data. Summary NNI results were also shown to be significantly correlated with DI results in the liver (R2=0.974) and in the brain (R2=0.915), and also correlated with established biological effects including fibrosis stage in the liver and age in the brain. Finally, repeatability error was lower in the brain using NNI compared to DI, and voxel-wise modeling using NNI stiffness maps detected larger effects than using DI maps with similar levels of smoothing. Conclusion: Artificial neural networks represent a new approach to inversion of MRE data. Summary results from NNI and DI are highly correlated and both are capable of detecting biologically relevant signals. Preliminary evidence suggests that NNI stiffness estimates may be more resistant to noise than an algebraic DI approach. Taken together, these results merit future investigation into NNIs to improve the estimation of stiffness in small regions.

Original languageEnglish (US)
JournalMagnetic Resonance in Medicine
DOIs
StateAccepted/In press - Jan 1 2017

Fingerprint

Elasticity Imaging Techniques
Noise
Brain
Liver
Fibrosis

Keywords

  • Inversion
  • MR elastography
  • Neural networks
  • Shear modulus
  • Stiffness

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

@article{e6356fc0a5f648a09a66f5a4345e3743,
title = "Artificial neural networks for stiffness estimation in magnetic resonance elastography",
abstract = "Purpose: To investigate the feasibility of using artificial neural networks to estimate stiffness from MR elastography (MRE) data. Methods: Artificial neural networks were fit using model-based training patterns to estimate stiffness from images of displacement using a patch size of ∼1cm in each dimension. These neural network inversions (NNIs) were then evaluated in a set of simulation experiments designed to investigate the effects of wave interference and noise on NNI accuracy. NNI was also tested in vivo, comparing NNI results against currently used methods. Results: In 4 simulation experiments, NNI performed as well or better than direct inversion (DI) for predicting the known stiffness of the data. Summary NNI results were also shown to be significantly correlated with DI results in the liver (R2=0.974) and in the brain (R2=0.915), and also correlated with established biological effects including fibrosis stage in the liver and age in the brain. Finally, repeatability error was lower in the brain using NNI compared to DI, and voxel-wise modeling using NNI stiffness maps detected larger effects than using DI maps with similar levels of smoothing. Conclusion: Artificial neural networks represent a new approach to inversion of MRE data. Summary results from NNI and DI are highly correlated and both are capable of detecting biologically relevant signals. Preliminary evidence suggests that NNI stiffness estimates may be more resistant to noise than an algebraic DI approach. Taken together, these results merit future investigation into NNIs to improve the estimation of stiffness in small regions.",
keywords = "Inversion, MR elastography, Neural networks, Shear modulus, Stiffness",
author = "Matthew Murphy and Armando Manduca and Trazasko, {Joshua D} and Glaser, {Kevin J.} and Huston, {John III} and Ehman, {Richard Lorne}",
year = "2017",
month = "1",
day = "1",
doi = "10.1002/mrm.27019",
language = "English (US)",
journal = "Magnetic Resonance in Medicine",
issn = "0740-3194",
publisher = "John Wiley and Sons Inc.",

}

TY - JOUR

T1 - Artificial neural networks for stiffness estimation in magnetic resonance elastography

AU - Murphy, Matthew

AU - Manduca, Armando

AU - Trazasko, Joshua D

AU - Glaser, Kevin J.

AU - Huston, John III

AU - Ehman, Richard Lorne

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Purpose: To investigate the feasibility of using artificial neural networks to estimate stiffness from MR elastography (MRE) data. Methods: Artificial neural networks were fit using model-based training patterns to estimate stiffness from images of displacement using a patch size of ∼1cm in each dimension. These neural network inversions (NNIs) were then evaluated in a set of simulation experiments designed to investigate the effects of wave interference and noise on NNI accuracy. NNI was also tested in vivo, comparing NNI results against currently used methods. Results: In 4 simulation experiments, NNI performed as well or better than direct inversion (DI) for predicting the known stiffness of the data. Summary NNI results were also shown to be significantly correlated with DI results in the liver (R2=0.974) and in the brain (R2=0.915), and also correlated with established biological effects including fibrosis stage in the liver and age in the brain. Finally, repeatability error was lower in the brain using NNI compared to DI, and voxel-wise modeling using NNI stiffness maps detected larger effects than using DI maps with similar levels of smoothing. Conclusion: Artificial neural networks represent a new approach to inversion of MRE data. Summary results from NNI and DI are highly correlated and both are capable of detecting biologically relevant signals. Preliminary evidence suggests that NNI stiffness estimates may be more resistant to noise than an algebraic DI approach. Taken together, these results merit future investigation into NNIs to improve the estimation of stiffness in small regions.

AB - Purpose: To investigate the feasibility of using artificial neural networks to estimate stiffness from MR elastography (MRE) data. Methods: Artificial neural networks were fit using model-based training patterns to estimate stiffness from images of displacement using a patch size of ∼1cm in each dimension. These neural network inversions (NNIs) were then evaluated in a set of simulation experiments designed to investigate the effects of wave interference and noise on NNI accuracy. NNI was also tested in vivo, comparing NNI results against currently used methods. Results: In 4 simulation experiments, NNI performed as well or better than direct inversion (DI) for predicting the known stiffness of the data. Summary NNI results were also shown to be significantly correlated with DI results in the liver (R2=0.974) and in the brain (R2=0.915), and also correlated with established biological effects including fibrosis stage in the liver and age in the brain. Finally, repeatability error was lower in the brain using NNI compared to DI, and voxel-wise modeling using NNI stiffness maps detected larger effects than using DI maps with similar levels of smoothing. Conclusion: Artificial neural networks represent a new approach to inversion of MRE data. Summary results from NNI and DI are highly correlated and both are capable of detecting biologically relevant signals. Preliminary evidence suggests that NNI stiffness estimates may be more resistant to noise than an algebraic DI approach. Taken together, these results merit future investigation into NNIs to improve the estimation of stiffness in small regions.

KW - Inversion

KW - MR elastography

KW - Neural networks

KW - Shear modulus

KW - Stiffness

UR - http://www.scopus.com/inward/record.url?scp=85036541230&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85036541230&partnerID=8YFLogxK

U2 - 10.1002/mrm.27019

DO - 10.1002/mrm.27019

M3 - Article

C2 - 29193306

AN - SCOPUS:85036541230

JO - Magnetic Resonance in Medicine

JF - Magnetic Resonance in Medicine

SN - 0740-3194

ER -