• Corpus ID: 235743110

Unsupervised learning of MRI tissue properties using MRI physics models

  title={Unsupervised learning of MRI tissue properties using MRI physics models},
  author={Divya Varadarajan and Katherine L. Bouman and Andr{\'e} J. W. van der Kouwe and Bruce R. Fischl and Adrian V. Dalca},
In neuroimaging, MRI tissue properties characterize underlying neurobiology, provide quantitative biomarkers for neurological disease detection and analysis, and can be used to synthesize arbitrary MRI contrasts. Estimating tissue properties from a single scan session using a protocol available on all clinical scanners promises to reduce scan time and cost, enable quantitative analysis in routine clinical scans and provide scan-independent biomarkers of disease. However, existing tissue… 

Figures and Tables from this paper

Breathing Freely: Self-supervised Liver T1rho Mapping from A Single T1rho-weighted Image
A self-supervised mapping method by taking only one T1rho-weighted image to do the mapping, which can achieve better mapping performance than the traditional fitting method, particularly in free-breathing scenarios.


Rapid MR relaxometry using deep learning: An overview of current techniques and emerging trends
The goal of this paper is to discuss the applications of deep learning for rapid MR relaxometry and to review emerging deep‐learning‐based techniques that can be applied to improve MR Relaxometry in terms of imaging speed, image quality, and quantification robustness.
Deep Learning for Fast and Spatially Constrained Tissue Quantification From Highly Accelerated Data in Magnetic Resonance Fingerprinting
A spatially constrained quantification method that uses the signals at multiple neighboring pixels to better estimate tissue properties at the central pixel is proposed and a unique two-step deep learning model is designed that learns the mapping from the observed signals to the desired properties for tissue quantification.
Advances in Imaging Brain Metabolism.
Metabolism is central to neuroimaging because it can reveal pathways by which neuronal and glial cells use nutrients to fuel their growth and function. We focus on advanced magnetic resonance imaging
Accelerated Dynamic MRI Exploiting Sparsity and Low-Rank Structure: k-t SLR
A novel algorithm to reconstruct dynamic magnetic resonance imaging data from under-sampled k-t space data using the compact representation of the data in the Karhunen Louve transform (KLT) domain to exploit the correlations in the dataset.
Model-Based Iterative Reconstruction for Radial Fast Spin-Echo MRI
Simulations and experimental results demonstrate that the proposed reconstruction method directly yields a spin-density and relaxivity map from only a single radial data set that is neither affected by the typical artifacts from TE mixing, nor by streaking artifacts from the incomplete k-space coverage at individual echo times.
In vivo determination of T1 and T2 in the brain of patients with severe but stable multiple sclerosis
T1 and T2 in plaques were found to cover a wide range, which could be explained only by inherent biophysical dis similarity of the plaques, possibly due to differences in disease activity, edema and gliosis.
Accelerating MR parameter mapping using sparsity‐promoting regularization in parametric dimension
The proposed p‐CS regularization strategy uses smoothness of signal evolution in the parametric dimension within compressed sensing framework (p‐CS) to provide accurate and precise estimation of parametric maps from undersampled data.
Quantitative mapping of T1 and T2* discloses nigral and brainstem pathology in early Parkinson's disease
Multidimensional correlation spectroscopic imaging of exponential decays: From theoretical principles to in vivo human applications
This article presents an overview of a recent multidimensional correlation spectroscopic imaging approach to the problem of relaxation or diffusion MR signal decays, and demonstrates an initial proof‐of‐principle application of this kind of approach to in vivo human MRI experiments.