Dictionary-Free MRI PERK: Parameter Estimation via Regression with Kernels

@article{Nataraj2018DictionaryFreeMP,
  title={Dictionary-Free MRI PERK: Parameter Estimation via Regression with Kernels},
  author={Gopal Nataraj and Jon-Fredrik Nielsen and Clayton D. Scott and Jeffrey A. Fessler},
  journal={IEEE Transactions on Medical Imaging},
  year={2018},
  volume={37},
  pages={2103-2114}
}
  • Gopal Nataraj, J. Nielsen, +1 author J. Fessler
  • Published 6 October 2017
  • Mathematics, Medicine, Engineering, Physics, Computer Science
  • IEEE Transactions on Medical Imaging
This paper introduces a fast, general method for dictionary-free parameter estimation in quantitative magnetic resonance imaging (QMRI) parameter estimation via regression with kernels (PERK). PERK first uses prior distributions and the nonlinear MR signal model to simulate many parameter-measurement pairs. Inspired by machine learning, PERK then takes these parameter-measurement pairs as labeled training points and learns from them a nonlinear regression function using kernel functions and… 
Whole Brain Myelin Water Mapping in One Minute Using Tensor Dictionary Learning With Low-Rank Plus Sparse Regularization
TLDR
A novel tensor dictionary learning algorithm with low-rank and sparse regularization (TDLLS) is proposed to reconstruct the myelin water content in the brain from the undersampled data to accelerate the long acquisition.
Compressive MRI quantification using convex spatiotemporal priors and deep auto-encoders
TLDR
Tested on a number of datasets, the proposed scheme for recovering accurate and consistent quantitative information from novel and aggressively subsampled 2D/3D quantitative MRI acquisition protocols demonstrates effectiveness.
Compressive MRI quantification using convex spatiotemporal priors and deep encoder-decoder networks
TLDR
Tested on a number of datasets, the proposed scheme for recovering accurate and consistent quantitative information from novel and aggressively subsampled 2D/3D quantitative MRI acquisition protocols demonstrates effectiveness.
Fast, Precise Myelin Water Quantification using DESS MRI and Kernel Learning
Purpose: To investigate the feasibility of myelin water content quantification using fast dual-echo steady-state (DESS) scans and machine learning with kernels. Methods: We optimized combinations of
Complex-valued Deep Learning with Applications to Magnetic Resonance Image Synthesis
TLDR
This dissertation presents the deep learning methods to speed up quantitative MR fingerprinting and synthesize the standard contrast-weighted images directly from the same MR fingerprinted scan, and proposes an empirical noise model that allows neural networks to successfully synthesize quantitative images.
Deep model-based magnetic resonance parameter mapping network (DOPAMINE) for fast T1 mapping using variable flip angle method
TLDR
DOPAMINE exhibited quantitatively and qualitatively superior performance to that of conventional methods in reconstructing MR parameter maps from undersampled multi-channel k-space data and can thus reduce the scan time of quantitative MR parameter mapping that uses a VFA model.
Accelerated reconstruction of dictionary-based T2 relaxation maps based on dictionary compression and gradient descent search algorithms.
TLDR
This work presents two alternative post-processing approaches for accelerating the reconstruction of EMC-based T2 relaxation maps using Dictionary compression using principal component analysis (PCA) and gradient-descent search algorithm.
Optimization methods for MR image reconstruction (long version).
TLDR
This review paper summarizes several key models and optimization algorithms for MR image reconstruction, including both the type of methods that have FDA approval for clinical use, as well as more recent methods being considered in the research community that use data-adaptive regularizers.
Optimization methods for MR image reconstruction
TLDR
This review paper summarizes several key models and optimization algorithms for MR image reconstruction, including both the type of methods that have FDA approval for clinical use, as well as more recent methods being considered in the research community that use data-adaptive regularizers.
Myelin water imaging data analysis in less than one minute
TLDR
The time for accurate MWF calculation can be dramatically reduced to less than 1 min by the proposed NN, addressing one of the barriers facing future clinical feasibility of MWI.
...
1
2
3
...

References

SHOWING 1-10 OF 81 REFERENCES
Dictionary-free MRI parameter estimation via kernel ridge regression
TLDR
Kernel-based estimation is applied to quantify six parameters per voxel describing the steady-state magnetization dynamics of two water compartments from simulated data and achieves comparable mean-squared error as grid search, with dramatically reduced computation.
Maximum Likelihood Estimation of T 1 Relaxation Parameters Using VARPRO
INTRODUCTION Accurate T1 relaxation time parameter estimation is essential for a wide variety of quantitative MR applications; for example, obtaining accurate T1 values can be crucial in problems
Model-Based MR Parameter Mapping With Sparsity Constraints: Parameter Estimation and Performance Bounds
TLDR
The proposed method utilizes a formulation that integrates the explicit signal model with sparsity constraints on the model parameters, enabling direct estimation of the parameters of interest from highly undersampled, noisy k-space data.
Optimizing MR Scan Design for Model-Based ${T}_{1}$ , ${T}_{2}$ Estimation From Steady-State Sequences
TLDR
A systematic framework for selecting scan types and optimizing their respective parameters is described, based on a Cramér-Rao Bound (CRB)-inspired min-max optimization that finds scan parameter combinations that robustly enable precise object parameter estimation.
Nonlinear GRAPPA: A kernel approach to parallel MRI reconstruction
TLDR
Experimental results using phantom and in vivo data demonstrate that the proposed nonlinear GRAPPA method can significantly improve the reconstruction quality overGRAPPA and its state‐of‐the‐art derivatives.
Accelerated High-Dimensional MR Imaging With Sparse Sampling Using Low-Rank Tensors
TLDR
The usefulness of the proposed low-rank tensor based method for accelerated high-dimensional MR imaging using sparse sampling is demonstrated in MRI applications; it may also have applications beyond MRI.
A Kernel-Based Low-Rank (KLR) Model for Low-Dimensional Manifold Recovery in Highly Accelerated Dynamic MRI
TLDR
A novel algorithm with a kernel-based low-rank model generalizing the conventional low rank formulation is developed, which surpasses the conventionalLow rank-modeled approaches for dMRI.
Multiatlas Segmentation as Nonparametric Regression
TLDR
A novel theoretical framework to model and analyze the statistical characteristics of a wide range of segmentation methods that incorporate a database of label maps or atlases and shows that the expected error for large database sizes is well predicted by models learned on small databases.
Kernel Methods for Riemannian Analysis of Robust Descriptors of the Cerebral Cortex
TLDR
New regularized nonlinear methods for detecting group differences, using a Mercer kernel with an implicit lifting map to a reproducing kernel Hilbert space, and regression against clinical variables, using kernel density estimation are proposed.
Deep Learning for Rapid Sparse MR Fingerprinting Reconstruction
TLDR
Reconstruction of MRF data with a NN is accurate, 300 fold faster and more robust to noise and undersampling than conventional MRF dictionary matching.
...
1
2
3
4
5
...