Demonstrate a novel fast method for reconstruction of multi-dimensional MR fingerprinting (MRF) data using deep learning methods.A neural network (NN) is defined using the TensorFlow framework and trained on simulated MRF data computed with the extended phase graph formalism. The NN reconstruction accuracy for noiseless and noisy data is compared to conventional MRF template matching as a function of training data size and is quantified in simulated numerical brain phantom data and International Society for Magnetic Resonance in Medicine/National Institute of Standards and Technology phantom data measured on 1.5T and 3T scanners with an optimized MRF EPI and MRF fast imaging with steady state precession (FISP) sequences with spiral readout. The utility of the method is demonstrated in a healthy subject in vivo at 1.5T.Network training required 10 to 74 minutes; once trained, data reconstruction required approximately 10 ms for the MRF EPI and 76 ms for the MRF FISP sequence. Reconstruction of simulated, noiseless brain data using the NN resulted in a RMS error (RMSE) of 2.6 ms for T1 and 1.9 ms for T2 . The reconstruction error in the presence of noise was less than 10% for both T1 and T2 for SNR greater than 25 dB. Phantom measurements yielded good agreement (R2  = 0.99/0.99 for MRF EPI T1 /T2 and 0.94/0.98 for MRF FISP T1 /T2 ) between the T1 and T2 estimated by the NN and reference values from the International Society for Magnetic Resonance in Medicine/National Institute of Standards and Technology phantom.Reconstruction of MRF data with a NN is accurate, 300- to 5000-fold faster, and more robust to noise and dictionary undersampling than conventional MRF dictionary-matching.

MR fingerprinting Deep RecOnstruction NEtwork (DRONE)

The incoherence between measurement and sparsifying transform matrices and the restricted isometry property (RIP) of measurement matrix are two of the key factors in determining the performance of compressive sensing (CS). In CS-MRI, the randomly under-sampled Fourier matrix is used as the measurement matrix and the wavelet transform is usually used as sparsifying transform matrix. However, the incoherence between the randomly under-sampled Fourier matrix and the wavelet matrix is not optimal, which can deteriorate the performance of CS-MRI. Using the mathematical result that noiselets are maximally incoherent with wavelets, this paper introduces the noiselet unitary bases as the measurement matrix to improve the incoherence and RIP in CS-MRI. Based on an empirical RIP analysis that compares the multichannel noiselet and multichannel Fourier measurement matrices in CS-MRI, we propose a multichannel compressive sensing (MCS) framework to take the advantage of multichannel data acquisition used in MRI scanners. Simulations are presented in the MCS framework to compare the performance of noiselet encoding reconstructions and Fourier encoding reconstructions at different acceleration factors. The comparisons indicate that multichannel noiselet measurement matrix has better RIP than that of its Fourier counterpart, and that noiselet encoded MCS-MRI outperforms Fourier encoded MCS-MRI in preserving image resolution and can achieve higher acceleration factors. To demonstrate the feasibility of the proposed noiselet encoding scheme, a pulse sequences with tailored spatially selective RF excitation pulses was designed and implemented on a 3T scanner to acquire the data in the noiselet domain from a phantom and a human brain. The results indicate that noislet encoding preserves image resolution better than Fouirer encoding.

/pdf/multichannel-compressive-sensing-mri-using-noiselet-encoding-3x4rbcp2wg.pdf

Multichannel compressive sensing MRI using noiselet encoding.

Magnetic resonance (MR) fingerprinting is a new quantitative imaging paradigm, which simultaneously acquires multiple MR tissue parameter maps in a single experiment. In this paper, we present an estimation-theoretic framework to perform experiment design for MR fingerprinting. Specifically, we describe a discrete-time dynamic system to model spin dynamics, and derive an estimation-theoretic bound, i.e., the Cramer-Rao bound, to characterize the signal-to-noise ratio (SNR) efficiency of an MR fingerprinting experiment. We then formulate an optimal experiment design problem, which determines a sequence of acquisition parameters to encode MR tissue parameters with the maximal SNR efficiency, while respecting the physical constraints and other constraints from the image decoding/reconstruction process. We evaluate the performance of the proposed approach with numerical simulations, phantom experiments, and in vivo experiments. We demonstrate that the optimized experiments substantially reduce data acquisition time and/or improve parameter estimation. For example, the optimized experiments achieve about a factor of two improvement in the accuracy of ${T}_{2}$ maps, while keeping similar or slightly better accuracy of ${T}_{1}$ maps. Finally, as a remarkable observation, we find that the sequence of optimized acquisition parameters appears to be highly structured rather than randomly/pseudo-randomly varying as is prescribed in the conventional MR fingerprinting experiments.

Optimal Experiment Design for Magnetic Resonance Fingerprinting: Cramér-Rao Bound Meets Spin Dynamics

Purpose
Origami-based biomedical device design is an emerging technology due to its ability to be deployed from a minimal foldable pattern to a larger volume. This paper aims to review state-of-the-art origami structures applied in the medical device field.

/pdf/fabricating-biomedical-origami-a-state-of-the-art-review-45a2d5kr3g.pdf

Fabricating biomedical origami: a state-of-the-art review

3D MR fingerprinting with accelerated stack-of-spirals and hybrid sliding-window and GRAPPA reconstruction.

Purpose
To develop a method for accelerated and robust MR fingerprinting (MRF) with improved image reconstruction and parameter matching processes.

Theory and Methods
A sliding-window (SW) strategy was applied to MRF, in which signal and dictionary matching was conducted between fingerprints consisting of mixed-contrast image series reconstructed from consecutive data frames segmented by a sliding window, and a precalculated mixed-contrast dictionary. The effectiveness and performance of this new method, dubbed SW-MRF, was evaluated in both phantom and in vivo. Error quantifications were conducted on results obtained with various settings of SW reconstruction parameters.

Results
Compared with the original MRF strategy, the results of both phantom and in vivo experiments demonstrate that the proposed SW-MRF strategy either provided similar accuracy with reduced acquisition time, or improved accuracy with equal acquisition time. Parametric maps of T1, T2, and proton density of comparable quality could be achieved with a two-fold or more reduction in acquisition time. The effect of sliding-window width on dictionary sensitivity was also estimated.

Conclusion
The novel SW-MRF recovers high quality image frames from highly undersampled MRF data, which enables more robust dictionary matching with reduced numbers of data frames. This time efficiency may facilitate MRF applications in time-critical clinical settings. Magn Reson Med, 2016. © 2016 International Society for Magnetic Resonance in Medicine.

Robust sliding-window reconstruction for Accelerating the acquisition of MR fingerprinting.

A novel image encoding approach based on linear frequency-swept excitation has been recently proposed to overcome artifacts induced by various field perturbations in single-shot echo planar imaging. In this article, we develop a new super-resolved reconstruction method for it using the concepts of local k-space and partial Fourier transform. This method is superior to the originally developed conjugate gradient algorithm in convenience, image quality, and stability of solution. Reduced field-of-view is applied to the phase encoding direction to further enhance the spatial resolution and field perturbation immunity of the image obtained. Effectiveness of this new combined reconstruction method is demonstrated with a series of experiments on biological samples. Two single-shot sequences with different encoding features are tested. The results show that this reconstruction method maintains excellent field perturbation immunity and improves fidelity of the images. In vivo experiments on rat indicate that this solution is favorable for ultrafast imaging applications in which severe susceptibility heterogeneities around the tissue–air or tissue–bone interfaces, motion and oblique plane effects usually compromise the echo planar imaging image quality. Magn Reson Med, 2013. © 2012 Wiley Periodicals, Inc.

https://onlinelibrary.wiley.com/doi/pdf/10.1002/mrm.24366

Partial Fourier transform reconstruction for single-shot MRI with linear frequency-swept excitation.

Image reconstruction with low-rankness and self-consistency of k-space data in parallel MRI.

PURPOSE To propose a novel reconstruction method using parallel imaging with low rank constraint to accelerate high resolution multishot spiral diffusion imaging. THEORY AND METHODS The undersampled high resolution diffusion data were reconstructed based on a low rank (LR) constraint using similarities between the data of different interleaves from a multishot spiral acquisition. The self-navigated phase compensation using the low resolution phase data in the center of k-space was applied to correct shot-to-shot phase variations induced by motion artifacts. The low rank reconstruction was combined with sensitivity encoding (SENSE) for further acceleration. The efficiency of the proposed joint reconstruction framework, dubbed LR-SENSE, was evaluated through error quantifications and compared with l1 regularized compressed sensing method and conventional iterative SENSE method using the same datasets. RESULTS It was shown that with a same acceleration factor, the proposed LR-SENSE method had the smallest normalized sum-of-squares errors among all the compared methods in all diffusion weighted images and DTI-derived index maps, when evaluated with different acceleration factors (R = 2, 3, 4) and for all the acquired diffusion directions. CONCLUSION Robust high resolution diffusion weighted image can be efficiently reconstructed from highly undersampled multishot spiral data with the proposed LR-SENSE method. Magn Reson Med 77:1359-1366, 2017. © 2016 International Society for Magnetic Resonance in Medicine.

Ying Chen

Papers

Robust sliding-window reconstruction for Accelerating the acquisition of MR fingerprinting.

Partial Fourier transform reconstruction for single-shot MRI with linear frequency-swept excitation.

Image reconstruction with low-rankness and self-consistency of k-space data in parallel MRI.

Efficient parallel reconstruction for high resolution multishot spiral diffusion data with low rank constraint

An efficient de-convolution reconstruction method for spatiotemporal- encoding single-scan 2D MRI