scholarly journals Streamlined Magnetic Resonance Fingerprinting: Fast Whole-brain Coverage with Deep-learning Based Parameter Estimation

2020 ◽  
Author(s):  
Mahdi Khajehim ◽  
Thomas Christen ◽  
Fred Tam ◽  
Simon J. Graham
Keyword(s):  
Parameter Estimation ◽  
Deep Learning ◽  
Magnetic Resonance ◽  
Quantitative Mri ◽  
Parameter Estimates ◽  
Single Shot ◽  
Whole Brain ◽  
Dictionary Matching ◽  
Scan Time

AbstractMagnetic resonance fingerprinting (MRF) is a novel quantitative MRI (qMRI) framework that provides simultaneous estimates of multiple relaxation parameters as well as metrics of field inhomogeneity in a single acquisition. However, current bottlenecks exist in the forms of (1) scan time; (2) need for custom image reconstruction; (3) large dictionary sizes; (4) long dictionary-matching time. The aim of this study is to introduce a novel streamlined magnetic-resonance fingerprinting (sMRF) framework that is based on a single-shot echo-planar imaging (EPI) sequence to simultaneously estimate tissue T1, T2, and T2* with integrated B1+ correction. Encouraged by recent work on EPI-based MRF, we developed a method that combines spin-echo EPI with gradient-echo EPI to achieve T2 in addition to T1 and T2* quantification. To this design, we add simultaneous multi-slice (SMS) acceleration to enable full-brain coverage in a few minutes. Moreover, in the parameter-estimation step, we use deep learning to train a deep neural network (DNN) to accelerate the estimation process by orders of magnitude. Notably, due to the high image quality of the EPI scans, the training process can rely simply on Bloch-simulated data. The DNN also removes the need for storing large dictionaries. Phantom scans along with in-vivo multi-slice scans from seven healthy volunteers were acquired with resolutions of 1.1×1.1×3 mm3 and 1.7×1.7×3 mm3, and the results were validated against ground truth measurements. Excellent correspondence was found between our T1, T2, and T2* estimates and results obtained from standard approaches. In the phantom scan, a strong linear relationship (R=1-1.04, R2>0.96) was found for all parameter estimates, with a particularly high agreement for T2 estimation (R2>0.99). Similar findings are reported for the in-vivo human data for all of our parameter estimates. Incorporation of DNN results in a reduction of parameter estimation time on the order of 1000 x and a reduction in storage requirements on the order of 2500 x while achieving highly similar results as conventional dictionary matching (%differences of 7.4±0.4%, 3.6±0.3% and 6.0±0.4% error in T1, T2, and T2* estimation). Thus, sMRF has the potential to be the method of choice for future MRF studies by providing ease of implementation, fast whole-brain coverage, and ultra-fast T1/T2/T2* estimation.

scholarly journals Streamlined magnetic resonance fingerprinting: Fast whole-brain coverage with deep-learning based parameter estimation

NeuroImage ◽  
2021 ◽  
Vol 238 ◽  
pp. 118237
Author(s):  
Mahdi Khajehim ◽  
Thomas Christen ◽  
Fred Tam ◽  
Simon J. Graham
Keyword(s):  
Parameter Estimation ◽  
Deep Learning ◽  
Magnetic Resonance ◽  
Whole Brain

scholarly journals Minimal Linear Networks for Magnetic Resonance Image Reconstruction

2019 ◽  
Vol 9 (1) ◽  
Author(s):  
Gilad Liberman ◽  
Benedikt A. Poser
Keyword(s):  
Deep Learning ◽  
Magnetic Resonance ◽  
Image Reconstruction ◽  
Simulated Data ◽  
Linear Networks ◽  
Learning Framework ◽  
Scan Time ◽  
Learning Techniques ◽  
Ill Posed ◽  
Magnetic Resonance Imaging Mri

AbstractModern sequences for Magnetic Resonance Imaging (MRI) trade off scan time with computational challenges, resulting in ill-posed inverse problems and the requirement to account for more elaborated signal models. Various deep learning techniques have shown potential for image reconstruction from reduced data, outperforming compressed sensing, dictionary learning and other advanced techniques based on regularization, by characterization of the image manifold. In this work we suggest a framework for reducing a “neural” network to the bare minimum required by the MR physics, reducing the network depth and removing all non-linearities. The networks performed well both on benchmark simulated data and on arterial spin labeling perfusion imaging, showing clear images while preserving sensitivity to the minute signal changes. The results indicate that the deep learning framework plays a major role in MR image reconstruction, and suggest a concrete approach for probing into the contribution of additional elements.

scholarly journals Whole-Brain Imaging of Subvoxel T1-Diffusion Correlation Spectra in Human Subjects

2021 ◽  
Vol 15 ◽  
Author(s):  
Alexandru V. Avram ◽  
Joelle E. Sarlls ◽  
Peter J. Basser
Keyword(s):  
Pulse Sequence ◽  
Mean Diffusivity ◽  
Inversion Recovery ◽  
Quantitative Mri ◽  
Single Shot ◽  
Myelinated Fiber ◽  
Whole Brain ◽  
Model Free ◽  
Wide Range ◽  
Biological Specificity

T1 relaxation and water mobility generate eloquent MRI tissue contrasts with great diagnostic value in many neuroradiological applications. However, conventional methods do not adequately quantify the microscopic heterogeneity of these important biophysical properties within a voxel, and therefore have limited biological specificity. We describe a new correlation spectroscopic (CS) MRI method for measuring how T1 and mean diffusivity (MD) co-vary in microscopic tissue environments. We develop a clinical pulse sequence that combines inversion recovery (IR) with single-shot isotropic diffusion encoding (IDE) to efficiently acquire whole-brain MRIs with a wide range of joint T1-MD weightings. Unlike conventional diffusion encoding, the IDE preparation ensures that all subvoxel water pools are weighted by their MDs regardless of the sizes, shapes, and orientations of their corresponding microscopic diffusion tensors. Accordingly, IR-IDE measurements are well-suited for model-free, quantitative spectroscopic analysis of microscopic water pools. Using numerical simulations, phantom experiments, and data from healthy volunteers we demonstrate how IR-IDE MRIs can be processed to reconstruct maps of two-dimensional joint probability density functions, i.e., correlation spectra, of subvoxel T1-MD values. In vivo T1-MD spectra show distinct cerebrospinal fluid and parenchymal tissue components specific to white matter, cortical gray matter, basal ganglia, and myelinated fiber pathways, suggesting the potential for improved biological specificity. The one-dimensional marginal distributions derived from the T1-MD correlation spectra agree well with results from other relaxation spectroscopic and quantitative MRI studies, validating the T1-MD contrast encoding and the spectral reconstruction. Mapping subvoxel T1-diffusion correlations in patient populations may provide a more nuanced, comprehensive, sensitive, and specific neuroradiological assessment of the non-specific changes seen on fluid-attenuated inversion recovery (FLAIR) and diffusion-weighted MRIs (DWIs) in cancer, ischemic stroke, or brain injury.

scholarly journals Multi-parametric quantitative spinal cord MRI with unified signal readout and image denoising

2019 ◽  
Author(s):  
Francesco Grussu ◽  
Marco Battiston ◽  
Jelle Veraart ◽  
Torben Schneider ◽  
Julien Cohen-Adad ◽  
...  
Keyword(s):  
Spinal Cord ◽  
Synthetic Data ◽  
Principal Component ◽  
Quantitative Mri ◽  
Single Shot ◽  
Planar Imaging ◽  
Physiological Noise ◽  
Parametric Maps ◽  
Signal Readout

AbstractMulti-parametric quantitative MRI (qMRI) of the spinal cord is a promising non-invasive tool to probe early microstructural damage in neurological disorders. It is usually performed by combining acquisitions with multiple signal readouts, which exhibit different thermal noise levels, geometrical distortions and susceptibility to physiological noise. This ultimately hinders joint multi-contrast modelling and makes the geometric correspondence of parametric maps challenging. We propose an approach to overcome these limitations, by implementing state-of-the-art microstructural MRI of the spinal cord with a unified signal readout. We base our acquisition on single-shot echo planar imaging with reduced field-of-view, and obtain data from two different vendors (vendor 1: Philips Achieva; vendor 2: Siemens Prisma). Importantly, the unified acquisition allows us to compare signal and noise across contrasts, thus enabling overall quality enhancement via Marchenko-Pastur (MP) Principal Component Analysis (PCA) denoising. MP-PCA is a recent method relying on redundant acquisitions, i.e. such that the number of measurements is much larger than the number of informative principal components. Here we used in vivo and synthetic data to test whether a unified readout enables more efficient denoising of less redundant acquisitions, since these can be denoised jointly with more redundant ones. We demonstrate that a unified readout provides robust multi-parametric maps, including diffusion and kurtosis tensors from diffusion MRI, myelin metrics from two-pool magnetisation transfer, and T1 and T2 from relaxometry. Moreover, we show that MP-PCA improves the quality of our multi-contrast acquisitions, since it reduces the coefficient of variation (i.e. variability) by up to 15% for mean kurtosis, 8% for bound pool fraction (BPF, myelin-sensitive), and 13% for T1, while enabling more efficient denoising of modalities limited in redundancy (e.g. relaxometry). In conclusion, multi-parametric spinal cord qMRI with unified readout is feasible and provides robust microstructural metrics with matched resolution and distortions, whose quality benefits from MP-PCA denoising, a useful pre-processing tool for spinal cord MRI.

Single Shot T1ρ Magnetic Resonance Imaging Of Metabolically Generated Water In Vivo

2009 ◽  
pp. 279-286 ◽  
Author(s):  
Eric A. Mellon ◽  
R. Shashank Beesam ◽  
Mallikarjunarao Kasam ◽  
James E. Baumgardner ◽  
Arijitt Borthakur ◽  
...  
Keyword(s):  
Magnetic Resonance Imaging ◽  
Magnetic Resonance ◽  
Single Shot ◽  
Resonance Imaging

Development of fast deep learning quantification for magnetic resonance fingerprinting in vivo

2020 ◽  
Vol 70 ◽  
pp. 81-90
Author(s):  
Peng Cao ◽  
Di Cui ◽  
Vince Vardhanabhuti ◽  
Edward S. Hui
Keyword(s):  
Deep Learning ◽  
Magnetic Resonance

scholarly journals Imaging Multiple Sclerosis Pathology at 160μm Isotropic Resolution by Human Whole-Brain Ex Vivo Magnetic Resonance Imaging at 3T

2021 ◽  
Author(s):  
Weigel Matthias ◽  
Dechent Peter ◽  
Galbusera Riccardo ◽  
Bahn Erik ◽  
Nair Govind ◽  
...  
Keyword(s):  
Magnetic Resonance Imaging ◽  
Multiple Sclerosis ◽  
Magnetic Resonance ◽  
Human Brain ◽  
Ex Vivo ◽  
Molecular Layer ◽  
Cortical Lesion ◽  
Resonance Imaging ◽  
Whole Brain

AbstractPostmortem magnetic resonance imaging (MRI) of the fixed healthy and diseased human brain facilitates spatial resolutions and image quality that is not achievable with in vivo MRI scans. Though challenging - and almost exclusively performed at 7T field strength - depicting the tissue architecture of the entire brain in fine detail is invaluable since it enables the study of neuroanatomy and uncovers important pathological features in neurological disorders. The objectives of the present work were (i) to develop a 3D isotropic ultra-high-resolution imaging approach for human whole-brain ex vivo acquisitions working on a standard clinical 3T MRI system, and (ii) to explore the sensitivity and specificity of this concept for specific pathoanatomical features of multiple sclerosis. The reconstructed images demonstrate unprecedented resolution and soft tissue contrast of the diseased human brain at 3T, thus allowing visualization of sub-millimetric lesions in the different cortical layers and in the cerebellar cortex, as well as unique cortical lesion characteristics such as the presence of incomplete / complete iron rims, and patterns of iron accumulation. Further details such as the subpial molecular layer, the line of Gennari, and some intrathalamic nuclei are also well distinguishable.

scholarly journals Lipid suppressed and tissue-fraction corrected metabolic distributions in human central brain structures using 2D 1H Magnetic Resonance Spectroscopic Imaging at 7 tesla

2020 ◽  
Author(s):  
Alex A. Bhogal ◽  
Tommy A.A. Broeders ◽  
Lisan Morsinkhof ◽  
Mirte Edens ◽  
Sahar Nassirpour ◽  
...  
Keyword(s):  
Magnetic Resonance ◽  
High Resolution ◽  
Spectroscopic Imaging ◽  
Magnetic Resonance Spectroscopic Imaging ◽  
Brain Diseases ◽  
7 Tesla ◽  
Partial Volume ◽  
Scan Time ◽  
Central Brain

ABSTRACTMagnetic resonance spectroscopic imaging (MRSI) has the potential to add a layer of understanding of the neurobiological mechanisms underlying brain diseases, disease progression, and treatment efficacy. Limitations related to metabolite fitting of low SNR data, signal variations due to partial volume effects, acquisition and extra-cranial lipid artefacts, along with clinically relevant aspects such as scan-time constraints, are among the factors that hinder the widespread implementation of in vivo MRSI. The aim of this work was to address these factors and to develop an acquisition, reconstruction and post-processing pipeline to derive lipid suppressed metabolite values based on Free Induction Decay (FID-MRSI) measurements made using a 7 tesla MR scanner. Anatomical images were used to perform high-resolution (1mm3) partial-volume correction to account for grey matter, white matter and cerebral-spinal fluid signal contributions. Implementation of automatic quality control thresholds and normalization of metabolic maps from 23 subjects to the MNI standard atlas facilitated the creation of high-resolution average metabolite maps of several clinically relevant metabolites in central brain regions, while accounting for macromolecular distributions. Reported metabolite values include glutamate, choline, (phospo)creatine, myo-inositol, glutathione, N-acetyl aspartyl glutamate(and glutamine) and N-acetyl aspartate. MNI-registered average metabolite maps facilitate group-based analysis; thus offering the possibility to mitigate uncertainty in variable MRSI.

Sign in / Sign up

Export Citation Format

Share Document