High-resolution 1H-MRSI at 9.4 T by integrating relaxation enhancement and subspace imaging.

Achieving high-resolution and high signal-to-noise ratio (SNR) in vivo metabolic imaging via fast magnetic resonance spectroscopic imaging (MRSI) has been a longstanding challenge. This study combines the methods of relaxation enhancement (RE) and subspace imaging for the first time, enabling high-resolution and high-SNR in vivo MRSI of rodent brains at 9.4 T. Specifically, an RE-based chemical shift imaging sequence, which combines a frequency-selective pulse to excite only the metabolite frequencies with minimum perturbation of the water spins and a pair of adiabatic pulses to spatially localize the slice of interest, is designed and evaluated in vivo. This strategy effectively shortens the apparent T1 of metabolites, thereby increasing the SNR during relatively short repetition time ((TR) compared with acquisitions with only spatially selective wideband excitations, and does not require water suppression. The SNR was further enhanced via a state-of-the-art subspace reconstruction method. A novel subspace learning strategy tailored for 9.4 T and RE acquisitions is developed. In vivo, high-resolution (e.g., voxel size of 0.6 × 0.6 × 1.5 mm3) MRSI of both healthy mouse brains and a glioma-bearing mouse brain in 12.5 min has been demonstrated.

Multiscale biochemical mapping of the brain through deep-learning-enhanced high-throughput mass spectrometry.

Spatial omics technologies can reveal the molecular intricacy of the brain. While mass spectrometry imaging (MSI) provides spatial localization of compounds, comprehensive biochemical profiling at a brain-wide scale in three dimensions by MSI with single-cell resolution has not been achieved. We demonstrate complementary brain-wide and single-cell biochemical mapping using MEISTER, an integrative experimental and computational mass spectrometry (MS) framework. Our framework integrates a deep-learning-based reconstruction that accelerates high-mass-resolving MS by 15-fold, multimodal registration creating three-dimensional (3D) molecular distributions and a data integration method fitting cell-specific mass spectra to 3D datasets. We imaged detailed lipid profiles in tissues with millions of pixels and in large single-cell populations acquired from the rat brain. We identified region-specific lipid contents and cell-specific localizations of lipids depending on both cell subpopulations and anatomical origins of the cells. Our workflow establishes a blueprint for future development of multiscale technologies for biochemical characterization of the brain.

Multi-Parametric Molecular Imaging of the Brain Using Optimized Multi-TE Subspace MRSI.

OBJECTIVE: To develop a novel multi-TE MR spectroscopic imaging (MRSI) approach to enable label-free, simultaneous, high-resolution mapping of several molecules and their biophysical parameters in the brain. METHODS: The proposed method uniquely integrated an augmented molecular-component-specific subspace model for multi-TE 1H-MRSI signals, an estimation-theoretic experiment optimization (nonuniform TE selection) for molecule separation and parameter estimation, a physics-driven subspace learning strategy for spatiospectral reconstruction and molecular quantification, and a new accelerated multi-TE MRSI acquisition for generating high-resolution data in clinically relevant times. Numerical studies, phantom and in vivo experiments were conducted to validate the optimized experiment design and demonstrate the imaging capability offered by the proposed method. RESULTS: The proposed TE optimization improved estimation of metabolites, neurotransmitters and their T2's over conventional TE choices, e.g., reducing variances of neurotransmitter concentration by  âˆ¼  40% and metabolite T2 by  âˆ¼  60%. Simultaneous metabolite and neurotransmitter mapping of the brain can be achieved at a nominal resolution of 3.4 × 3.4 × 6.4 mm 3. High-resolution, 3D metabolite T2 mapping was made possible for the first time. The translational potential of the proposed method was demonstrated by mapping biochemical abnormality in a post-traumatic epilepsy (PTE) patient. CONCLUSION: The feasibility for high-resolution mapping of metabolites/neurotransmitters and metabolite T2's within clinically relevant time was demonstrated. We expect our method to offer richer information for revealing and understanding metabolic alterations in neurological diseases. SIGNIFICANCE: A novel multi-TE MRSI approach was presented that enhanced the technological capability of multi-parametric molecular imaging of the brain. The proposed method presents new technology development and application opportunities for providing richer molecular level information to uncover and comprehend metabolic changes relevant in various neurological applications.

High-Dimensional MR Reconstruction Integrating Subspace and Adaptive Generative Models.

OBJECTIVE: To develop a new method that integrates subspace and generative image models for high-dimensional MR image reconstruction. METHODS: We proposed a formulation that synergizes a low-dimensional subspace model of high-dimensional images, an adaptive generative image prior serving as spatial constraints on the sequence of "contrast-weighted" images or spatial coefficients of the subspace model, and a conventional sparsity regularization. A special pretraining plus subject-specific network adaptation strategy was proposed to construct an accurate generative-network-based representation for images with varying contrasts. An iterative algorithm was introduced to jointly update the subspace coefficients and the multi-resolution latent space of the generative image model that leveraged an recently proposed intermediate layer optimization technique for network inversion. RESULTS: We evaluated the utility of the proposed method for two high-dimensional imaging applications: accelerated MR parameter mapping and high-resolution MR spectroscopic imaging. Improved performance over state-of-the-art subspace-based methods was demonstrated in both cases. CONCLUSION: The proposed method provided a new way to address high-dimensional MR image reconstruction problems by incorporating an adaptive generative model as a data-driven spatial prior for constraining subspace reconstruction. SIGNIFICANCE: Our work demonstrated the potential of integrating data-driven and adaptive generative priors with canonical low-dimensional modeling for high-dimensional imaging problems.

High-Dimensional MR Spatiospectral Imaging by Integrating Physics-Based Modeling and Data-Driven Machine Learning: Current progress and future directions.

Magnetic resonance spectroscopic imaging (MRSI) offers a unique molecular window into the physiological and pathological processes in the human body. However, the applications of MRSI have been limited by a number of long-standing technical challenges due to high dimensionality and low signal-to-noise ratio (SNR). Recent technological developments integrating physics-based modeling and data-driven machine learning that exploit unique physical and mathematical properties of MRSI signals have demonstrated impressive performance in addressing these challenges for rapid, high-resolution, quantitative MRSI. This paper provides a systematic review of these progresses in the context of MRSI physics and offers perspectives on promising future directions.

Integrative Multiscale Biochemical Mapping of the Brain via Deep-Learning-Enhanced High-Throughput Mass Spectrometry.

Elucidating the spatial-biochemical organization of the brain across different scales produces invaluable insight into the molecular intricacy of the brain. While mass spectrometry imaging (MSI) provides spatial localization of compounds, comprehensive chemical profiling at a brain-wide scale in three dimensions by MSI with single-cell resolution has not been achieved. We demonstrate complementary brain-wide and single-cell biochemical mapping via MEISTER, an integrative experimental and computational mass spectrometry framework. MEISTER integrates a deep-learning-based reconstruction that accelerates high-mass-resolving MS by 15-fold, multimodal registration creating 3D molecular distributions, and a data integration method fitting cell-specific mass spectra to 3D data sets. We imaged detailed lipid profiles in tissues with data sets containing millions of pixels, and in large single-cell populations acquired from the rat brain. We identified region-specific lipid contents, and cell-specific localizations of lipids depending on both cell subpopulations and anatomical origins of the cells. Our workflow establishes a blueprint for future developments of multiscale technologies for biochemical characterization of the brain.

Ultra-High-Field MRI in the Diagnosis and Management of Gliomas: A Systematic Review.

Importance: Gliomas, tumors of the central nervous system, are classically diagnosed through invasive surgical biopsy and subsequent histopathological study. Innovations in ultra-high field (UHF) imaging, namely 7-Tesla magnetic resonance imaging (7T MRI) are advancing preoperative tumor grading, visualization of intratumoral structures, and appreciation of small brain structures and lesions. Objective: Summarize current innovative uses of UHF imaging techniques in glioma diagnostics and treatment. Methods: A systematic review in accordance with PRISMA guidelines was performed utilizing PubMed. Case reports and series, observational clinical trials, and randomized clinical trials written in English were included. After removing unrelated studies and those with non-human subjects, only those related to 7T MRI were independently reviewed and summarized for data extraction. Some preclinical animal models are briefly described to demonstrate future usages of ultra-high-field imaging. Results: We reviewed 46 studies (43 human and 3 animal models) which reported clinical usages of UHF MRI in the diagnosis and management of gliomas. Current literature generally supports greater resolution imaging from 7T compared to 1.5T or 3T MRI, improving visualization of cerebral microbleeds and white and gray matter, and providing more precise localization for radiotherapy targeting. Additionally, studies found that diffusion or susceptibility-weighted imaging techniques applied to 7T MRI, may be used to predict tumor grade, reveal intratumoral structures such as neovasculature and microstructures like axons, and indicate isocitrate dehydrogenase 1 mutation status in preoperative imaging. Similarly, newer imaging techniques such as magnetic resonance spectroscopy and chemical exchange saturation transfer imaging can be performed on 7T MRI to predict tumor grading and treatment efficacy. Geometrical distortion, a known challenge of 7T MRI, was at a tolerable level in all included studies. Conclusion: UHF imaging has the potential to preoperatively and non-invasively grade gliomas, provide precise therapy target areas, and visualize lesions not seen on conventional MRI.

Enhancing the Throughput of FT Mass Spectrometry Imaging Using Joint Compressed Sensing and Subspace Modeling.

Mass spectrometry imaging (MSI) allows for untargeted mapping of the chemical composition of tissues with attomole detection limits. MSI using Fourier transform (FT)-based mass spectrometers, such as FT-ion cyclotron resonance (FT-ICR), grants the ability to examine the chemical space with unmatched mass resolution and mass accuracy. However, direct imaging of large tissue samples using FT-ICR is slow. In this work, we present an approach that combines the subspace modeling of ICR temporal signals with compressed sensing to accelerate high-resolution FT-ICR MSI. A joint subspace and spatial sparsity constrained model computationally reconstructs high-resolution MSI data from the sparsely sampled transients with reduced duration, allowing a significant reduction in imaging time. Simulation studies and experimental implementation of the proposed method in investigation of brain tissues demonstrate a 10-fold enhancement in throughput of FT-ICR MSI, without the need for instrumental or hardware modifications.

SNR Enhancement for Multi-TE MRSI Using Joint Low-Dimensional Model and Spatial Constraints.

We present a novel method to enhance the SNR for multi-TE MR spectroscopic imaging (MRSI) data by integrating learned nonlinear low-dimensional model and spatial constraints. A deep complex convolutional autoencoder (DCCAE) was developed to learn a nonlinear low-dimensional representation of the high-dimensional multi-TE 1H spectroscopy signals. The learned model significantly reduces the data dimension thus serving as an effective constraint for noise reduction. A reconstruction formulation was proposed to integrate the spatiospectral encoding model, the learned model, and a spatial constraint for an SNR-enhancing reconstruction from multi-TE data. The proposed method has been evaluated using both numerical simulations and in vivo brain MRSI experiments. The superior denoising performance of the proposed over alternative methods was demonstrated, both qualitatively and quantitatively. In vivo multi-TE data was used to assess the improved metabolite quantification reproducibility and accuracy achieved by the proposed method. We expect the proposed SNR-enhancing reconstruction to enable faster and/or higher-resolution multi-TE 1H-MRSI of the brain, potentially useful for various clinical applications.

Epigenetic MRI: Noninvasive imaging of DNA methylation in the brain.

Both neuronal and genetic mechanisms regulate brain function. While there are excellent methods to study neuronal activity in vivo, there are no nondestructive methods to measure global gene expression in living brains. Here, we present a method, epigenetic MRI (eMRI), that overcomes this limitation via direct imaging of DNA methylation, a major gene-expression regulator. eMRI exploits the methionine metabolic pathways for DNA methylation to label genomic DNA through 13C-enriched diets. A 13C magnetic resonance spectroscopic imaging method then maps the spatial distribution of labeled DNA. We validated eMRI using pigs, whose brains have stronger similarity to humans in volume and anatomy than rodents, and confirmed efficient 13C-labeling of brain DNA. We also discovered strong regional differences in global DNA methylation. Just as functional MRI measurements of regional neuronal activity have had a transformational effect on neuroscience, we expect that the eMRI signal, both as a measure of regional epigenetic activity and as a possible surrogate for regional gene expression, will enable many new investigations of human brain function, behavior, and disease.

DeepSENSE: Learning coil sensitivity functions for SENSE reconstruction using deep learning.

PURPOSE: To improve the estimation of coil sensitivity functions from limited auto-calibration signals (ACS) in SENSE-based reconstruction for brain imaging. METHODS: We propose to use deep learning to estimate coil sensitivity functions by leveraging information from previous scans obtained using the same RF receiver system. Specifically, deep convolutional neural networks were designed to learn an end-to-end mapping from the initial sensitivity to the high-resolution counterpart. Sensitivity alignment was further proposed to reduce the geometric variation caused by different subject positions and imaging FOVs. Cross-validation with a small set of datasets was performed to validate the learned neural network. Iterative SENSE reconstruction was adopted to evaluate the utility of the sensitivity functions from the proposed and conventional methods. RESULTS: The proposed method produced improved sensitivity estimates and SENSE reconstructions compared to the conventional methods in terms of aliasing and noise suppression with very limited ACS data. Cross-validation with a small set of data demonstrated the feasibility of learning coil sensitivity functions for brain imaging. The network learned on the spoiled GRE data can be applied to predict sensitivity functions for spin-echo and MPRAGE datasets. CONCLUSION: A deep learning-based method has been proposed for improving the estimation of coil sensitivity functions. Experimental results have demonstrated the feasibility and potential of the proposed method for improving SENSE-based reconstructions especially when the ACS data are limited.

High-resolution, 3D multi-TE 1 H MRSI using fast spatiospectral encoding and subspace imaging.

PURPOSE: To develop a novel method to achieve fast, high-resolution, 3D multi-TE 1 H-MRSI of the brain. METHODS: A new multi-TE MRSI acquisition strategy was developed that integrates slab selective excitation with adiabatic refocusing for better volume coverage, rapid spatiospectral encoding, sparse multi-TE sampling, and interleaved water navigators for field mapping and calibration. Special data processing strategies were developed to interpolate the sparsely sampled data, remove nuisance signals, and reconstruct multi-TE spatiospectral distributions with high SNR. Phantom and in vivo experiments have been carried out to demonstrate the capability of the proposed method. RESULTS: The proposed acquisition can produce multi-TE 1 H-MRSI data with three TEs at a nominal spatial resolution of 3.4 × 3.4 × 5.3 mm3 in around 20 min. High-SNR brain metabolite spatiospectral reconstructions can be obtained from both a metabolite phantom and in vivo experiments by the proposed method. CONCLUSION: High-resolution, 3D multi-TE 1 H-MRSI of the brain can be achieved within clinically feasible time. This capability, with further optimizations, could be translated to clinical applications and neuroscience studies where simultaneously mapping metabolites and neurotransmitters and TE-dependent molecular spectral changes are of interest.

Separation of Metabolites and Macromolecules for Short-TE 1H-MRSI Using Learned Component-Specific Representations.

Short-echo-time (TE) proton magnetic resonance spectroscopic imaging (MRSI) allows for simultaneously mapping a number of molecules in the brain, and has been recognized as an important tool for studying in vivo biochemistry in various neuroscience and disease applications. However, separation of the metabolite and macromolecule (MM) signals present in the short-TE data with significant spectral overlaps remains a major technical challenge. This work introduces a new approach to solve this problem by integrating imaging physics and representation learning. Specifically, a mixed unsupervised and supervised learning-based strategy was developed to learn the metabolite and MM-specific low-dimensional representations using deep autoencoders. A constrained reconstruction formulation is proposed to integrate the MRSI spatiospectral encoding model and the learned representations as effective constraints for signal separation. An efficient algorithm was developed to solve the resulting optimization problem with provable convergence. Simulation and experimental results have been obtained to demonstrate the component-specific representation power of the learned models and the capability of the proposed method in separating metabolite and MM signals for practical short-TE [Formula: see text]-MRSI data.

High-Resolution 3D Spin-Echo MRSI Using Interleaved Water Navigators, Sparse Sampling and Subspace-Based Processing.

This work presents a new method to achieve accelerated, high-resolution magnetic resonance spectroscopic imaging (MRSI) with spin-echo excitations. A new data acquisition strategy is proposed that integrates adiabatic refocusing, elimination of lipid suppression, rapid spatiospectral encoding with sparse (k,t)-space sampling, and interleaved water navigators. This integration leads to a significantly improved combination of volume coverage, spatial resolution (approximately 3 × 3.4 × 4 mm3) and speed (< 10 minutes), while eliminating additional scans for field mapping and coil sensitivity estimation. A data processing strategy that integrates parallel imaging reconstruction and subspace-based processing is devised to produce high-SNR spatiospectral reconstruction from the sparsely sampled, noisy and highresolution MRSI data. Promising in vivo results have been obtained to demonstrate the potential of the proposed method.Clinical relevance- The proposed method enabled volumetric MRSI with a nominal resolution of 3 × 3.4 × 4 mm3 in less than 10 minutes. With further developments and optimizations, the proposed method is expected to be useful for providing molecular-level information of brain functions and diseases, and has the potential to provide new biomarkers for disease diagnosis and treatment monitoring.

Accelerating Fourier Transform-Ion Cyclotron Resonance Mass Spectrometry Imaging Using a Subspace Approach.

We present a subspace method that accelerates data acquisition using Fourier transform-ion cyclotron resonance (FT-ICR) mass spectrometry imaging (MSI). For MSI of biological tissue samples, there is a finite number of heterogeneous tissue types with distinct chemical profiles that introduce redundancy in the high-dimensional measurements. Our subspace model exploits the redundancy in data measured from whole-slice tissue samples by decomposing the transient signals into linear combinations of a set of basis transients with the desired spectral resolution. This decomposition allowed us to design a strategy that acquires a subset of long transients for basis determination and short transients for the remaining pixels, drastically reducing the acquisition time. The computational reconstruction strategy can maintain high-mass-resolution and spatial-resolution MSI while providing a 10-fold improvement in throughput. We validated the capability of the subspace model using a rat sagittal brain slice imaging data set. Comprehensive evaluation of the quality of the mass spectral and ion images demonstrated that the reconstructed data produced by the reported method required only 15% of the typical acquisition time and exhibited both qualitative and quantitative consistency when compared to the original data. Our method enables either higher sample throughput or higher-resolution images at similar acquisition lengths, providing greater flexibility in obtaining FT-ICR MSI measurements.

Biodegradable MRI Visible Drug Eluting Stent Reinforced by Metal Organic Frameworks.

Metal-organic frameworks (MOFs) have applications in numerous fields. However, the development of MOF-based "theranostic" macroscale devices is not achieved. Here, heparin-coated biocompatible MOF/poly(ε-caprolactone) (PCL) "theranostic" stents are developed, where NH2 -Materials of Institute Lavoisier (MIL)-101(Fe) encapsulates and releases rapamycin (an immunosuppressive drug). These stents also act as a remarkable source of contrast in ex vivo magnetic resonance imaging (MRI) compared to the invisible polymeric stent. The in vitro release patterns of heparin and rapamycin respectively can ensure a type of programmed model to prevent blood coagulation immediately after stent placement in the artery and stenosis over a longer term. Due to the presence of hydrolysable functionalities in MOFs, the stents are shown to be highly biodegradable in degradation tests under various conditions. Furthermore, there is no compromise of mechanical strength or flexibility with MOF compositing. The system described here promises many biomedical applications in macroscale theranostic devices. The use of MOF@PCL can render a medical device MRI-visible while simultaneously acting as a carrier for therapeutic agents.

High-Resolution Dynamic 31P-MR Spectroscopic Imaging for Mapping Mitochondrial Function.

OBJECTIVE: To enable non-invasive dynamic metabolic mapping in rodent model studies of mitochondrial function using 31P-MR spectroscopic imaging (MRSI). METHODS: We developed a novel method for high-resolution dynamic 31P-MRSI. The method synergistically integrates physics-based models of spectral structures, biochemical modeling of molecular dynamics, and subspace learning to capture spatiospectral variations. Fast data acquisition was achieved using rapid spiral trajectories and sparse sampling of (k, t, T)-space; image reconstruction was accomplished using a low-rank tensor-based framework. RESULTS: The proposed method provided high-resolution dynamic metabolic mapping in rat hindlimb at spatial and temporal resolutions of 4[Formula: see text]2 mm3 and 1.28 s, respectively. This allowed for in vivo mapping of the time-constant of phosphocreatine resynthesis, a well established index of mitochondrial oxidative capacity. Multiple rounds of in vivo experiments were performed to demonstrate reproducibility, and in vitro experiments were used to validate the accuracy of the estimated metabolite maps. CONCLUSIONS: A new model-based method is proposed to achieve high-resolution dynamic 31P-MRSI. The proposed method's ability to delineate metabolic heterogeneity was demonstrated in rat hindlimb. SIGNIFICANCE: Abnormal mitochondrial metabolism is a key cellular dysfunction in many prevalent diseases such as diabetes and heart disease; however, current understanding of mitochondrial function is mostly gained from studies on isolated mitochondria under nonphysiological conditions. The proposed method has the potential to open new avenues of research by allowing in vivo and longitudinal studies of mitochondrial dysfunction in disease development and progression.

Intravoxel B0 inhomogeneity corrected reconstruction using a low-rank encoding operator.

PURPOSE: To present a general and efficient method for macroscopic intravoxel B0 inhomogeneity corrected reconstruction from multi-TE acquisitions. THEORY AND METHODS: A signal encoding model for multi-TE gradient echo (GRE) acquisitions that incorporates 3D intravoxel B0 field variations is derived, and a low-rank approximation to the encoding operator is introduced under piecewise linear B0 assumption. The low-rank approximation enables very efficient computation and memory usage, and allows the proposed signal model to be integrated into general inverse problem formulations that are compatible with multi-coil and undersampling acquisitions as well as different regularization functions. RESULTS: Experimental multi-echo GRE data were acquired to evaluate the proposed method. Effective reduction of macroscopic intravoxel B0 inhomogeneity induced artifacts was demonstrated. Improved R2∗ estimation from the corrected reconstruction over standard Fourier reconstruction has also been obtained. CONCLUSIONS: The proposed method can effectively correct the effects of intravoxel B0 inhomogeneity, and can be useful for various imaging applications involving GRE-based acquisitions, including fMRI, quantitative R2∗ and susceptibility mapping, and MR spectroscopic imaging.

Ultrafast magnetic resonance spectroscopic imaging using SPICE with learned subspaces.

PURPOSE: To develop a subspace learning method for the recently proposed subspace-based MRSI approach known as SPICE, and achieve ultrafast 1 H-MRSI of the brain. THEORY AND METHODS: A novel strategy is formulated to learn a low-dimensional subspace representation of MR spectra from specially acquired training data and use the learned subspace for general MRSI experiments. Specifically, the subspace learning problem is formulated as learning "empirical" distributions of molecule-specific spectral parameters (e.g., concentrations, lineshapes, and frequency shifts) by integrating physics-based model and the training data. The learned spectral parameters and quantum mechanical simulation basis can then be combined to construct acquisition-specific subspace for spatiospectral encoding and processing. High-resolution MRSI acquisitions combining ultrashort-TE/short-TR excitation, sparse sampling, and the elimination of water suppression have been performed to evaluate the feasibility of the proposed method. RESULTS: The accuracy of the learned subspace and the capability of the proposed method in producing high-resolution 3D 1 H metabolite maps and high-quality spatially resolved spectra (with a nominal resolution of ∼2.4 × 2.4 × 3 mm3 in 5 minutes) were demonstrated using phantom and in vivo studies. By eliminating water suppression, we are also able to extract valuable information from the water signals for data processing ( B0 map, frequency drift, and coil sensitivity) as well as for mapping tissue susceptibility and relaxation parameters. CONCLUSIONS: The proposed method enables ultrafast 1 H-MRSI of the brain using a learned subspace, eliminating the need of acquiring subject-dependent navigator data (known as D1 ) in the original SPICE technique. It represents a new way to perform MRSI experiments and an important step toward practical applications of high-resolution MRSI.

Constrained Magnetic Resonance Spectroscopic Imaging by Learning Nonlinear Low-Dimensional Models.

Magnetic resonance spectroscopic imaging (MRSI) is a powerful molecular imaging modality but has very limited speed, resolution, and SNR tradeoffs. Construction of a low-dimensional model to effectively reduce the dimensionality of the imaging problem has recently shown great promise in improving these tradeoffs. This paper presents a new approach to model and reconstruct the spectroscopic signals by learning a nonlinear low-dimensional representation of the general MR spectra. Specifically, we trained a deep neural network to capture the low-dimensional manifold, where the high-dimensional spectroscopic signals reside. A regularization formulation is proposed to effectively integrate the learned model and physics-based data acquisition model for MRSI reconstruction with the capability to incorporate additional spatiospectral constraints. An efficient numerical algorithm was developed to solve the associated optimization problem involving back-propagating the trained network. Simulation and experimental results were obtained to demonstrate the representation power of the learned model and the ability of the proposed formulation in producing SNR-enhancing reconstruction from the practical MRSI data.