Search

We present a novel method for partial Fourier reconstruction using a zero-shot unsupervised deep learning approach. Specifically, this method enhances partial Fourier reconstruction by integrating the traditional phase constraint with recent advancements in zero-shot deep learning techniques. Recently, a zero-shot deep learning method based on an untrained generative prior has been proposed, demonstrating effectiveness in multi-echo/contrast reconstruction. This approach is based on the assumption that MR images can be nonlinearly represented through an untrained artificial neural network, allowing for simultaneous image reconstruction and prior learning without requiring any training data. In this study, we integrate this framework with the virtual conjugate coil (VCC) phase constraint to enable robust partial Fourier reconstruction. We evaluated the proposed method on the fastMRI dataset, the QALAS multi-contrast dataset, and a low-field dataset. Our experiment results confirm that the proposed method yields improved reconstruction quality compared to existing methods, with approximately $1.15 \times $ to $2.5\times $ reductions in NRMSE. The proposed technique enables robust partial Fourier reconstruction, proving valuable in numerous applications. Moreover, it adopts a zero-shot approach, eliminating the need for training data, which enhances its utility in scenarios where data collection is challenging.

Abstract Diffusion magnetic resonance imaging is an important tool for mapping tissue microstructure and structural connectivity non‐invasively in the in vivo human brain. Numerous diffusion signal models are proposed to quantify microstructural properties. Nonetheless, accurate estimation of model parameters is computationally expensive and impeded by image noise. Supervised deep learning‐based estimation approaches exhibit efficiency and superior performance but require additional training data and may be not generalizable. A new DIffusion Model OptimizatioN framework using physics‐informed and self‐supervised Deep learning entitled “DIMOND” is proposed to address this problem. DIMOND employs a neural network to map input image data to model parameters and optimizes the network by minimizing the difference between the input acquired data and synthetic data generated via the diffusion model parametrized by network outputs. DIMOND produces accurate diffusion tensor imaging results and is generalizable across subjects and datasets. Moreover, DIMOND outperforms conventional methods for fitting sophisticated microstructural models including the kurtosis and NODDI model. Importantly, DIMOND reduces NODDI model fitting time from hours to minutes, or seconds by leveraging transfer learning. In summary, the self‐supervised manner, high efficacy, and efficiency of DIMOND increase the practical feasibility and adoption of microstructure and connectivity mapping in clinical and neuroscientific applications.

Gilles de la Tourette syndrome (GTS) is a neuropsychiatric movement disorder with reported abnormalities in various neurotransmitter systems. Considering the integral role of iron in neurotransmitter synthesis and transport, it is hypothesized that iron exhibits a role in GTS pathophysiology. As a surrogate measure of brain iron, quantitative susceptibility mapping (QSM) was performed in 28 patients with GTS and 26 matched controls. Significant susceptibility reductions in the patients, consistent with reduced local iron content, were obtained in subcortical regions known to be implicated in GTS. Regression analysis revealed a significant negative association of tic scores and striatal susceptibility. To interrogate genetic mechanisms that may drive these reductions, spatially specific relationships between susceptibility and gene-expression patterns from the Allen Human Brain Atlas were assessed. Correlations in the striatum were enriched for excitatory, inhibitory, and modulatory neurochemical signaling mechanisms in the motor regions, mitochondrial processes driving ATP production and iron‑sulfur cluster biogenesis in the executive subdivision, and phosphorylation-related mechanisms affecting receptor expression and long-term potentiation in the limbic subdivision. This link between susceptibility reductions and normative transcriptional profiles suggests that disruptions in iron regulatory mechanisms are involved in GTS pathophysiology and may lead to pervasive abnormalities in mechanisms regulated by iron-containing enzymes.

Purpose: To develop a high-fidelity diffusion MRI acquisition and reconstruction framework with reduced echo-train-length for less T2* image blurring compared to typical highly accelerated echo-planar imaging (EPI) acquisitions at sub-millimeter isotropic resolution. Methods: We first proposed a circular-EPI trajectory with partial Fourier sampling on both the readout and phase-encoding directions to minimize the echo-train-length and echo time. We then utilized this trajectory in an interleaved two-shot EPI acquisition with reversed phase-encoding polarity, to aid in the correction of off-resonance-induced image distortions and provide complementary k-space coverage in the missing partial Fourier regions. Using model-based reconstruction with structured low-rank constraint and smooth phase prior, we corrected the shot-to-shot phase variations across the two shots and recover the missing k-space data. Finally, we combined the proposed acquisition/reconstruction framework with an SNR-efficient RF-encoded simultaneous multi-slab technique, termed gSlider, to achieve high-fidelity 720 µm and 500 µm isotropic resolution in-vivo diffusion MRI. Results: Both simulation and in-vivo results demonstrate the effectiveness of the proposed acquisition and reconstruction framework to provide distortion-corrected diffusion imaging at the mesoscale with markedly reduced T2*-blurring. The in-vivo results of 720 µm and 500 µm datasets show high-fidelity diffusion images with reduced image blurring and echo time using the proposed approaches. Conclusions: The proposed method provides high-quality distortion-corrected diffusion-weighted images with ∼40% reduction in the echo-train-length and T2* blurring at 500µm-isotropic-resolution compared to standard multi-shot EPI.

The objective of this study is to evaluate the efficacy of deep learning (DL) techniques in improving the quality of diffusion MRI (dMRI) data in clinical applications. The study aims to determine whether the use of artificial intelligence (AI) methods in medical images may result in the loss of critical clinical information and/or the appearance of false information. To assess this, the focus was on the angular resolution of dMRI and a clinical trial was conducted on migraine, specifically between episodic and chronic migraine patients. The number of gradient directions had an impact on white matter analysis results, with statistically significant differences between groups being drastically reduced when using 21 gradient directions instead of the original 61. Fourteen teams from different institutions were tasked to use DL to enhance three diffusion metrics (FA, AD and MD) calculated from data acquired with 21 gradient directions and a b-value of 1000 s/mm2. The goal was to produce results that were comparable to those calculated from 61 gradient directions. The results were evaluated using both standard image quality metrics and Tract-Based Spatial Statistics (TBSS) to compare episodic and chronic migraine patients. The study results suggest that while most DL techniques improved the ability to detect statistical differences between groups, they also led to an increase in false positive. The results showed that there was a constant growth rate of false positives linearly proportional to the new true positives, which highlights the risk of generalization of AI-based tasks when assessing diverse clinical cohorts and training using data from a single group. The methods also showed divergent performance when replicating the original distribution of the data and some exhibited significant bias. In conclusion, extreme caution should be exercised when using AI methods for harmonization or synthesis in clinical studies when processing heterogeneous data in clinical studies, as important information may be altered, even when global metrics such as structural similarity or peak signal-to-noise ratio appear to suggest otherwise.

Tremendous efforts have been made in the last decade to advance cutting-edge MRI technology in pursuit of mapping structural connectivity in the living human brain with unprecedented sensitivity and speed. The first Connectom 3T MRI scanner equipped with a 300 mT/m whole-body gradient system was installed at the Massachusetts General Hospital in 2011 and was specifically constructed as part of the Human Connectome Project. Since that time, numerous technological advances have been made to enable the broader use of the Connectom high gradient system for diffusion tractography and tissue microstructure studies and leverage its unique advantages and sensitivity to resolving macroscopic and microscopic structural information in neural tissue for clinical and neuroscientific studies. The goal of this review article is to summarize the technical developments that have emerged in the last decade to support and promote large-scale and scientific studies of the human brain using the Connectom scanner. We provide a brief historical perspective on the development of Connectom gradient technology and the efforts that led to the installation of three other Connectom 3T MRI scanners worldwide – one in the United Kingdom in Cardiff, Wales, another in continental Europe in Leipzig, Germany, and the latest in Asia in Shanghai, China. We summarize the key developments in gradient hardware and image acquisition technology that have formed the backbone of Connectom-related research efforts, including the rich array of high-sensitivity receiver coils, pulse sequences, image artifact correction strategies and data preprocessing methods needed to optimize the quality of high-gradient strength diffusion MRI data for subsequent analyses. Finally, we review the scientific impact of the Connectom MRI scanner, including advances in diffusion tractography, tissue microstructural imaging, ex vivo validation, and clinical investigations that have been enabled by Connectom technology. We conclude with brief insights into the unique value of strong gradients for diffusion MRI and where the field is headed in the coming years.

Diffusion tensor magnetic resonance imaging (DTI) is a widely adopted neuroimaging method for the in vivo mapping of brain tissue microstructure and white matter tracts. Nonetheless, the noise in the diffusion-weighted images (DWIs) decreases the accuracy and precision of DTI derived microstructural parameters and leads to prolonged acquisition time for achieving improved signal-to-noise ratio (SNR). Deep learning-based image denoising using convolutional neural networks (CNNs) has superior performance but often requires additional high-SNR data for supervising the training of CNNs, which reduces the feasibility of supervised learning-based denoising in practice. In this work, we develop a self-supervised deep learning-based method entitled “SDnDTI” for denoising DTI data, which does not require additional high-SNR data for training. Specifically, SDnDTI divides multi-directional DTI data into many subsets of six DWI volumes and transforms DWIs from each subset to along the same diffusion-encoding directions through the diffusion tensor model, generating multiple repetitions of DWIs with identical image contrasts but different noise observations. SDnDTI removes noise by first denoising each repetition of DWIs using a deep 3-dimensional CNN with the average of all repetitions with higher SNR as the training target, following the same approach as normal supervised learning based denoising methods, and then averaging CNN-denoised images for achieving higher SNR. The denoising efficacy of SDnDTI is demonstrated in terms of the similarity of output images and resultant DTI metrics compared to the ground truth generated using substantially more DWI volumes on two datasets with different spatial resolutions, b-values and numbers of input DWI volumes provided by the Human Connectome Project (HCP) and the Lifespan HCP in Aging. The SDnDTI results preserve image sharpness and textural details and substantially improve upon those from the raw data. The results of SDnDTI are comparable to those from supervised learning-based denoising and outperform those from state-of-the-art conventional denoising algorithms including BM4D, AONLM and MPPCA. By leveraging domain knowledge of diffusion MRI physics, SDnDTI makes it easier to use CNN-based denoising methods in practice and has the potential to benefit a wider range of research and clinical applications that require accelerated DTI acquisition and high-quality DTI data for mapping of tissue microstructure, fiber tracts and structural connectivity in the living human brain.

While many useful microstructural indices, as well as orientation distribution functions, can be obtained from multi-shell dMRI data, there is growing interest in exploring the richer set of microstructural features that can be extracted from the full ensemble average propagator (EAP). The EAP can be readily computed from diffusion spectrum imaging (DSI) data, at the cost of a very lengthy acquisition. Compressed sensing (CS) has been used to make DSI more practical by reducing its acquisition time. CS applied to DSI (CS-DSI) attempts to reconstruct the EAP from significantly undersampled q-space data. We present a post mortem validation study where we evaluate the ability of CS-DSI to approximate not only fully sampled DSI but also multi-shell acquisitions with high fidelity. Human brain samples are imaged with high-resolution DSI at 9.4T and with polarization-sensitive optical coherence tomography (PSOCT). The latter provides direct measurements of axonal orientations at microscopic resolutions, allowing us to evaluate the mesoscopic orientation estimates obtained from diffusion MRI, in terms of their angular error and the presence of spurious peaks. We test two fast, dictionary-based, L2-regularized algorithms for CS-DSI reconstruction. We find that, for a CS acceleration factor of R=3, i.e., an acquisition with 171 gradient directions, one of these methods is able to achieve both low angular error and low number of spurious peaks. With a scan length similar to that of high angular resolution multi-shell acquisition schemes, this CS-DSI approach is able to approximate both fully sampled DSI and multi-shell data with high accuracy. Thus it is suitable for orientation reconstruction and microstructural modeling techniques that require either grid- or shell-based acquisitions. We find that the signal-to-noise ratio (SNR) of the training data used to construct the dictionary can have an impact on the accuracy of CS-DSI, but that there is substantial robustness to loss of SNR in the test data. Finally, we show that, as the CS acceleration factor increases beyond R=3, the accuracy of these reconstruction methods degrade, either in terms of the angular error, or in terms of the number of spurious peaks. Our results provide useful benchmarks for the future development of even more efficient q-space acceleration techniques.

The first phase of the Human Connectome Project pioneered advances in MRI technology for mapping the macroscopic structural connections of the living human brain through the engineering of a whole-body human MRI scanner equipped with maximum gradient strength of 300 mT/m, the highest ever achieved for human imaging. While this instrument has made important contributions to the understanding of macroscale connectional topology, it has also demonstrated the potential of dedicated high-gradient performance scanners to provide unparalleled in vivo assessment of neural tissue microstructure. Building on the initial groundwork laid by the original Connectome scanner, we have now embarked on an international, multi-site effort to build the next-generation human 3T Connectome scanner (Connectome 2.0) optimized for the study of neural tissue microstructure and connectional anatomy across multiple length scales. In order to maximize the resolution of this in vivo microscope for studies of the living human brain, we will push the diffusion resolution limit to unprecedented levels by (1) nearly doubling the current maximum gradient strength from 300 mT/m to 500 mT/m and tripling the maximum slew rate from 200 T/m/s to 600 T/m/s through the design of a one-of-a-kind head gradient coil optimized to minimize peripheral nerve stimulation; (2) developing high-sensitivity multi-channel radiofrequency receive coils for in vivo and ex vivo human brain imaging; (3) incorporating dynamic field monitoring to minimize image distortions and artifacts; (4) developing new pulse sequences to integrate the strongest diffusion encoding and highest spatial resolution ever achieved in the living human brain; and (5) calibrating the measurements obtained from this next-generation instrument through systematic validation of diffusion microstructural metrics in high-fidelity phantoms and ex vivo brain tissue at progressively finer scales with accompanying diffusion simulations in histology-based micro-geometries. We envision creating the ultimate diffusion MRI instrument capable of capturing the complex multi-scale organization of the living human brain – from the microscopic scale needed to probe cellular geometry, heterogeneity and plasticity, to the mesoscopic scale for quantifying the distinctions in cortical structure and connectivity that define cyto- and myeloarchitectonic boundaries, to improvements in estimates of macroscopic connectivity.

A recently introduced model-based deep learning (MoDL) technique successfully incorporates convolutional neural network (CNN)-based regularizers into physics-based parallel imaging reconstruction using a small number of network parameters. Wave-controlled aliasing in parallel imaging (CAIPI) is an emerging parallel imaging method that accelerates imaging acquisition by employing sinusoidal gradients in the phase- and slice/partition-encoding directions during the readout to take better advantage of 3D coil sensitivity profiles. We propose wave-encoded MoDL (wave-MoDL) combining the wave-encoding strategy with unrolled network constraints for highly accelerated 3D imaging while enforcing data consistency. We extend wave-MoDL to reconstruct multicontrast data with CAIPI sampling patterns to leverage similarity between multiple images to improve the reconstruction quality. We further exploit this to enable rapid quantitative imaging using an interleaved look-locker acquisition sequence with T2 preparation pulse (3D-QALAS). Wave-MoDL enables a 40 s MPRAGE acquisition at 1 mm resolution at 16-fold acceleration. For quantitative imaging, wave-MoDL permits a 1:50 min acquisition for T1, T2, and proton density mapping at 1 mm resolution at 12-fold acceleration, from which contrast-weighted images can be synthesized as well. In conclusion, wave-MoDL allows rapid MR acquisition and high-fidelity image reconstruction and may facilitate clinical and neuroscientific applications by incorporating unrolled neural networks into wave-CAIPI reconstruction.

Automatic cerebral cortical surface reconstruction is a useful tool for cortical anatomy quantification, analysis and visualization. Recently, the Human Connectome Project and several studies have shown the advantages of using T1-weighted magnetic resonance (MR) images with sub-millimeter isotropic spatial resolution instead of the standard 1-mm isotropic resolution for improved accuracy of cortical surface positioning and thickness estimation. Nonetheless, sub-millimeter resolution images are noisy by nature and require averaging multiple repetitions to increase the signal-to-noise ratio for precisely delineating the cortical boundary. The prolonged acquisition time and potential motion artifacts pose significant barriers to the wide adoption of cortical surface reconstruction at sub-millimeter resolution for a broad range of neuroscientific and clinical applications. We address this challenge by evaluating the cortical surface reconstruction resulting from denoised single-repetition sub-millimeter T1-weighted images. We systematically characterized the effects of image denoising on empirical data acquired at 0.6 mm isotropic resolution using three classical denoising methods, including denoising convolutional neural network (DnCNN), block-matching and 4-dimensional filtering (BM4D) and adaptive optimized non-local means (AONLM). The denoised single-repetition images were found to be highly similar to 6-repetition averaged images, with a low whole-brain averaged mean absolute difference of ~0.016, high whole-brain averaged peak signal-to-noise ratio of ~33.5 dB and structural similarity index of ~0.92, and minimal gray matter–white matter contrast loss (2% to 9%). The whole-brain mean absolute discrepancies in gray matter–white matter surface placement, gray matter–cerebrospinal fluid surface placement and cortical thickness estimation were lower than 165 μm, 155 μm and 145 μm—sufficiently accurate for most applications. These discrepancies were approximately one third to half of those from 1-mm isotropic resolution data. The denoising performance was equivalent to averaging ~2.5 repetitions of the data in terms of image similarity, and 1.6–2.2 repetitions in terms of the cortical surface placement accuracy. The scan-rescan variability of the cortical surface positioning and thickness estimation was lower than 170 μm. Our unique dataset and systematic characterization support the use of denoising methods for improved cortical surface reconstruction at sub-millimeter resolution.

Myelin water imaging techniques based on multi-compartment relaxometry have been developed as an important tool to measure myelin concentration in vivo, but are limited by the long scan time of multi-contrast multi-echo acquisition. In this work, a fast imaging technique, termed variable flip angle Echo Planar Time-Resolved Imaging (vFA-EPTI), is developed to acquire multi-echo and multi-flip-angle gradient-echo data with significantly reduced acquisition time, providing rich information for multi-compartment analysis of gradient-echo myelin water imaging (GRE-MWI). The proposed vFA-EPTI method achieved 26 folds acceleration with good accuracy by utilizing an efficient continuous readout, optimized spatiotemporal encoding across echoes and flip angles, as well as a joint subspace reconstruction. An approach to estimate off-resonance field changes between different flip-angle acquisitions was also developed to ensure high-quality joint reconstruction across flip angles. The accuracy of myelin water fraction (MWF) estimate under high acceleration was first validated by a retrospective undersampling experiment using a lengthy fully-sampled data as reference. Prospective experiments were then performed where whole-brain MWF and multi-compartment quantitative maps were obtained in 5 min at 1.5 mm isotropic resolution and 24 min at 1 mm isotropic resolution at 3T. Additionally, ultra-high resolution data at 600 µm isotropic resolution were acquired at 7T, which show detailed structures within the cortex such as the line of Gennari, demonstrating the ability of the proposed method for submillimeter GRE-MWI that can be used to study cortical myeloarchitecture in vivo.

Diffusion tensor magnetic resonance imaging (DTI) is unsurpassed in its ability to map tissue microstructure and structural connectivity in the living human brain. Nonetheless, the angular sampling requirement for DTI leads to long scan times and poses a critical barrier to performing high-quality DTI in routine clinical practice and large-scale research studies. In this work we present a new processing framework for DTI entitled DeepDTI that minimizes the data requirement of DTI to six diffusion-weighted images (DWIs) required by conventional voxel-wise fitting methods for deriving the six unique unknowns in a diffusion tensor using data-driven supervised deep learning. DeepDTI maps the input non-diffusion-weighted (b ​= ​0) image and six DWI volumes sampled along optimized diffusion-encoding directions, along with T1-weighted and T2-weighted image volumes, to the residuals between the input and high-quality output b = 0 image and DWI volumes using a 10-layer three-dimensional convolutional neural network (CNN). The inputs and outputs of DeepDTI are uniquely formulated, which not only enables residual learning to boost CNN performance but also enables tensor fitting of resultant high-quality DWIs to generate orientational DTI metrics for tractography. The very deep CNN used by DeepDTI leverages the redundancy in local and non-local spatial information and across diffusion-encoding directions and image contrasts in the data. The performance of DeepDTI was systematically quantified in terms of the quality of the output images, DTI metrics, DTI-based tractography and tract-specific analysis results. We demonstrate rotationally-invariant and robust estimation of DTI metrics from DeepDTI that are comparable to those obtained with two b ​= ​0 images and 21 DWIs for the primary eigenvector derived from DTI and two b ​= ​0 images and 26–30 DWIs for various scalar metrics derived from DTI, achieving 3.3–4.6 × ​acceleration, and twice as good as those of a state-of-the-art denoising algorithm at the group level. The twenty major white-matter tracts can be accurately identified from the tractography of DeepDTI results. The mean distance between the core of the major white-matter tracts identified from DeepDTI results and those from the ground-truth results using 18 ​b ​= ​0 images and 90 DWIs measures around 1–1.5 ​mm. DeepDTI leverages domain knowledge of diffusion MRI physics and power of deep learning to render DTI, DTI-based tractography, major white-matter tracts identification and tract-specific analysis more feasible for a wider range of neuroscientific and clinical studies.