RF coil design strategies for improving SNR at the ultrahigh magnetic field of 10.5 Tesla.

Purpose: To develop multichannel transmit and receive arrays towards capturing the ultimate-intrinsic-SNR (uiSNR) at 10.5 Tesla (T) and to demonstrate the feasibility and potential of whole-brain, high-resolution human brain imaging at this high field strength. Methods: A dual row 16-channel self-decoupled transmit (Tx) array was converted to a 16Tx/Rx transceiver using custom transmit/receive switches. A 64-channel receive-only (64Rx) array was built to fit into the 16Tx/Rx array. Electromagnetic modeling and experiments were employed to define safe operation limits of the resulting 16Tx/80Rx array and obtain FDA approval for human use. Results: The 64Rx array alone captured approximately 50% of the central uiSNR at 10.5T while the identical 7T 64Rx array captured ∼76% of uiSNR at this lower field strength. The 16Tx/80Rx configuration brought the fraction of uiSNR captured at 10.5T to levels comparable to the performance of the 64Rx array at 7T. SNR data obtained at the two field strengths with these arrays displayed dependent increases over a large central region. Whole-brain high resolution T 2 * and T 1 weighted anatomical and gradient-recalled echo EPI BOLD fMRI images were obtained at 10.5T for the first time with such an advanced array, illustrating the promise of >10T fields in studying the human brain. Conclusion: We demonstrated the ability to approach the uiSNR at 10.5T over the human brain with a novel, high channel count array, achieving large SNR gains over 7T, currently the most commonly employed ultrahigh field platform, and demonstrate high resolution and high contrast anatomical and functional imaging at 10.5T.

Performance of receive head arrays versus ultimate intrinsic SNR at 7 T and 10.5 T.

PURPOSE: We examined magnetic field dependent SNR gains and ability to capture them with multichannel receive arrays for human head imaging in going from 7 T, the most commonly used ultrahigh magnetic field (UHF) platform at the present, to 10.5 T, which represents the emerging new frontier of >10 T in UHFs. METHODS: Electromagnetic (EM) models of 31-channel and 63-channel multichannel arrays built for 10.5 T were developed for 10.5 T and 7 T simulations. A 7 T version of the 63-channel array with an identical coil layout was also built. Array performance was evaluated in the EM model using a phantom mimicking the size and electrical properties of the human head and a digital human head model. Experimental data was obtained at 7 T and 10.5 T with the 63-channel array. Ultimate intrinsic SNR (uiSNR) was calculated for the two field strengths using a voxelized cloud of dipoles enclosing the phantom or the digital human head model as a reference to assess the performance of the two arrays and field depended SNR gains. RESULTS: uiSNR calculations in both the phantom and the digital human head model demonstrated SNR gains at 10.5 T relative to 7 T of 2.6 centrally, ˜2 at the location corresponding to the edge of the brain, ˜1.4 at the periphery. The EM models demonstrated that, centrally, both arrays captured ˜90% of the uiSNR at 7 T, but only ˜65% at 10.5 T, leading only to ˜2-fold gain in array SNR in going from 7 to 10.5 T. This trend was also observed experimentally with the 63-channel array capturing a larger fraction of the uiSNR at 7 T compared to 10.5 T, although the percentage of uiSNR captured were slightly lower at both field strengths compared to EM simulation results. CONCLUSIONS: Major uiSNR gains are predicted for human head imaging in going from 7 T to 10.5 T, ranging from ˜2-fold at locations corresponding to the edge of the brain to 2.6-fold at the center, corresponding to approximately quadratic increase with the magnetic field. Realistic 31- and 63-channel receive arrays, however, approach the central uiSNR at 7 T, but fail to do so at 10.5 T, suggesting that more coils and/or different type of coils will be needed at 10.5 T and higher magnetic fields.

The possible influence of third-order shim coils on gradient-magnet interactions: an inter-field and inter-site study.

OBJECTIVE: To assess the possible influence of third-order shim coils on the behavior of the gradient field and in gradient-magnet interactions at 7 T and above. MATERIALS AND METHODS: Gradient impulse response function measurements were performed at 5 sites spanning field strengths from 7 to 11.7 T, all of them sharing the same exact whole-body gradient coil design. Mechanical fixation and boundary conditions of the gradient coil were altered in several ways at one site to study the impact of mechanical coupling with the magnet on the field perturbations. Vibrations, power deposition in the He bath, and field dynamics were characterized at 11.7 T with the third-order shim coils connected and disconnected inside the Faraday cage. RESULTS: For the same whole-body gradient coil design, all measurements differed greatly based on the third-order shim coil configuration (connected or not). Vibrations and gradient transfer function peaks could be affected by a factor of 2 or more, depending on the resonances. Disconnecting the third-order shim coils at 11.7 T also suppressed almost completely power deposition peaks at some frequencies. DISCUSSION: Third-order shim coil configurations can have major impact in gradient-magnet interactions with consequences on potential hardware damage, magnet heating, and image quality going beyond EPI acquisitions.

Unsupervised deep learning with convolutional neural networks for static parallel transmit design: A retrospective study.

PURPOSE: To mitigate B 1 + $$ {B}_1^{+} $$ inhomogeneity at 7T for multi-channel transmit arrays using unsupervised deep learning with convolutional neural networks (CNNs). METHODS: Deep learning parallel transmit (pTx) pulse design has received attention, but such methods have relied on supervised training and did not use CNNs for multi-channel B 1 + $$ {B}_1^{+} $$ maps. In this work, we introduce an alternative approach that facilitates the use of CNNs with multi-channel B 1 + $$ {B}_1^{+} $$ maps while performing unsupervised training. The multi-channel B 1 + $$ {B}_1^{+} $$ maps are concatenated along the spatial dimension to enable shift-equivariant processing amenable to CNNs. Training is performed in an unsupervised manner using a physics-driven loss function that minimizes the discrepancy of the Bloch simulation with the target magnetization, which eliminates the calculation of reference transmit RF weights. The training database comprises 3824 2D sagittal, multi-channel B 1 + $$ {B}_1^{+} $$ maps of the healthy human brain from 143 subjects. B 1 + $$ {B}_1^{+} $$ data were acquired at 7T using an 8Tx/32Rx head coil. The proposed method is compared to the unregularized magnitude least-squares (MLS) solution for the target magnetization in static pTx design. RESULTS: The proposed method outperformed the unregularized MLS solution for RMS error and coefficient-of-variation and had comparable energy consumption. Additionally, the proposed method did not show local phase singularities leading to distinct holes in the resulting magnetization unlike the unregularized MLS solution. CONCLUSION: Proposed unsupervised deep learning with CNNs performs better than unregularized MLS in static pTx for speed and robustness.

Locally low-rank denoising in transform domains.

Purpose: To develop an extension to locally low rank (LLR) denoising techniques based on transform domain processing that reduces the number of images required in the MR image series for high-quality denoising. Theory and Methods: LLR methods with random matrix theory-based thresholds are successfully used in the denoising of MR image series in a number of applications. The performance of these methods depend on how well the LLR assumption is satisfied, which deteriorates with few numbers of images, as is commonly encountered in quantitative MRI applications. We propose a transform-domain approach for denoising of MR image series to represent the underlying signal with higher fidelity when using a locally low rank approximation. The efficacy of the method is demonstrated for fully-sampled k-space, undersampled k-space, DICOM images, and complex-valued SENSE-1 images in quantitative MRI applications with as few as 4 images. Results: For both MSK and brain applications, the transform domain denoising preserves local subtle variability, whereas the quantitative maps based on image domain LLR methods tend to be locally more homogeneous. Conclusion: A transform domain extension to LLR denoising produces high quality images and is compatible with both raw k-space data and vendor reconstructed data. This allows for improved imaging and more accurate quantitative analyses and parameters obtained therefrom.

DENOISING DIFFUSION MRI: CONSIDERATIONS AND IMPLICATIONS FOR ANALYSIS.

Development of diffusion MRI (dMRI) denoising approaches has experienced considerable growth over the last years. As noise can inherently reduce accuracy and precision in measurements, its effects have been well characterised both in terms of uncertainty increase in dMRI-derived features and in terms of biases caused by the noise floor, the smallest measurable signal given the noise level. However, gaps in our knowledge still exist in objectively characterising dMRI denoising approaches in terms of both of these effects and assessing their efficacy. In this work, we reconsider what a denoising method should and should not do and we accordingly define criteria to characterise the performance. We propose a comprehensive set of evaluations, including i) benefits in improving signal quality and reducing noise variance, ii) gains in reducing biases and the noise floor and improving, iii) preservation of spatial resolution, iv) agreement of denoised data against a gold standard, v) gains in downstream parameter estimation (precision and accuracy), vi) efficacy in enabling noise-prone applications, such as ultra-high-resolution imaging. We further provide newly acquired complex datasets (magnitude and phase) with multiple repeats that sample different SNR regimes to highlight performance differences under different scenarios. Without loss of generality, we subsequently apply a number of exemplar patch-based denoising algorithms to these datasets, including Non-Local Means, Marchenko-Pastur PCA (MPPCA) in the magnitude and complex domain and NORDIC, and compare them with respect to the above criteria and against a gold standard complex average of multiple repeats. We demonstrate that all tested denoising approaches reduce noise-related variance, but not always biases from the elevated noise floor. They all induce a spatial resolution penalty, but its extent can vary depending on the method and the implementation. Some denoising approaches agree with the gold standard more than others and we demonstrate challenges in even defining such a standard. Overall, we show that dMRI denoising performed in the complex domain is advantageous to magnitude domain denoising with respect to all the above criteria.

Mapping the evolution of regional brain network efficiency and its association with cognitive abilities during the first twenty-eight months of life.

Human brain undergoes rapid growth during the first few years of life. While previous research has employed graph theory to study early brain development, it has mostly focused on the topological attributes of the whole brain. However, examining regional graph-theory features may provide unique insights into the development of cognitive abilities. Utilizing a large and longitudinal rsfMRI dataset from the UNC/UMN Baby Connectome Project, we investigated the developmental trajectories of regional efficiency and evaluated the relationships between these changes and cognitive abilities using Mullen Scales of Early Learning during the first twenty-eight months of life. Our results revealed a complex and spatiotemporally heterogeneous development pattern of regional global and local efficiency during this age period. Furthermore, we found that the trajectories of the regional global efficiency at the left temporal occipital fusiform and bilateral occipital fusiform gyri were positively associated with cognitive abilities, including visual reception, expressive language, receptive language, and early learning composite scores (P < 0.05, FDR corrected). However, these associations were weakened with age. These findings offered new insights into the regional developmental features of brain topologies and their associations with cognition and provided evidence of ongoing optimization of brain networks at both whole-brain and regional levels.

A 32-Channel Sleeve Antenna Receiver Array for Human Head MRI Applications at 10.5 T.

For human brain magnetic resonance imaging (MRI), high channel count ( ≥ 32 ) radiofrequency receiver coil arrays are utilized to achieve maximum signal-to-noise ratio (SNR) and to accelerate parallel imaging techniques. With ultra-high field (UHF) MRI at 7 tesla (T) and higher, dipole antenna arrays have been shown to generate high SNR in the deep regions of the brain, however the array elements exhibit increased electromagnetic coupling with one another, making array construction more difficult with the increasing number of elements. Compared to a classical dipole antenna array, a sleeve antenna array incorporates the coaxial ground into the feed-point, resulting in a modified asymmetric antenna structure with improved intra-element decoupling. Here, we extended our previous 16-channel sleeve transceiver work and developed a 32-channel azimuthally arranged sleeve antenna receive-only array for 10.5 T human brain imaging. We experimentally compared the achievable SNR of the sleeve antenna array at 10.5 T to a more traditional 32-channel loop array bult onto a human head-shaped former. The results obtained with a head shaped phantom clearly demonstrated that peripheral intrinsic SNR can be significantly improved compared to a loop array with the same number of elements- except for the superior part of the phantom where sleeve antenna elements are not located.

Default mode network failure and neurodegeneration across aging and amnestic and dysexecutive Alzheimer's disease.

From a complex systems perspective, clinical syndromes emerging from neurodegenerative diseases are thought to result from multiscale interactions between aggregates of misfolded proteins and the disequilibrium of large-scale networks coordinating functional operations underpinning cognitive phenomena. Across all syndromic presentations of Alzheimer's disease, age-related disruption of the default mode network is accelerated by amyloid deposition. Conversely, syndromic variability may reflect selective neurodegeneration of modular networks supporting specific cognitive abilities. In this study, we leveraged the breadth of the Human Connectome Project-Aging cohort of non-demented individuals (N = 724) as a normative cohort to assess the robustness of a biomarker of default mode network dysfunction in Alzheimer's disease, the network failure quotient, across the aging spectrum. We then examined the capacity of the network failure quotient and focal markers of neurodegeneration to discriminate patients with amnestic (N = 8) or dysexecutive (N = 10) Alzheimer's disease from the normative cohort at the patient level, as well as between Alzheimer's disease phenotypes. Importantly, all participants and patients were scanned using the Human Connectome Project-Aging protocol, allowing for the acquisition of high-resolution structural imaging and longer resting-state connectivity acquisition time. Using a regression framework, we found that the network failure quotient related to age, global and focal cortical thickness, hippocampal volume, and cognition in the normative Human Connectome Project-Aging cohort, replicating previous results from the Mayo Clinic Study of Aging that used a different scanning protocol. Then, we used quantile curves and group-wise comparisons to show that the network failure quotient commonly distinguished both dysexecutive and amnestic Alzheimer's disease patients from the normative cohort. In contrast, focal neurodegeneration markers were more phenotype-specific, where the neurodegeneration of parieto-frontal areas associated with dysexecutive Alzheimer's disease, while the neurodegeneration of hippocampal and temporal areas associated with amnestic Alzheimer's disease. Capitalizing on a large normative cohort and optimized imaging acquisition protocols, we highlight a biomarker of default mode network failure reflecting shared system-level pathophysiological mechanisms across aging and dysexecutive and amnestic Alzheimer's disease and biomarkers of focal neurodegeneration reflecting distinct pathognomonic processes across the amnestic and dysexecutive Alzheimer's disease phenotypes. These findings provide evidence that variability in inter-individual cognitive impairment in Alzheimer's disease may relate to both modular network degeneration and default mode network disruption. These results provide important information to advance complex systems approaches to cognitive aging and degeneration, expand the armamentarium of biomarkers available to aid diagnosis, monitor progression and inform clinical trials.

Evaluating increases in sensitivity from NORDIC for diverse fMRI acquisition strategies.

As the neuroimaging field moves towards detecting smaller effects at higher spatial resolutions, and faster sampling rates, there is increased attention given to the deleterious contribution of unstructured, thermal noise. Here, we critically evaluate the performance of a recently developed reconstruction method, termed NORDIC, for suppressing thermal noise using datasets acquired with various field strengths, voxel sizes, sampling rates, and task designs. Following minimal preprocessing, statistical activation (t-values) of NORDIC processed data was compared to the results obtained with alternative denoising methods. Additionally, we examined the consistency of the estimates of task responses at the single-voxel, single run level, using a finite impulse response (FIR) model. To examine the potential impact on effective image resolution, the overall smoothness of the data processed with different methods was estimated. Finally, to determine if NORDIC alters or removes temporal information important for modeling responses, we employed an exhaustive leave-p-out cross validation approach, using FIR task responses to predict held out timeseries, quantified using R2. After NORDIC, the t-values are increased, an improvement comparable to what could be achieved by 1.5 voxels smoothing, and task events are clearly visible and have less cross-run error. These advantages are achieved with smoothness estimates increasing by less than 4%, while 1.5 voxel smoothing is associated with increases of over 140%. Cross-validated R2s based on the FIR models show that NORDIC is not measurably distorting the temporal structure of the data under this approach and is the best predictor of non-denoised time courses. The results demonstrate that analyzing 1 run of data after NORDIC produces results equivalent to using 2 to 3 original runs and that NORDIC performs equally well across a diverse array of functional imaging protocols. Significance Statement: For functional neuroimaging, the increasing availability of higher field strengths and ever higher spatiotemporal resolutions has led to concomitant increase in concerns about the deleterious effects of thermal noise. Historically this noise source was suppressed using methods that reduce spatial precision such as image blurring or averaging over a large number of trials or sessions, which necessitates large data collection efforts. Here, we critically evaluate the performance of a recently developed reconstruction method, termed NORDIC, which suppresses thermal noise. Across datasets varying in field strength, voxel sizes, sampling rates, and task designs, NORDIC produces substantial gains in data quality. Both conventional t-statistics derived from general linear models and coefficients of determination for predicting unseen data are improved. These gains match or even exceed those associated with 1 voxel Full Width Half Max image smoothing, however, even such small amounts of smoothing are associated with a 52% reduction in estimates of spatial precision, whereas the measurable difference in spatial precision is less than 4% following NORDIC.

Reconstruction for 7T high-resolution whole-brain diffusion MRI using two-stage N/2 ghost correction and L1-SPIRiT without single-band reference.

PURPOSE: To combine a new two-stage N/2 ghost correction and an adapted L1-SPIRiT method for reconstruction of 7T highly accelerated whole-brain diffusion MRI (dMRI) using only autocalibration scans (ACS) without the need of additional single-band reference (SBref) scans. METHODS: The proposed ghost correction consisted of a 3-line reference approach in stage 1 and the reference-free entropy method in stage 2. The adapted L1-SPIRiT method was formulated within the 3D k-space framework. Its efficacy was examined by acquiring two dMRI data sets at 1.05-mm isotropic resolutions with a total acceleration of 6 or 9 (i.e., 2-fold or 3-fold slice and 3-fold in-plane acceleration). Diffusion analysis was performed to derive DTI metrics and estimate fiber orientation distribution functions (fODFs). The results were compared with those of 3D k-space GRAPPA using only ACS, all in reference to 3D k-space GRAPPA using both ACS and SBref (serving as a reference). RESULTS: The proposed ghost correction eliminated artifacts more robustly than conventional approaches. Our adapted L1-SPIRiT method outperformed 3D k-space GRAPPA when using only ACS, improving image quality to what was achievable with 3D k-space GRAPPA using both ACS and SBref scans. The improvement in image quality further resulted in an improvement in estimation performances for DTI and fODFs. CONCLUSION: The combination of our new ghost correction and adapted L1-SPIRiT method can reliably reconstruct 7T highly accelerated whole-brain dMRI without the need of SBref scans, increasing acquisition efficiency and reducing motion sensitivity.

Motion robust magnetic resonance imaging via efficient Fourier aggregation.

We present a method for suppressing motion artifacts in anatomical magnetic resonance acquisitions. Our proposed technique, termed MOTOR-MRI, can recover and salvage images which are otherwise heavily corrupted by motion induced artifacts and blur which renders them unusable. Contrary to other techniques, MOTOR-MRI operates on the reconstructed images and not on k-space data. It relies on breaking the standard acquisition protocol into several shorter ones (while maintaining the same total acquisition time) and subsequent efficient aggregation in Fourier space of locally sharp and consistent information among them, producing a sharp and motion mitigated image. We demonstrate the efficacy of the technique on T2-weighted turbo spin echo magnetic resonance brain scans with severe motion corruption from both 3 T and 7 T scanners and show significant qualitative and quantitative improvement in image quality. MOTOR-MRI can operate independently, or in conjunction with additional motion correction methods.

Heritability of brain neurovascular coupling.

The moment-to-moment variation of neurovascular coupling in the brain was determined by computing the moment-to-moment turnover of the blood-oxygen-level-dependent signal (TBOLD) at resting state. Here we show that 1) TBOLD is heritable, 2) its heritability estimates are highly correlated between left and right hemispheres, and 3) the degree of its heritability is determined, in part, by the anatomical proximity of the brain areas involved. We also show that the regional distribution of TBOLD in the cortex is significantly associated with that of the vesicular acetylcholine transporter. These findings establish that TBOLD as a key heritable measure of local cortical brain function captured by neurovascular coupling.NEW & NOTEWORTHY Here we show that the sample-to-sample turnover of the resting state fMRI blood-oxygen-level-dependent turnover (TBOLD) is heritable, the left and right hemisphere TBOLD heritabilities are highly correlated, and TBOLD heritability varies among cortical areas. Moreover, we documented that TBOLD is associated with the regional cortical distribution of the vesicular acetylcholine transporter.

Revisiting [Formula: see text]-wavelet compressed-sensing MRI in the era of deep learning.

Following their success in numerous imaging and computer vision applications, deep-learning (DL) techniques have emerged as one of the most prominent strategies for accelerated MRI reconstruction. These methods have been shown to outperform conventional regularized methods based on compressed sensing (CS). However, in most comparisons, CS is implemented with two or three hand-tuned parameters, while DL methods enjoy a plethora of advanced data science tools. In this work, we revisit [Formula: see text]-wavelet CS reconstruction using these modern tools. Using ideas such as algorithm unrolling and advanced optimization methods over large databases that DL algorithms utilize, along with conventional insights from wavelet representations and CS theory, we show that [Formula: see text]-wavelet CS can be fine-tuned to a level close to DL reconstruction for accelerated MRI. The optimized [Formula: see text]-wavelet CS method uses only 128 parameters compared to >500,000 for DL, employs a convex reconstruction at inference time, and performs within <1% of a DL approach that has been used in multiple studies in terms of quantitative quality metrics.

Multi-mask self-supervised learning for physics-guided neural networks in highly accelerated magnetic resonance imaging.

Self-supervised learning has shown great promise because of its ability to train deep learning (DL) magnetic resonance imaging (MRI) reconstruction methods without fully sampled data. Current self-supervised learning methods for physics-guided reconstruction networks split acquired undersampled data into two disjoint sets, where one is used for data consistency (DC) in the unrolled network, while the other is used to define the training loss. In this study, we propose an improved self-supervised learning strategy that more efficiently uses the acquired data to train a physics-guided reconstruction network without a database of fully sampled data. The proposed multi-mask self-supervised learning via data undersampling (SSDU) applies a holdout masking operation on the acquired measurements to split them into multiple pairs of disjoint sets for each training sample, while using one of these pairs for DC units and the other for defining loss, thereby more efficiently using the undersampled data. Multi-mask SSDU is applied on fully sampled 3D knee and prospectively undersampled 3D brain MRI datasets, for various acceleration rates and patterns, and compared with the parallel imaging method, CG-SENSE, and single-mask SSDU DL-MRI, as well as supervised DL-MRI when fully sampled data are available. The results on knee MRI show that the proposed multi-mask SSDU outperforms SSDU and performs as well as supervised DL-MRI. A clinical reader study further ranks the multi-mask SSDU higher than supervised DL-MRI in terms of signal-to-noise ratio and aliasing artifacts. Results on brain MRI show that multi-mask SSDU achieves better reconstruction quality compared with SSDU. The reader study demonstrates that multi-mask SSDU at R = 8 significantly improves reconstruction compared with single-mask SSDU at R = 8, as well as CG-SENSE at R = 2.

Magnetic field strength dependent SNR gain at the center of a spherical phantom and up to 11.7T.

PURPOSE: The SNR at the center of a spherical phantom of known electrical properties was measured in quasi-identical experimental conditions as a function of magnetic field strength between 3 T and 11.7 T. METHODS: The SNR was measured at the center of a spherical water saline phantom with a gradient-recalled echo sequence. Measurements were performed at NeuroSpin at 3, 7, and 11.7 T. The phantom was then shipped to Maastricht University and then to the University of Minnesota for additional data points at 7, 9.4, and 10.5 T. Experiments were carried out with the exact same type of birdcage volume coil (except at 3 T, where a similar coil was used) to attempt at isolating the evolution of SNR with field strength alone. Phantom electrical properties were characterized over the corresponding frequency range. RESULTS: Electrical properties were found to barely vary over the frequency range. Removing the influence of the flip-angle excitation inhomogeneity was crucial, as expected. After such correction, measurements revealed a gain of SNR growing as B0 1.94 ± 0.16 compared with B0 2.13 according to ultimate intrinsic SNR theory. CONCLUSIONS: By using quasi-identical experimental setups (RF volume coil, phantom, electrical properties, and protocol), this work reports experimental data between 3 T and 11.7 T, enabling the comparison with SNR theories in which conductivity and permittivity can be assumed to be constant with respect to field strength. According to ultimate SNR theory, these results can be reasonably extrapolated to the performance of receive arrays with greater than about 32 elements for central SNR in the same spherical phantom.

Resting-state functional connectivity identifies individuals and predicts age in 8-to-26-month-olds.

Resting-state functional connectivity (rsFC) measured with fMRI has been used to characterize functional brain maturation in typically and atypically developing children and adults. However, its reliability and utility for predicting development in infants and toddlers is less well understood. Here, we use fMRI data from the Baby Connectome Project study to measure the reliability and uniqueness of rsFC in infants and toddlers and predict age in this sample (8-to-26 months old; n = 170). We observed medium reliability for within-session infant rsFC in our sample, and found that individual infant and toddler's connectomes were sufficiently distinct for successful functional connectome fingerprinting. Next, we trained and tested support vector regression models to predict age-at-scan with rsFC. Models successfully predicted novel infants' age within ± 3.6 months error and a prediction R2 = .51. To characterize the anatomy of predictive networks, we grouped connections into 11 infant-specific resting-state functional networks defined in a data-driven manner. We found that connections between regions of the same network-i.e. within-network connections-predicted age significantly better than between-network connections. Looking ahead, these findings can help characterize changes in functional brain organization in infancy and toddlerhood and inform work predicting developmental outcome measures in this age range.

Residual RAKI: A hybrid linear and non-linear approach for scan-specific k-space deep learning.

Parallel imaging is the most clinically used acceleration technique for magnetic resonance imaging (MRI) in part due to its easy inclusion into routine acquisitions. In k-space based parallel imaging reconstruction, sub-sampled k-space data are interpolated using linear convolutions. At high acceleration rates these methods have inherent noise amplification and reduced image quality. On the other hand, non-linear deep learning methods provide improved image quality at high acceleration, but the availability of training databases for different scans, as well as their interpretability hinder their adaptation. In this work, we present an extension of Robust Artificial-neural-networks for k-space Interpolation (RAKI), called residual-RAKI (rRAKI), which achieves scan-specific machine learning reconstruction using a hybrid linear and non-linear methodology. In rRAKI, non-linear CNNs are trained jointly with a linear convolution implemented via a skip connection. In effect, the linear part provides a baseline reconstruction, while the non-linear CNN that runs in parallel provides further reduction of artifacts and noise arising from the linear part. The explicit split between the linear and non-linear aspects of the reconstruction also help improve interpretability compared to purely non-linear methods. Experiments were conducted on the publicly available fastMRI datasets, as well as high-resolution anatomical imaging, comparing GRAPPA and its variants, compressed sensing, RAKI, Scan Specific Artifact Reduction in K-space (SPARK) and the proposed rRAKI. Additionally, highly-accelerated simultaneous multi-slice (SMS) functional MRI reconstructions were also performed, where the proposed rRAKI was compred to Read-out SENSE-GRAPPA and RAKI. Our results show that the proposed rRAKI method substantially improves the image quality compared to conventional parallel imaging, and offers sharper images compared to SPARK and ℓ1-SPIRiT. Furthermore, rRAKI shows improved preservation of time-varying dynamics compared to both parallel imaging and RAKI in highly-accelerated SMS fMRI.