Evaluation of principal component analysis image denoising on multi-exponential MRI relaxometry

High-resolution perfusion imaging in rodents using pCASL at 9.4 T

Adequate perfusion is critical for maintaining normal brain function and aberrations thereof are hallmarks of many diseases. Pseudo-Continuous Arterial Spin Labeling (pCASL) MRI enables noninvasive quantitative perfusion mapping without contrast agent injection and with a higher signal-to-noise ratio (SNR) than alternative methods. Despite its great potential, pCASL remains challenging, unstable, and relatively low-resolution in rodents - especially in mice - thereby limiting the investigation of perfusion properties in many transgenic or other relevant rodent models of disease. Here, we address this gap by developing a novel experimental setup for high-resolution pCASL imaging in mice and combining it with the enhanced SNR of cryogenic probes. We show that our new experimental setup allows for optimal positioning of the carotids within the cryogenic coil, rendering labeling reproducible. With the proposed methodology, we managed to increase the spatial resolution of pCASL perfusion images by a factor of 15 in mice; a factor of 6 in rats is gained compared to the state of the art just by virtue of the cryogenic coil. We also show that the improved pCASL perfusion imaging allows much better delineation of specific brain areas, both in healthy animals as well as in rat and mouse models of stroke. Our results bode well for future high-definition pCASL perfusion imaging in rodents.

Microstructure-Informed Myelin Mapping (MIMM) from routine multi-echo gradient echo data using multiscale physics modeling of iron and myelin effects and QSM

PURPOSE

Myelin quantification is used in the study of demyelination in neurodegenerative diseases. A novel noninvasive MRI method, Microstructure-Informed Myelin Mapping (MIMM), is proposed to quantify the myelin volume fraction (MVF) from a routine multi-gradient echo sequence (mGRE) using a multiscale biophysical signal model of the effects of microstructural myelin and iron.

THEORY AND METHODS

In MIMM, the effects of myelin are modeled based on the Hollow Cylinder Fiber Model accounting for anisotropy, while iron is considered as an isotropic paramagnetic point source. This model is used to create a dictionary of mGRE magnitude signal evolution and total voxel susceptibility using finite elements of size 0.2 μm. Next, voxel-by-voxel stochastic matching pursuit between acquired mGRE data (magnitude+QSM) and the pre-computed dictionary generates quantitative MVF and iron susceptibility maps. Dictionary matching was evaluated under three conditions: (1) without fiber orientation (basic), (2) with fiber orientation obtained using DTI, and (3) with fiber orientation obtained using an atlas (atlas). MIMM was compared with the three-pool complex fitting (3PCF) using T2-relaxometry myelin water fraction (MWF) map as reference.

RESULTS

The DTI MIMM and atlas MIMM approaches were equally effective in reducing the overestimation of MVF in certain white matter tracts observed in the basic MIMM approach, and they both showed good agreement with T2-relaxometry MWF. MIMM MVF reduced myelin overestimation of globus pallidus observed in 3PCF MWF.

CONCLUSION

MIMM processing of mGRE data can provide MVF maps from routine clinical scans without requiring special sequences.

Self-supervised learning for denoising of multidimensional MRI data

PURPOSE

To develop a fast denoising framework for high-dimensional MRI data based on a self-supervised learning scheme, which does not require ground truth clean image.

THEORY AND METHODS

Quantitative MRI faces limitations in SNR, because the variation of signal amplitude in a large set of images is the key mechanism for quantification. In addition, the complex non-linear signal models make the fitting process vulnerable to noise. To address these issues, we propose a fast deep-learning framework for denoising, which efficiently exploits the redundancy in multidimensional MRI data. A self-supervised model was designed to use only noisy images for training, bypassing the challenge of clean data paucity in clinical practice. For validation, we used two different datasets of simulated magnetization transfer contrast MR fingerprinting (MTC-MRF) dataset and in vivo DWI image dataset to show the generalizability.

RESULTS

The proposed method drastically improved denoising performance in the presence of mild-to-severe noise regardless of noise distributions compared to previous methods of the BM3D, tMPPCA, and Patch2self. The improvements were even pronounced in the following quantification results from the denoised images.

CONCLUSION

The proposed MD-S2S (Multidimensional-Self2Self) denoising technique could be further applied to various multi-dimensional MRI data and improve the quantification accuracy of tissue parameter maps.

NOise Reduction with DIstribution Corrected (NORDIC) principal component analysis improves brain activity detection across rodent and human functional MRI contexts

NOise Reduction with DIstribution Corrected (NORDIC) principal component analysis (PCA) has been shown to selectively suppress thermal noise and improve the temporal signal-to-noise ratio (tSNR) in human functional magnetic resonance imaging (fMRI). However, the feasibility to improve data quality for rodent fMRI using NORDIC PCA remains uncertain. NORDIC PCA may also be particularly beneficial for improving topological brain mapping, as conventional mapping requires precise spatiotemporal signals from large datasets (ideally ~1 hour acquisition) for individual representations. In this study, we evaluated the effects of NORDIC PCA compared with "Standard" processing in various rodent fMRI contexts that range from task-evoked optogenetic fMRI to resting-state fMRI. We also evaluated the effects of NORDIC PCA on human resting-state and retinotopic mapping fMRI via population receptive field (pRF) modeling. In rodent optogenetic fMRI, apart from doubling the tSNR, NORDIC PCA resulted in a larger number of activated voxels and a significant decrease in the variance of evoked brain responses without altering brain morphology. In rodent resting-state fMRI, we found that NORDIC PCA induced a nearly threefold increase in tSNR and preserved task-free relative cerebrovascular reactivity (rCVR) across cortical depth. NORDIC PCA further improved the detection of TGN020-induced aquaporin-4 inhibition on rCVR compared with Standard processing without NORDIC PCA. NORDIC PCA also increased the tSNR for both human resting-state and pRF fMRI, and for the latter also increased activation cluster sizes while retaining retinotopic organization. This suggests that NORDIC PCA preserves the spatiotemporal precision of fMRI signals needed for pRF analysis, and effectively captures small activity changes with high sensitivity. Taken together, these results broadly demonstrate the value of NORDIC PCA for the enhanced detection of neural dynamics across various rodent and human fMRI contexts. This can in turn play an important role in improving fMRI image quality and sensitivity for translational and preclinical neuroimaging research.

Tensor denoising of quantitative multi-parameter mapping

PURPOSE

Quantitative multi-parameter mapping (MPM) provides maps of physical quantities representing physiologically meaningful tissue characteristics, which allows to investigate microstructure-function relationships reflecting normal or pathologic processes in the brain. However, the achievable spatial resolution and stability of MPM for basic research or clinical applications is severely constrained by SNR limits of the MR acquisition process, resulting in relatively long acquisition times. To increase SNR, we denoise MPM acquisitions using principal component analysis along tensors exploiting the Marchenko-Pastur law (tMPPCA).

METHODS

tMPPCA denoising was applied to three sets of MPM raw data before the quantification of maps of proton density, magnetization transfer saturation, R1, and R2*. The regional SNR gain for high-resolution MPM was investigated as well as reproducibility gains for clinically optimized protocols with moderate and high acceleration factors at different image resolutions.

RESULTS

Substantial noise reduction in raw data was achieved, resulting in reduced noise for quantitative mapping up to sixfold without introducing bias of mean values (below 1%). Scan-rescan fluctuations were reduced up to threefold. Denoising allowed to decrease the voxel volume fourfold at the same scan time or reduce the scan time twofold at same voxel volume without loss of sensitivity.

CONCLUSIONS

tMPPCA denoising can (a) improve of fine spatial and temporal patterns, (b) considerably reduce scan time for clinical applications, or (c) increase resolution to potentially push cutting-edge MPM protocols from the upper to the lower limit of the mesoscopic scale.

A comprehensive protocol for quantitative magnetic resonance imaging of the brain at 3 Tesla

Quantitative MRI (qMRI) has been shown to be clinically useful for numerous applications in the brain and body. The development of rapid, accurate, and reproducible qMRI techniques offers access to new multiparametric data, which can provide a comprehensive view of tissue pathology. This work introduces a multiparametric qMRI protocol along with full postprocessing pipelines, optimized for brain imaging at 3 Tesla and using state-of-the-art qMRI tools. The total scan time is under 50 minutes and includes eight pulse-sequences, which produce range of quantitative maps including T1, T2, and T2* relaxation times, magnetic susceptibility, water and macromolecular tissue fractions, mean diffusivity and fractional anisotropy, magnetization transfer ratio (MTR), and inhomogeneous MTR. Practical tips and limitations of using the protocol are also provided and discussed. Application of the protocol is presented on a cohort of 28 healthy volunteers and 12 brain regions-of-interest (ROIs). Quantitative values agreed with previously reported values. Statistical analysis revealed low variability of qMRI parameters across subjects, which, compared to intra-ROI variability, was x4.1 ± 0.9 times higher on average. Significant and positive linear relationship was found between right and left hemispheres' values for all parameters and ROIs with Pearson correlation coefficients of r>0.89 (P<0.001), and mean slope of 0.95 ± 0.04. Finally, scan-rescan stability demonstrated high reproducibility of the measured parameters across ROIs and volunteers, with close-to-zero mean difference and without correlation between the mean and difference values (across map types, mean P value was 0.48 ± 0.27). The entire quantitative data and postprocessing scripts described in the manuscript are publicly available under dedicated GitHub and Figshare repositories. The quantitative maps produced by the presented protocol can promote longitudinal and multi-center studies, and improve the biological interpretability of qMRI by integrating multiple metrics that can reveal information, which is not apparent when examined using only a single contrast mechanism.

In vivo disentanglement of diffusion frequency-dependence, tensor shape, and relaxation using multidimensional MRI

Diffusion MRI with free gradient waveforms, combined with simultaneous relaxation encoding, referred to as multidimensional MRI (MD-MRI), offers microstructural specificity in complex biological tissue. This approach delivers intravoxel information about the microstructure, local chemical composition, and importantly, how these properties are coupled within heterogeneous tissue containing multiple microenvironments. Recent theoretical advances incorporated diffusion time dependency and integrated MD-MRI with concepts from oscillating gradients. This framework probes the diffusion frequency,ω $$ \omega $$ , in addition to the diffusion tensor,D $$ \mathbf{D} $$ , and relaxation,R 1 $$ {R}_1 $$ ,R 2 $$ {R}_2 $$ , correlations. AD ω - R 1 - R 2 $$ \mathbf{D}\left(\omega \right)-{R}_1-{R}_2 $$ clinical imaging protocol was then introduced, with limited brain coverage and 3 mm3 voxel size, which hinder brain segmentation and future cohort studies. In this study, we introduce an efficient, sparse in vivo MD-MRI acquisition protocol providing whole brain coverage at 2 mm3 voxel size. We demonstrate its feasibility and robustness using a well-defined phantom and repeated scans of five healthy individuals. Additionally, we test different denoising strategies to address the sparse nature of this protocol, and show that efficient MD-MRI encoding design demands a nuanced denoising approach. The MD-MRI framework provides rich information that allows resolving the diffusion frequency dependence into intravoxel components based on theirD ω - R 1 - R 2 $$ \mathbf{D}\left(\omega \right)-{R}_1-{R}_2 $$ distribution, enabling the creation of microstructure-specific maps in the human brain. Our results encourage the broader adoption and use of this new imaging approach for characterizing healthy and pathological tissues.

A user independent denoising method for x-nuclei MRI and MRS

PURPOSE

X-nuclei (also called non-proton MRI) MRI and spectroscopy are limited by the intrinsic low SNR as compared to conventional proton imaging. Clinical translation of x-nuclei examination warrants the need of a robust and versatile tool improving image quality for diagnostic use. In this work, we compare a novel denoising method with fewer inputs to the current state-of-the-art denoising method.

METHODS

Denoising approaches were compared on human acquisitions of sodium (23 Na) brain, deuterium (2 H) brain, carbon (13 C) heart and brain, and simulated dynamic hyperpolarized 13 C brain scans, with and without additional noise. The current state-of-the-art denoising method Global-local higher order singular value decomposition (GL-HOSVD) was compared to the few-input method tensor Marchenko-Pastur principal component analysis (tMPPCA). Noise-removal was quantified by residual distributions, and statistical analyses evaluated the differences in mean-square-error and Bland-Altman analysis to quantify agreement between original and denoised results of noise-added data.

RESULTS

GL-HOSVD and tMPPCA showed similar performance for the variety of x-nuclei data analyzed in this work, with tMPPCA removing ˜5% more noise on average over GL-HOSVD. The mean ratio between noise-added and denoising reproducibility coefficients of the Bland-Altman analysis when compared to the original are also similar for the two methods with 3.09 ± 1.03 and 2.83 ± 0.79 for GL-HOSVD and tMPPCA, respectively.

CONCLUSION

The strength of tMPPCA lies in the few-input approach, which generalizes well to different data sources. This makes the use of tMPPCA denoising a robust and versatile tool in x-nuclei imaging improvements and the preferred denoising method.

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.

The CALIPR framework for highly accelerated myelin water imaging with improved precision and sensitivity

Quantitative magnetic resonance imaging (MRI) techniques are powerful tools for the study of human tissue, but, in practice, their utility has been limited by lengthy acquisition times. Here, we introduce the Constrained, Adaptive, Low-dimensional, Intrinsically Precise Reconstruction (CALIPR) framework in the context of myelin water imaging (MWI); a quantitative MRI technique generally regarded as the most rigorous approach for noninvasive, in vivo measurement of myelin content. The CALIPR framework exploits data redundancy to recover high-quality images from a small fraction of an imaging dataset, which allowed MWI to be acquired with a previously unattainable sequence (fully sampled acquisition 2 hours:57 min:20 s) in 7 min:26 s (4.2% of the dataset, acceleration factor 23.9). CALIPR quantitative metrics had excellent precision (myelin water fraction mean coefficient of variation 3.2% for the brain and 3.0% for the spinal cord) and markedly increased sensitivity to demyelinating disease pathology compared to a current, widely used technique. The CALIPR framework facilitates drastically improved MWI and could be similarly transformative for other quantitative MRI applications.

In vivo disentanglement of diffusion frequency-dependence, tensor shape, and relaxation using multidimensional MRI

Diffusion MRI with free gradient waveforms, combined with simultaneous relaxation encoding, referred to as multidimensional MRI (MD-MRI), offers microstructural specificity in complex biological tissue. This approach delivers intravoxel information about the microstructure, local chemical composition, and importantly, how these properties are coupled within heterogeneous tissue containing multiple microenvironments. Recent theoretical advances incorporated diffusion time dependency and integrated MD-MRI with concepts from oscillating gradients. This framework probes the diffusion frequency, ω , in addition to the diffusion tensor, D , and relaxation, R 1 , R 2 , correlations. A D ( ω ) - R 1 - R 2 clinical imaging protocol was then introduced, with limited brain coverage and 3 mm3 voxel size, which hinder brain segmentation and future cohort studies. In this study, we introduce an efficient, sparse in vivo MD-MRI acquisition protocol providing whole brain coverage at 2 mm3 voxel size. We demonstrate its feasibility and robustness using a well-defined phantom and repeated scans of five healthy individuals. Additionally, we test different denoising strategies to address the sparse nature of this protocol, and show that efficient MD-MRI encoding design demands a nuanced denoising approach. The MD-MRI framework provides rich information that allows resolving the diffusion frequency dependence into intravoxel components based on their D ( ω ) - R 1 - R 2 distribution, enabling the creation of microstructure-specific maps in the human brain. Our results encourage the broader adoption and use of this new imaging approach for characterizing healthy and pathological tissues.

Feasibility of free-breathing 19 F MRI image acquisition to characterize ventilation defects in CF and healthy volunteers at wash-in

PURPOSE

To explore the feasibility of measuring ventilation defect percentage (VDP) using 19 F MRI during free-breathing wash-in of fluorinated gas mixture with postacquisition denoising and to compare these results with those obtained through traditional Cartesian breath-hold acquisitions.

METHODS

Eight adults with cystic fibrosis and 5 healthy volunteers completed a single MR session on a Siemens 3T Prisma. 1 H Ultrashort-TE MRI sequences were used for registration and masking, and ventilation images with 19 F MRI were obtained while the subjects breathed a normoxic mixture of 79% perfluoropropane and 21% oxygen (O2 ). 19 F MRI was performed during breath holds and while free breathing with one overlapping spiral scan at breath hold for VDP value comparison. The 19 F spiral data were denoised using a low-rank matrix recovery approach.

RESULTS

VDP measured using 19 F VIBE and 19 F spiral images were highly correlated (r = 0.84) at 10 wash-in breaths. Second-breath VDPs were also highly correlated (r = 0.88). Denoising greatly increased SNR (pre-denoising spiral SNR, 2.46 ± 0.21; post-denoising spiral SNR, 33.91 ± 6.12; and breath-hold SNR, 17.52 ± 2.08).

CONCLUSION

Free-breathing 19 F lung MRI VDP analysis was feasible and highly correlated with breath-hold measurements. Free-breathing methods are expected to increase patient comfort and extend ventilation MRI use to patients who are unable to perform breath holds, including younger subjects and those with more severe lung disease.

Tuned bipolar oscillating gradients for mapping frequency dispersion of diffusion kurtosis in the human brain

PURPOSE

Oscillating gradient spin-echo (OGSE) sequences have demonstrated an ability to probe time-dependent microstructural features, although they often suffer from low SNR due to increased TEs. In this work we introduce frequency-tuned bipolar (FTB) gradients as a variation of oscillating gradients with reduced TE and demonstrate their utility by mapping the frequency dispersion of kurtosis in human subjects.

METHODS

An FTB oscillating gradient waveform is presented that provides encoding of 1.5 net oscillation periods, thereby reducing the TE of the acquisition. Simulations were performed to determine an optimal protocol based on the SNR of kurtosis frequency dispersion-defined as the difference in kurtosis between pulsed and oscillating gradient acquisitions. Healthy human subjects were scanned at 7T using pulsed gradient and an optimized 23 Hz FTB protocol, which featured a maximum b-value of 2500 s/mm² . In addition, to directly compare existing methods, measurements using traditional cosine OGSE were also acquired.

RESULTS

FTB oscillating gradients demonstrated equivalent frequency-dependent diffusion measurements compared with cosine-modulated OGSE while enabling a significant reduction in TE. Optimization and in vivo results suggest that FTB gradients provide increased SNR of kurtosis dispersion maps compared with traditional cosine OGSE. The optimized FTB gradient protocol demonstrated consistent reductions in apparent kurtosis values and increased diffusivity in generated frequency dispersion maps.

CONCLUSIONS

This work presents an alternative to traditional cosine OGSE sequences, enabling more time-efficient acquisitions of frequency-dependent diffusion quantities as demonstrated through in vivo kurtosis frequency dispersion maps.

Metabolic activity diffusion imaging (MADI): II. Noninvasive, high-resolution human brain mapping of sodium pump flux and cell metrics

We introduce a new ¹ H₂ O magnetic resonance approach: metabolic activity diffusion imaging (MADI). Numerical diffusion-weighted imaging decay simulations characterized by the mean cellular water efflux (unidirectional) rate constant (k_io ), mean cell volume (V), and cell number density (ρ) are produced from Monte Carlo random walks in virtual stochastically sized/shaped cell ensembles. Because of active steady-state trans-membrane water cycling (AWC), k_io reflects the cytolemmal Na⁺ , K⁺ ATPase (NKA) homeostatic cellular metabolic rate (^c MR_NKA ). A digital 3D "library" contains thousands of simulated single diffusion-encoded (SDE) decays. Library entries match well with disparate, animal, and human experimental SDE decays. The V and ρ values are consistent with estimates from pertinent in vitro cytometric and ex vivo histopathological literature: in vivo V and ρ values were previously unavailable. The library allows noniterative pixel-by-pixel experimental SDE decay library matchings that can be used to advantage. They yield proof-of-concept MADI parametric mappings of the awake, resting human brain. These reflect the tissue morphology seen in conventional MRI. While V is larger in gray matter (GM) than in white matter (WM), the reverse is true for ρ. Many brain structures have k_io values too large for current, invasive methods. For example, the median WM k_io is 22s^-1 ; likely reflecting mostly exchange within myelin. The k_io •V product map displays brain tissue ^c MR_NKA variation. The GM activity correlates, quantitatively and qualitatively, with the analogous resting-state brain ¹⁸ FDG-PET tissue glucose consumption rate (^t MR_glucose ) map; but noninvasively, with higher spatial resolution, and no pharmacokinetic requirement. The cortex, thalamus, putamen, and caudate exhibit elevated metabolic activity. MADI accuracy and precision are assessed. The results are contextualized with literature overall homeostatic brain glucose consumption and ATP production/consumption measures. The MADI/PET results suggest different GM and WM metabolic pathways. Preliminary human prostate results are also presented.

Mapping of magnetic resonance imaging's transverse relaxation time at low signal-to-noise ratio using Bloch simulations and principal component analysis image denoising

High-resolution mapping of magnetic resonance imaging (MRI)'s transverse relaxation time (T₂ ) can benefit many clinical applications by offering improved anatomic details, enhancing the ability to probe tissues' microarchitecture, and facilitating the identification of early pathology. Increasing spatial resolutions, however, decreases data's signal-to-noise ratio (SNR), particularly at clinical scan times. This impairs imaging quality, and the accuracy of subsequent radiological interpretation. Recently, principal component analysis (PCA) was employed for denoising diffusion-weighted MR images and was shown to be effective for improving parameter estimation in multiexponential relaxometry. This study combines the Marchenko-Pastur PCA (MP-PCA) signal model with the echo modulation curve (EMC) algorithm for denoising multiecho spin-echo (MESE) MRI data and improving the precision of EMC-generated single T₂ relaxation maps. The denoising technique was validated on simulations, phantom scans, and in vivo brain and knee data. MESE scans were performed on a 3-T Siemens scanner. The acquired images were denoised using the MP-PCA algorithm and were then provided as input for the EMC T₂ -fitting algorithm. Quantitative analysis of the denoising quality included comparing the standard deviation and coefficient of variation of T₂ values, along with gold standard SNR estimation of the phantom scans. The presented denoising technique shows an increase in T₂ maps' precision and SNR, while successfully preserving the morphological features of the tissue. Employing MP-PCA denoising as a preprocessing step decreases the noise-related variability of T₂ maps produced by the EMC algorithm and thus increases their precision. The proposed method can be useful for a wide range of clinical applications by facilitating earlier detection of pathologies and improving the accuracy of patients' follow-up.

Variability and reproducibility of multi-echo T2 relaxometry: Insights from multi-site, multi-session and multi-subject MRI acquisitions

Quantitative magnetic resonance imaging (qMRI) can increase the specificity and sensitivity of conventional weighted MRI to underlying pathology by comparing meaningful physical or chemical parameters, measured in physical units, with normative values acquired in a healthy population. This study focuses on multi-echo T₂ relaxometry, a qMRI technique that probes the complex tissue microstructure by differentiating compartment-specific T₂ relaxation times. However, estimation methods are still limited by their sensitivity to the underlying noise. Moreover, estimating the model's parameters is challenging because the resulting inverse problem is ill-posed, requiring advanced numerical regularization techniques. As a result, the estimates from distinct regularization strategies are different. In this work, we aimed to investigate the variability and reproducibility of different techniques for estimating the transverse relaxation time of the intra- and extra-cellular space (T2IE) in gray (GM) and white matter (WM) tissue in a clinical setting, using a multi-site, multi-session, and multi-run T₂ relaxometry dataset. To this end, we evaluated three different techniques for estimating the T₂ spectra (two regularized non-negative least squares methods and a machine learning approach). Two independent analyses were performed to study the effect of using raw and denoised data. For both the GM and WM regions, and the raw and denoised data, our results suggest that the principal source of variance is the inter-subject variability, showing a higher coefficient of variation (CoV) than those estimated for the inter-site, inter-session, and inter-run, respectively. For all reconstruction methods studied, the CoV ranged between 0.32 and 1.64%. Interestingly, the inter-session variability was close to the inter-scanner variability with no statistical differences, suggesting that T2IE is a robust parameter that could be employed in multi-site neuroimaging studies. Furthermore, the three tested methods showed consistent results and similar intra-class correlation (ICC), with values superior to 0.7 for most regions. Results from raw data were slightly more reproducible than those from denoised data. The regularized non-negative least squares method based on the L-curve technique produced the best results, with ICC values ranging from 0.72 to 0.92.

Machine Learning Based on Diffusion Kurtosis Imaging Histogram Parameters for Glioma Grading

Glioma grading plays an important role in surgical resection. We investigated the ability of different feature reduction methods in support vector machine (SVM)-based diffusion kurtosis imaging (DKI) histogram parameters to distinguish glioma grades. A total of 161 glioma patients who underwent magnetic resonance imaging (MRI) from January 2017 to January 2020 were included retrospectively. The patients were divided into low-grade (n = 61) and high-grade (n = 100) groups. Parametric DKI maps were derived, and 45 features from the DKI maps were extracted semi-automatically for analysis. Three feature selection methods [principal component analysis (PCA), recursive feature elimination (RFE) and least absolute shrinkage and selection operator (LASSO)] were used to establish the glioma grading model with an SVM classifier. To evaluate the performance of SVM models, the receiver operating characteristic (ROC) curves of SVM models for distinguishing glioma grades were compared with those of conventional statistical methods. The conventional ROC analysis showed that mean diffusivity (MD) variance, MD skewness and mean kurtosis (MK) C50 could effectively distinguish glioma grades, particularly MD variance. The highest classification distinguishing AUC was found using LASSO at 0.904 ± 0.069. In comparison, classification AUC by PCA was 0.866 ± 0.061, and 0.899 ± 0.079 by RFE. The SVM-PCA model with the lowest AUC among the SVM models was significantly better than the conventional ROC analysis (z = 1.947, p = 0.013). These findings demonstrate the superiority of DKI histogram parameters by LASSO analysis and SVM for distinguishing glioma grades.

Improved Multi-Echo Gradient-Echo-Based Myelin Water Fraction Mapping Using Dimensionality Reduction

Multi-echo gradient-echo (mGRE)-based myelin water fraction (MWF) mapping is a promising myelin water imaging (MWI) modality but is vulnerable to noise and artifact corruption. The linear dimensionality reduction (LDR) method has recently shown improvements with regard to these challenges. However, the magnitude value based low rank operators have been shown to misestimate the MWF for regions with [Formula: see text] anisotropy. This paper presents a nonlinear dimensionality reduction (NLDR) method to estimate the MWF map better by encouraging nonlinear low dimensionality of mGRE signal sources. Specifically, we implemented a fully connected deep autoencoder to extract the low-dimensional features of complex-valued signals and incorporated a sparse regularization to separate the anomaly sources that do not reside in the low-dimensional manifold. Simulations and in vivo experiments were performed to evaluate the accuracy of the MWF map under various situations. The proposed NLDR-based MWF improves the accuracy of the MWF map over the conventional nonlinear least-squares method and the LDR-based MWF and maintains robustness against noise and artifact corruption.

A data-driven T2 relaxation analysis approach for myelin water imaging: Spectrum analysis for multiple exponentials via experimental condition oriented simulation (SAME-ECOS)

PURPOSE

The decomposition of multi-exponential decay data into a T₂ spectrum poses substantial challenges for conventional fitting algorithms, including non-negative least squares (NNLS). Based on a combination of the resolution limit constraint and machine learning neural network algorithm, a data-driven and highly tailorable analysis method named spectrum analysis for multiple exponentials via experimental condition oriented simulation (SAME-ECOS) was proposed.

THEORY AND METHODS

The theory of SAME-ECOS was derived. Then, a paradigm was presented to demonstrate the SAME-ECOS workflow, consisting of a series of calculation, simulation, and model training operations. The performance of the trained SAME-ECOS model was evaluated using simulations and six in vivo brain datasets. The code is available at https://github.com/hanwencat/SAME-ECOS.

RESULTS

Using NNLS as the baseline, SAME-ECOS achieved over 15% higher overall cosine similarity scores in producing the T₂ spectrum, and more than 10% lower mean absolute error in calculating the myelin water fraction (MWF), as well as demonstrated better robustness to noise in the simulation tests. Applying to in vivo data, MWF from SAME-ECOS and NNLS was highly correlated among all study participants. However, a distinct separation of the myelin water peak and the intra/extra-cellular water peak was only observed in the mean T₂ spectra determined using SAME-ECOS. In terms of data processing speed, SAME-ECOS is approximately 30 times faster than NNLS, achieving a whole-brain analysis in 3 min.

CONCLUSION

Compared with NNLS, the SAME-ECOS method yields much more reliable T₂ spectra in a dramatically shorter time, increasing the feasibility of multi-component T₂ decay analysis in clinical settings.

Quantitative platform for accurate and reproducible assessment of transverse (T2 ) relaxation time

MRI's transverse relaxation time (T₂ ) is sensitive to tissues' composition and pathological state. While variations in T₂ values can be used as clinical biomarkers, it is challenging to quantify this parameter in vivo due to the complexity of the MRI signal model, differences in protocol implementations, and hardware imperfections. Herein, we provide a detailed analysis of the echo modulation curve (EMC) platform, offering accurate and reproducible mapping of T₂ values, from 2D multi-slice multi-echo spin-echo (MESE) protocols. Computer simulations of the full Bloch equations are used to generate an advanced signal model, which accounts for stimulated echoes and transmit field (B₁⁺ ) inhomogeneities. In addition to quantifying T₂ values, the EMC platform also provides proton density (PD) maps, and fat-water fraction maps. The algorithm's accuracy, reproducibility, and insensitivity to T₁ values are validated on a phantom constructed by the National Institute of Standards and Technology and on in vivo human brains. EMC-derived T₂ maps show excellent agreement with ground truth values for both in vitro and in vivo models. Quantitative values are accurate and stable across scan settings and for the physiological range of T₂ values, while showing robustness to main field (B₀ ) inhomogeneities, to variations in T₁ relaxation time, and to magnetization transfer. Extension of the algorithm to two-component fitting yields accurate fat and water T₂ maps along with their relative fractions, similar to a reference three-point Dixon technique. Overall, the EMC platform allows to generate accurate and stable T₂ maps, with a full brain coverage using a standard MESE protocol and at feasible scan times. The utility of EMC-based T₂ maps was demonstrated on several clinical applications, showing robustness to variations in other magnetic properties. The algorithm is available online as a full stand-alone package, including an intuitive graphical user interface.

Improved estimation of myelin water fractions with learned parameter distributions

PURPOSE

To improve estimation of myelin water fraction (MWF) in the brain from multi-echo gradient-echo imaging data.

METHODS

A systematic sensitivity analysis was first conducted to characterize the conventional exponential models used for MWF estimation. A new estimation method was then proposed for improved estimation of MWF from practical gradient-echo imaging data. The proposed method uses an extended signal model that includes a finite impulse response filter to compensate for practical signal variations. This new model also enables the use of prelearned parameter distributions as well as low-rank signal structures to improve parameter estimation. The resulting parameter estimation problem was solved optimally in the Bayesian sense.

RESULTS

Our sensitivity analysis results showed that the conventional exponential models were very sensitive to measurement noise and modeling errors. Our simulation and experimental results showed that our proposed method provided a substantial improvement in reliability, reproducibility, and robustness of MWF estimates over the conventional methods. Clinical results obtained from stroke patients indicated that the proposed method, with its improved capability, could reveal the loss of myelin in lesions, demonstrating its translational potentials.

CONCLUSION

This paper addressed the problem of robust MWF estimation from gradient-echo imaging data. A new method was proposed to provide improved MWF estimation in the presence of significant noise and modeling errors. The performance of the proposed method has been evaluated using both simulated and experimental data, showing significantly improved robustness over the existing methods. The proposed method may prove useful for quantitative myelin imaging in clinical applications.

Super Resolution of Magnetic Resonance Images

In this work, novel denoising and super resolution (SR) approaches for magnetic resonance (MR) images are addressed, and are integrated in a unified framework, which do not require example low resolution (LR)/high resolution (HR)/cross-modality/noise-free images and prior information of noise-noise variance. The proposed method categorizes the patches as either smooth or textured and then denoises them by deploying different denoising strategies for efficient denoising. The denoising algorithm is integrated into the SR approach, which uses a gradient profile-based constraint in a sparse representation-based framework to improve the resolution of MR images with reduced smearing of image details. This constraint regularizes the estimation of HR images such that the estimated HR image has gradient profiles similar to the gradient profiles of the original HR image. For this, the gradient profile sharpness (GPS) values of an unknown HR image are estimated using an approximated piece-wise linear relation among GPS values of LR and upsampled LR images. The experiments are performed on three different publicly available datasets. The proposed SR approach outperforms the existing unsupervised SR approach addressed for real MR images that exploits low rank and total variation (LRTV) regularization, by an average peak signal to noise ratio (PSNR) of 0.73 dB and 0.38 dB for upsampling factors 2 and 3, respectively. For the super resolution of noisy real MR images (degraded with 2% noise), the proposed approach outperforms the LRTV approach by an average PSNR of 0.54 dB and 0.46 dB for upsampling factors 2 and 3, respectively. The qualitative analysis is shown for real MR images from healthy subjects and subjects with Alzheimer's disease and structural deformity, i.e., cavernoma.

A model-based framework for correcting B 1 + inhomogeneity effects in magnetization transfer saturation and inhomogeneous magnetization transfer saturation maps

PURPOSE

In this work, we propose that Δ B 1 + -induced errors in magnetization transfer (MT) saturation (MT_sat ) maps can be corrected with use of an R₁ and B 1 + map and through numerical simulations of the sequence.

THEORY AND METHODS

One healthy subject was scanned at 3.0T using a partial quantitative MT protocol to estimate the relationship between observed R₁ (R_1,obs ) and apparent bound pool size ( M 0 , a p p B ) in the brain. MT_sat values were simulated for a range of B 1 + , R_1,obs , and M 0 , a p p B . An equation was fit to the simulated MT_sat , then a linear relationship between R_1,obs and M 0 , a p p B was generated. These results were used to generate correction factor maps for the MT_sat acquired from single-point data. The proposed correction was compared to an empirical correction factor with different MT-preparation schemes.

RESULTS

M 0 , a p p B was highly correlated with R_1,obs (r > 0.96), permitting the use of R_1,obs to estimate M 0 , a p p B for B 1 + correction. All B 1 + corrected MT_sat maps displayed a decreased correlation with B 1 + compared to uncorrected MT_sat and MT_sat corrected with an empirical factor in the corpus callosum. There was good agreement between the proposed approach and the empirical correction with radiofrequency saturation at 2 kHz, with larger deviations seen when using saturation pulses further off-resonance and in inhomogeneous (ih) MT_sat maps.

CONCLUSION

The proposed correction decreases the dependence of MT_sat on B 1 + inhomogeneities. Furthermore, this flexible framework permits the use of different saturation protocols, making it useful for correcting B 1 + inhomogeneities in ihMT.

Data-Driven multi-Contrast spectral microstructure imaging with InSpect: INtegrated SPECTral component estimation and mapping

•

Unsupervised learning technique for spectroscopic analysis of quantitative MRI.

•

Shares information across voxels to improve estimation of multi-dimensional or single-dimensional spectra.

•

Spectral maps are dramatically improved compared to existing approaches.

•

Can potentially identify and map tissue environments; in placental diffusion-relaxometry MRI we demonstrate that it identifies components that correspond to distinct tissue types.

We introduce and demonstrate an unsupervised machine learning technique for spectroscopic analysis of quantitative MRI experiments. Our algorithm supports estimation of one-dimensional spectra from single-contrast data, and multidimensional correlation spectra from simultaneous multi-contrast data. These spectrum-based approaches allow model-free investigation of tissue properties, but require regularised inversion of a Laplace transform or Fredholm integral, which is an ill-posed calculation. Here we present a method that addresses this limitation in a data-driven way. The algorithm simultaneously estimates a canonical basis of spectral components and voxelwise maps of their weightings, thereby pooling information across whole images to regularise the ill-posed problem. We show in simulations that our algorithm substantially outperforms current voxelwise spectral approaches. We demonstrate the method on multi-contrast diffusion-relaxometry placental MRI scans, revealing anatomically-relevant sub-structures, and identifying dysfunctional placentas. Our algorithm vastly reduces the data required to reliably estimate spectra, opening up the possibility of quantitative MRI spectroscopy in a wide range of new applications. Our InSpect code is available at github.com/paddyslator/inspect.

Whole Brain Myelin Water Mapping in One Minute Using Tensor Dictionary Learning With Low-Rank Plus Sparse Regularization

The quantification of myelin water content in the brain can be obtained by the multi-echo [Formula: see text] weighted images ( [Formula: see text]WIs). To accelerate the long acquisition, a novel tensor dictionary learning algorithm with low-rank and sparse regularization (TDLLS) is proposed to reconstruct the [Formula: see text]WIs from the undersampled data. The proposed algorithm explores the local and nonlocal similarity and the global temporal redundancy in the real and imaginary parts of the complex relaxation signals. The joint application of the low-rank constraints on the dictionaries and the sparse constraints on the core coefficient tensors improves the performance of the tensor-based recovery. Parallel imaging is incorporated into the TDLLS algorithm (pTDLLS) for further acceleration. A pulse sequence is proposed to prospectively undersample the Ky-t space to obtain the whole brain high-quality myelin water fraction (MWF) maps within 1 minute at an undersampling rate (R) of 6.

PIRACY: An Optimized Pipeline for Functional Connectivity Analysis in the Rat Brain

Resting state functional MRI (rs-fMRI) is a widespread and powerful tool for investigating functional connectivity (FC) and brain disorders. However, FC analysis can be seriously affected by random and structured noise from non-neural sources, such as physiology. Thus, it is essential to first reduce thermal noise and then correctly identify and remove non-neural artifacts from rs-fMRI signals through optimized data processing methods. However, existing tools that correct for these effects have been developed for human brain and are not readily transposable to rat data. Therefore, the aim of the present study was to establish a data processing pipeline that can robustly remove random and structured noise from rat rs-fMRI data. It includes a novel denoising approach based on the Marchenko-Pastur Principal Component Analysis (MP-PCA) method, FMRIB's ICA-based Xnoiseifier (FIX) for automatic artifact classification and cleaning, and global signal regression (GSR). Our results show that: (I) MP-PCA denoising substantially improves the temporal signal-to-noise ratio, (II) the pre-trained FIX classifier achieves a high accuracy in artifact classification, and (III) both independent component analysis (ICA) cleaning and GSR are essential steps in correcting for possible artifacts and minimizing the within-group variability in control animals while maintaining typical connectivity patterns. Reduced within-group variability also facilitates the exploration of potential between-group FC changes, as illustrated here in a rat model of sporadic Alzheimer's disease.

Comparison of non-parametric T2 relaxometry methods for myelin water quantification

Multi-component T₂ relaxometry allows probing tissue microstructure by assessing compartment-specific T₂ relaxation times and water fractions, including the myelin water fraction. Non-negative least squares (NNLS) with zero-order Tikhonov regularization is the conventional method for estimating smooth T₂ distributions. Despite the improved estimation provided by this method compared to non-regularized NNLS, the solution is still sensitive to the underlying noise and the regularization weight. This is especially relevant for clinically achievable signal-to-noise ratios. In the literature of inverse problems, various well-established approaches to promote smooth solutions, including first-order and second-order Tikhonov regularization, and different criteria for estimating the regularization weight have been proposed, such as L-curve, Generalized Cross-Validation, and Chi-square residual fitting. However, quantitative comparisons between the available reconstruction methods for computing the T₂ distribution, and between different approaches for selecting the optimal regularization weight, are lacking. In this study, we implemented and evaluated ten reconstruction algorithms, resulting from the individual combinations of three penalty terms with three criteria to estimate the regularization weight, plus non-regularized NNLS. Their performance was evaluated both in simulated data and real brain MRI data acquired from healthy volunteers through a scan-rescan repeatability analysis. Our findings demonstrate the need for regularization. As a result of this work, we provide a list of recommendations for selecting the optimal reconstruction algorithms based on the acquired data. Moreover, the implemented methods were packaged in a freely distributed toolbox to promote reproducible research, and to facilitate further research and the use of this promising quantitative technique in clinical practice.

Accuracy and precision of statistical descriptors obtained from multidimensional diffusion signal inversion algorithms

In biological tissues, typical MRI voxels comprise multiple microscopic environments, the local organization of which can be captured by microscopic diffusion tensors. The measured diffusion MRI signal can, therefore, be written as the multidimensional Laplace transform of an intravoxel diffusion tensor distribution (DTD). Tensor-valued diffusion encoding schemes have been designed to probe specific features of the DTD, and several algorithms have been introduced to invert such data and estimate statistical descriptors of the DTD, such as the mean diffusivity, the variance of isotropic diffusivities, and the mean squared diffusion anisotropy. However, the accuracy and precision of these estimations have not been assessed systematically and compared across methods. In this article, we perform and compare such estimations in silico for a one-dimensional Gamma fit, a generalized two-term cumulant approach, and two-dimensional and four-dimensional Monte-Carlo-based inversion techniques, using a clinically feasible tensor-valued acquisition scheme. In particular, we compare their performance at different signal-to-noise ratios (SNRs) for voxel contents varying in terms of the aforementioned statistical descriptors, orientational order, and fractions of isotropic and anisotropic components. We find that all inversion techniques share similar precision (except for a lower precision of the two-dimensional Monte Carlo inversion) but differ in terms of accuracy. While the Gamma fit exhibits infinite-SNR biases when the signal deviates strongly from monoexponentiality and is unaffected by orientational order, the generalized cumulant approach shows infinite-SNR biases when this deviation originates from the variance in isotropic diffusivities or from the low orientational order of anisotropic diffusion components. The two-dimensional Monte Carlo inversion shows remarkable accuracy in all systems studied, given that the acquisition scheme possesses enough directions to yield a rotationally invariant powder average. The four-dimensional Monte Carlo inversion presents no infinite-SNR bias, but suffers significantly from noise in the data, while preserving good contrast in most systems investigated.

Non-negative least squares computation for in vivo myelin mapping using simulated multi-echo spin-echo T2 decay data

Multi-compartment T₂ mapping has gained particular relevance for the study of myelin water in the brain. As a facilitator of rapid saltatory axonal signal transmission, myelin is a cornerstone indicator of white matter development and function. Regularized non-negative least squares fitting of multi-echo T₂ data has been widely employed for the computation of the myelin water fraction (MWF), and the obtained MWF maps have been histopathologically validated. MWF measurements depend upon the quality of the data acquisition, B₁⁺ homogeneity and a range of fitting parameters. In this special issue article, we discuss the relevance of these factors for the accurate computation of multi-compartment T₂ and MWF maps. We generated multi-echo spin-echo T₂ decay curves following the Carr-Purcell-Meiboom-Gill approach for various myelin concentrations and myelin T₂ scenarios by simulating the evolution of the magnetization vector between echoes based on the Bloch equations. We demonstrated that noise and imperfect refocusing flip angles yield systematic underestimations in MWF and intra-/extracellular water geometric mean T₂ (gmT₂ ). MWF estimates were more stable than myelin water gmT₂ time across different settings of the T₂ analysis. We observed that the lower limit of the T₂ distribution grid should be slightly shorter than TE₁ . Both TE₁ and the acquisition echo spacing also have to be sufficiently short to capture the rapidly decaying myelin water T₂ signal. Among all parameters of interest, the estimated MWF and intra-/extracellular water gmT₂ differed by approximately 0.13-4 percentage points and 3-4 ms, respectively, from the true values, with larger deviations observed in the presence of greater B₁⁺ inhomogeneities and at lower signal-to-noise ratio. Tailoring acquisition strategies may allow us to better characterize the T₂ distribution, including the myelin water, in vivo.

Multi-parametric quantitative in vivo spinal cord MRI with unified signal readout and image denoising

Multi-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 in vivo 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 in vivo (i.e. with matched spatial encoding parameters across a range of imaging contrasts). 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 multi-contrast image denoising methods. As a proof-of-concept, here we provide a demonstration with one such method, known as Marchenko-Pastur (MP) Principal Component Analysis (PCA) denoising. MP-PCA is a singular value (SV) decomposition truncation approach that relies on redundant acquisitions, i.e. such that the number of measurements is large compared to the number of components that are maintained in the truncated SV decomposition. Here we used in vivo and synthetic data to test whether a unified readout enables more efficient MP-PCA 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 17% for mean kurtosis, 8% for bound pool fraction (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 multi-contrast denoising methods such as MP-PCA.

A field-monitoring-based approach for correcting eddy-current-induced artifacts of up to the 2nd spatial order in human-connectome-project-style multiband diffusion MRI experiment at 7T: A pilot study

Over the recent years, significant advances in Spin-Echo (SE) Echo-Planar (EP) Diffusion MRI (dMRI) have enabled improved fiber tracking conspicuity in the human brain. At the same time, pushing the spatial resolution and using higher b-values inherently expose the acquired images to further eddy-current-induced distortion and blurring. Recently developed data-driven correction techniques, capable of significantly mitigating these defects, are included in the reconstruction pipelines developed for the Human Connectome Project (HCP) driven by the NIH BRAIN initiative. In this case, however, corrections are derived from the original diffusion-weighted (DW) magnitude images affected by distortion and blurring. Considering the complexity of k-space deviations in the presence of time varying high spatial order eddy currents, distortion and blurring may not be fully reversed when relying on magnitude DW images only. An alternative approach, consisting of iteratively reconstructing DW images based on the actual magnetic field spatiotemporal evolution measured with a magnetic field monitoring camera, has been successfully implemented at 3T in single band dMRI (Wilm et al., 2017, 2015). In this study, we aim to demonstrate the efficacy of this eddy current correction method in the challenging context of HCP-style multiband (MB ​= ​2) dMRI protocol. The magnetic field evolution was measured during the EP-dMRI readout echo train with a field monitoring camera equipped with 16 ¹⁹F NMR probes. The time variation of 0^th, 1^st and 2^nd order spherical field harmonics were used to reconstruct DW images. Individual DW images reconstructed with and without field correction were compared. The impact of eddy current correction was evaluated by comparing the corresponding direction-averaged DW images and fractional anisotropy (FA) maps. ¹⁹F field monitoring data confirmed the existence of significant field deviations induced by the diffusion-encoding gradients, with variations depending on diffusion gradient amplitude and direction. In DW images reconstructed with the field correction, residual aliasing artifacts were reduced or eliminated, and when high b-values were applied, better gray/white matter delineation and sharper gyri contours were observed, indicating reduced signal blurring. The improvement in image quality further contributed to sharper contours and better gray/white matter delineation in mean DW images and FA maps. In conclusion, we demonstrate that up-to-2^nd-order-eddy-current-induced field perturbation in multiband, in-plane accelerated HCP-style dMRI acquisition at 7T can be corrected by integrating the measured field evolution in image reconstruction.