Magnetic resonance fingerprinting

B1-MRF: Large dynamic range MRF-based absolute B 1 + mapping in the human body at 7T

PURPOSE

This study aims to map the transmit magnetic field (B 1 + $$ {B}_1^{+} $$ ) in the human body at 7T using MR fingerprinting (MRF), with a focus on achieving high accuracy and precision across a large dynamic range, particularly at low flip angles (FAs).

METHODS

A FLASH-based MRF sequence (B1-MRF) with highB 1 + $$ {B}_1^{+} $$ sensitivity was developed. Phantom and in vivo abdominal imaging were performed at 7T, and the results were compared with established reference methods, including a slow but precise preparation-based method (PEX), saturated TurboFLASH (satTFL), actual flip angle imaging (AFI) and Bloch-Siegert shift (BSS).

RESULTS

The MRF signal curve was highly sensitive toB 1 + $$ {B}_1^{+} $$ , while T1 sensitivity was comparatively low. The phantom experiment showed good agreement ofB 1 + $$ {B}_1^{+} $$ to PEX for a T1 range of 204-1691 ms evaluated at FAs from 0° to 70°. Compared to the references, a dynamic range increase larger than a factor of two was determined experimentally. In vivo liver scans showed a strong correlation between B1-MRF, satTFL, and RPE-AFI in a low body mass index (BMI) subject (18.1 kg/m2). However, in larger BMI subjects (≥25.5 kg/m2), inconsistencies were observed in lowB 1 + $$ {B}_1^{+} $$ regions for satTFL and RPE-AFI, while B1-MRF still provided consistent results in these regions.

CONCLUSION

B1-MRF provides accurate and preciseB 1 + $$ {B}_1^{+} $$ maps over a wide range of FAs, surpassing the capabilities of existing methods in the FA range < 60°. Its enhanced sensitivity at low FAs is advantageous for various applications requiring preciseB 1 + $$ {B}_1^{+} $$ estimates, potentially advancing the frontiers of ultra-high field (UHF) body imaging at 7T and beyond.

DeepEMC-T2 mapping: Deep learning-enabled T2 mapping based on echo modulation curve modeling

PURPOSE

Echo modulation curve (EMC) modeling enables accurate quantification of T2 relaxation times in multi-echo spin-echo (MESE) imaging. The standard EMC-T2 mapping framework, however, requires sufficient echoes and cumbersome pixel-wise dictionary-matching steps. This work proposes a deep learning version of EMC-T2 mapping, called DeepEMC-T2 mapping, to efficiently estimate accurate T2 maps from fewer echoes.

METHODS

DeepEMC-T2 mapping was developed using a modified U-Net to estimate both T2 and proton density (PD) maps directly from MESE images. The network implements several new features to improve the accuracy of T2/PD estimation. A total of 67 MESE datasets acquired in axial orientation were used for network training and evaluation. An additional 57 datasets acquired in coronal orientation with different scan parameters were used to evaluate the generalizability of the framework. The performance of DeepEMC-T2 mapping was evaluated in seven experiments.

RESULTS

Compared to the reference, DeepEMC-T2 mapping achieved T2 estimation errors from 1% to 11% and PD estimation errors from 0.4% to 1.5% with ten/seven/five/three echoes, which are more accurate than standard EMC-T2 mapping. By incorporating datasets acquired with different scan parameters and orientations for joint training, DeepEMC-T2 exhibits robust generalizability across varying imaging protocols. Increasing the echo spacing and including longer echoes improve the accuracy of parameter estimation. The new features proposed in DeepEMC-T2 mapping all enabled more accurate T2 estimation.

CONCLUSIONS

DeepEMC-T2 mapping enables simplified, efficient, and accurate T2 quantification directly from MESE images without dictionary matching. Accurate T2 estimation from fewer echoes allows for increased volumetric coverage and/or higher slice resolution without prolonging total scan times.

Unraveling contributions to the Z-spectrum signal at 3.5 ppm of human brain tumors

PURPOSE

To evaluate the influence of the confounding factors, direct water saturation (DWS), and magnetization transfer contrast (MTC) effects on measured Z-spectra and amide proton transfer (APT) contrast in brain tumors.

METHODS

High-grade glioma patients were scanned using an RF saturation-encoded 3D MR fingerprinting (MRF) sequence at 3 T. For MRF reconstruction, a recurrent neural network was designed to learn free water and semisolid macromolecule parameter mappings of the underlying multiple tissue properties from saturation-transfer MRF signals. The DWS spectra and MTC spectra were synthesized by solving Bloch-McConnell equations and evaluated in brain tumors.

RESULTS

The dominant contribution to the saturation effect at 3.5 ppm was from DWS and MTC effects, but 25%-33% of the saturated signal in the gadolinium-enhancing tumor (13%-20% for normal tissue) was due to the APT effect. The APT# signal of the gadolinium-enhancing tumor was significantly higher than that of the normal-appearing white matter (10.1% vs. 8.3% at 1 μT and 11.2% vs. 7.8% at 1.5 μT).

CONCLUSION

The RF saturation-encoded MRF allowed us to separate contributions to the saturation signal at 3.5 ppm in the Z-spectrum. Although free water and semisolid MTC are the main contributors, significant APT contrast between tumor and normal tissues was observed.

Physics-guided self-supervised learning for retrospective T1 and T2 mapping from conventional weighted brain MRI: Technical developments and initial validation in glioblastoma

PURPOSE

To develop a self-supervised learning method to retrospectively estimate T1 and T2 values from clinical weighted MRI.

METHODS

A self-supervised learning approach was constructed to estimate T1, T2, and proton density maps from conventional T1- and T2-weighted images. MR physics models were employed to regenerate the weighted images from the network outputs, and the network was optimized based on loss calculated between the synthesized and input weighted images, alongside additional constraints based on prior information. The method was evaluated on healthy volunteer data, with conventional mapping as references. The reproducibility was examined on two 3.0T scanners. Performance in tumor characterization was inspected by applying the method to a public glioblastoma dataset.

RESULTS

For T1 and T2 estimation from three weighted images (T1 MPRAGE, T1 gradient echo sequences, and T2 turbo spin echo), the deep learning method achieved global voxel-wise error ≤9% in brain parenchyma and regional error ≤12.2% in six types of brain tissues. The regional measurements obtained from two scanners showed mean differences ≤2.4% and correlation coefficients >0.98, demonstrating excellent reproducibility. In the 50 glioblastoma patients, the retrospective quantification results were in line with literature reports from prospective methods, and the T2 values were found to be higher in tumor regions, with sensitivity of 0.90 and specificity of 0.92 in a voxel-wise classification task between normal and abnormal regions.

CONCLUSION

The self-supervised learning method is promising for retrospective T1 and T2 quantification from clinical MR images, with the potential to improve the availability of quantitative MRI and facilitate brain tumor characterization.

Improved gradient echo magnitude- and phase-based mapping of T 2 $$ {\mathrm{T}}_2 $$ using multiple RF spoiling increments at 3T and 7T

PURPOSE

The transverse relaxation time T  2 $$ {}_2 $$ holds significant relevance in clinical applications and research studies. Conventional T  2 $$ {}_2 $$ mapping approaches rely on spin-echo sequences, which require lengthy acquisition times and involve high radiofrequency (RF) power deposition. An alternative gradient echo (GRE) phase-based T  2 $$ {}_2 $$ mapping method, utilizing steady-state acquisitions at one small RF spoil phase increment, was recently demonstrated. Here, a modified magnitude- and phase-based T  2 $$ {}_2 $$ mapping approach is proposed, which improvesT 2 $$ {\mathrm{T}}_2 $$ estimations by simultaneous fitting ofT 1 $$ {\mathrm{T}}_1 $$ and signal amplitude (A ∝ P D $$ A\propto PD $$ ) at three or more RF spoiling phase increments, instead of assuming a fixedT 1 $$ {\mathrm{T}}_1 $$ value.

METHODS

The feasibility of the magnitude-phase-based method was assessed by simulations, in phantom and in vivo experiments using skipped-CAIPI three-dimensional-echo-planar imaging (3D-EPI) for rapid GRE imaging.T 2 $$ {\mathrm{T}}_2 $$ ,T 1 $$ {\mathrm{T}}_1 $$ and PD estimations obtained by our method were compared toT 2 $$ {\mathrm{T}}_2 $$ of the phase-based method andT 1 $$ {\mathrm{T}}_1 $$ and PD of spoiled GRE-based multi-parameter mapping using a multi-echo version of the same sequence.

RESULTS

The agreement of the proposedT 2 $$ {\mathrm{T}}_2 $$ with ground truth and referenceT 2 $$ {\mathrm{T}}_2 $$ values was higher than that of phase-basedT 2 $$ {\mathrm{T}}_2 $$ in simulations and in phantom data. While phase-basedT 2 $$ {\mathrm{T}}_2 $$ overestimation increases with actualT 2 $$ {\mathrm{T}}_2 $$ andT 1 $$ {\mathrm{T}}_1 $$ , the proposed method is accurate over a large range of physiologically meaningfulT 2 $$ {\mathrm{T}}_2 $$ andT 1 $$ {\mathrm{T}}_1 $$ values. At the same time, precision is improved. In vivo results were in line with these observations.

CONCLUSION

Accurate magnitude-phase-based T  2 $$ {}_2 $$ mapping is feasible in less than 5 min scan time for 1 mm nominal isotropic whole-head coverage at 3T and 7T.

Simultaneous 3D T 1 $$ {\mathrm{T}}_1 $$ , T 2 $$ {\mathrm{T}}_2 $$ , and fat-signal-fraction mapping with respiratory-motion correction for comprehensive liver tissue characterization at 0.55 T

PURPOSE

To develop a framework for simultaneous three-dimensional (3D) mapping ofT 1 $$ {\mathrm{T}}_1 $$ ,T 2 $$ {\mathrm{T}}_2 $$ , and fat signal fraction in the liver at 0.55 T.

METHODS

The proposed sequence acquires four interleaved 3D volumes with a two-echo Dixon readout.T 1 $$ {\mathrm{T}}_1 $$ andT 2 $$ {\mathrm{T}}_2 $$ are encoded into each volume via preparation modules, and dictionary matching allows simultaneous estimation ofT 1 $$ {\mathrm{T}}_1 $$ ,T 2 $$ {\mathrm{T}}_2 $$ , andM 0 $$ {M}_0 $$ for water and fat separately. 2D image navigators permit respiratory binning, and motion fields from nonrigid registration between bins are used in a nonrigid respiratory-motion-corrected reconstruction, enabling 100% scan efficiency from a free-breathing acquisition. The integrated nature of the framework ensures the resulting maps are always co-registered.

RESULTS

T 1 $$ {\mathrm{T}}_1 $$ ,T 2 $$ {\mathrm{T}}_2 $$ , and fat-signal-fraction measurements in phantoms correlated strongly (adjustedr 2 > 0 . 98 $$ {r}^2>0.98 $$ ) with reference measurements. Mean liver tissue parameter values in 10 healthy volunteers were427 ± 22 $$ 427\pm 22 $$ ,47 . 7 ± 3 . 3   ms $$ 47.7\pm 3.3\;\mathrm{ms} $$ , and7 ± 2 % $$ 7\pm 2\% $$ forT 1 $$ {\mathrm{T}}_1 $$ ,T 2 $$ {\mathrm{T}}_2 $$ , and fat signal fraction, giving biases of71 $$ 71 $$ ,- 30 . 0   ms $$ -30.0\;\mathrm{ms} $$ , and- 5 $$ -5 $$ percentage points, respectively, when compared to conventional methods.

CONCLUSION

A novel sequence for comprehensive characterization of liver tissue at 0.55 T was developed. The sequence provides co-registered 3DT 1 $$ {\mathrm{T}}_1 $$ ,T 2 $$ {\mathrm{T}}_2 $$ , and fat-signal-fraction maps with full coverage of the liver, from a single nine-and-a-half-minute free-breathing scan. Further development is needed to achieve accurate proton-density fat fraction (PDFF) estimation in vivo.

Rotary excitation of non-sinusoidal pulsed magnetic fields: Towards non-invasive direct detection of cardiac conduction

PURPOSE

There is a need for high resolution non-invasive imaging methods of physiologic magnetic fields. The purpose of this work is to develop a MRI detection approach for non-sinusoidal magnetic fields based on the rotary excitation (REX) mechanism which was previously successfully applied for the detection of oscillating magnetic fields in the sub-nT range.

METHODS

The new detection concept was examined by means of Bloch simulations, evaluating the interaction effect of spin-locked magnetization and low-frequency pulsed magnetic fields. The REX detection approach was validated under controlled conditions in phantom experiments at 3 T. Gaussian and sinc-shaped stimuli were investigated. In addition, the detection of artificial fields resembling a cardiac QRS complex, which is the most prominent peak visible on a magnetocardiogram, was tested.

RESULTS

Bloch simulations demonstrated that the REX method has a high sensitivity to pulsed fields in the resonance case, which is met when the spin-lock frequency coincides with a non-zero Fourier component of the stimulus field. In the experiments, we found that magnetic stimuli of different durations and waveforms can be distinguished by their characteristic REX response spectrum. The detected REX amplitude was proportional to the stimulus peak amplitude (R2 > 0.98) and the lowest field detection was 1 nT. Furthermore, the detection of QRS-like fields with varying QRS durations yielded significant results in a phantom setup (p < 0.001).

CONCLUSION

REX detection can be transferred to non-sinusoidal pulsed magnetic fields and could provide a non-invasive, quantitative tool for spatially resolved assessment of cardiac biomagnetism. Potential applications include the direct detection and characterization of cardiac conduction.

Accelerated 2D radial Look-Locker T1 mapping using a deep learning-based rapid inversion recovery sampling technique

Efficient abdominal coverage with T1-mapping methods currently available in the clinic is limited by the breath hold period (BHP) and the time needed for T1 recovery. This work develops a T1-mapping framework for efficient abdominal coverage based on rapid T1 recovery curve (T1RC) sampling, slice-selective inversion, optimized slice interleaving, and a convolutional neural network (CNN)-based T1 estimation. The effect of reducing the T1RC sampling was evaluated by comparing T1 estimates from T1RC ranging from 0.63 to 2.0 s with reference T1 values obtained from T1RC = 2.5-5 s. Slice interleaving methodologies were evaluated by comparing the T1 variation in abdominal organs across slices. The repeatability of the proposed framework was demonstrated by performing acquisition on test subjects across imaging sessions. Analysis of in vivo data based on retrospectively shortening the T1RC showed that with the CNN framework, a T1RC = 0.84 s yielded T1 estimates without significant changes in mean T1 (p > 0.05) or significant increase in T1 variability (p > 0.48) compared to the reference. Prospectively acquired data using T1RC = 0.84 s, an optimized slice interleaving scheme, and the CNN framework enabled 21 slices in a 20 s BHP. Analyses across abdominal organs produced T1 values within 2% of the reference. Repeatability experiments yielded Pearson's correlation, repeatability coefficient, and coefficient of variation of 0.99, 2.5%, and 0.12%, respectively. The proposed T1 mapping framework provides full abdominal coverage within a single BHP.

Engineering Principles and Bioengineering in Global Health

Medical technology plays a significant role in the reduction of disability and mortality due to the global burden of disease. The lack of diagnostic technology has been identified as the largest gap in the global health care pathway, and the cost of this technology is a driving factor for its lack of proliferation. Technology developed in high-income countries is often focused on producing high-quality, patient-specific data at a cost high-income markets can pay. While machine learning plays an important role in this process, great care must be taken to ensure appropriate translation to clinical practice.

Advancing the Collaboration Between Imaging and Radiation Oncology

The fusion of cutting-edge imaging technologies with radiation therapy (RT) has catalyzed transformative breakthroughs in cancer treatment in recent decades. It is critical for us to review our achievements and preview into the next phase for future synergy between imaging and RT. This paper serves as a review and preview for fostering collaboration between these two domains in the forthcoming decade. Firstly, it delineates ten prospective directions ranging from technological innovations to leveraging imaging data in RT planning, execution, and preclinical research. Secondly, it presents major directions for infrastructure and team development in facilitating interdisciplinary synergy and clinical translation. We envision a future where seamless integration of imaging technologies into RT will not only meet the demands of RT but also unlock novel functionalities, enhancing accuracy, efficiency, safety, and ultimately, the standard of care for patients worldwide.

Emerging and anticipated innovations in prostate cancer MRI and their impact on patient care

Prostate cancer (PCa) remains the leading malignancy affecting men, with over 3 million men living with the disease in the US, and an estimated 288,000 new cases and almost 35,000 deaths in 2023 in the United States alone. Over the last few decades, imaging has been a cornerstone in PCa care, with a crucial role in the detection, staging, and assessment of PCa recurrence or by guiding diagnostic or therapeutic interventions. To improve diagnostic accuracy and outcomes in PCa care, remarkable advancements have been made to different imaging modalities in recent years. This paper focuses on reviewing the main innovations in the field of PCa magnetic resonance imaging, including MRI protocols, MRI-guided procedural interventions, artificial intelligence algorithms and positron emission tomography, which may impact PCa care in the future.

Efficient pulse sequence design framework for high-dimensional MR fingerprinting scans using systematic error index

PURPOSE

For effective optimization of MR fingerprinting (MRF) pulse sequences, estimating and minimizing errors from actual scan conditions are crucial. Although virtual-scan simulations offer an approximation to these errors, their computational demands become expensive for high-dimensional MRF frameworks, where interactions between more than two tissue properties are considered. This complexity makes sequence optimization impractical. We introduce a new mathematical model, the systematic error index (SEI), to address the scalability challenges for high-dimensional MRF sequence design.

METHODS

By eliminating the need to perform dictionary matching, the SEI model approximates quantification errors with low computational costs. The SEI model was validated in comparison with virtual-scan simulations. The SEI model was further applied to optimize three high-dimensional MRF sequences that quantify two to four tissue properties. The optimized scans were examined in simulations and healthy subjects.

RESULTS

The proposed SEI model closely approximated the virtual-scan simulation outcomes while achieving hundred- to thousand-times acceleration in the computational speed. In both simulation and in vivo experiments, the optimized MRF sequences yield higher measurement accuracy with fewer undersampling artifacts at shorter scan times than the heuristically designed sequences.

CONCLUSION

We developed an efficient method for estimating real-world errors in MRF scans with high computational efficiency. Our results illustrate that the SEI model could approximate errors both qualitatively and quantitatively. We also proved the practicality of the SEI model of optimizing sequences for high-dimensional MRF frameworks with manageable computational power. The optimized high-dimensional MRF scans exhibited enhanced robustness against undersampling and system imperfections with faster scan times.

Evaluating T1, T2 Relaxation, and Proton Density in Normal Brain Using Synthetic MRI with Fast Imaging Protocol

Synthetic MRI is being increasingly used for the quantification of brain longitudinal relaxation time (T1), transverse relaxation time (T2), and proton density (PD) values. However, the effect of fast imaging protocols on these quantitative values has not been fully estimated. The purpose of this study was to investigate the effect of fast scan parameters on T1, T2, and PD measured with a multi-dynamic multi-echo (MDME) sequence of normal brain at 3.0T. Thirty-four volunteers were scanned using 3 MDME sequences with different scan times (named Fast, 2 min, 29 sec; Routine, 4 min, 07 sec; and Research, 7 min, 46 sec, respectively). The measured T1, T2, and PD in 18 volumes of interest (VOI) of brain were compared between the 3 sequences using rank sum test, t test, coefficients of variation (CVs) analysis, correlation analysis, and Bland-Altman analysis. We found that even though T1, T2, and PD were significantly different between the 3 sequences in most of the brain regions, the intersequence CVs were relatively low and linear correlation were high. Bland-Altman plots showed that most of the values fall within the 95% prediction limits. We concluded that fast imaging protocols of MDME sequence used in our study can potentially be used for quantitative evaluation of brain tissues. Since changing scan parameters can affect the measured T1, T2, and PD values, it is necessary to use consistent scan parameter for comparing or following up cases quantitatively.

Quantification of water exchange across the blood-brain barrier using noncontrast MR fingerprinting

PURPOSE

A method is proposed to quantify cerebral blood volume (v b $$ {v}_b $$ ) and intravascular water residence time (τ b $$ {\tau}_b $$ ) using MR fingerprinting (MRF), applied using a spoiled gradient echo sequence without the need for contrast agent.

METHODS

An in silico study optimized an acquisition protocol to maximize the sensitivity of the measurement tov b $$ {v}_b $$ andτ b $$ {\tau}_b $$ changes. Its accuracy in the presence of variations inT 1 , t $$ {\mathrm{T}}_{1,t} $$ ,T 1 , b $$ {\mathrm{T}}_{1,b} $$ , andB 1 $$ {\mathrm{B}}_1 $$ was evaluated. The optimized protocol (scan time of 19 min) was then tested in a exploratory healthy volunteer study (10 volunteers, mean age 24± $$ \pm $$ 3, six males) at 3 T with a repeat scan taken after repositioning to allow estimation of repeatability.

RESULTS

Simulations show that assuming literature values forT 1 , b $$ {\mathrm{T}}_{1,b} $$ andT 1 , t $$ {\mathrm{T}}_{1,t} $$ , no variation inB 1 $$ {\mathrm{B}}_1 $$ , while fitting onlyv b $$ {v}_b $$ andτ b $$ {\tau}_b $$ , leads to large errors in quantification ofv b $$ {v}_b $$ andτ b $$ {\tau}_b $$ , regardless of noise levels. However, simulations also show that matchingT 1 , t $$ {\mathrm{T}}_{1,t} $$ ,T 1 , b $$ {\mathrm{T}}_{1,b} $$ ,B 1 + $$ {\mathrm{B}}_1^{+} $$ ,v b $$ {v}_b $$ andτ b $$ {\tau}_b $$ , simultaneously is feasible at clinically achievable noise levels. Across the healthy volunteers, all parameter quantifications fell within the expected literature range. In addition, the maps show good agreement between hemispheres suggesting physiologically relevant information is being extracted. Expected differences between white and gray matterT 1 , t $$ {\mathrm{T}}_{1,t} $$ (p < 0.0001) andv b $$ {v}_b $$ (p < 0.0001) are observed,T 1 , b $$ {\mathrm{T}}_{1,b} $$ andτ b $$ {\tau}_b $$ show no significant differences, p = 0.4 and p = 0.6, respectively. Moderate to excellent repeatability was seen between repeat scans: mean intra-class correlation coefficient ofT 1 , t : 0 . 91 $$ {\mathrm{T}}_{1,t}:0.91 $$ ,T 1 , b : 0 . 58 $$ {\mathrm{T}}_{1,b}:0.58 $$ ,v b : 0 . 90 $$ {v}_b:0.90 $$ , andτ b : 0 . 96 $$ {\tau}_b:0.96 $$ .

CONCLUSION

We demonstrate that regional simultaneous quantification ofv b $$ {v}_b $$ ,τ b $$ {\tau}_b $$ ,T 1 , b , T 1 , t $$ {\mathrm{T}}_{1,b},{T}_{1,t} $$ , andB 1 + $$ {\mathrm{B}}_1^{+} $$ using MRF is feasible in vivo.

Three-dimensional simultaneous T1 and T2* relaxation times and quantitative susceptibility mapping at 3 T: A multicenter validation study

We aimed to determine the intra-site repeatability and cross-site reproducibility of T1 and T2* relaxation times and quantitative susceptibility (χ) values obtained through quantitative parameter mapping (QPM) at 3 T. This prospective study included three 3-T scanners with the same hardware and software platform at three sites. The brains of twelve healthy volunteers were scanned three times using QPM at three sites. Intra-site repeatability and cross-site reproducibility were evaluated based on voxel-wise and region-of-interest analyses. The within-subject coefficient of variation (wCV), within-subject standard deviation (wSD), linear regression, Bland-Altman plot, and intraclass correlation coefficient (ICC) were used for evaluation. The intra-site repeatability wCV was 11.9 ± 6.86% for T1 and 3.15 ± 0.03% for T2*, and wSD of χ at 3.35 ± 0.10 parts per billion (ppb). Intra-site ICC(1,k) values for T1, T2*, and χ were 0.878-0.904, 0.972-0.976, and 0.966-0.972, respectively, indicating high consistency within the same scanner. Linear regression analysis revealed a strong agreement between measurements from each site and the site-average measurement, with R-squared values ranging from 0.79 to 0.83 for T1, 0.94-0.95 for T2*, and 0.95-0.96 for χ. The cross-site wCV was 13.4 ± 5.47% for T1 and 3.69 ± 2.25% for T2*, and cross-site wSD of χ at 4.08 ± 3.22 ppb. The cross-site ICC(2,1) was 0.707, 0.913, and 0.902 for T1, T2*, and χ, respectively. QPM provides T1, T2*, and χ values with an intra-site repeatability of <12% and cross-site reproducibility of <14%. These findings may contribute to the development of multisite studies.

Bias-reduced neural networks for parameter estimation in quantitative MRI

PURPOSE

To develop neural network (NN)-based quantitative MRI parameter estimators with minimal bias and a variance close to the Cramér-Rao bound.

THEORY AND METHODS

We generalize the mean squared error loss to control the bias and variance of the NN's estimates, which involves averaging over multiple noise realizations of the same measurements during training. Bias and variance properties of the resulting NNs are studied for two neuroimaging applications.

RESULTS

In simulations, the proposed strategy reduces the estimates' bias throughout parameter space and achieves a variance close to the Cramér-Rao bound. In vivo, we observe good concordance between parameter maps estimated with the proposed NNs and traditional estimators, such as nonlinear least-squares fitting, while state-of-the-art NNs show larger deviations.

CONCLUSION

The proposed NNs have greatly reduced bias compared to those trained using the mean squared error and offer significantly improved computational efficiency over traditional estimators with comparable or better accuracy.

T1 mapping from routine 3D T1-weighted inversion recovery sequences in clinical practice: comparison against reference inversion recovery fast field echo T1 scans and feasibility in multiple sclerosis

BACKGROUND AND PURPOSE

Quantitative T1 mapping can be an essential tool for assessing tissue injury in multiple sclerosis (MS). We introduce T1-REQUIRE, a method that converts a single high-resolution anatomical 3D T1-weighted Turbo Field Echo (3DT1TFE) scan into a parametric T1 map that could be used for quantitative assessment of tissue damage. We present the accuracy and feasibility of this method in MS.

METHODS

14 subjects with relapsing-remitting MS and 10 healthy subjects were examined. T1 maps were generated from 3DT1TFE images using T1-REQUIRE, which estimates T1 values using MR signal equations and internal tissue reference T1 values. Estimated T1 of lesions, white, and gray matter regions were compared with reference Inversion-Recovery Fast Field Echo T1 values and analyzed via correlation and Bland-Altman (BA) statistics.

RESULTS

159 T1-weighted (T1W) hypointense MS lesions and 288 gray matter regions were examined. T1 values for MS lesions showed a Pearson's correlation of r = 0.81 (p < 0.000), R2 = 0.65, and Bias = 4.18%. BA statistics showed a mean difference of -53.95 ms and limits of agreement (LOA) of -344.20 and 236.30 ms. Non-lesional normal-appearing white matter had a correlation coefficient of r = 0.82 (p < 0.000), R2 = 0.67, Bias = 8.78%, mean difference of 73.87 ms, and LOA of -55.67 and 203.41 ms.

CONCLUSIONS

We demonstrate the feasibility of retroactively derived high-resolution T1 maps from routinely acquired anatomical images, which could be used to quantify tissue pathology in MS. The results of this study will set the stage for testing this method in larger clinical studies for examining MS disease activity and progression.

Advanced MRI techniques in abdominal imaging

Magnetic resonance imaging (MRI) is a crucial modality for abdominal imaging evaluation of focal lesions and tissue properties. However, several obstacles, such as prolonged scan times, limitations in patients' breath-hold capacity, and contrast agent-associated artifacts, remain in abdominal MR images. Recent techniques, including parallel imaging, three-dimensional acquisition, compressed sensing, and deep learning, have been developed to reduce the scan time while ensuring acceptable image quality or to achieve higher resolution without extending the scan duration. Quantitative measurements using MRI techniques enable the noninvasive evaluation of specific materials. A comprehensive understanding of these advanced techniques is essential for accurate interpretation of MRI sequences. Herein, we therefore review advanced abdominal MRI techniques.

Quantifying brain development in the HEALthy Brain and Child Development (HBCD) Study: The magnetic resonance imaging and spectroscopy protocol

The HEALthy Brain and Child Development (HBCD) Study, a multi-site prospective longitudinal cohort study, will examine human brain, cognitive, behavioral, social, and emotional development beginning prenatally and planned through early childhood. The acquisition of multimodal magnetic resonance-based brain development data is central to the study's core protocol. However, application of Magnetic Resonance Imaging (MRI) methods in this population is complicated by technical challenges and difficulties of imaging in early life. Overcoming these challenges requires an innovative and harmonized approach, combining age-appropriate acquisition protocols together with specialized pediatric neuroimaging strategies. The HBCD MRI Working Group aimed to establish a core acquisition protocol for all 27 HBCD Study recruitment sites to measure brain structure, function, microstructure, and metabolites. Acquisition parameters of individual modalities have been matched across MRI scanner platforms for harmonized acquisitions and state-of-the-art technologies are employed to enable faster and motion-robust imaging. Here, we provide an overview of the HBCD MRI protocol, including decisions of individual modalities and preliminary data. The result will be an unparalleled resource for examining early neurodevelopment which enables the larger scientific community to assess normative trajectories from birth through childhood and to examine the genetic, biological, and environmental factors that help shape the developing brain.

POSE: POSition Encoding for accelerated quantitative MRI

Quantitative MRI utilizes multiple acquisitions with varying sequence parameters to sufficiently characterize a biophysical model of interest, resulting in undesirable scan times. Here we propose, validate and demonstrate a new general strategy for accelerating MRI using subvoxel shifting as a source of encoding called POSition Encoding (POSE). The POSE framework applies unique subvoxel shifts along the acquisition parameter dimension, thereby creating an extra source of encoding. Combining with a biophysical signal model of interest, accelerated and enhanced resolution maps of biophysical parameters are obtained. This has been validated and demonstrated through numerical Bloch equation simulations, phantom experiments and in vivo experiments using the variable flip angle signal model in 3D acquisitions as an application example. Monte Carlo simulations were performed using in vivo data to investigate our method's noise performance. POSE quantification results from numerical Bloch equation simulations of both a numerical phantom and realistic digital brain phantom concur well with the reference method, validating our method both theoretically and for realistic situations. NIST phantom experiment results show excellent overall agreement with the reference method, confirming our method's applicability for a wide range of T1 values. In vivo results not only exhibit good agreement with the reference method, but also show g-factors that significantly outperforms conventional parallel imaging methods with identical acceleration. Furthermore, our results show that POSE can be combined with parallel imaging to further accelerate while maintaining superior noise performance over parallel imaging that uses lower acceleration factors.

Magnetization transfer explains most of the T 1 variability in the MRI literature

Purpose

To identify the predominant source of theT 1 variability described in the literature, which ranges from 0.6-1.1s for brain white matter at 3T.

Methods

25T 1 -mapping methods from the literature were simulated with a mono-exponential and magnetization-transfer (MT) models, each followed by mono-exponential fitting. A single set of model parameters was assumed for the simulation of all methods, and these parameters were estimated by fitting the simulation-based to the corresponding literatureT 1 values of white matter at 3T.

Results

Mono-exponential simulations suggest good inter-method reproducibility and fail to explain the highly variableT 1 estimates in the literature. In contrast, MT simulations suggest that a mono-exponential fit results in a variableT 1 and explain up to 62% of the literature's variability.

Conclusion

The results suggest that a mono-exponential model does not adequately describe longitudinal relaxation in biological tissue. Therefore,T 1 in biological tissue should be considered only a semi-quantitative metric that is inherently contingent upon the imaging methodology; and comparisons between differentT 1 -mapping methods and the use of simplistic spin systems-such as doped-water phantoms-for validation should be viewed with caution.

Synthetic MRI and MR Fingerprinting-Derived Relaxometry of Antenatal Human Brainstem Myelination: A Postmortem-Based Quantitative Imaging Study

BACKGROUND AND PURPOSE

The radiologic evaluation of ongoing myelination is currently limited prenatally. Novel quantitative MR imaging modalities provide relaxometric properties that are linked to myelinogenesis. In this retrospective postmortem imaging study, the capability of Synthetic MR imaging and MR fingerprinting-derived relaxometry for tracking fetal myelin development was investigated. Moreover, the consistency of results for both MR approaches was analyzed.

MATERIALS AND METHODS

In 26 cases, quantitative postmortem fetal brain MR data were available (gestational age range, 15 + 1 to 32 + 1; female/male ratio, 14/12). Relaxometric measurements (T1-/T2-relexation times) were determined in the medulla oblongata and the midbrain using Synthetic MR imaging/MR fingerprinting-specific postprocessing procedures (Synthetic MR imaging and MR Robust Quantitative Tool for MR fingerprinting). The Pearson correlations were applied to detect relationships between T1-relaxation times/T2-relaxation times metrics and gestational age at MR imaging. Intraclass correlation coefficients were calculated to assess the consistency of the results provided by both modalities.

RESULTS

Both modalities provided quantitative data that revealed negative correlations with gestational age at MR imaging: Synthetic MR imaging-derived relaxation times (medulla oblongata [r = -0.459; P = .021]; midbrain [r = -0.413; P = .040]), T2-relaxation times (medulla oblongata [r = -0.625; P < .001]; midbrain [r = -0.571; P = .003]), and MR fingerprinting-derived T1-relaxation times (medulla oblongata [r = -0.433; P = .035]; midbrain [r = -0.386; P = .062]), and T2-relaxation times (medulla oblongata [r =-0.883; P < .001]; midbrain [r = -0.890; P < .001]).The intraclass correlation coefficient analysis for result consistency between both MR approaches ranged between 0.661 (95% CI, 0.351-0.841) (T2-relaxation times: medulla oblongata) and 0.920 (95% CI, 0.82-0.965) (T1-relaxation times: midbrain).

CONCLUSIONS

There is a good-to-excellent consistency between postmortem Synthetic MR imaging and MR fingerprinting myelin quantifications in fetal brains older than 15 + 1 gestational age. The strong correlations between quantitative myelin metrics and gestational age indicate the potential of quantitative MR imaging to identify delayed or abnormal states of myelination at prenatal stages of cerebral development.

Quantitative assessment of gadolinium deposition in dentate nuclei with MR fingerprinting

RATIONALE AND OBJECTIVES

Gadolinium deposition in the dentate nucleus (DN) has been evaluated by T1-weighted imaging (T1WI) and T1 (R1) mapping, but not MR fingerprinting (MRF). This study investigated associations between T1 and T2 values of DN and gadolinium-based contrast agents (GBCAs) using 2-dimensional MRF.

MATERIALS AND METHODS

This study included 101 patients. Region of interest analysis was performed for T1 and T2 values of DN on MRF (T1-MRF, T2-MRF) and T1-weighted images (T1WI ratio). T1 and T2 ratios compared to normal cerebellar white matter (T1-MRF ratio, T2-MRF ratio) were calculated. The type of previous GBCA was confirmed in 79 patients, and linear regressions were performed between T1, T2 values and number of GBCAs.

RESULTS

Good correlations were observed between T1-MRF and T1WI ratio (ρ = -0.69, P < 0.001) and between T1-MRF ratio and T1WI ratio (ρ = -0.76, P < 0.001). Mild correlations were observed between T2-MRF and T1WI ratio (ρ = -0.32, P < 0.001) and between T2-MRF ratio and T1WI ratio (ρ = -0.44, P < 0.001). The number of linear-type GBCAs was associated with T1-MRF (β = -0.62, P < 0.001) and T1-MRF ratio (β = -0.54, P < 0.001) in univariate linear regression analyses, and with T1-MRF (β = -0.61, P < 0.001) and T1-MRF ratio (β = -0.53, P < 0.001) in multivariate analysis. The number of linear-type GBCAs was associated with T2-MRF (β = -0.30, P < 0.001) and T2-MRF ratio (β = -0.29, P < 0.001) in univariate analyses, and with T2-MRF (β = -0.31, P < 0.001) and T2-MRF ratio (β = -0.32, P < 0.001) in multivariate analyses. No associations were observed between number of macrocyclic GBCAs and T1-MRF (ratio) or T2-MRF (ratio).

CONCLUSION

The number of linear-type GBCA administrations was associated with lower T1 and T2 values (ratios) in DN.

QRAGE-Simultaneous multiparametric quantitative MRI of water content, T1, T2*, and magnetic susceptibility at ultrahigh field strength

PURPOSE

To introduce quantitative rapid gradient-echo (QRAGE), a novel approach for the simultaneous mapping of multiple quantitative MRI parameters, including water content, T1, T2*, and magnetic susceptibility at ultrahigh field strength.

METHODS

QRAGE leverages a newly developed multi-echo MPnRAGE sequence, facilitating the acquisition of 171 distinct contrast images across a range of inversion and TE points. To maintain a short acquisition time, we introduce MIRAGE2, a novel model-based reconstruction method that exploits prior knowledge of temporal signal evolution, represented as damped complex exponentials. MIRAGE2 minimizes local Block-Hankel and Casorati matrices. Parameter maps are derived from the reconstructed contrast images through postprocessing steps. We validate QRAGE through extensive simulations, phantom studies, and in vivo experiments, demonstrating its capability for high-precision imaging.

RESULTS

In vivo brain measurements show the promising performance of QRAGE, with test-retest SDs and deviations from reference methods of < 0.8% for water content, < 17 ms for T1, and < 0.7 ms for T2*. QRAGE achieves whole-brain coverage at a 1-mm isotropic resolution in just 7 min and 15 s, comparable to the acquisition time of an MP2RAGE scan. In addition, QRAGE generates a contrast image akin to the UNI image produced by MP2RAGE.

CONCLUSION

QRAGE is a new, successful approach for simultaneously mapping multiple MR parameters at ultrahigh field.

Fast quantitative MRI: Spiral Acquisition Matching-Based Algorithm (SAMBA) for Robust T1 and T2 Mapping

Conventional diagnostic images from Magnetic Resonance Imaging (MRI) are typically qualitative and require subjective interpretation. Alternatively, quantitative MRI (qMRI) methods have become more prevalent in recent years with multiple clinical and preclinical imaging applications. Quantitative MRI studies on preclinical MRI scanners are being used to objectively assess tissues and pathologies in animal models and to evaluate new molecular MRI contrast agents. Low-field preclinical MRI scanners (≤3.0T) are particularly important in terms of evaluating these new MRI contrast agents at human MRI field strengths. Unfortunately, these low-field preclinical qMRI methods are challenged by long acquisition times, intrinsically low MRI signal levels, and susceptibility to motion artifacts. In this study, we present a new rapid qMRI method for a preclinical 3.0T MRI scanner that combines a Spiral Acquisition with a Matching-Based Algorithm (SAMBA) to rapidly and quantitatively evaluate MRI contrast agents. In this initial development, we compared SAMBA with gold-standard Spin Echo MRI methods using Least Squares Fitting (SELSF) in vitro phantoms and demonstrated shorter scan times without compromising measurement accuracy or repeatability. These initial results will pave the way for future in vivo qMRI studies using state-of-the-art chemical probes.

Fast and High-Resolution T2 Mapping Based on Echo Merging Plus k-t Undersampling with Reduced Refocusing Flip Angles (TEMPURA) as Methods for Human Renal MRI

PURPOSE

To develop a highly accelerated multi-echo spin-echo method, TEMPURA, for reducing the acquisition time and/or increasing spatial resolution for kidney T2 mapping.

METHODS

TEMPURA merges several adjacent echoes into one k-space by either combining independent echoes or sharing one echo between k-spaces. The combined k-space is reconstructed based on compressed sensing theory. Reduced flip angles are used for the refocusing pulses, and the extended phase graph algorithm is used to correct the effects of indirect echoes. Two sequences were developed: a fast breath-hold sequence; and a high-resolution sequence. The performance was evaluated prospectively on a phantom, 16 healthy subjects, and two patients with different types of renal tumors.

RESULTS

The fast TEMPURA method reduced the acquisition time from 3-5 min to one breath-hold (18 s). Phantom measurements showed that fast TEMPURA had a mean absolute percentage error (MAPE) of 8.2%, which was comparable to a standardized respiratory-triggered sequence (7.4%), but much lower than a sequence accelerated by purely k-t undersampling (21.8%). High-resolution TEMPURA reduced the in-plane voxel size from 3 × 3 to 1 × 1 mm2, resulting in improved visualization of the detailed anatomical structure. In vivo T2 measurements demonstrated good agreement (fast: MAPE = 1.3%-2.5%; high-resolution: MAPE = 2.8%-3.3%) and high correlation coefficients (fast: R = 0.85-0.98; high-resolution: 0.82-0.96) with the standardized method, outperforming k-t undersampling alone (MAPE = 3.3-4.5%, R = 0.57-0.59).

CONCLUSION

TEMPURA provides fast and high-resolution renal T2 measurements. It has the potential to improve clinical throughput and delineate intratumoral heterogeneity and tissue habitats at unprecedented spatial resolution.

Cartilage compositional MRI-a narrative review of technical development and clinical applications over the past three decades

Articular cartilage damage and degeneration are among hallmark manifestations of joint injuries and arthritis, classically osteoarthritis. Cartilage compositional MRI (Cart-C MRI), a quantitative technique, which aims to detect early-stage cartilage matrix changes that precede macroscopic alterations, began development in the 1990s. However, despite the significant advancements over the past three decades, Cart-C MRI remains predominantly a research tool, hindered by various technical and clinical hurdles. This paper will review the technical evolution of Cart-C MRI, delve into its clinical applications, and conclude by identifying the existing gaps and challenges that need to be addressed to enable even broader clinical application of Cart-C MRI.

T1 and T2 measurements across multiple 0.55T MRI systems using open-source vendor-neutral sequences

PURPOSE

To compare T1 and T2 measurements across commercial and prototype 0.55T MRI systems in both phantom and healthy participants using the same vendor-neutral pulse sequences, reconstruction, and analysis methods.

METHODS

Standard spin echo measurements and abbreviated protocol measurements of T1, B1, and T2 were made on two prototype 0.55 T systems and two commercial 0.55T systems using an ISMRM/NIST system phantom. Additionally, five healthy participants were imaged at each system using the abbreviated protocol for T1, B1, and T2 measurement. The phantom measurements were compared to NMR-based reference measurements to determine accuracy, and both phantom and in vivo measurements were compared to assess reproducibility and differences between the prototype and commercial systems.

RESULTS

Vendor-neutral sequences were implemented across all four systems, and the code for pulse sequences and reconstruction is freely available. For participants, there was no difference in the mean T1 and T2 relaxation times between the prototype and commercial systems. In the phantom, there were no significant differences between the prototype and commercial systems for T1 and T2 measurements using the abbreviated protocol.

CONCLUSION

Quantitative T1 and T2 measurements at 0.55T in phantom and healthy participants are not statistically different across the prototype and commercial systems.

MyoFold: Joint myocardial tissue composition and wall motion quantification via a highly folded sequence

PURPOSE

This study aims to develop and evaluate a novel cardiovascular MR sequence, MyoFold, designed for the simultaneous quantifications of myocardial tissue composition and wall motion.

METHODS

MyoFold is designed as a 2D single breathing-holding sequence, integrating joint T1/T2 mapping and cine imaging. The sequence uses a 2-fold accelerated balanced SSFP (bSSFP) for data readout and incorporates electrocardiogram synchronization to align with the cardiac cycle. MyoFold initially acquires six single-shot inversion-recovery images, completed during the diastole of six successive heartbeats. T2 preparation (T2-prep) is applied to introduce T2 weightings for the last three images. Subsequently, over the following six heartbeats, segmented bSSFP is performed for the movie of the entire cardiac cycle, synchronized with an electrocardiogram. A neural network trained using numerical simulations of MyoFold is used for T1 and T2 calculations. MyoFold was validated through phantom and in vivo experiments, with comparisons made against MOLLI, SASHA, T2-prep bSSFP, and the conventional cine.

RESULTS

In phantom studies, MyoFold exhibited a 10% overestimation in T1 measurements, whereas T2 measurements demonstrated high accuracy. In vivo experiments revealed that MyoFold T1 had comparable accuracy to SASHA and precision similar to MOLLI. MyoFold demonstrated good agreement with T2-prep bSSFP in myocardial T2 measurements. No significant differences were observed in the quantification of left-ventricle wall thickness and function between MyoFold and the conventional cine.

CONCLUSION

MyoFold presents as a rapid, simple, and multitasking approach for quantitative cardiovascular MR examinations, offering simultaneous assessment of tissue composition and wall motion. The sequence's multitasking capabilities make it a promising tool for comprehensive cardiac evaluations in clinical settings.

Assessment of accuracy and repeatability of quantitative parameter mapping in MRI

We aimed to evaluate the accuracy and repeatability of the T1, T2*, and proton density (PD) values obtained by quantitative parameter mapping (QPM) using the ISMRM/NIST MRI system phantom and compared them with computer simulations. We compared the relaxation times and PD obtained through QPM with the reference values of the ISMRM/NIST MRI system phantom and conventional methods. Furthermore, we evaluated the presence or absence of influences other than noise in T1 and T2* values obtained by QPM by comparing the obtained coefficient of variation (CV) with simulation results. The T1, T2*, and PD values by QPM showed a strong correlation with the measured values and the referenced values. The simulated CVs of QPM calculated for each sphere showed similar trends to those of the actual scans.

T1 Mapping of the Abdomen, From the AJR "How We Do It" Special Series

By exploiting different tissues' characteristic T1 relaxation times, T1-weighted images help distinguish normal and abnormal tissues, aiding assessment of diffuse and local pathologies. However, such images do not provide quantitative T1 values. Advances in abdominal MRI techniques have enabled measurement of abdominal organs' T1 relaxation times, which can be used to create color-coded quantitative maps. T1 mapping is sensitive to tissue microenvironments including inflammation and fibrosis and has received substantial interest for noninvasive imaging of abdominal organ pathology. In particular, quantitative mapping provides a powerful tool for evaluation of diffuse disease by making apparent changes in T1 occurring across organs that may otherwise be difficult to identify. Quantitative measurement also facilitates sensitive monitoring of longitudinal T1 changes. Increased T1 in liver helps to predict parenchymal fibro-inflammation, in pancreas is associated with reduced exocrine function from chronic or autoimmune pancreatitis, and in kidney is associated with impaired renal function and aids diagnosis of chronic kidney disease. In this review, we describe the acquisition, postprocessing, and analysis of T1 maps in the abdomen and explore applications in liver, spleen, pancreas, and kidney. We highlight practical aspects of implementation and standardization, technical pitfalls and confounding factors, and areas of likely greatest clinical impact.

Beyond the Conventional Structural MRI: Clinical Application of Deep Learning Image Reconstruction and Synthetic MRI of the Brain

ABSTRACT

Recent technological advancements have revolutionized routine brain magnetic resonance imaging (MRI) sequences, offering enhanced diagnostic capabilities in intracranial disease evaluation. This review explores 2 pivotal breakthrough areas: deep learning reconstruction (DLR) and quantitative MRI techniques beyond conventional structural imaging. DLR using deep neural networks facilitates accelerated imaging with improved signal-to-noise ratio and spatial resolution, enhancing image quality with short scan times. DLR focuses on supervised learning applied to clinical implementation and applications. Quantitative MRI techniques, exemplified by 2D multidynamic multiecho, 3D quantification using interleaved Look-Locker acquisition sequences with T2 preparation pulses, and magnetic resonance fingerprinting, enable precise calculation of brain-tissue parameters and further advance diagnostic accuracy and efficiency. Potential DLR instabilities and quantification and bias limitations will be discussed. This review underscores the synergistic potential of DLR and quantitative MRI, offering prospects for improved brain imaging beyond conventional methods.

3D MR fingerprinting for dynamic contrast-enhanced imaging of whole mouse brain

PURPOSE

Quantitative MRI enables direct quantification of contrast agent concentrations in contrast-enhanced scans. However, the lengthy scan times required by conventional methods are inadequate for tracking contrast agent transport dynamically in mouse brain. We developed a 3D MR fingerprinting (MRF) method for simultaneous T1 and T2 mapping across the whole mouse brain with 4.3-min temporal resolution.

METHOD

We designed a 3D MRF sequence with variable acquisition segment lengths and magnetization preparations on a 9.4T preclinical MRI scanner. Model-based reconstruction approaches were employed to improve the accuracy and speed of MRF acquisition. The method's accuracy for T1 and T2 measurements was validated in vitro, while its repeatability of T1 and T2 measurements was evaluated in vivo (n = 3). The utility of the 3D MRF sequence for dynamic tracking of intracisternally infused gadolinium-diethylenetriamine pentaacetic acid (Gd-DTPA) in the whole mouse brain was demonstrated (n = 5).

RESULTS

Phantom studies confirmed accurate T1 and T2 measurements by 3D MRF with an undersampling factor of up to 48. Dynamic contrast-enhanced MRF scans achieved a spatial resolution of 192 × 192 × 500 μm3 and a temporal resolution of 4.3 min, allowing for the analysis and comparison of dynamic changes in concentration and transport kinetics of intracisternally infused Gd-DTPA across brain regions. The sequence also enabled highly repeatable, high-resolution T1 and T2 mapping of the whole mouse brain (192 × 192 × 250 μm3) in 30 min.

CONCLUSION

We present the first dynamic and multi-parametric approach for quantitatively tracking contrast agent transport in the mouse brain using 3D MRF.

Fast Whole-Brain MR Multi-Parametric Mapping with Scan-Specific Self-Supervised Networks

Quantification of tissue parameters using MRI is emerging as a powerful tool in clinical diagnosis and research studies. The need for multiple long scans with different acquisition parameters prohibits quantitative MRI from reaching widespread adoption in routine clinical and research exams. Accelerated parameter mapping techniques leverage parallel imaging, signal modelling and deep learning to offer more practical quantitative MRI acquisitions. However, the achievable acceleration and the quality of maps are often limited. Joint MAPLE is a recent state-of-the-art multi-parametric and scan-specific parameter mapping technique with promising performance at high acceleration rates. It synergistically combines parallel imaging, model-based and machine learning approaches for joint mapping ofT 1 , T 2 * , proton density and the field inhomogeneity. However, Joint MAPLE suffers from prohibitively long reconstruction time to estimate the maps from a multi-echo, multi-flip angle (MEMFA) dataset at high resolution in a scan-specific manner. In this work, we propose a faster version of Joint MAPLE which retains the mapping performance of the original version. Coil compression, random slice selection, parameter-specific learning rates and transfer learning are synergistically combined in the proposed framework. It speeds-up the reconstruction time up to 700 times than the original version and processes a whole-brain MEMFA dataset in 21 minutes on average, which originally requires ~260 hours for Joint MAPLE. The mapping performance of the proposed framework is ~2-fold better than the standard and the state-of-the-art evaluated reconstruction techniques on average in terms of the root mean squared error.

3D MR Fingerprinting for Dynamic Contrast-Enhanced Imaging of Whole Mouse Brain

Purpose

Quantitative MRI enables direct quantification of contrast agent concentrations in contrast-enhanced scans. However, the lengthy scan times required by conventional methods are inadequate for tracking contrast agent transport dynamically in mouse brain. We developed a 3D MR fingerprinting (MRF) method for simultaneous T1 and T2 mapping across the whole mouse brain with 4.3-min temporal resolution.

Method

We designed a 3D MRF sequence with variable acquisition segment lengths and magnetization preparations on a 9.4T preclinical MRI scanner. Model-based reconstruction approaches were employed to improve the accuracy and speed of MRF acquisition. The method's accuracy for T1 and T2 measurements was validated in vitro, while its repeatability of T1 and T2 measurements was evaluated in vivo (n=3). The utility of the 3D MRF sequence for dynamic tracking of intracisternally infused Gd-DTPA in the whole mouse brain was demonstrated (n=5).

Results

Phantom studies confirmed accurate T1 and T2 measurements by 3D MRF with an undersampling factor up to 48. Dynamic contrast-enhanced (DCE) MRF scans achieved a spatial resolution of 192 ✕ 192 ✕ 500 μm3 and a temporal resolution of 4.3 min, allowing for the analysis and comparison of dynamic changes in concentration and transport kinetics of intracisternally infused Gd-DTPA across brain regions. The sequence also enabled highly repeatable, high-resolution T1 and T2 mapping of the whole mouse brain (192 ✕ 192 ✕ 250 μm3) in 30 min.

Conclusion

We present the first dynamic and multi-parametric approach for quantitatively tracking contrast agent transport in the mouse brain using 3D MRF.

Simultaneous whole-liver water T 1 $$ {\mathrm{T}}_1 $$ and T 2 $$ {\mathrm{T}}_2 $$ mapping with isotropic resolution during free-breathing

PURPOSE

To develop and validate a data acquisition scheme combined with a motion-resolved reconstruction and dictionary-matching-based parameter estimation to enable free-breathing isotropic resolution self-navigated whole-liver simultaneous water-specificT 1 $$ {\mathrm{T}}_1 $$ (wT 1 $$ {\mathrm{wT}}_1 $$ ) andT 2 $$ {\mathrm{T}}_2 $$ (wT 2 $$ {\mathrm{wT}}_2 $$ ) mapping for the characterization of diffuse and oncological liver diseases.

METHODS

The proposed data acquisition consists of a magnetization preparation pulse and a two-echo gradient echo readout with a radial stack-of-stars trajectory, repeated with different preparations to achieve differentT 1 $$ {\mathrm{T}}_1 $$ andT 2 $$ {\mathrm{T}}_2 $$ contrasts in a fixed acquisition time of 6 min. Regularized reconstruction was performed using self-navigation to account for motion during the free-breathing acquisition, followed by water-fat separation. Bloch simulations of the sequence were applied to optimize the sequence timing forB 1 $$ {B}_1 $$ insensitivity at 3 T, to correct for relaxation-induced blurring, and to mapT 1 $$ {\mathrm{T}}_1 $$ andT 2 $$ {\mathrm{T}}_2 $$ using a dictionary. The proposed method was validated on a water-fat phantom with varying relaxation properties and in 10 volunteers against imaging and spectroscopy reference values. The performance and robustness of the proposed method were evaluated in five patients with abdominal pathologies.

RESULTS

Simulations demonstrate goodB 1 $$ {B}_1 $$ insensitivity of the proposed method in measuringT 1 $$ {\mathrm{T}}_1 $$ andT 2 $$ {\mathrm{T}}_2 $$ values. The proposed method produces co-registeredwT 1 $$ {\mathrm{wT}}_1 $$ andwT 2 $$ {\mathrm{wT}}_2 $$ maps with a good agreement with reference methods (phantom:wT 1 = 1 . 02 wT 1,ref - 8 . 93 ms , R 2 = 0 . 991 $$ {\mathrm{wT}}_1=1.02\kern0.1em {\mathrm{wT}}_{1,\mathrm{ref}}-8.93\kern0.1em \mathrm{ms},{R}^2=0.991 $$ ;wT 2 = 1 . 03 wT 2,ref + 0 . 73 ms , R 2 = 0 . 995 $$ {\mathrm{wT}}_2=1.03\kern0.1em {\mathrm{wT}}_{2,\mathrm{ref}}+0.73\kern0.1em \mathrm{ms},{R}^2=0.995 $$ ). The proposedwT 1 $$ {\mathrm{wT}}_1 $$ andwT 2 $$ {\mathrm{wT}}_2 $$ mapping exhibits good repeatability and can be robustly performed in patients with pathologies.

CONCLUSIONS

The proposed method allows whole-liverwT 1 $$ {\mathrm{wT}}_1 $$ andwT 2 $$ {\mathrm{wT}}_2 $$ quantification with high accuracy at isotropic resolution in a fixed acquisition time during free-breathing.

Multiparametric Characterization of Focal Cortical Dysplasia Using 3D MR Fingerprinting

OBJECTIVES

To develop a multiparametric machine-learning (ML) framework using high-resolution 3 dimensional (3D) magnetic resonance (MR) fingerprinting (MRF) data for quantitative characterization of focal cortical dysplasia (FCD).

MATERIALS

We included 119 subjects, 33 patients with focal epilepsy and histopathologically confirmed FCD, 60 age- and gender-matched healthy controls (HCs), and 26 disease controls (DCs). Subjects underwent whole-brain 3 Tesla MRF acquisition, the reconstruction of which generated T1 and T2 relaxometry maps. A 3D region of interest was manually created for each lesion, and z-score normalization using HC data was performed. We conducted 2D classification with ensemble models using MRF T1 and T2 mean and standard deviation from gray matter and white matter for FCD versus controls. Subtype classification additionally incorporated entropy and uniformity of MRF metrics, as well as morphometric features from the morphometric analysis program (MAP). We translated 2D results to individual probabilities using the percentage of slices above an adaptive threshold. These probabilities and clinical variables were input into a support vector machine for individual-level classification. Fivefold cross-validation was performed and performance metrics were reported using receiver-operating-characteristic-curve analyses.

RESULTS

FCD versus HC classification yielded mean sensitivity, specificity, and accuracy of 0.945, 0.980, and 0.962, respectively; FCD versus DC classification achieved 0.918, 0.965, and 0.939. In comparison, visual review of the clinical magnetic resonance imaging (MRI) detected 48% (16/33) of the lesions by official radiology report. In the subgroup where both clinical MRI and MAP were negative, the MRF-ML models correctly distinguished FCD patients from HCs and DCs in 98.3% of cross-validation trials for the magnetic resonance imaging negative group and MAP negative group. Type II versus non-type-II classification exhibited mean sensitivity, specificity, and accuracy of 0.835, 0.823, and 0.83, respectively; type IIa versus IIb classification showed 0.85, 0.9, and 0.87. In comparison, the transmantle sign was present in 58% (7/12) of the IIb cases.

INTERPRETATION

The MRF-ML framework presented in this study demonstrated strong efficacy in noninvasively classifying FCD from normal cortex and distinguishing FCD subtypes. ANN NEUROL 2024.

Repeatability of Quantitative Knee Cartilage T1, T2, and T1ρ Mapping With 3D-MRI Fingerprinting

BACKGROUND

Three-dimensional MR fingerprinting (3D-MRF) techniques have been recently described for simultaneous multiparametric mapping of knee cartilage. However, investigation of repeatability remains limited.

PURPOSE

To assess the intra-day and inter-day repeatabilities of knee cartilage T1, T2, and T1ρ maps using a 3D-MRF sequence for simultaneous measurement.

STUDY TYPE

Prospective.

SUBJECTS

Fourteen healthy subjects (35.4 ± 9.3 years, eight males), scanned on Day 1 and Day 7.

FIELD STRENGTH/SEQUENCE

3 T/3D-MRF, T1, T2, and T1ρ maps.

ASSESSMENT

The acquisition of 3D-MRF cartilage (simultaneous acquisition of T1, T2, and T1ρ maps) were acquired using a dictionary pattern-matching approach. Conventional cartilage T1, T2, and T1ρ maps were acquired using variable flip angles and a modified 3D-Turbo-Flash sequence with different echo and spin-lock times, respectively, and were fitted using mono-exponential models. Each sequence was acquired on Day 1 and Day 7 with two scans on each day.

STATISTICAL TESTS

The mean and SD for cartilage T1, T2, and T1ρ were calculated in five manually segmented regions of interest (ROIs), including lateral femur, lateral tibia, medial femur, medial tibia, and patella cartilages. Intra-subject and inter-subject repeatabilities were assessed using coefficient of variation (CV) and intra-class correlation coefficient (ICC), respectively, on the same day and among different days. Regression and Bland-Altman analysis were performed to compare maps between the conventional and 3D-MRF sequences.

RESULTS

The CV in all ROIs was lower than 7.4%, 8.4%, and 7.5% and the ICC was higher than 0.56, 0.51, and 0.52 for cartilage T1, T2, and T1ρ, respectively. The MRF results had a good agreement with the conventional methods with a linear regression slope >0.61 and R2 > 0.59.

CONCLUSION

The 3D-MRF sequence had high intra-subject and inter-subject repeatabilities for simultaneously measuring knee cartilage T1, T2, and T1ρ with good agreement with conventional sequences.

EVIDENCE LEVEL

1 TECHNICAL EFFICACY: Stage 1.

In vivo T2 measurements of the fetal brain using single-shot fast spin echo sequences

PURPOSE

We propose a quantitative framework for motion-corrected T2 fetal brain measurements in vivo and validate the single-shot fast spin echo (SS-FSE) sequence to perform these measurements.

METHODS

Stacks of two-dimensional SS-FSE slices are acquired with different echo times (TE) and motion-corrected with slice-to-volume reconstruction (SVR). The quantitative T2 maps are obtained by a fit to a dictionary of simulated signals. The sequence is selected using simulated experiments on a numerical phantom and validated on a physical phantom scanned on a 1.5T system. In vivo quantitative T2 maps are obtained for five fetuses with gestational ages (GA) 21-35 weeks on the same 1.5T system.

RESULTS

The simulated experiments suggested that a TE of 400 ms combined with the clinically utilized TEs of 80 and 180 ms were most suitable for T2 measurements in the fetal brain. The validation on the physical phantom confirmed that the SS-FSE T2 measurements match the gold standard multi-echo spin echo measurements. We measured average T2s of around 200 and 280 ms in the fetal brain grey and white matter, respectively. This was slightly higher than fetal T2* and the neonatal T2 obtained from previous studies.

CONCLUSION

The motion-corrected SS-FSE acquisitions with varying TEs offer a promising practical framework for quantitative T2 measurements of the moving fetus.

Manifold Regularizer for High-Resolution fMRI Joint Reconstruction and Dynamic Quantification

Oscillating Steady-State Imaging (OSSI) is a recently developed fMRI acquisition method that can provide 2 to 3 times higher SNR than standard fMRI approaches. However, because the OSSI signal exhibits a nonlinear oscillation pattern, one must acquire and combine nc (e.g., 10) OSSI images to get an image that is free of oscillation for fMRI, and fully sampled acquisitions would compromise temporal resolution. To improve temporal resolution and accurately model the nonlinearity of OSSI signals, instead of using subspace models that are not well suited for the data, we build the MR physics for OSSI signal generation as a regularizer for the undersampled reconstruction. Our proposed physics-based manifold model turns the disadvantages of OSSI acquisition into advantages and enables joint reconstruction and quantification. OSSI manifold model (OSSIMM) outperforms subspace models and reconstructs high-resolution fMRI images with a factor of 12 acceleration and without spatial or temporal smoothing. Furthermore, OSSIMM can dynamically quantify important physics parameters, including R2∗ maps, with a temporal resolution of 150 ms.

GPU-accelerated Bloch simulations and MR-STAT reconstructions using the Julia programming language

PURPOSE

MR-STAT is a relatively new multiparametric quantitative MRI technique in which quantitative paramater maps are obtained by solving a large-scale nonlinear optimization problem. Managing reconstruction times is one of the main challenges of MR-STAT. In this work we leverage GPU hardware to reduce MR-STAT reconstruction times. A highly optimized, GPU-compatible Bloch simulation toolbox is developed as part of this work that can be utilized for other quantitative MRI techniques as well.

METHODS

The Julia programming language was used to develop a flexible yet highly performant and GPU-compatible Bloch simulation toolbox called BlochSimulators.jl. The runtime performance of the toolbox is benchmarked against other Bloch simulation toolboxes. Furthermore, a (partially matrix-free) modification of a previously presented (matrix-free) MR-STAT reconstruction algorithm is proposed and implemented using the Julia language on GPU hardware. The proposed algorithm is combined with BlochSimulators.jl and the resulting MR-STAT reconstruction times on GPU hardware are compared to previously presented MR-STAT reconstruction times.

RESULTS

The BlochSimulators.jl package demonstrates superior runtime performance on both CPU and GPU hardware when compared to other existing Bloch simulation toolboxes. The GPU-accelerated partially matrix-free MR-STAT reconstruction algorithm, which relies on BlochSimulators.jl, allows for reconstructions of 68 seconds per two-dimensional (2D slice).

CONCLUSION

By combining the proposed Bloch simulation toolbox and the partially matrix-free reconstruction algorithm, 2D MR-STAT reconstructions can be performed in the order of one minute on a modern GPU card. The Bloch simulation toolbox can be utilized for other quantitative MRI techniques as well, for example for online dictionary generation for MR Fingerprinting.

Recurrent neural network-based simultaneous cardiac T1, T2, and T1ρ mapping

The purpose of the current study was to explore the feasibility of training a deep neural network to accelerate the process of generating T1, T2, and T1ρ maps for a recently proposed free-breathing cardiac multiparametric mapping technique, where a recurrent neural network (RNN) was utilized to exploit the temporal correlation among the multicontrast images. The RNN-based model was developed for rapid and accurate T1, T2, and T1ρ estimation. Bloch simulation was performed to simulate a dataset of more than 10 million signals and time correspondences with different noise levels for network training. The proposed RNN-based method was compared with a dictionary-matching method and a conventional mapping method to evaluate the model's effectiveness in phantom and in vivo studies at 3 T, respectively. In phantom studies, the RNN-based method and the dictionary-matching method achieved similar accuracy and precision in T1, T2, and T1ρ estimations. In in vivo studies, the estimated T1, T2, and T1ρ values obtained by the two methods achieved similar accuracy and precision for 10 healthy volunteers (T1: 1228.70 ± 53.80 vs. 1228.34 ± 52.91 ms, p > 0.1; T2: 40.70 ± 2.89 vs. 41.19 ± 2.91 ms, p > 0.1; T1ρ: 45.09 ± 4.47 vs. 45.23 ± 4.65 ms, p > 0.1). The RNN-based method can generate cardiac multiparameter quantitative maps simultaneously in just 2 s, achieving 60-fold acceleration compared with the dictionary-matching method. The RNN-accelerated method offers an almost instantaneous approach for reconstructing accurate T1, T2, and T1ρ maps, being much more efficient than the dictionary-matching method for the free-breathing multiparametric cardiac mapping technique, which may pave the way for inline mapping in clinical applications.

Improved quantitative parameter estimation for prostate T2 relaxometry using convolutional neural networks

OBJECTIVE

Quantitative parameter mapping conventionally relies on curve fitting techniques to estimate parameters from magnetic resonance image series. This study compares conventional curve fitting techniques to methods using neural networks (NN) for measuring T2 in the prostate.

MATERIALS AND METHODS

Large physics-based synthetic datasets simulating T2 mapping acquisitions were generated for training NNs and for quantitative performance comparisons. Four combinations of different NN architectures and training corpora were implemented and compared with four different curve fitting strategies. All methods were compared quantitatively using synthetic data with known ground truth, and further compared on in vivo test data, with and without noise augmentation, to evaluate feasibility and noise robustness.

RESULTS

In the evaluation on synthetic data, a convolutional neural network (CNN), trained in a supervised fashion using synthetic data generated from naturalistic images, showed the highest overall accuracy and precision amongst the methods. On in vivo data, this best performing method produced low-noise T2 maps and showed the least deterioration with increasing input noise levels.

DISCUSSION

This study showed that a CNN, trained with synthetic data in a supervised manner, may provide superior T2 estimation performance compared to conventional curve fitting, especially in low signal-to-noise regions.

Reducing femoral flow artefacts in radial magnetic resonance fingerprinting of the prostate using region-optimised virtual coils

High acceleration factors in radial magnetic resonance fingerprinting (MRF) of the prostate lead to strong streak-like artefacts from flow in the femoral blood vessels, possibly concealing important anatomical information. Region-optimised virtual (ROVir) coils is a beamforming-based framework to create virtual coils that maximise signal in a region of interest while minimising signal in a region of interference. In this study, the potential of removing femoral flow streak artefacts in prostate MRF using ROVir coils is demonstrated in silico and in vivo. The ROVir framework was applied to radial MRF k-space data in an automated pipeline designed to maximise prostate signal while minimising signal from the femoral vessels. The method was tested in 15 asymptomatic volunteers at 3 T. The presence of streaks was visually assessed and measurements of whole prostate T1, T2 and signal-to-noise ratio (SNR) with and without streak correction were examined. In addition, a purpose-built simulation framework in which blood flow through the femoral vessels can be turned on and off was used to quantitatively evaluate ROVir's ability to suppress streaks in radial prostate MRF. In vivo it was shown that removing selected ROVir coils visibly reduces streak-like artefacts from the femoral blood flow, without increasing the reconstruction time. On average, 80% of the prostate SNR was retained. A similar reduction of streaks was also observed in silico, while the quantitative accuracy of T1 and T2 mapping was retained. In conclusion, ROVir coils efficiently suppress streaking artefacts from blood flow in radial MRF of the prostate, thereby improving the visual clarity of the images, without significant sacrifices to acquisition time, reconstruction time and accuracy of quantitative values. This is expected to help enable T1 and T2 mapping of prostate cancer in clinically viable times, aiding differentiation between prostate cancer from noncancer and healthy prostate tissue.

Musculoskeletal perturbations of deep space radiation: Assessment using a Gateway MRI

Human space exploration expansion from Low-Earth Orbit to deep space is accelerating the need to monitor and address the known health concerns related to deep space radiation. The human musculoskeletal system is vulnerable to these risks (alongside microgravity) and its health reflects the well-being of other body systems. Multiparametric magnetic resonance imaging (MRI) is an important approach for assessing temporal physiological changes in the musculoskeletal system. We propose that ultra-low-field MRI provides an optimal low Size Weight and Power (SwaP) solution for non-invasively monitoring muscle and bone changes on the planned Gateway lunar space station. Our proposed ultra-low-field Gateway MRI meets low SWaP design specifications mandated by limited room in the lunar space station. This review summarizes the current state of our knowledge on musculoskeletal consequences of spaceflight, especially with respect to radiation, and then elaborates how MRI can be used to monitor the deleterious effects of space travel and the efficacy of putative countermeasures. We argue that an ultra-low-field MRI in cis-lunar space on the Gateway can provide valuable research and medical insights into the effects of deep space radiation exposure on astronauts. Such an MRI would also allow the development of imaging protocols that would facilitate Earth-bound teams to monitor space personnel musculoskeletal changes during future interplanetary spaceflight. It will especially have a role in monitoring countermeasures, such as the use of melanin, in protecting space explorers.

Free-breathing three-dimensional simultaneous myocardial T1 and T2 mapping based on multi-parametric SAturation-recovery and Variable-flip-Angle

BACKGROUND

Quantitative myocardial tissue characterization with T1 and T2 parametric mapping can provide an accurate and complete assessment of tissue abnormalities across a broad range of cardiomyopathies. However, current clinical T1 and T2 mapping tools rely predominantly on two-dimensional (2D) breath-hold sequences. Clinical adoption of three-dimensional (3D) techniques is limited by long scan duration. The aim of this study is to develop and validate a time-efficient 3D free-breathing simultaneous T1 and T2 mapping sequence using multi-parametric SAturation-recovery and Variable-flip-Angle (mSAVA).

METHODS

mSAVA acquires four volumes for simultaneous whole-heart T1 and T2 mapping. We validated mSAVA using simulations, phantoms, and in-vivo experiments at 3T in 11 healthy subjects and 11 patients with diverse cardiomyopathies. T1 and T2 values by mSAVA were compared with modified Look-Locker inversion recovery (MOLLI) and gradient and spin echo (GraSE), respectively. The clinical performance of mSAVA was evaluated against late gadolinium enhancement (LGE) imaging in patients.

RESULTS

Phantom T1 and T2 by mSAVA showed a strong correlation to reference sequences (R2 = 0.98 and 0.99). In-vivo imaging with an imaging resolution of 1.5 × 1.5 × 8 mm3 could be achieved. Myocardial T1 and T2 of healthy subjects by mSAVA were 1310 ± 46 and 44.6 ± 2.0 ms, respectively, with T1 standard deviation higher than MOLLI (105 ± 12 vs 60 ± 16 ms) and T2 standard deviation lower than GraSE (4.5 ± 0.8 vs 5.5 ± 1.0 ms). mSAVA T1 and T2 maps presented consistent findings in patients undergoing LGE. Myocardial T1 and T2 of all patients by mSAVA were 1421 ± 79 and 47.2 ± 3.3 ms, respectively.

CONCLUSION

mSAVA is a fast 3D technique promising for clinical whole-heart T1 and T2 mapping.

Fast multi-compartment Microstructure Fingerprinting in brain white matter

We proposed two deep neural network based methods to accelerate the estimation of microstructural features of crossing fascicles in the white matter. Both methods focus on the acceleration of a multi-dictionary matching problem, which is at the heart of Microstructure Fingerprinting, an extension of Magnetic Resonance Fingerprinting to diffusion MRI. The first acceleration method uses efficient sparse optimization and a dedicated feed-forward neural network to circumvent the inherent combinatorial complexity of the fingerprinting estimation. The second acceleration method relies on a feed-forward neural network that uses a spherical harmonics representation of the DW-MRI signal as input. The first method exhibits a high interpretability while the second method achieves a greater speedup factor. The accuracy of the results and the speedup factors of several orders of magnitude obtained on in vivo brain data suggest the potential of our methods for a fast quantitative estimation of microstructural features in complex white matter configurations.

Artificial intelligence for neuro MRI acquisition: a review

OBJECT

To review recent advances of artificial intelligence (AI) in enhancing the efficiency and throughput of the MRI acquisition workflow in neuroimaging, including planning, sequence design, and correction of acquisition artifacts.

MATERIALS AND METHODS

A comprehensive analysis was conducted on recent AI-based methods in neuro MRI acquisition. The study focused on key technological advances, their impact on clinical practice, and potential risks associated with these methods.

RESULTS

The findings indicate that AI-based algorithms have a substantial positive impact on the MRI acquisition process, improving both efficiency and throughput. Specific algorithms were identified as particularly effective in optimizing acquisition steps, with reported improvements in workflow efficiency.

DISCUSSION

The review highlights the transformative potential of AI in neuro MRI acquisition, emphasizing the technological advances and clinical benefits. However, it also discusses potential risks and challenges, suggesting areas for future research to mitigate these concerns and further enhance AI integration in MRI acquisition.

Bringing MRI to low- and middle-income countries: Directions, challenges and potential solutions

The global disparity of magnetic resonance imaging (MRI) is a major challenge, with many low- and middle-income countries (LMICs) experiencing limited access to MRI. The reasons for limited access are technological, economic and social. With the advancement of MRI technology, we explore why these challenges still prevail, highlighting the importance of MRI as the epidemiology of disease changes in LMICs. In this paper, we establish a framework to develop MRI with these challenges in mind and discuss the different aspects of MRI development, including maximising image quality using cost-effective components, integrating local technology and infrastructure and implementing sustainable practices. We also highlight the current solutions-including teleradiology, artificial intelligence and doctor and patient education strategies-and how these might be further improved to achieve greater access to MRI.

Deep magnetic resonance fingerprinting based on Local and Global Vision Transformer

To mitigate systematic errors in magnetic resonance fingerprinting (MRF), the precomputed dictionary is usually computed with minimal granularity across the entire range of tissue parameters. However, the dictionary grows exponentially with the number of parameters increase, posing significant challenges to the computational efficiency and matching accuracy of pattern-matching algorithms. Existing works, primarily based on convolutional neural networks (CNN), focus solely on local information to reconstruct multiple parameter maps, lacking in-depth investigations on the MRF mechanism. These methods may not exploit long-distance redundancies and the contextual information within voxel fingerprints introduced by the Bloch equation dynamics, leading to limited reconstruction speed and accuracy. To overcome these limitations, we propose a novel end-to-end neural network called the Local and Global Vision Transformer (LG-ViT) for MRF parameter reconstruction. Our proposed LG-ViT employs a multi-stage architecture that effectively reduces the computational overhead associated with the high-dimensional MRF data and the transformer model. Specifically, a local Transformer encoder is proposed to capture contextual information embedded within voxel fingerprints and local correlations introduced by the interconnected human tissues. Additionally, a global Transformer encoder is proposed to leverage long-distance dependencies arising from shared characteristics among different tissues across various spatial regions. By incorporating MRF physics-based data priors and effectively capturing local and global correlations, our proposed LG-ViT can achieve fast and accurate MRF parameter reconstruction. Experiments on both simulation and in vivo data demonstrate that the proposed method enables faster and more accurate MRF parameter reconstruction compared to state-of-the-art deep learning-based methods.