Search

ImUnity is an original 2.5D deep-learning model designed for efficient and flexible MR image harmonization. A VAE-GAN network, coupled with a confusion module and an optional biological preservation module, uses multiple 2D slices taken from different anatomical locations in each subject of the training database, as well as image contrast transformations for its training. It eventually generates 'corrected' MR images that can be used for various multi-center population studies. Using 3 open source databases (ABIDE, OASIS and SRPBS), which contain MR images from multiple acquisition scanner types or vendors and a large range of subjects ages, we show that ImUnity: (1) outperforms state-of-the-art methods in terms of quality of images generated using traveling subjects; (2) removes sites or scanner biases while improving patients classification; (3) harmonizes data coming from new sites or scanners without the need for an additional fine-tuning and (4) allows the selection of multiple MR reconstructed images according to the desired applications. Tested here on T1-weighted images, ImUnity could be used to harmonize other types of medical images., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023. Published by Elsevier B.V.)

Background: Recent advancements in simultaneous multi-slice (SMS) imaging techniques have enabled whole-brain resting-state fMRI (rs-fMRI) scanning at sub-second temporal resolution, providing spectral ranges much wider than the typically used range of 0.01-0.1 Hz. However, the advantages of this accelerated acquisition for rs-fMRI have not been evaluated., New Method: In this study, we used SMS Echo Planar Imaging (EPI) to probe whole-brain functional connectivity with a short repetition time (TR = 350 ms) and compared it with standard EPI with a longer TR of 2000 ms. We determined the effect of scan length and investigated the temporal filtration strategies that optimize results based on metrics of signal-noise separation and test-retest reliability using both seed-based and independent component analysis (ICA)., Results: We found that use of either the entire frequency range of 0.01-1.4 Hz or the entire frequency range with the exclusion of typical cardiac and respiratory frequency values tended to provide the best functional connectivity maps., Comparison With Existing Methods: We found that the SMS-acquired rs-fMRI scans had improved the signal-noise separation, while preserving the same level of test-retest reliability compared to conventional EPI, and enabled the detection of reliable functional connectivity networks with scan times as short as 3 min., Conclusions: Our findings suggest that whole-brain rs-fMRI studies may benefit from the increased temporal resolution enabled by the SMS-EPI acquisition, leading to drastic scan time reductions, which in turn should enable the more widespread use of rs-fMRI in clinical research protocols., (Copyright © 2018 Elsevier B.V. All rights reserved.)

In resting-state functional MRI (rs-fMRI), functional networks are assessed utilizing the temporal correlation between spontaneous blood oxygen level-dependent signal fluctuations of spatially remote brain regions. Recently, several groups have shown that temporal shifts are present in rs-fMRI maps in patients with cerebrovascular disease due to spatial differences in arterial arrival times, and that this can be exploited to map arrival times in the brain. This suggests that rs-fMRI connectivity mapping may be similarly sensitive to such temporal shifts, and that standard rs-fMRI analysis methods may fail to identify functional connectivity networks. To investigate this, we studied the default mode network (DMN) in Moyamoya disease patients and compared it with normal healthy volunteers. Our results show that using standard independent component analysis (ICA) and seed-based approaches, arterial arrival delays lead to inaccurate incomplete characterization of functional connectivity within the DMN in Moyamoya disease patients. Furthermore, we propose two techniques to correct these errors, for seed-based and ICA methods, respectively. Using these methods, we demonstrate that it is possible to mitigate the deleterious effects of arterial arrival time on the assessment of functional connectivity of the DMN. As these corrections have not been applied to the vast majority of >200 prior rs-fMRI studies in patients with cerebrovascular disease, we suggest that they be interpreted with great caution. Correction methods should be applied in any rs-fMRI connectivity study of subjects expected to have abnormally delayed arterial arrival times.

Purpose: To simultaneously assess reproducibility of three MRI transverse relaxation parameters ( R2', R2*, and R 2 ) for brain tissue oxygenation mapping and to assess changes in these parameters with inhalation of gases that increase and decrease oxygenation, to identify the most sensitive parameter for imaging brain oxygenation., Materials and Methods: Forty-eight healthy subjects (25 male, ages 35 ± 8 years) were scanned at 3.0 Tesla, each with one of four gases (mildly and strongly hypercapnic and hypoxic) administered in a challenge paradigm, using a gas delivery setup designed for patient use. Cerebral blood flow mapping with arterial spin labeling, and simultaneous R2', R2*, and R 2 mapping with gradient-echo sampling of free induction decay and echo (GESFIDE) were performed. Reproducibility in air and gas-induced changes were evaluated using nonparametric analysis with correction for multiple comparisons., Results: Our gas delivery setup achieved stable gas challenges as shown by physiological monitoring. Test-retest variability of R2', R2*, and R 2 were found to be 0.24 s -1 (8.6% of mean), 0.24 s -1 (1.3% of mean), and 0.15 s -1 (1.0% of mean), respectively. Strong hypoxia produced the most conclusive oxygenation-driven relaxation change, inducing increases in R2' (25 ± 13%, P = 0.03), R2* (5 ± 2%, P = 0.02), and R 2 (2 ± 2%, NS)., Conclusion: We benchmarked the intra-scan test-retest variability in GESFIDE-based transverse relaxation rate mapping. Using a reliable framework for gas challenge paradigms, we recommend strong hypoxia for validating oxygenation mapping methods, and the use of tissue R2' change, instead of R2* or R 2 , as a metric for studying brain tissue oxygenation using transverse relaxation methods., Level of Evidence: 1 Technical Efficacy: Stage 1 J. MAGN. RESON. IMAGING 2017;46:704-714., (© 2017 International Society for Magnetic Resonance in Medicine.)

Measurement of the ability of blood vessels to dilate and constrict, known as vascular reactivity, is often performed with breath-holding tasks that transiently raise arterial blood carbon dioxide (P a CO 2 ) levels. However, following the proper commands for a breath-holding experiment may be difficult or impossible for many patients. In this study, we evaluated two approaches for obtaining vascular reactivity information using blood oxygenation level-dependent signal fluctuations obtained from resting-state functional magnetic resonance imaging data: physiological fluctuation regression and coefficient of variation of the resting-state functional magnetic resonance imaging signal. We studied a cohort of 28 older adults (69 ± 7 years) and found that six of them (21%) could not perform the breath-holding protocol, based on an objective comparison with an idealized respiratory waveform. In the subjects that could comply, we found a strong linear correlation between data extracted from spontaneous resting-state functional magnetic resonance imaging signal fluctuations and the blood oxygenation level-dependent percentage signal change during breath-holding challenge ( R 2  = 0.57 and 0.61 for resting-state physiological fluctuation regression and resting-state coefficient of variation methods, respectively). This technique may eliminate the need for subject cooperation, thus allowing the evaluation of vascular reactivity in a wider range of clinical and research conditions in which it may otherwise be impractical.

Advanced imaging techniques including computed tomography (CT) angiography, CT perfusion, magnetic resonance (MR) angiography, MR with diffusion- and perfusion-weighted imaging, and, more recently, resting-state BOLD (Blood Oxygen Level Dependent) functional MRI (rs-fMRI) are increasingly used to evaluate patients with acute ischemic stroke. Advanced imaging allows for identification of patients with ischemic stroke and determination of the size of infarcted and potentially salvageable tissue, all of which yield crucial information for proper stroke management. The addition of rs-fMRI for ischemia adds information at the microvascular level, thereby improving the understanding of pathophysiologic mechanisms of impaired cerebral perfusion and tissue oxygenation beyond the known concepts at the macrovascular level. As such, it may further delineate functional and dysfunctional neuronal networks, guide stroke interventions, and improve prognosis and monitoring of patient outcomes.

The aim of this study was to determine the ability of multiparametric MRI to identify the early effects of individual treatment, during combined chemo-radiotherapy on brain tumours. Eighty male rats bearing 9L gliosarcomas were randomized into four groups: untreated, anti-angiogenic therapy (SORA group), microbeam radiation therapy (MRT group) and both treatments (MRT+SORA group). Multiparametric MRI (tumour volume, diffusion-weighted MR imaging (ADC), blood volume fraction (BVf), microvessel index (VSI), vessel wall integrity (AUC(P846)) and tissue oxygen saturation (StO2)) was performed 1 day before and 2, 5 and 8 days after treatment initiation. Unpaired t-tests and one-way ANOVA were used for statistical analyses. Each MR parameter measured in our protocol was revealed to be sensitive to tumour changes induced by any of the therapies used (individually or combined). When compared with untreated tumours, SORA induced a decrease in BVf, VSI, StO2 and AUC(P846), MRT generated an increase in ADC and AUC(P846) and combined therapies yielded mixed effects: an increase in ADC and AUC(P846) and a decrease in BVf, StO2 and AUC(P846). MRT and MRT+SORA significantly slowed tumour growth. Despite these two groups presenting with similar tumour sizes, the information yielded from MR multiparameter assessment indicated that, when used concomitantly, each therapy induced distinguishable and appreciable physiological changes in the tumour. Our results suggest that multiparametric MRI can monitor the effects of individual treatments, used concomitantly, on brain tumours. Such monitoring would be useful for the detection of tumour resistance to drug/radiotherapy in patients undergoing concomitant therapies., (Copyright © 2015 John Wiley & Sons, Ltd.)

Purpose: R2', the reversible component of transverse relaxation, is an important susceptibility measurement for studies of brain physiology and pathologies. In existing literature, different R2' measurement methods are used with assumption of equivalency. This study explores the choice of measurement method in healthy, young subjects at 3T., Methods: In this study, a modified gradient-echo sampling of free induction decay and echo (GESFIDE) sequence was used to compare four standard R2' measurement methods: asymmetric spin echo (ASE), standard GESFIDE, gradient echo sampling of the spin echo (GESSE), and separate R2 and R2* mapping., Results: GESSE returned lower R2' measurements than other methods (P < 0.05). Intersubject mean R2' in gray matter was found to be 2.7 s(-1) using standard GESFIDE and GESSE, versus 3.4-3.8 s(-1) using other methods. In white matter, mean R2' from GESSE was 2.3 s(-1) while other methods produced 3.7-4.3 s(-1) . R2 correction was applied to partially reduce the discrepancies between the methods, but significant differences remained, likely due to violation of the fundamental assumption of a single-compartmental tissue model, and hence monoexponential decay., Conclusion: R2' measurements are influenced significantly by the choice of method. Awareness of this issue is important when designing and interpreting studies that involve R2' measurements., (© 2014 Wiley Periodicals, Inc.)

Purpose: To investigate if delays in resting-state spontaneous fluctuations of the BOLD (sfBOLD) signal can be used to create maps similar to time-to-maximum of the residue function (Tmax) in Moyamoya patients and to determine whether sfBOLD delays affect the results of brain connectivity mapping., Materials and Methods: Ten patients were scanned at 3 Tesla using a gradient-echo echo planar imaging sequence for sfBOLD imaging. Cross correlation analysis was performed between each brain voxel signal and a reference signal comprised of either the superior sagittal sinus (SSS) or whole brain (WB) average time course. sfBOLD delay maps were created based on the time shift necessary to maximize the correlation coefficient, and compared with dynamic susceptibility contrast Tmax maps. Standard and time-shifted resting-state BOLD connectivity analyses of the default mode network were compared., Results: Good linear correlations were found between sfBOLD delays and Tmax using the SSS as reference (r(2)  = 0.8, slope = 1.4, intercept = -4.6) or WB (r(2)  = 0.7, slope = 0.8, intercept = -3.2). New nodes of connectivity were found in delayed regions when accounting for delays in the analysis., Conclusion: Resting-state sfBOLD imaging can create delay maps similar to Tmax maps without the use of contrast agents in Moyamoya patients. Accounting for these delays may affect the results of functional connectivity maps., (© 2014 Wiley Periodicals, Inc.)

A quantitative estimate of cerebral blood oxygen saturation is of critical importance in the investigation of cerebrovascular disease. While positron emission tomography can map in vivo the oxygen level in blood, it has limited availability and requires ionizing radiation. Magnetic resonance imaging (MRI) offers an alternative through the blood oxygen level-dependent contrast. Here, we describe an in vivo and non-invasive approach to map brain tissue oxygen saturation (StO2) with high spatial resolution. StO2 obtained with MRI correlated well with results from blood gas analyses for various oxygen and hematocrit challenges. In a stroke model, the hypoxic areas delineated in vivo by MRI spatially matched those observed ex vivo by pimonidazole staining. In a model of diffuse traumatic brain injury, MRI was able to detect even a reduction in StO2 that was too small to be detected by histology. In a F98 glioma model, MRI was able to map oxygenation heterogeneity. Thus, the MRI technique may improve our understanding of the pathophysiology of several brain diseases involving impaired oxygenation.

Spontaneous fluctuations in blood oxygenation level-dependent (BOLD) images are the basis of resting-state fMRI and frequently used for functional connectivity studies. However, there may be intrinsic information in the amplitudes of these fluctuations. We investigated the possibility of using the amplitude of spontaneous BOLD signal fluctuations as a biomarker for cerebral vasomotor reactivity. We compared the coefficient of variation (CV) of the time series (defined as the temporal standard deviation of the time series divided by the mean signal intensity) in two populations: 1) Ten young healthy adults and 2) Ten hypertensive elderly subjects with chronic kidney disease (CKD). We found a statistically significant increase (P<0.01) in the CV values for the CKD patients compared with the young healthy adults in both gray matter (GM) and white matter (WM). The difference was independent of the exact segmentation method, became more significant after correcting for physiological signals using RETROICOR, and mainly arose from very low frequency components of the BOLD signal fluctuation (f<0.025 Hz). Furthermore, there was a strong relationship between WM and GM signal fluctuation CV's (R(2)= 0.87) in individuals, with a ratio of about 1:3. These results suggest that amplitude of the spontaneous BOLD signal fluctuations may be used to assess the cerebrovascular reactivity mechanisms and provide valuable information about variations with age and different disease states.

Cerebral blood volume maps are usually acquired using dynamic susceptibility contrast imaging which inherently limits the spatial resolution and signal to noise ratio of the images. In this study, we used ferumoxytol (AMAG Pharmaceuticals, Inc., Cambridge, MA), an FDA-approved compound, to obtain high-resolution cerebral blood volume maps with a steady-state approach in seven healthy volunteers. R2* maps (0.8 × 0.8 × 1 mm(3)) were acquired before and after injection of ferumoxytol and an intraindividual normalization protocol was used to obtain quantitative values. The results show excellent contrast between white and gray matter as well as fine highly detailed vascular structures. An average blood volume of 4% was found in the brain of all volunteers, consistent with prior literature values. A linear relationship was found between ferumoxytol dose (mg/kg) and ΔR2* (1/s) in gray (R(2) = 0.98) and white matter (R(2) = 0.98). A quadratic relationship was found in the sagittal sinus (R(2) = 0.98). The cerebral blood volume maps compare well with lower resolution dynamic susceptibility contrast-MRI and their use should improve the evaluation of small and heterogeneous lesions and facilitate intrapatient and interpatient comparisons., (Copyright © 2012 Wiley Periodicals, Inc.)

The quantification of bolus-tracking MRI techniques remains challenging. The acquisition usually relies on one contrast and the analysis on a simplified model of the various phenomena that arise within a voxel, leading to inaccurate perfusion estimates. To evaluate how simplifications in the interstitial model impact perfusion estimates, we propose a numerical tool to simulate the MR signal provided by a dynamic contrast enhanced (DCE) MRI experiment. Our model encompasses the intrinsic R1 and R2 relaxations, the magnetic field perturbations induced by susceptibility interfaces (vessels and cells), the diffusion of the water protons, the blood flow, the permeability of the vessel wall to the the contrast agent (CA) and the constrained diffusion of the CA within the voxel. The blood compartment is modeled as a uniform compartment. The different blocks of the simulation are validated and compared to classical models. The impact of the CA diffusivity on the permeability and blood volume estimates is evaluated. Simulations demonstrate that the CA diffusivity slightly impacts the permeability estimates (< 5% for classical blood flow and CA diffusion). The effect of long echo times is investigated. Simulations show that DCE-MRI performed with an echo time TE = 5 ms may already lead to significant underestimation of the blood volume (up to 30% lower for brain tumor permeability values). The potential and the versatility of the proposed implementation are evaluated by running the simulation with realistic vascular geometry obtained from two photons microscopy and with impermeable cells in the extravascular environment. In conclusion, the proposed simulation tool describes DCE-MRI experiments and may be used to evaluate and optimize acquisition and processing strategies.

Purpose: To assess the reproducibility of the magnetic resonance (MR) estimate of blood oxygen saturation (sO(2)) in the rat brain, to evaluate the relationship between low MR estimate of sO(2) values and tissue hypoxia in a hypoxic and necrotic glioscarcoma model (9L gliosarcoma cells), and to evaluate the capability of the MR estimate of sO(2) parameter to help identify modifications induced by an antiangiogenic treatment (sorafenib) in 9L gliosarcoma tumors., Materials and Methods: Experiments were performed with permits from the French Ministry of Agriculture. Forty-eight male rats bearing a 9L gliosarcoma were randomized in untreated and treated (sorafenib) groups. MR blood volume fraction and MR estimate of sO(2) parameters were estimated 1 day before and 1, 3, 5, and 8 days after the start of the treatment. The in vivo MR estimate of sO(2) measurement was correlated with the ex vivo hypoxia assessment by using pimonidazole staining. Paired and unpaired t tests, as well as parametric Pearson tests, were used for the statistical analyses., Results: In healthy tissues, MR estimate of sO(2) measurements were comparable to literature values and were reproducible (mean across all animals, 68.0% ± 6.5 [standard deviation]). In untreated tumors, MR estimate of sO(2) and immunohistochemical analysis yielded correlated fractional hypoxic-necrotic areas (R(2) = 0.81). In tumors treated with antiangiogenic therapy, tumor MR estimate of sO(2) was decreased with respect to the healthy tissue (P< .001)., Conclusion: Results of this study suggest that the MR estimate of sO(2) is a reproducible estimate that could be used as an in vivo probe of hypoxia in brain tumors and as a sensitive reporter of the hypoxic effects of antiangiogenic therapies., (© RSNA, 2012.)

Quantitative blood oxygenation level dependent approaches have been designed to obtain quantitative oxygenation information using MRI. A mathematical model is usually fitted to the time signal decay of a gradient-echo and spin-echo measurements to derive hemodynamic parameters such as the blood oxygen saturation or the cerebral blood volume. Although the results in rats and human brain have been encouraging, recent studies have pointed out the need for independent estimation of one or more variables to increase the accuracy of the method. In this study, a multiparametric quantitative blood oxygenation level dependent approach is proposed. A combination of arterial spin labeling and dynamic susceptibility contrast methods were used to obtain quantitative estimates of cerebral blood volume and cerebral blood flow. These results were combined with T 2 and T(2) measurements to derive maps of blood oxygen saturation or cerebral metabolic rate of oxygen. In 12 normal subjects, a mean cerebral blood volume of 4.33 ± 0.7%, cerebral blood flow of 43.8 ± 5.7 mL/min/100 g, blood oxygen saturation of 60 ± 6% and cerebral metabolic rate of oxygen 157 ± 23 μmol/100 g/min were found, which are in agreement with literature values. The results obtained in this study suggest that this methodology could be applied to study brain hypoxia in the setting of pathology., (Copyright © 2011 Wiley Periodicals, Inc.)

Functional MRI (fMRI) brain studies performed in the presence of a steady-state or "blood pool" contrast agent yields activation maps that are weighted for cerebral blood volume (CBV). Previous animal experiments suggest significant contrast-to-noise ratio (CNR) improvements, but these studies have not yet been performed in humans due to the lack of availability of a suitable agent. Here we report the use of the USPIO ferumoxytol (AMAG Pharmaceuticals, Inc., Cambridge, MA) for functional brain activation in humans, termed contrast enhanced functional blood volume imaging (CE-fBVI). Four subjects were scanned during a unilateral finger tapping task with standard blood-oxygen level dependent (BOLD) imaging before contrast and CE-fBVI after contrast injection. The CE-fBVI response showed both a fast (5.8±1.3 s) and a slow (75.3±27.5 s) component of CBV response to stimuli. A significant CNR gain of approximately 2-3 was found for CE-fBVI compared to BOLD fMRI. Interestingly, less susceptibility-related signal dropouts were observed in the inferior frontal and temporal lobes with CE-fBVI. The combination of higher CNR and better spatial specificity, enabled by CE-fBVI using blood pool USPIO contrast agent opens the door to higher resolution brain mapping., (Copyright © 2012 Elsevier Inc. All rights reserved.)

Vessel size index (VSI), a parameter related to the distribution of vessel diameters, may be estimated using two MRI approaches: (i) dynamic susceptibility contrast (DSC) MRI following the injection of a bolus of Gd-chelate. This technique is routinely applied in the clinic to assess intracranial tissue perfusion in patients; (ii) steady-state susceptibility contrast with USPIO contrast agents, which is considered here as the standard method. Such agents are not available for human yet and the steady-state approach is currently limited to animal studies. The aim is to compare VSI estimates obtained with these two approaches on rats bearing C6 glioma (n = 7). In a first session, VSI was estimated from two consecutive injections of Gd-Chelate (Gd(1) and Gd(2)). In a second session (4 hours later), VSI was estimated using USPIO. Our findings indicate that both approaches yield comparable VSI estimates both in contralateral (VSI{USPIO} = 7.5 ± 2.0 µm, VSI{Gd(1)} = 6.5 ± 0.7 µm) and in brain tumour tissues (VSI{USPIO} = 19.4 ± 7.1 µm, VSI{Gd(1)} = 16.6 ± 4.5 µm). We also observed that, in the presence of BBB leakage (as it occurs typically in brain tumours), applying a preload of Gd-chelate improves the VSI estimate with the DSC approach both in contralateral (VSI{Gd(2)} = 7.1 ± 0.4 µm) and in brain tumour tissues (VSI{Gd(2)} = 18.5 ± 4.3 µm) but is not mandatory. VSI estimates do not appear to be sensitive to T(1) changes related to Gd extravasation. These results suggest that robust VSI estimates may be obtained in patients at 3 T or higher magnetic fields with the DSC approach., (Copyright © 2011 John Wiley & Sons, Ltd.)

Purpose: To analyze the contribution of the transverse relaxation parameter (T2), macroscopic field inhomogeneities (B0), and blood volume fraction (BVf) to blood oxygen level-dependent (BOLD)-based magnetic resonance (MR) measurements of blood oxygen saturation (SO2) obtained in a brain tumor model., Materials and Methods: This study was approved by the local committee for animal care and use. Experiments were performed in accordance with permit 380 820 from the French Ministry of Agriculture. The 9L gliosarcoma cells were implanted in the brain of eight rats. Fifteen days later, 4.7-T MR examinations were performed to estimate T2*, T2, BVf, and T2*ΔB0corrected in the tumor and contralateral regions. MR estimates of SO2 were derived by combining T2, BVf, and T2*ΔB0corrected according to a recently described quantitative BOLD approach. Scatterplots and linear regression analysis were used to identify correlation between parameters. Paired Student t tests were used to compare the tumor region with the contralateral region., Results: No significant correlations were found between T2* and any parameter in either tumor tissue or healthy tissue. T2* in the tumor and T2* in the uninvolved contralateral brain were the same (36 msec±4 [standard deviation] vs 36 msec±5, respectively), which might suggest similar oxygenation. Adding T2 information (98 msec±7 vs 68 msec±2, respectively) alone yields results that suggest apparent hypo-oxygenation of the tumor, while incorporating BVf (5.3%±0.6 vs 2.6%±0.3, respectively) alone yields results that suggest apparent hyperoxygenation. MR estimates of SO2 obtained with a complete quantitative BOLD analysis, although not correlated with T2* values, suggest normal oxygenation (68%±3 vs 65%±4, respectively). MR estimates of SO2 obtained in the contralateral tissue agree with previously reported values., Conclusion: Additional measurements, such as BVf, T2, and B0, are needed to obtain reliable information on oxygenation with BOLD MR imaging. The proposed quantitative BOLD approach, which includes these measurements, appears to be a promising tool with which to map tumor oxygenation., (© RSNA, 2011)

Early imaging or blood biomarkers of tumor response is needed to customize anti-tumor therapy on an individual basis. This study evaluates the sensitivity and relevance of five potential MRI biomarkers. Sixty nude rats were implanted with human glioma cells (U-87 MG) and randomized into three groups: one group received an anti-angiogenic treatment (Sorafenib), a second a cytotoxic drug [1,3-bis(2-chloroethyl)-1-nitrosourea, BCNU (Carmustine)] and a third no treatment. The tumor volume, apparent diffusion coefficient (ADC) of water, blood volume fraction (BVf), microvessel diameter (vessel size index, VSI) and vessel wall integrity (contrast enhancement, CE) were monitored before and during treatment. Sorafenib reduced tumor CE as early as 1 day after treatment onset. By 4 days after treatment onset, tumor BVf was reduced and tumor VSI was increased. By 14 days after treatment onset, ADC was increased and the tumor growth rate was reduced. With BCNU, ADC was increased and the tumor growth rate was reduced 14 days after treatment onset. Thus, the estimated MRI parameters were sensitive to treatment at different times after treatment onset and in a treatment-dependent manner. This study suggests that multiparametric MR monitoring could allow the assessment of new anti-tumor drugs and the optimization of combined therapies., (Copyright © 2010 John Wiley & Sons, Ltd.)

Magnetic resonance fingerprinting (MRF) is a quantitative MRI (qMRI) framework that provides simultaneous estimates of multiple relaxation parameters as well as metrics of field inhomogeneity in a single acquisition. However, current challenges exist in the forms of (1) scan time; (2) need for custom image reconstruction; (3) large dictionary sizes; (4) long dictionary-matching time. This study aims to introduce a novel streamlined magnetic-resonance fingerprinting (sMRF) framework based on a single-shot echo-planar imaging (EPI) sequence to simultaneously estimate tissue T1, T2, and T2* with integrated B1+ correction. Encouraged by recent work on EPI-based MRF, we developed a method that combines spin-echo EPI with gradient-echo EPI to achieve T2 in addition to T1 and T2* quantification. To this design, we add simultaneous multi-slice (SMS) acceleration to enable full-brain coverage in a few minutes. Moreover, in the parameter-estimation step, we use deep learning to train a deep neural network (DNN) to accelerate the estimation process by orders of magnitude. Notably, due to the high image quality of the EPI scans, the training process can rely simply on Bloch-simulated data. The DNN also removes the need for storing large dictionaries. Phantom scans along with in-vivo multi-slice scans from seven healthy volunteers were acquired with resolutions of 1.1×1.1×3 mm3 and 1.7×1.7×3 mm3, and the results were validated against ground truth measurements. Excellent correspondence was found between our T1, T2, and T2* estimates and results obtained from standard approaches. In the phantom scan, a strong linear relationship (R = 1–1.04, R2>0.96) was found for all parameter estimates, with a particularly high agreement for T2 estimation (R2>0.99). Similar findings are reported for the in-vivo human data for all of our parameter estimates. Incorporation of DNN results in a reduction of parameter estimation time on the order of 1000 x and a reduction in storage requirements on the order of 2500 x while achieving highly similar results as conventional dictionary matching (%differences of 7.4 ± 0.4%, 3.6 ± 0.3% and 6.0 ± 0.4% error in T1, T2, and T2* estimation). Thus, sMRF has the potential to be the method of choice for future MRF studies by providing ease of implementation, fast whole-brain coverage, and ultra-fast T1/T2/T2* estimation. [ABSTRACT FROM AUTHOR]