Multi-contrast Imaging Research Articles

LSTM-Based RNN Framework to Remove Motion Artifacts in Dynamic Multicontrast MR Images with Registration Model

Today, many people under the age of 10 are being examined for brain-related issues, including tumours, without displaying any symptoms. It is not unusual for children to develop brain-related concerns such as tumours and central nervous system disorders, which may affect 15% of the population. Medical experts believe that the irregular eating habits (junk food) and the consumption of pesticide-tainted fruits and vegetables are to blame. The human body is naturally resistant to harmful gears, but only up to a point. If it exceeds the limit, a cell manipulation process is automatically initiated that can remove dangerous inactive tissues from the cell membrane and later grows into tumour blockage in the human body. Thus, the adoption of an advanced computer-based diagnostic system is highly recommended in order to generate visually enhanced images for anomaly identification and infectious tissue segmentation. In most cases, an MR image is chosen since it is easier to distinguish between affected and nonaffected tissue. Conventional convolution neural network (CCNN) mapping and feature extraction are difficult because of the vast volume of data. In addition, it takes a lengthy time for the MRI scanning process to obtain diverse positions for anomaly identification. Aside from the discomfort, the patient may experience motion abnormalities. Recurrent neural network (RNN) classifies tumour regions into several isolated portions much faster and more accurately, so that it can be prevented. To remove motion artefacts from dynamic multicontrast MR images, a novel long short-term memory- (LSTM-) based RNN framework is introduced in this research. With this method, the MR image’s visual quality is improved over CCNN while simultaneously mapping a larger volume and extracting more quiet characteristics than CCNN can. DC-CNN, SMSR-CNN, FMSI-CNN, and DRCA-CNN results are compared. For both low and high signal-to-noise ratios, the suggested LSTM-based RNN framework has gained reasonable feature intelligibility (SNRs). In comparison to previous approaches, it requires less computing and has higher accuracy when it comes to detecting infected portions.

Frequency importance analysis for chemical exchange saturation transfer magnetic resonance imaging using permuted random forest.

Chemical exchange saturation transfer magnetic resonance imaging (CEST MRI) is a promising molecular imaging tool that allows sensitive detection of endogenous metabolic changes. However, because the CEST spectrum does not display a clear peak like MR spectroscopy, its signal interpretation is challenging, especially under 3-T field strength or with a large saturation B1 . Herein, as an alternative to conventional Z-spectral fitting approaches, a permuted random forest (PRF) method is developed to determine featured saturation frequencies for lesion identification, so-called CEST frequency importance analysis. Briefly, voxels in the CEST dataset were labeled as lesion and control according to multicontrast MR images. Then, by considering each voxel's saturation signal series as a sample, a permutation importance algorithm was employed to rank the contribution of saturation frequency offsets in the differentiation of lesion and normal tissue. Simulations demonstrated that PRF could correctly determine the frequency offsets (3.5 or -3.5ppm) for classifying two groups of Z-spectra, under a range of B0 , B1 conditions and sample sizes. For ischemic rat brains, PRF only displayed high feature importance around amide frequency at 2h postischemia, reflecting that the pH changes occurred at an early stage. By contrast, the data acquired at 24 h postischemia exhibited high feature importance at multiple frequencies (amide, water, and lipids), which suggested the complex tissue changes that occur during the later stages. Finally, PRF was assessed using 3-T CEST data from four brain tumor patients. By defining the tumor region on amide proton transfer-weighted images, PRF analysis identified different CEST frequency importance for two types of tumors (glioblastoma and metastatic tumor) (p < 0.05, with each image slice as a subject). In conclusion, the PRF method was able to rank and interpret the contribution of all acquired saturation offsets to lesion identification; this may facilitate CEST analysis in clinical applications, and open up new doors for comprehensive CEST analysis tools other than model-based approaches.

Characterizing the Correlation Between Angiogenesis and Osteogenesis In Vivo Using Multicontrast Functional Imaging in a Calvarial Defect Model

Angiogenesis is a key factor in bone healing that allows the delivery of oxygen, nutrients, inflammatory and bone precursor cells to the defect site. It induces dramatic structural and functional remodeling of the vasculature during the first 4 weeks of bone healing. Although recent advances in optical imaging have elucidated the in vivo relationship between angiogenesis and osteogenesis in a calvarial defect model, these efforts were mostly limited to structural imaging of the vasculature and bone. Therefore, to better characterize changes in vascular function (i.e. blood flow, oxygenation, etc.) during the bone healing cascade we developed a multicontrast optical imaging framework to assess in vivo changes in microvascular architecture using intrinsic optical signal (IOS) imaging; changes in blood flow with laser speckle contrast (LSC) imaging; and bone formation with bright‐field imaging at high spatial (5 µm) and temporal (200 ms) resolutions. With this system we acquired multicontrast images from 10 animals with calvarial defects every 2 days, over 4 weeks (Fig. 1a, d). The bone and vasculature were then digitally segmented from the bright‐field and IOS images for quantitative analysis (Fig. 1b, c). We also euthanized an animal each week for assessing 3D changes in bone volume and vascular architecture using an ex vivo, high‐resolution (10 µm) CT imaging workflow we recently developed called VascuViz (Fig. 2a, b, d). Using this framework, we demonstrated that angiogenic remodeling within the bone defect microenvironment was most robust from post injury day (D) 4 to D10, during which blood vessel length, volume and density all showed significant increase (Fig. 2c). The blood flow exhibited a sudden increase at D6 when angiogenesis started, and the angiogenic vessels were most perfused at D12 (Fig. 1d). These vascular changes peaked by the end of week 2 and plateaued during the next 2 weeks which were correlated with a large increase in bone volume during the first 2 weeks and smaller increase during the last 2 weeks (Fig. 2e). These results indicate that both structural and functional changes of the microvascular system strongly correlate with bone growth during the early stages of bone healing.Next, we plan to include additional indices of vascular function such as vessel maturity and intravascular oxygenation, as well as image‐based hemodynamic models to further characterize the role of the vascular microenvironment during osteogenesis. We believe that this novel imaging framework for characterizing the defect microenvironment can be utilized to inform the design of novel tissue engineering (TE) constructs for craniofacial bone regeneration. Collectively, these advances herald a new era of “image‐based TE”.

Motion-corrected 3D-EPTI with efficient 4D navigator acquisition for fast and robust whole-brain quantitative imaging.

To develop a motion estimation and correction method for motion-robust three-dimensional (3D) quantitative imaging with 3D-echo-planar time-resolved imaging. The 3D-echo-planar time-resolved imaging technique was designed with additional four-dimensional navigator acquisition (x-y-z-echoes) to achieve fast and motion-robust quantitative imaging of the human brain. The four-dimensional-navigator is inserted into the relaxation-recovery deadtime of the sequence in every pulse TR (∼2s) to avoid extra scan time, and to provide continuous tracking of the 3D head motion and B0 -inhomogeneity changes. By using an optimized spatiotemporal encoding combined with a partial-Fourier scheme, the navigator acquires a large central k-t data block for accurate motion estimation using only four small-flip-angle excitations and readouts, resulting in negligible signal-recovery reduction to the 3D-echo-planar time-resolved imaging acquisition. By incorporating the estimated motion and B0 -inhomogeneity changes into the reconstruction, multi-contrast images can be recovered with reduced motion artifacts. Simulation shows the cost to the SNR efficiency from the added navigator acquisitions is <1%. Both simulation and in vivo retrospective experiments were conducted, that demonstrate the four-dimensional navigator provided accurate estimation of the 3D motion and B0 -inhomogeneity changes, allowing effective reduction of image artifacts in quantitative maps. Finally, in vivo prospective undersampling acquisition was performed with and without head motion, in which the motion corrupted data after correction show close image quality and consistent quantifications to the motion-free scan, providing reliable quantitative measurements even with head motion. The proposed four-dimensional navigator acquisition provides reliable tracking of the head motion and B0 change with negligible SNR cost, equips the 3D-echo-planar time-resolved imaging technique for motion-robust and efficient quantitative imaging.

Analysis of refraction and scattering image artefacts in x-ray analyzer-based imaging

X-ray analyzer-based imaging (ABI) is a powerful phase-sensitive technique that can provide a wide dynamic range of density and extract useful physical properties of the sample. It derives contrast from x-ray absorption, refraction, and scattering properties of the investigated sample. However, x-ray ABI setups can be susceptible to external vibrations, and mechanical imprecisions of system components, e.g., the precision of motor, which are unavoidable in practical experiments. Those factors will provoke deviations of analyzer angular positions and hence errors in the acquired image data. Consequently, those errors will introduce artefacts in the retrieved refraction and scattering images. These artefacts are disadvantageous for further image interpretation and tomographic reconstruction. For this purpose, this work aims to analyze image artefacts resulting from deviations of analyzer angular positions. Analytical expressions of the refraction and scattering image artefacts are derived theoretically and validated by synchrotron radiation experiments. The results show that for the refraction image, the artefact is independent of the sample’s absorption and scattering signals. By contrast, artefact of the scattering image is dependent on both the sample’s refraction and scattering signals, but not on absorption signal. Furthermore, the effect of deviations of analyzer angular positions on the accuracy of the retrieved images is investigated, which can be of use for optimization of data acquisition. This work offers the possibility to develop advanced multi-contrast image retrieval algorithms that suppress artefacts in the retrieved refraction and scattering images in x-ray analyzer-based imaging.

Blip up-down acquisition for spin- and gradient-echo imaging (BUDA-SAGE) with self-supervised denoising enables efficient T2 , T2 *, para- and dia-magnetic susceptibility mapping.

To rapidly obtain high resolution T2 , T2 *, and quantitative susceptibility mapping (QSM) source separation maps with whole-brain coverage and high geometric fidelity. We propose Blip Up-Down Acquisition for Spin And Gradient Echo imaging (BUDA-SAGE), an efficient EPI sequence for quantitative mapping. The acquisition includes multiple T2 *-, T2 '-, and T2 -weighted contrasts. We alternate the phase-encoding polarities across the interleaved shots in this multi-shot navigator-free acquisition. A field map estimated from interim reconstructions was incorporated into the joint multi-shot EPI reconstruction with a structured low rank constraint to eliminate distortion. A self-supervised neural network (NN), MR-Self2Self (MR-S2S), was used to perform denoising to boost SNR. Using Slider encoding allowed us to reach 1 mm isotropic resolution by performing super-resolution reconstruction on volumes acquired with 2 mm slice thickness. Quantitative T2 (=1/R2 ) and T2 * (=1/R2 *) maps were obtained using Bloch dictionary matching on the reconstructed echoes. QSM was estimated using nonlinear dipole inversion on the gradient echoes. Starting from the estimated R2 /R2 * maps, R2 ' information was derived and used in source separation QSM reconstruction, which provided additional para- and dia-magnetic susceptibility maps. In vivo results demonstrate the ability of BUDA-SAGE to provide whole-brain, distortion-free, high-resolution, multi-contrast images and quantitative T2 /T2 * maps, as well as yielding para- and dia-magnetic susceptibility maps. Estimated quantitative maps showed comparable values to conventional mapping methods in phantom and in vivo measurements. BUDA-SAGE acquisition with self-supervised denoising and Slider encoding enables rapid, distortion-free, whole-brain T2 /T2 * mapping at 1 mm isotropic resolution under 90 s.

Three-dimensional multicontrast blood imaging with a single acquisition: Simultaneous non-contrast-enhanced MRA and vessel wall imaging in the thoracic aorta.

To evaluate MRA and vessel wall imaging (VWI) image quality in the thoracic aorta using a novel method named BRIDGE (bright and dark blood images with multishot gradient-echo EPI). The BRIDGE method consists of 3D multishot gradient-echo EPI acquisition using pulse gating, navigator gating, and magnetization preparation with a T2 -preparation pulse and a nonselective inversion-recovery pulse. The BRIDGE and conventional methods (noncontrast MRA based on 3D turbo-field-echo [TFE] and VWI based on 3D turbo spin echo with variable refocusing flip angle [VRFA-TSE]) were performed in 10 healthy volunteers and 10 patients. The SNR, contrast-to-noise ratio (CNR), and sharpness in the thoracic aorta were compared for MRA evaluation. The values of SNRlumen , SNRwall , CNRwall-lumen , contrast ratio (CR)lumen-muscle , coefficient of variation, sharpness, lumen area, and wall area in the thoracic aorta were compared for VWI evaluation. Two radiologists independently performed qualitative image-analysis assessments. When MRA and VWI were acquired, the acquisition time was 26.6% to 27.8% shorter with BRIDGE than the conventional method. In the MRA evaluation, BRIDGE and TFE methods were comparable. In the VWI evaluation, BRIDGE was superior to the VRFA-TSE method in blood suppression and evaluation of the ascending aorta. Because the blood signal suppression of BRIDGE is based on the T1 value of blood, the blood signal can be suppressed more uniformly than with the VRFA-TSE method, regardless of age, blood flow velocity, or vascular anatomy. The BRIDGE method can provide both MRA, to assess vascular anatomy and luminal changes, and VWI, to assess the vessel wall and detect vulnerable plaques, in a single scan.

Multi-Modal X-ray Imaging and Analysis for Characterization of Urinary Stones

Backgound: The composition of stones formed in the urinary tract plays an important role in their management over time. The most common imaging method for the non-invasive evaluation of urinary stones is radiography and computed tomography (CT). However, CT is not very sensitive, and cannot differentiate between all critical stone types. In this study, we propose the application, and evaluate the potential, of a multi-modal (or multi-contrast) X-ray imaging technique called speckle-based imaging (SBI) to differentiate between various types of urinary stones. Methods: Three different stone samples were extracted from animal and human urinary tracts and examined in a laboratory-based speckle tracking setup. The results were discussed based on an X-ray diffraction analysis and a comparison with X-ray microtomography and grating-based interferometry. Results: The stones were classified through compositional analysis by X-ray diffraction. The multi-contrast images obtained using the SBI method provided detailed information about the composition of various urinary stone types, and could differentiate between them. X-ray SBI could provide highly sensitive and high-resolution characterizations of different urinary stones in the radiography mode, comparable to those by grating interferometry. Conclusions: This investigation demonstrated the capability of the SBI technique for the non-invasive classification of urinary stones through radiography in a simple and cost-effective laboratory setting. This opens the possibility for further studies concerning full-field in vivo SBI for the clinical imaging of urinary stones.

Multi-Contrast MRI Image Synthesis Using Switchable Cycle-Consistent Generative Adversarial Networks.

Multi-contrast MRI images use different echo and repetition times to highlight different tissues. However, not all desired image contrasts may be available due to scan-time limitations, suboptimal signal-to-noise ratio, and/or image artifacts. Deep learning approaches have brought revolutionary advances in medical image synthesis, enabling the generation of unacquired image contrasts (e.g., T1-weighted MRI images) from available image contrasts (e.g., T2-weighted images). Particularly, CycleGAN is an advanced technique for image synthesis using unpaired images. However, it requires two separate image generators, demanding more training resources and computations. Recently, a switchable CycleGAN has been proposed to address this limitation and successfully implemented using CT images. However, it remains unclear if switchable CycleGAN can be applied to cross-contrast MRI synthesis. In addition, whether switchable CycleGAN is able to outperform original CycleGAN on cross-contrast MRI image synthesis is still an open question. In this paper, we developed a switchable CycleGAN model for image synthesis between multi-contrast brain MRI images using a large set of publicly accessible pediatric structural brain MRI images. We conducted extensive experiments to compare switchable CycleGAN with original CycleGAN both quantitatively and qualitatively. Experimental results demonstrate that switchable CycleGAN is able to outperform CycleGAN model on pediatric MRI brain image synthesis.

Elevated Hemoglobin A1c Is Associated With Leaky Plaque Neovasculature as Detected by Dynamic Contrast-Enhanced Magnetic Resonance Imaging.

Patients with diabetes have accelerated atherosclerosis progression, but the underlying mechanisms are not fully understood. Dynamic contrast-enhanced magnetic resonance imaging has allowed in vivo characterization of plaque neovasculature, which plays a critical role in plaque progression. We aimed to evaluate the impact of diabetes on carotid plaque neovasculature as assessed by dynamic contrast-enhanced magnetic resonance imaging. Patients with recent ischemic stroke and ipsilateral carotid plaque underwent multicontrast magnetic resonance imaging for characterizing plaque morphology and dynamic contrast-enhanced magnetic resonance imaging for pharmacokinetic parameters of plaque neovasculature, including transfer constant (Ktrans, reflecting flow, endothelial surface area, and permeability) and fractional plasma volume (νp). Sixty-five patients were enrolled, including 30 patients with diabetes (years since diagnosis: median 5.0 [interquartile range, [3.0-12.0]) and 35 patients without diabetes. Subjects with diabetes had a greater plaque burden and a higher prevalence of high-risk characteristics. Additionally, carotid plaques in the subjects with diabetes showed higher Ktrans than those in the subjects without diabetes (0.100±0.048 min-1 versus 0.067±0.042 min-1, P=0.005) but νp was numerically lower in the subjects with diabetes (5.2±3.7% versus 6.2±4.3%, P=0.31). The association of diabetes with high Ktrans (β=0.033, P=0.005) was independent of patient and plaque characteristics and remained largely intact after adjusting for serum lipids, glucose, or hs-CRP (high-sensitivity C-reactive protein). However, it became nonexistent after adjusting for hemoglobin A1c (β=-0.010, P=0.49). Dynamic contrast-enhanced magnetic resonance imaging of carotid plaques suggested that plaque neovasculature in patients with diabetes is leaky, indicating enhanced capability of bringing blood constituents and facilitating extravasation of inflammatory cells, erythrocytes, and plasma proteins. Leaky plaque neovasculature correlated with hemoglobin A1c and may play a role in accelerated atherosclerosis progression in diabetes.

X-ray speckle-based dark-field imaging of water transport in porous ceramics

The evaluation of liquid transport through porous ceramics are of high importance in numerous applications of these materials, ranging from chemical and physical filters to biomaterials. We present a proof-of-concept of the capability of speckle-based Xray dark-field imaging (XDFI) for studying the water transport through porous materials with high sensitivity and sub-pixel resolution in a laboratory. Speckle-based imaging (SBI) takes advantage of a simple and flexible setup, with only an additional and inexpensive textured mask, to provide complementary multi-contrast images. Porous ceramic samples with different pore size ranges were imaged in dry and different pure water-saturated states, via an X-ray speckle-tracking setup. The retrieved darkfield images revealed a high sensitivity to (1) the pore size range and to (2) the local water saturation degree. Independently of the pore size range, the dark-field signal decreased upon water saturation. Compared with previously reported laboratory-scale XDFI results for water transport through porous materials, the speckle-tracking approach allows achieving higher temporal and spatial resolutions, thus broadening the range of (water) transport processes which can be investigated without using any contrast agent.

Inverted Hartmann mask made by deep X-ray lithography for single-shot multi-contrast X-ray imaging with laboratory setup.

This paper reports on the fabrication and characterization of an inverted Hartmann mask and its application for multi-contrast X-ray imaging of polymer composite material in a laboratory setup. Hartmann masks open new possibilities for high-speed X-ray imaging, obtaining orientation-independent information on internal structures without rotating the object. The mask was manufactured with deep X-ray lithography and gold electroplating on a low-absorbing polyimide substrate. Such an approach allows us to produce gratings with a small period and high aspect ratio, leading to a higher spatial resolution and extension towards higher X-ray energies. Tuning the manufacturing process, we achieved a homogeneous patterned area without supporting structures, thus avoiding losses on visibility. We tested mask performance in a laboratory setup with a conventional flat panel detector and assessed mask imaging capabilities using a tailored phantom sample of various sizes. We performed multi-modal X-ray imaging of epoxy matrix polymer composites reinforced with glass fibers and containing microcapsules filled with a healing agent. Hartmann masks made by X-ray lithography enabled fast-tracking of structural changes in low absorbing composite materials and of a self-healing mechanism triggered by mechanical stress.

Evaluation of Multisource Adaptive MRI Fusion for Gross Tumor Volume Delineation of Hepatocellular Carcinoma.

PurposeTumor delineation plays a critical role in radiotherapy for hepatocellular carcinoma (HCC) patients. The incorporation of MRI might improve the ability to correctly identify tumor boundaries and delineation consistency. In this study, we evaluated a novel Multisource Adaptive MRI Fusion (MAMF) method in HCC patients for tumor delineation.MethodsTen patients with HCC were included in this study retrospectively. Contrast-enhanced T1-weighted MRI at portal-venous phase (T1WPP), contrast-enhanced T1-weighted MRI at 19-min delayed phase (T1WDP), T2-weighted (T2W), and diffusion-weighted MRI (DWI) were acquired on a 3T MRI scanner and imported to in-house-developed MAMF software to generate synthetic MR fusion images. The original multi-contrast MR image sets were registered to planning CT by deformable image registration (DIR) using MIM. Four observers independently delineated gross tumor volumes (GTVs) on the planning CT, four original MR image sets, and the fused MRI for all patients. Tumor contrast-to-noise ratio (CNR) and Dice similarity coefficient (DSC) of the GTVs between each observer and a reference observer were measured on the six image sets. Inter-observer and inter-patient mean, SD, and coefficient of variation (CV) of the DSC were evaluated.ResultsFused MRI showed the highest tumor CNR compared to planning CT and original MR sets in the ten patients. The mean ± SD tumor CNR was 0.72 ± 0.73, 3.66 ± 2.96, 4.13 ± 3.98, 4.10 ± 3.17, 5.25 ± 2.44, and 9.82 ± 4.19 for CT, T1WPP, T2W, DWI, T1WDP, and fused MRI, respectively. Fused MRI has the minimum inter-observer and inter-patient variations as compared to original MR sets and planning CT sets. GTV delineation inter-observer mean DSC across the ten patients was 0.81 ± 0.09, 0.85 ± 0.08, 0.88 ± 0.04, 0.89 ± 0.08, 0.90 ± 0.04, and 0.95 ± 0.02 for planning CT, T1WPP, T2W, DWI, T1WDP, and fused MRI, respectively. The patient mean inter-observer CV of DSC was 3.3%, 3.2%, 1.7%, 2.6%, 1.5%, and 0.9% for planning CT, T1WPP, T2W, DWI, T1WDP, and fused MRI, respectively.ConclusionThe results demonstrated that the fused MRI generated using the MAMF method can enhance tumor CNR and improve inter-observer consistency of GTV delineation in HCC as compared to planning CT and four commonly used MR image sets (T1WPP, T1WDP, T2W, and DWI). The MAMF method holds great promise in MRI applications in HCC radiotherapy treatment planning.

Multi-Contrast Magnetic Resonance Imaging of Visual White Matter Pathways in Patients With Glaucoma.

PurposeGlaucoma is a disorder that involves visual field loss caused by retinal ganglion cell damage. Previous diffusion magnetic resonance imaging (dMRI) studies have demonstrated that retinal ganglion cell damage affects tissues in the optic tract (OT) and optic radiation (OR). However, because previous studies have used a simple diffusion tensor model to analyze dMRI data, the microstructural interpretation of white matter tissue changes remains uncertain. In this study, we used a multi-contrast MRI approach to further clarify the type of microstructural damage that occurs in patients with glaucoma.MethodsWe collected dMRI data from 17 patients with glaucoma and 30 controls using 3-tesla (3T) MRI. Using the dMRI data, we estimated three types of tissue property metrics: intracellular volume fraction (ICVF), orientation dispersion index (ODI), and isotropic volume fraction (IsoV). Quantitative T1 (qT1) data, which may be relatively specific to myelin, were collected from all subjects.ResultsIn the OT, all four metrics showed significant differences between the glaucoma and control groups. In the OR, only the ICVF showed significant between-group differences. ICVF was significantly correlated with qT1 in the OR of the glaucoma group, although qT1 did not show any abnormality at the group level.ConclusionsOur results suggest that, at the group level, tissue changes in OR caused by glaucoma might be explained by axonal damage, which is reflected in the intracellular diffusion signals, rather than myelin damage. The significant correlation between ICVF and qT1 suggests that myelin damage might also occur in a smaller number of severe cases.

Isotropic multichannel total variation framework for joint reconstruction of multicontrast parallel MRI.

Purpose: To develop a synergistic image reconstruction framework that exploits multicontrast (MC), multicoil, and compressed sensing (CS) redundancies in magnetic resonance imaging (MRI). Approach: CS, MC acquisition, and parallel imaging (PI) have been individually well developed, but the combination of the three has not been equally well studied, much less the potential benefits of isotropy within such a setting. Inspired by total variation theory, we introduce an isotropic MC image regularizer and attain its full potential by integrating it into compressed MC multicoil MRI. A convex optimization problem is posed to model the new variational framework and a first-order algorithm is developed to solve the problem. Results: It turns out that the proposed isotropic regularizer outperforms many of the state-of-the-art reconstruction methods not only in terms of rotation-invariance preservation of symmetrical features, but also in suppressing noise or streaking artifacts, which are normally encountered in PI methods at aggressive undersampling rates. Moreover, the new framework significantly prevents intercontrast leakage of contrast-specific details, which seems to be a difficult situation to handle for some variational and low-rank MC reconstruction approaches. Conclusions: The new framework is a viable option for image reconstruction in fast protocols of MC parallel MRI, potentially reducing patient discomfort in otherwise long and time-consuming scans.

Tailored magnetic resonance fingerprinting for simultaneous non-synthetic and quantitative imaging: A repeatability study.

The goals of this study include: (a) generating tailored magnetic resonance fingerprinting (TMRF) based non-synthetic imaging; (b) assessing the repeatability of TMRF and deep learning-based mapping of in vitro ISMRM/NIST phantom and in vivo brain data of healthy human subjects. We have acquired qualitative images obtained from the vendor-supplied gold standard (GS), MRF (synthetic), and TMRF (non-synthetic) on one representative healthy human brain. We also acquired 30 datasets on the ISMRM/NIST phantom for the in vitro repeatability study on a GE Discovery 3T MR750w scanner using the TMRF sequence. We compared T1 and T2 maps generated from 30 ISMRM/NIST phantom datasets to the spin-echo (SE) based GS method as part of the in vitro repeatability study. R-squared coefficient of determination in a simple linear regression and Bland-Altman analysis were computed for 30 datasets of ISMRM/NIST phantom to assess the accuracy of in vitro quantitative TMRF data. The repeatability of T1 and T2 estimates by TMRF was evaluated by calculating the standard deviation (SD) divided by the average of 30 datasets for each sphere, respectively. We acquired 10 volunteers for the in vivo repeatability study on the same scanner using the same TMRF sequence. These volunteers were imaged five times with two runs per repetition, resulting in 100 in vivo datasets. Five contrasts, T1 and T2 maps of 10 human volunteers acquired over five repetitions, were evaluated in the in vivo repeatability study. We computed the intraclass correlation coefficient (ICC) of the signal-to-noise ratio (SNR), signal intensities, T1 and T2 relaxation times in white matter (WM), and gray matter (GM). The synthetic images generated from MRF show partial volume and flow artifacts compared to non-synthetic images obtained from TMRF images and the GS. In vitro studies show that TMRF estimates have less than 5% variations except sphere 14 in the T2 array (6.36%). TMRF and SE relaxometry measurements were strongly correlated; R2 values were 0.9958 and 0.9789 for T1 and T2 estimates, respectively. Based on the ICC values, SNR, mean intensity values, and relaxation times of WM and GM for the in vivo studies were consistent. T1 and T2 values of WM and GM were similar to previously published values. The mean ± SD of T1 and T2 for WM for ten subjects and five repeats are 992 ± 41ms and 99 ± 6ms, while the corresponding values for T1 and T2 for GM are 1598 ± 73ms and 152 ± 14ms. TMRF and deep learning-based reconstruction produce repeatable, non-synthetic multi-contrast images, and parametric maps simultaneously.

3D Echo Planar Time-resolved Imaging (3D-EPTI) for ultrafast multi-parametric quantitative MRI

Multi-parametric quantitative MRI has shown great potential to improve the sensitivity and specificity of clinical diagnosis and to enhance our understanding of complex brain processes, but suffers from long scan time especially at high spatial resolution. To address this longstanding challenge, we introduce a novel approach, termed 3D Echo Planar Time-resolved Imaging (3D-EPTI), which significantly increases the acceleration capacity of MRI sampling, and provides high acquisition efficiency for multi-parametric MRI. This is achieved by exploiting the spatiotemporal correlation of MRI data at multiple timescales through new encoding strategies within and between efficient continuous readouts. Specifically, an optimized spatiotemporal CAIPI encoding within the readouts combined with a radial-block sampling strategy across the readouts enables an acceleration rate of 800 fold in the k-t space. A subspace reconstruction was employed to resolve thousands of high-quality multi-contrast images. We have demonstrated the ability of 3D-EPTI to provide robust and repeatable whole-brain simultaneous T1, T2, T2*, PD and B1+ mapping at high isotropic resolution within minutes (e.g., 1-mm isotropic resolution in 3 minutes), and to enable submillimeter multi-parametric imaging to study detailed brain structures.

MRI-Guided Online Adaptive Stereotactic Body Radiation Therapy of Liver and Pancreas Tumors on an MR-Linac System.

Simple SummaryThe hybrid magnetic resonance imaging and medical linear accelerator (MR-Linac) systems are expected to revolutionize radiation therapy, uniquely offering high quality soft-tissue contrast and fast imaging to facilitate the online re-planning and guidance of the treatment delivery. While the clinical procedures for stereotactic body radiotherapy are well-established for conventional linacs (with their strengths and weaknesses), they still require significant development and refinement for the MR-Linac systems. Adjustment of fractionation schemes including clinical goals, patient selection, organ motion management, treatment session length, staff logistics, and overall complexity of the online re-planning sessions are key factors that drive the adoption of MR-Linac technologies. In this report, we present the clinical implementation of an MRI-guided radiation therapy workflow, which was used to treat 16 upper gastro-intestinal cancer patients on a 1.5 T MR-Linac platform. The workflow was proven to be feasible for a wide range of clinical scenarios, and the overall treatment session time was significantly reduced as tasks were optimized and the clinical team gradually gained expertise.Purpose: To describe a comprehensive workflow for MRI-guided online adaptive stereotactic body radiation therapy (SBRT) specific to upper gastrointestinal cancer patients with abdominal compression on a 1.5T MR-Linac system. Additionally, we discuss the workflow’s clinical feasibility and early experience in the case of 16 liver and pancreas patients. Methods: Eleven patients with liver cancer and five patients with pancreas cancer were treated with online adaptive MRI-guidance under abdominal compression. Two liver patients received single-fraction treatments; the remainder plus all pancreas cancer patients received five fractions. A total of 65 treatment sessions were investigated to provide analytics relevant to the online adaptive processes. The quantification of target and organ motion as well as definition and validation of internal target volume (ITV) margins were performed via multi-contrast imaging provided by three different 2D cine sequences. The plan generation was driven by full re-optimization strategies and using T2-weighted 3D image series acquired by means of a respiratory-triggered exhale phase or a time-averaged imaging protocol. As a pre-requisite for the clinical development of the procedure, the image quality was thoroughly investigated via phantom measurements and numerical simulations specific to upper abdominal sites. The delivery of the online adaptive treatments was facilitated by real-time monitoring with 2D cine imaging. Results: Liver 1-fraction and 5-fraction online adaptive session time were on average 80 and 67.5 min, respectively. The total session length varied between 70–90 min for a single fraction and 55–90 min for five fractions. The pancreas sessions were 54–85 min long with an average session time of 68.2 min. Target visualization on the 2D cine image data varied per patient, with at least one of the 2D cine sequences providing sufficient contrast to confidently identify its location and confirm reproducibility of ITV margins. The mean/range of absolute and relative dose values for all treatment sessions evaluated with ArcCheck were 90.6/80.9–96.1% and 99/95.4–100%, respectively. Conclusion: MR-guidance is feasible for liver and pancreas tumors when abdominal compression is used to reduce organ motion, improve imaging quality, and achieve a robust intra- and inter-fraction patient setup. However, the treatment length is significantly longer than for the conventional linac, and patient compliance is paramount for the successful completion of the treatment. Opportunities for reducing the online adaptive session time should be explored. As the next steps, dose-of-the-day and dose accumulation analysis and tools are needed to enhance the workflow and to help further refine the online re-planning processes.

Multicolor Photoacoustic Volumetric Imaging of Subcellular Structures.

Photoacoustic imaging (PAI) has been widely used in multiscale and multicontrast imaging of biological structures and functions. Optical resolution photoacoustic microscopy (OR-PAM), an emerging submodality of PAI, features high lateral resolution and rich optical contrast, indicating great potential in visualizing cellular and subcellular structures. However, three-dimensional (3D) imaging of subcellular structures using OR-PAM has remained a challenge due to the limited axial resolution. In this study, we propose a multicolor 3D photoacoustic microscopy with high lateral/axial resolutions of 0.42/2 and 0.5/2.5 μm at 532 and 780 nm excitation, respectively. Owing to the significantly increased axial resolution, we could visualize the volumetric subcellular structures of melanoma cells using intrinsic contrast. In addition, we carried out multicolor imaging of labeled microtubules/clathrin-coated pits (CCP) and microtubules/mitochondria, respectively, with one scanning by using two different excitation wavelengths. The internal connections between different subcellular structures are revealed by quantitatively comparing the spatial distributions of microtubules/CCP and microtubules/mitochondria in a single cell. Current results suggest that the proposed OR-PAM may serve as an efficient tool for subcellular and cytophysiological studies.

Deep learning-based convolutional neural network for intramodality brain MRI synthesis.

PurposeThe existence of multicontrast magnetic resonance (MR) images increases the level of clinical information available for the diagnosis and treatment of brain cancer patients. However, acquiring the complete set of multicontrast MR images is not always practically feasible. In this study, we developed a state‐of‐the‐art deep learning convolutional neural network (CNN) for image‐to‐image translation across three standards MRI contrasts for the brain.MethodsBRATS’2018 MRI dataset of 477 patients clinically diagnosed with glioma brain cancer was used in this study, with each patient having T1‐weighted (T1), T2‐weighted (T2), and FLAIR contrasts. It was randomly split into 64%, 16%, and 20% as training, validation, and test set, respectively. We developed a U‐Net model to learn the nonlinear mapping of a source image contrast to a target image contrast across three MRI contrasts. The model was trained and validated with 2D paired MR images using a mean‐squared error (MSE) cost function, Adam optimizer with 0.001 learning rate, and 120 epochs with a batch size of 32. The generated synthetic‐MR images were evaluated against the ground‐truth images by computing the MSE, mean absolute error (MAE), peak signal‐to‐noise ratio (PSNR), and structural similarity index (SSIM).ResultsThe generated synthetic‐MR images with our model were nearly indistinguishable from the real images on the testing dataset for all translations, except synthetic FLAIR images had slightly lower quality and exhibited loss of details. The range of average PSNR, MSE, MAE, and SSIM values over the six translations were 29.44–33.25 dB, 0.0005–0.0012, 0.0086–0.0149, and 0.932–0.946, respectively. Our results were as good as the best‐reported results by other deep learning models on BRATS datasets.ConclusionsOur U‐Net model exhibited that it can accurately perform image‐to‐image translation across brain MRI contrasts. It could hold great promise for clinical use for improved clinical decision‐making and better diagnosis of brain cancer patients due to the availability of multicontrast MRIs. This approach may be clinically relevant and setting a significant step to efficiently fill a gap of absent MR sequences without additional scanning.