Search

Compressed sensing magnetic resonance imaging (CS-MRI) seeks to recover visual information from subsampled measurements for diagnostic tasks. Traditional CS-MRI methods often separately address measurement subsampling, image reconstruction, and task prediction, resulting in suboptimal end-to-end performance. In this work, we propose TACKLE as a unified framework for designing CS-MRI systems tailored to specific tasks. Leveraging recent co-design techniques, TACKLE jointly optimizes subsampling, reconstruction, and prediction strategies to enhance the performance on the downstream task. Our results on multiple public MRI datasets show that the proposed framework achieves improved performance on various tasks over traditional CS-MRI methods. We also evaluate the generalization ability of TACKLE by experimentally collecting a new dataset using different acquisition setups from the training data. Without additional fine-tuning, TACKLE functions robustly and leads to both numerical and visual improvements.

Retrospective motion correction (RMC) can substantially reduce motion artifacts in 3D brain MRI. However, for extensive motion, RMC performance is limited. We evaluate RMC with selective reacquisition (RMC+reacq) to expand the range of correctable motion, while directly comparing to prospective motion correction (PMC) for volumetric brain MRI, using external motion tracking. Both approaches lead to significant image quality improvement and the performance of RMC+reacq and PMC was only found statistically significant in 1 of 9 comparisons. These results suggest that RMC with selective reacquisition can match the performance of PMC for 3D-MPRAGE and 3D-FLAIR sequences.

International audience; Head-motion is a main cause of artifacts in MRI. The performance of a markerless optical system, which records the subject's face, estimates head-motion, and allows real-time repositioning of the FOV, is evaluated for neuroanatomical MRI. Sets of T1W/T2W-images were collected from 2 subjects instructed to perform different head-motion protocols during the acquisitions. The System performance is evaluated by comparing images with/without motion-correction. The optical system ensures good corrections of head-motions. However, the correction quality depends on the amplitude of the movement, its location in k-space and its nature. Residual effects of large amplitudes/amounts movements may persist on corrected images. Summary of main findings:-Good performance of prospective motion correction for anatomical brain MRI-Correction quality depends on the magnitude of the motion, its location in k-space and its nature-Optical system is more effective at correcting discontinuous than continuous large motions, especially for T2W-MRI

In neuroimaging, MRI tissue properties characterize underlying neurobiology, provide quantitative biomarkers for neurological disease detection and analysis, and can be used to synthesize arbitrary MRI contrasts. Estimating tissue properties from a single scan session using a protocol available on all clinical scanners promises to reduce scan time and cost, enable quantitative analysis in routine clinical scans and provide scan-independent biomarkers of disease. However, existing tissue properties estimation methods - most often $\mathbf{T_1}$ relaxation, $\mathbf{T_2^*}$ relaxation, and proton density ($\mathbf{PD}$) - require data from multiple scan sessions and cannot estimate all properties from a single clinically available MRI protocol such as the multiecho MRI scan. In addition, the widespread use of non-standard acquisition parameters across clinical imaging sites require estimation methods that can generalize across varying scanner parameters. However, existing learning methods are acquisition protocol specific and cannot estimate from heterogenous clinical data from different imaging sites. In this work we propose an unsupervised deep-learning strategy that employs MRI physics to estimate all three tissue properties from a single multiecho MRI scan session, and generalizes across varying acquisition parameters. The proposed strategy optimizes accurate synthesis of new MRI contrasts from estimated latent tissue properties, enabling unsupervised training, we also employ random acquisition parameters during training to achieve acquisition generalization. We provide the first demonstration of estimating all tissue properties from a single multiecho scan session. We demonstrate improved accuracy and generalizability for tissue property estimation and MRI synthesis., 11 Pages, 9 figures

The Adolescent Brain Cognitive Development (ABCD) Study is an ongoing, nationwide study of the effects of environmental influences on behavioral and brain development in adolescents. The main objective of the study is to recruit and assess over eleven thousand 9-10-year-olds and follow them over the course of 10 years to characterize normative brain and cognitive development, the many factors that influence brain development, and the effects of those factors on mental health and other outcomes. The study employs state-of-the-art multimodal brain imaging, cognitive and clinical assessments, bioassays, and careful assessment of substance use, environment, psychopathological symptoms, and social functioning. The data is a resource of unprecedented scale and depth for studying typical and atypical development. The aim of this manuscript is to describe the baseline neuroimaging processing and subject-level analysis methods used by ABCD. Processing and analyses include modality-specific corrections for distortions and motion, brain segmentation and cortical surface reconstruction derived from structural magnetic resonance imaging (sMRI), analysis of brain microstructure using diffusion MRI (dMRI), task-related analysis of functional MRI (fMRI), and functional connectivity analysis of resting-state fMRI. This manuscript serves as a methodological reference for users of publicly shared neuroimaging data from the ABCD Study.

Skullstripping is defined as the task of segmenting brain tissue from a full head magnetic resonance image~(MRI). It is a critical component in neuroimage processing pipelines. Downstream deformable registration and whole brain segmentation performance is highly dependent on accurate skullstripping. Skullstripping is an especially challenging task for infant~(age range 0--18 months) head MRI images due to the significant size and shape variability of the head and the brain in that age range. Infant brain tissue development also changes the $T_1$-weighted image contrast over time, making consistent skullstripping a difficult task. Existing tools for adult brain MRI skullstripping are ill equipped to handle these variations and a specialized infant MRI skullstripping algorithm is necessary. In this paper, we describe a supervised skullstripping algorithm that utilizes three trained fully convolutional neural networks~(CNN), each of which segments 2D $T_1$-weighted slices in axial, coronal, and sagittal views respectively. The three probabilistic segmentations in the three views are linearly fused and thresholded to produce a final brain mask. We compared our method to existing adult and infant skullstripping algorithms and showed significant improvement based on Dice overlap metric~(average Dice of 0.97) with a manually labeled ground truth data set. Label fusion experiments on multiple, unlabeled data sets show that our method is consistent and has fewer failure modes. In addition, our method is computationally very fast with a run time of 30 seconds per image on NVidia P40/P100/Quadro 4000 GPUs.

The human thalamus is a brain structure that comprises numerous, highly specific nuclei. Since these nuclei are known to have different functions and to be connected to different areas of the cerebral cortex, it is of great interest for the neuroimaging community to study their volume, shape and connectivity in vivo with MRI. In this study, we present a probabilistic atlas of the thalamic nuclei built using ex vivo brain MRI scans and histological data, as well as the application of the atlas to in vivo MRI segmentation. The atlas was built using manual delineation of 26 thalamic nuclei on the serial histology of 12 whole thalami from six autopsy samples, combined with manual segmentations of the whole thalamus and surrounding structures (caudate, putamen, hippocampus, etc.) made on in vivo brain MR data from 39 subjects. The 3D structure of the histological data and corresponding manual segmentations was recovered using the ex vivo MRI as reference frame, and stacks of blockface photographs acquired during the sectioning as intermediate target. The atlas, which was encoded as an adaptive tetrahedral mesh, shows a good agreement with previous histological studies of the thalamus in terms of volumes of representative nuclei. When applied to segmentation of in vivo scans using Bayesian inference, the atlas shows excellent test-retest reliability, robustness to changes in input MRI contrast, and ability to detect differential thalamic effects in subjects with Alzheimer's disease. The probabilistic atlas and companion segmentation tool are publicly available as part of the neuroimaging package FreeSurfer., Graphical abstract Image 1, Highlights • A probabilistic atlas of the human thalamus (26 nuclei) derived from histology. • 3D histology reconstruction assisted by ex vivo MRI and blockface photographs. • Companion Bayesian method segments the nuclei from in vivo scans of any MRI contrast. • Method has excellent test-retest reliability and detects differential effects in AD. • The method is publicly available as part of FreeSurfer.

We present a technique for automatically assigning a neuroanatomical label to each voxel in an MRI volume based on probabilistic information automatically estimated from a manually labeled training set. In contrast to existing segmentation procedures that only label a small number of tissue classes, the current method assigns one of 37 labels to each voxel, including left and right caudate, putamen, pallidum, thalamus, lateral ventricles, hippocampus, and amygdala. The classification technique employs a registration procedure that is robust to anatomical variability, including the ventricular enlargement typically associated with neurological diseases and aging. The technique is shown to be comparable in accuracy to manual labeling, and of sufficient sensitivity to robustly detect changes in the volume of noncortical structures that presage the onset of probable Alzheimer's disease.

The Magnetization-Prepared 2 Rapid Acquisition Gradient Echo (MP2RAGE) method achieves spatially uniform contrast across the entire brain between gray matter and surrounding white matter tissue and cerebrospinal fluid by rapidly acquiring data at two points during an inversion recovery, and then combining the two volumes so as to cancel out sources of intensity and contrast bias, making it useful for neuroimaging studies at ultrahigh field strengths ( 7 T). To quantify the effectiveness of the MP2RAGE method for quantitative morphometric neuroimaging, we performed tissue segmentation and cerebral cortical surface reconstruction of the MP2RAGE data and compared the results with those generated from conventional multi-echo MPRAGE (MEMPRAGE) data across a group of healthy subjects. To do so, we developed a preprocessing scheme for the MP2RAGE image data to allow for automatic cortical segmentation and surface reconstruction using FreeSurfer and analysis methods to compare the positioning of the surface meshes. Using image volumes with 1 mm isotropic voxels we found a scan-rescan reproducibility of cortical thickness estimates to be 0.15 mm (or 6%) for the MEMPRAGE data and a slightly lower reproducibility of 0.19 mm (or 8%) for the MP2RAGE data. We also found that the thickness estimates were systematically smaller in the MP2RAGE data, and that both the interior and exterior cortical boundaries estimated from the MP2RAGE data were consistently positioned within the corresponding boundaries estimated from the MEMPRAGE data. Therefore several measureable differences exist in the appearance of cortical gray matter and its effect on automatic segmentation methods that must be considered when choosing an acquisition or segmentation method for studies requiring cortical surface reconstructions. We propose potential extensions to the MP2RAGE method that may help to reduce or eliminate these discrepancies. (C) 2013 Elsevier Inc. All rights reserved.

Magnetic Resonance Imaging (MRI) is remarkable in that it is possible to obtain image resolutions much smaller than the wavelength of the radiated signal. This is achieved through the use of specialised gradient coils that linearly manipulate the magnitude of the magnetic field within the imaging volume. The instantaneous signal received from the subject represents a periodically varying map based on the duration and magnitude (moment) of the previously applied gradient fields. Representing an object as the sum of periodic maps is difficult and as a result many unique gradient moments are required to form an image. When the subject moves the periodic maps are no longer coherent and the constructive/destructive interference becomes invalid. The artefacts are dependent on how and when motion occurred, and manifest as ghosting, ringing and blurring of the image. This thesis describes a novel approach to measuring and correcting for motion as the data are acquired. A small device was constructed that combines observations from a magnetometer (static magnetic field [z]) and an accelerometer (earth’s gravitational field [y]) with an angular rate sensor to determine its orientation with respect to the imaging coordinate frame (VectOrient). It was precise enough to track the subject’s heart beat and breathing and accurate to within one degree. A gradient field probe was then designed for position encoding. The probe measured the rate of change of the gradient magnetic fields using three mutually orthogonal pickup coils. Assuming linear gradients and using Maxwell’s equations, with negligible rates of change of curl and divergence, it was possible to accurately model the three dimensional vector fields that the gradients produce, eliminating the need for a laborious manual calibration. Sub-microsecond synchronisation was achieved by detecting radio frequency pulses in the imaging sequence with a small resonant circuit. This combined with a 2.4 GHz radio link enabled the probe to be wireless. Finally, the pickup coil observations were combined with the vector based orientation estimates and the gradient field model to achieve efficient multidimensional position, orientation and inter-gradient-delay encoding with a 880 µs pulse sequence insert. The Wireless Radio frequency triggered Acquisition Device (WRAD) tracks involuntary and deliberate subject motion, improving image quality without scanner specific calibration.

Imaging the fetal brain in utero is challenging due to the unpredictable motion of the fetus. Although ultra-fast MRI sequences are able to image a 2D slice in under a second, thus limiting the time in which fetal motion can corrupt images, Cartesian sampling makes these sequences sensitive to signal misregistration and motion-corruption. Corruption of a single 2D slice renders it impossible to reconstruct 3D volumes from these slices without complex slice-to-volume registration. There is a need for motion-robust sequences that can produce high-resolution 3D volumes of the fetal brain. The Siemens Cardiovascular sequence was edited to produce a new radial readout that sampled a 3D spherical volume of k-space with successive diametric spokes. The diameter end points map a spiral trajectory on the surface of a sphere. The trajectory was modified so that multiple sub-volumes of data are sampled during a single acquisition where M is the number of sub-spirals and N is the number of diametric spokes per sub-spiral. This allows reconstruction of individual sub-volumes of data to produce a series of low-resolution navigator images that can be co-registered to provide information on motion during the acquisition. In this way, a segmented sequence suited to self-navigation was developed. Imaging parameters for the 3D radial sequence were optimised based on theoretical calculations and scans performed in adult brains and abdomens. Optimum values for M and N needed to be determined. Increasing M for a constant total number of projections improves the temporal accuracy of motion tracking at the expense of decreased signal to noise ratio in the navigator images. The effects of breathing and rigid body motion on image quality were also compared between 3D radial and equivalent 3D Cartesian acquisitions. Custom reconstruction code was written to separate the incoming scan data according to the sub-spiral trajectories described within the sequence such that individual navigator images could be reconstructed. Successive sub-spiral images were co-registered to the first navigator image to quantify motion during the acquisition. The resulting transformation matrices were then applied to each sub-spiral image after reconstruction and co-registered sub-spiral images combined in image space to generate the final 3D volume. To improve the quality of navigator images, a method is presented to perform navigator image reconstruction at a lower base resolution, thus reducing streaking artifacts and improving the accuracy of image co-registrations. Finally, the methods developed were applied to two fetal scans. The radial sequence was shown to be more motion-robust than an equivalent Cartesian sequence. The minimum number of diametric spokes that provided navigator images that could be accurately co-registered when scanning an adult brain was N=256, which could be acquired in 1.25 s. For abdominal scans, the minimum number of spokes was N=1024, which could be acquired in about 6 s when water excitation is applied. However, the latter could potentially be reduced by reconstructing navigator images at a lower base resolution. Although fetal scans demonstrated poor image contrast, navigator images were able to track motion during the acquisition demonstrating the potential use of this method for self-navigation. In conclusion, a motion-robust radial sequence is presented with potential applications for prospective navigation during fetal MRI.

Includes bibliographical references., Prospective motion correction methods using an optical system, diffusion-weighted prospective acquisition correction, or a free induction decay navigator have recently been applied to correct for motion in diffusion tensor imaging. These methods have some limitations and drawbacks. This article describes a novel technique using a three-dimensional-echo planar imaging navigator, of which the contrast is independent of the b-value, to perform prospective motion correction in diffusion weighted images, without having to reacquire volumes during which motion occurred, unless motion exceeded some preset thresholds. Water phantom and human brain data were acquired using the standard and navigated diffusion sequences, and the mean and whole brain histogram of the fractional anisotropy and mean diffusivity were analyzed.

In population groups where subjects do not lie still during Magnetic Resonance Spectroscopy (MRS) scans, real-time volume of interest (VOI), frequency, and main magnetic field (B0) shim correction may be necessary. This work demonstrates firstly that head movement causes significant B0 disruption in both single voxel spectroscopy and spectroscopic imaging.