Effects of prospective motion correction on perivascular spaces at 7T MRI evaluated using motion artifact simulation.

PURPOSE: The effectiveness of prospective motion correction (PMC) is often evaluated by comparing artifacts in images acquired with and without PMC (NoPMC). However, such an approach is not applicable in clinical setting due to unavailability of NoPMC images. We aim to develop a simulation approach for demonstrating the ability of fat-navigator-based PMC in improving perivascular space (PVS) visibility in T2-weighted MRI. METHODS: MRI datasets from two earlier studies were used for motion artifact simulation and evaluating PMC, including T2-weighted NoPMC and PMC images. To simulate motion artifacts, k-space data at motion-perturbed positions were calculated from artifact-free images using nonuniform Fourier transform and misplaced onto the Cartesian grid before inverse Fourier transform. The simulation's ability to reproduce motion-induced blurring, ringing, and ghosting artifacts was evaluated using sharpness at lateral ventricle/white matter boundary, ringing artifact magnitude in the Fourier spectrum, and background noise, respectively. PVS volume fraction in white matter was employed to reflect its visibility. RESULTS: In simulation, sharpness, PVS volume fraction, and background noise exhibited significant negative correlations with motion score. Significant correlations were found in sharpness, ringing artifact magnitude, and PVS volume fraction between simulated and real NoPMC images (p ≤ 0.006). In contrast, such correlations were reduced and nonsignificant between simulated and real PMC images (p ≥ 0.48), suggesting reduction of motion effects with PMC. CONCLUSIONS: The proposed simulation approach is an effective tool to study the effects of motion and PMC on PVS visibility. PMC may reduce the systematic bias of PVS volume fraction caused by motion artifacts.

Mild Traumatic Brain Injury and Career Stage Associate with Visible Perivascular Spaces in Special Operations Forces Soldiers.

Mild traumatic brain injury (mTBI) and occupational blast exposure in military Service Members may lead to impaired brain waste clearance which increases neurological disease risk. Perivascular spaces (PVS) are a key part of the glymphatic system which supports brain waste clearance, preferentially during sleep. Visible PVS on clinical magnetic resonance imaging have been previously observed in patients with neurodegenerative diseases and animal neurotrauma models. The purpose of this study was to determine associations between PVS morphological characteristics, military career stage, and mTBI history in Special Operations Forces (SOF) Soldiers. Participants underwent T2-weighed neuroimaging to capture three-dimensional whole brain volumes. Segmentation was performed using a previously validated, multi-scale deep convolutional encoder-decoder neural network. Only PVS clusters within the white matter mask were quantified for analyses. Due to non-normal PVS metric distribution, non-parametric Mann-Whitney U tests were used to determine group differences in PVS outcomes. In total, 223 healthy SOF combat Soldiers (age = 33.1 ± 4.3yrs) were included, 217 reported career stage. Soldiers with mTBI history had greater PVS number (z = 2.51, P = 0.013) and PVS volume (z = 2.42, P = 0.016). In-career SOF combat Soldiers had greater PVS number (z = 2.56, P = 0.01) and PVS volume (z = 2.28, P = 0.02) compared to a baseline cohort. Mild TBI history is associated with increased PVS burden in SOF combat Soldiers that are clinically recovered from mTBI. This may indicate ongoing physiological changes that could lead to impaired waste clearance via the glymphatic system. Future studies should determine if PVS number and volume are meaningful neurobiological outcomes for neurodegenerative disease risk and if clinical interventions such as improving sleep can reduce PVS burden.

Diet-Stimulated Marrow Adiposity Fails to Worsen Early, Age-Related Bone Loss.

INTRODUCTION: Longitudinal effect of diet-induced obesity on bone is uncertain. Prior work showed both no effect and a decrement in bone density or quality when obesity begins prior to skeletal maturity. We aimed to quantify long-term effects of obesity on bone and bone marrow adipose tissue (BMAT) in adulthood. METHODS: Skeletally mature, female C57BL/6 mice (n = 70) aged 12 weeks were randomly allocated to low-fat diet (LFD; 10% kcal fat; n = 30) or high-fat diet (HFD; 60% kcal fat; n = 30), with analyses at 12, 15, 18, and 24 weeks (n = 10/group). Tibial microarchitecture was analyzed by µCT, and volumetric BMAT was quantified via 9.4T MRI/advanced image analysis. Histomorphometry of adipocytes and osteoclasts, and qPCR were performed. RESULTS: Body weight and visceral white adipose tissue accumulated in response to HFD started in adulthood. Trabecular bone parameters declined with advancing experimental age. BV/TV declined 22% in LFD (p = 0.0001) and 17% in HFD (p = 0.0022) by 24 weeks. HFD failed to appreciably alter BV/TV and had negligible impact on other microarchitecture parameters. Both dietary intervention and age accounted for variance in BMAT, with regional differences: distal femoral BMAT was more responsive to diet, while proximal femoral BMAT was more attenuated by age. BMAT increased 60% in the distal metaphysis in HFD at 18 and 24 weeks (p = 0.0011). BMAT in the proximal femoral diaphysis, unchanged by diet, decreased 45% due to age (p = 0.0002). Marrow adipocyte size via histomorphometry supported MRI quantification. Osteoclast number did not differ between groups. Tibial qPCR showed attenuation of some adipose, metabolism, and bone genes. A regulator of fatty acid ß-oxidation, cytochrome C (CYCS), was 500% more abundant in HFD bone (p < 0.0001; diet effect). CYCS also increased due to age, but to a lesser extent. HFD mildly increased OCN, TRAP, and SOST. CONCLUSIONS: Long-term high fat feeding after skeletal maturity, despite upregulation of visceral adiposity, body weight, and BMAT, failed to attenuate bone microarchitecture. In adulthood, we found aging to be a more potent regulator of microarchitecture than diet-induced obesity.

Improving motion robustness of 3D MR fingerprinting with a fat navigator.

PURPOSE: To develop a 3D MR fingerprinting (MRF) method in combination with fat navigators to improve its motion robustness for neuroimaging. METHODS: A rapid fat navigator was developed using the stack-of-spirals acquisition and non-Cartesian spiral GRAPPA. The fat navigator module was implemented in the 3D MRF sequence with high scan efficiency. The developed method was first validated in phantoms and five healthy subjects with intentional head motion. The method was further applied to infants with neonatal opioid withdrawal symptoms. The 3D MRF scans with fat navigators acquired with and without acceleration along the partition-encoding direction were both examined in the study. RESULTS: Both phantom and in vivo results demonstrated that the added fat navigator modules did not influence the quantification accuracy in MRF. In combination with non-Cartesian spiral GRAPPA, a rapid fat navigator sampling with whole-brain coverage was achieved in ˜0.5 s at 3T, reducing its sensitivity to potential motion. Based on the motion waveforms extracted from fat navigators, the motion robustness of the 3D MRF was largely improved. With the proposed method, the motion-corrupted MRF datasets yielded T1 and T2 maps with significantly reduced artifacts and high correlations with measurements from the reference motion-free MRF scans. CONCLUSION: We developed a 3D MRF method coupled with rapid fat navigators to improve its motion robustness for quantitative neuroimaging. Our results demonstrate that (1) accurate tissue quantification was preserved with the fat navigator modules and (2) the motion robustness for quantitative tissue mapping was largely improved with the developed method.

In vivo detection of penetrating arteriole alterations in cerebral white matter in patients with diabetes with 7 T MRI.

Cerebral small vessel disease (SVD) is responsible for primary intracerebral hemorrhages, lacunar infarcts and white matter hyperintensity in T2 weighted images. While the brain lesions attributed to small vessel disease can be characterized by conventional MRI, it remains challenging to noninvasively measure the early pathological changes of the small underlying vessels. We evaluated the feasibility of detecting alterations in white matter penetrating arterioles (PA) in patients with diabetes with ultra-high field 7 T MRI. 19 participants with diabetes mellitus (DM) and 19 age- and sex-matched healthy controls were scanned with whole brain T2 and susceptibility weighted MRI and a single slice phase contrast MRI 15 mm above the corpus callosum. The PC-MRI scans were repeated three times. PA masks were manually drawn on the first images after anonymization or automatically segmented on all three images. For each PA, lumen diameter, flow velocity and volume flow rate were derived by model-based analyses of complex difference images. Quasi-Poisson regression was performed for PA count using disease condition, age, and sex as independent variables. Linear mixed effect model analyses were performed for the other measurements using disease condition and age as fixed effect and participant pair specific disease condition as random effect. No severe radiological features of SVD were observed in T2 and susceptibility weighted images in any of the participants except for white matter hyperintensities with Fazekas score of 1 or 2 in 68% and 26% of patients and controls, respectively. The minimum diameter of visible PA was 78 µm and the majority had diameters <250 µm. Among the manually segmented PA with tilt angle less than 30o from the slice normal direction, flow velocities were lower in the DM group (1.9 ± 0.6 vs. 2.2 ± 0.6; p = 0.022), while no significant difference was observed in count, diameter, or volume flow rate. Similar results were observed in the automatically segmented PA. We also observed significantly increased diameter or decreased velocity with age in some of the scans. This study suggests that early PA alterations that are discriminative of disease state and age might be detectable in human cerebral white matter with 7 T MRI in vivo.

Prospective motion correction and automatic segmentation of penetrating arteries in phase contrast MRI at 7 T.

PURPOSE: To develop a prospective motion correction (MC) method for phase contrast (PC) MRI of penetrating arteries (PAs) in centrum semiovale at 7 T and to evaluate its performance using automatic PA segmentation. METHODS: Head motion was monitored and corrected during the scan based on fat navigator images. Two convolutional neural networks (CNN) were developed to automatically segment PAs and exclude surface vessels. Real-life scans with MC and without MC (NoMC) were performed to evaluate the MC performance. Motion score was calculated from the ranges of translational and rotational motion parameters. MC versus NoMC pairs with similar motion scores during MC and NoMC scans were compared. Data corrupted by motion were reacquired to further improve PA visualization. RESULTS: PA counts (NPA ) and PC and magnitude contrasts (MgC) relative to neighboring tissue were significantly correlated with motion score and were higher in MC than NoMC images at motion scores above 0.5-0.8 mm. Data reacquisition further increased PC but had no significant effect on NPA and MgC. CNNs had higher sensitivity and Dice similarity coefficient for detecting PAs than a threshold-based method. CONCLUSIONS: Prospective MC can improve the count and contrast of segmented PAs in the presence of severe motion. CNN-based PA segmentation has improved performance in delineating PAs than the threshold-based method.

Visual Motion Coherence Responses in Human Visual Cortex.

Random dot kinematograms (RDKs) have recently been used to train subjects with cortical scotomas to perform direction of motion discrimination, partially restoring visual motion perception. To study the recovery of visual perception, it is important to understand how visual areas in normal subjects and subjects with cortical scotomas respond to RDK stimuli. Studies in normal subjects have shown that blood oxygen level-dependent (BOLD) responses in human area hV5/MT+ increase monotonically with coherence, in general agreement with electrophysiology studies in primates. However, RDK responses in prior studies were obtained while the subject was performing fixation, not a motion discrimination condition. Furthermore, BOLD responses were gauged against a baseline condition of uniform illumination or static dots, potentially decreasing the specificity of responses for the spatial integration of local motion signals (motion coherence). Here, we revisit this question starting from a baseline RDK condition of no coherence, thereby isolating the component of BOLD response due specifically to the spatial integration of local motion signals. In agreement with prior studies, we found that responses in the area hV5/MT+ of healthy subjects were monotonically increasing when subjects fixated without performing a motion discrimination task. In contrast, when subjects were performing an RDK direction of motion discrimination task, responses in the area hV5/MT+ remained flat, changing minimally, if at all, as a function of motion coherence. A similar pattern of responses was seen in the area hV5/MT+ of subjects with dense cortical scotomas performing direction of motion discrimination for RDKs presented inside the scotoma. Passive RDK presentation within the scotoma elicited no significant hV5/MT+ responses. These observations shed further light on how visual cortex responses behave as a function of motion coherence, helping to prepare the ground for future studies using these methods to study visual system recovery after injury.

Effects of motion and retrospective motion correction on the visualization and quantification of perivascular spaces in ultrahigh resolution T2-weighted images at 7T.

PURPOSE: Motion can strongly affect MRI image quality and derived imaging measures. We studied the effects of motion and retrospective motion correction (MC) on the visualization and quantitative measures of the perivascular space and penetrating vessel (PVSV) complex, an essential part of the glymphatic system, on high-resolution T2 -weighted MRI images at 7T. METHODS: MC was achieved by adjusting k-space data based on head positions measured using fat navigator images. PVSV visibility and quantitative measures including diameter, volume fraction (VF), count, and contrast were compared between images with and without MC. RESULTS: Without MC, VF, and count decreased significantly with increasing head rotation. MC improved PVSV visualization in all cases with severe motion artifacts. MC decreased diameter in white matter (WM) and increased VF, count, and contrast in basal ganglia and WM. The changes of VF, count, and contrast after MC strongly correlated with motion severity. MC eliminated the significant dependences of VF and count on rotation and reduced the inter-subject variations of VF and count. The effect sizes of age and breathing gas effects on VF and count, and contrast increased in most cases after MC, while those on diameter exhibited inconsistent behavior. CONCLUSIONS: Motion affects PVSV quantification without MC. MC improves PVSV visibility and increases the statistical power of detecting physiological PVSV VF, count, and contrast changes but may have limited benefits for increasing the power for detecting diameter changes.

Changes in Infrapatellar Fat Pad Volume 6 to 12 Months After Anterior Cruciate Ligament Reconstruction and Associations With Patient-Reported Knee Function.

CONTEXT: Hypertrophy of the infrapatellar fat pad (IFP) in idiopathic knee osteoarthritis has been linked to deleterious synovial changes and joint pain related to mechanical tissue impingement. Yet little is known regarding the IFP's volumetric changes after anterior cruciate ligament reconstruction (ACLR). OBJECTIVES: To examine changes in IFP volume between 6 and 12 months after ACLR and determine associations between patient-reported outcomes and IFP volume at each time point as well as the volume change over time. In a subset of individuals, we examined interlimb IFP volume differences 12 months post-ACLR. STUDY DESIGN: Prospective cohort study. SETTING: Laboratory. PATIENTS OR OTHER PARTICIPANTS: We studied 26 participants (13 women, 13 men, age = 21.88 ± 3.58 years, body mass index = 23.82 ± 2.21 kg/m2) for our primary aims and 13 of those participants (8 women, 5 men, age = 21.15 ± 3.85 years, body mass index = 23.01 ± 2.01 kg/m2) for our exploratory aim. MAIN OUTCOME MEASURE(S): Using magnetic resonance imaging, we evaluated the IFP volume change between 6 and 12 months post-ACLR in the ACLR limb and between-limbs differences at 12 months in a subset of participants. International Knee Documentation Committee subjective knee evaluation (IKDC) scores were collected at 6-month and 12-month follow-ups, and associations between IFP volume and patient-reported outcomes were determined. RESULTS: The IFP volume in the ACLR limb increased from 6 months (19.67 ± 6.30 cm3) to 12 months (21.26 ± 6.91 cm3) post-ACLR. Greater increases of IFP volume between 6 and 12 months were significantly associated with better 6-month IKDC scores (r = .44, P = .03). The IFP volume was greater in the uninjured limb (22.71 ± 7.87 cm3) than in the ACLR limb (20.75 ± 9.03 cm3) 12 months post-ACLR. CONCLUSIONS: The IFP volume increased between 6 and 12 months post-ACLR; however, the IFP volume of the ACLR limb remained smaller than that of the uninjured limb at 12 months. In addition, those with better knee function 6 months post-ACLR demonstrated greater increases in IFP volume between 6 and 12 months post-ACLR. This suggests that greater IFP volumes may play a role in long-term joint health after ACLR.

Multi-Site Infant Brain Segmentation Algorithms: The iSeg-2019 Challenge.

To better understand early brain development in health and disorder, it is critical to accurately segment infant brain magnetic resonance (MR) images into white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF). Deep learning-based methods have achieved state-of-the-art performance; h owever, one of the major limitations is that the learning-based methods may suffer from the multi-site issue, that is, the models trained on a dataset from one site may not be applicable to the datasets acquired from other sites with different imaging protocols/scanners. To promote methodological development in the community, the iSeg-2019 challenge (http://iseg2019.web.unc.edu) provides a set of 6-month infant subjects from multiple sites with different protocols/scanners for the participating methods. T raining/validation subjects are from UNC (MAP) and testing subjects are from UNC/UMN (BCP), Stanford University, and Emory University. By the time of writing, there are 30 automatic segmentation methods participated in the iSeg-2019. In this article, 8 top-ranked methods were reviewed by detailing their pipelines/implementations, presenting experimental results, and evaluating performance across different sites in terms of whole brain, regions of interest, and gyral landmark curves. We further pointed out their limitations and possible directions for addressing the multi-site issue. We find that multi-site consistency is still an open issue. We hope that the multi-site dataset in the iSeg-2019 and this review article will attract more researchers to address the challenging and critical multi-site issue in practice.

Exercise Increases Bone in SEIPIN Deficient Lipodystrophy, Despite Low Marrow Adiposity.

Exercise, typically beneficial for skeletal health, has not yet been studied in lipodystrophy, a condition characterized by paucity of white adipose tissue, with eventual diabetes, and steatosis. We applied a mouse model of global deficiency of Bscl2 (SEIPIN), required for lipid droplet formation. Male twelve-week-old B6 knockouts (KO) and wild type (WT) littermates were assigned six-weeks of voluntary, running exercise (E) versus non-exercise (N=5-8). KO weighed 14% less than WT (p=0.01) and exhibited an absence of epididymal adipose tissue; KO liver Plin1 via qPCR was 9-fold that of WT (p=0.04), consistent with steatosis. Bone marrow adipose tissue (BMAT), unlike white adipose, was measurable, although 40.5% lower in KO vs WT (p=0.0003) via 9.4T MRI/advanced image analysis. SEIPIN ablation's most notable effect marrow adiposity was in the proximal femoral diaphysis (-56% KO vs WT, p=0.005), with relative preservation in KO-distal-femur. Bone via µCT was preserved in SEIPIN KO, though some quality parameters were attenuated. Running distance, speed, and time were comparable in KO and WT. Exercise reduced weight (-24% WT-E vs WT p<0.001) but not in KO. Notably, exercise increased trabecular BV/TV in both (+31%, KO-E vs KO, p=0.004; +14%, WT-E vs WT, p=0.006). The presence and distribution of BMAT in SEIPIN KO, though lower than WT, is unexpected and points to a uniqueness of this depot. That trabecular bone increases were achievable in both KO and WT, despite a difference in BMAT quantity/distribution, points to potential metabolic flexibility during exercise-induced skeletal anabolism.

Morphology of perivascular spaces and enclosed blood vessels in young to middle-aged healthy adults at 7T: Dependences on age, brain region, and breathing gas.

Perivascular spaces (PVSs) are fluid-filled spaces surrounding penetrating blood vessels in the brain and are an integral pathway of the glymphatic system. A PVS and the enclosed blood vessel are commonly visualized as a single vessel-like complex (denoted as PVSV) in high-resolution MRI images. Quantitative characterization of the PVSV morphology in MRI images in healthy subjects may serve as a reference for detecting disease related PVS and/or blood vessel alterations in patients with brain diseases. To this end, we evaluated the age dependences, spatial heterogeneities, and dynamic properties of PVSV morphological features in 45 healthy subjects (21-55 years old), using an ultra-high-resolution three-dimensional transverse relaxation time weighted MRI sequence (0.41 â€‹× â€‹0.41 â€‹× â€‹0.4 â€‹mm3) at 7T. Quantitative PVSV parameters, including apparent diameter, count, volume fraction (VF), and relative contrast to noise ratio (rCNR) were calculated in the white matter and subcortical structures. Dynamic changes were induced by carbogen breathing which are known to induce vasodilation and increase the blood oxygenation level in the brain. PVSV count and VF significantly increased with age in basal ganglia (BG), so did rCNR in BG, midbrain, and white matter (WM). Apparent PVSV diameter also showed a positive association with age in the three brain regions, although it did not reach statistical significance. The PVSV VF and count showed large inter-subject variations, with coefficients of variation ranging from 0.17 to 0.74 after regressing out age and gender effects. Both apparent diameter and VF exhibited significant spatial heterogeneity, which cannot be explained solely by radio-frequency field inhomogeneities. Carbogen breathing significantly increased VF in BG and WM, and rCNR in thalamus, BG, and WM compared to air breathing. Our results are consistent with gradual dilation of PVSs with age in healthy adults. The PVSV morphology exhibited spatial heterogeneity and large inter-subject variations and changed during carbogen breathing compared to air breathing.

Exercise Degrades Bone in Caloric Restriction, Despite Suppression of Marrow Adipose Tissue (MAT).

Marrow adipose tissue (MAT) and its relevance to skeletal health during caloric restriction (CR) is unknown: It remains unclear whether exercise, which is anabolic to bone in a calorie-replete state, alters bone or MAT in CR. We hypothesized that response of bone and MAT to exercise in CR differs from the calorie-replete state. Ten-week-old female B6 mice fed a regular diet (RD) or 30% CR diet were allocated to sedentary (RD, CR, n = 10/group) or running exercise (RD-E, CR-E, n = 7/group). After 6 weeks, CR mice weighed 20% less than RD, p < 0.001; exercise did not affect weight. Femoral bone volume (BV) via 3D MRI was 20% lower in CR versus RD (p < 0.0001). CR was associated with decreased bone by µCT: Tb.Th was 16% less in CR versus RD, p < 0.003, Ct.Th was 5% less, p < 0.07. In CR-E, Tb.Th was 40% less than RD-E, p < 0.0001. Exercise increased Tb.Th in RD (+23% RD-E versus RD, p < 0.003) but failed to do so in CR. Cortical porosity increased after exercise in CR (+28%, p = 0.04), suggesting exercise during CR is deleterious to bone. In terms of bone fat, metaphyseal MAT/ BV rose 159% in CR versus RD, p = 0.003 via 3D MRI. Exercise decreased MAT/BV by 52% in RD, p < 0.05, and also suppressed MAT in CR (-121%, p = 0.047). Histomorphometric analysis of adipocyte area correlated with MAT by MRI (R2 = 0.6233, p < 0.0001). With respect to bone, TRAP and Sost mRNA were reduced in CR. Intriguingly, the repressed Sost in CR rose with exercise and may underlie the failure of CR-bone quantity to increase in response to exercise. Notably, CD36, a marker of fatty acid uptake, rose 4088% in CR (p < 0.01 versus RD), suggesting that basal increases in MAT during calorie restriction serve to supply local energy needs and are depleted during exercise with a negative impact on bone. © 2019 The Authors. Journal of Bone and Mineral Research published by American Society for Bone and Mineral Research.

Hippocampal Sulcus Remnant: Common Finding in Nonelderly Adults on Ultra-High-Resolution 7T Magnetic Resonance Imaging.

OBJECTIVE: The objective of this study was to investigate the frequency of hippocampal sulcus remnants (HSRs) in nonelderly adults using ultra-high-resolution 7T magnetic resonance (MR) images and their imaging features. METHODS: A total of 33 healthy adults underwent 7T MR, and multiplanar images of 66 temporal lobes were reviewed independently by 2 neuroradiologists. The detectability of the HSR was calculated. In addition, the interobserver agreement on the rating scale was evaluated using the κ statistic. RESULTS: Both observers identified HSRs with 7T MR images in all subjects. Excellent interobserver agreement was shown (κ = 1.0). The shape of HSRs was variable (spot-like, curvilinear, ovoid, or beaded appearance). Volumes of the HSRs were not correlated with age. CONCLUSIONS: Hippocampal sulcus remnants are commonly seen in healthy nonelderly adults using 7T MR imaging. Accurate diagnosis of HSR based on the microanatomy of hippocampus makes it easier to differentiate them from lesions, and it may help prevent unnecessary treatment.

Acceleration of High-Resolution 3D MR Fingerprinting via a Graph Convolutional Network.

Magnetic resonance fingerprinting (MRF) is a novel imaging framework for fast and simultaneous quantification of multiple tissue properties. Recently, 3D MRF methods have been developed, but the acquisition speed needs to be improved before they can be adopted for clinical use. The purpose of this study is to develop a novel deep learning approach to accelerate 3D MRF acquisition along the slice-encoding direction in k-space. We introduce a graph-based convolutional neural network that caters to non-Cartesian spiral trajectories commonly used for MRF acquisition. We improve tissue quantification accuracy compared with the state of the art. Our method enables fast 3D MRF with high spatial resolution, allowing whole-brain coverage within 5min, making MRF more feasible in clinical settings.

FRNET: FLATTENED RESIDUAL NETWORK FOR INFANT MRI SKULL STRIPPING.

Skull stripping for brain MR images is a basic segmentation task. Although many methods have been proposed, most of them focused mainly on the adult MR images. Skull stripping for infant MR images is more challenging due to the small size and dynamic intensity changes of brain tissues during the early ages. In this paper, we propose a novel CNN based framework to robustly extract brain region from infant MR image without any human assistance. Specifically, we propose a simplified but more robust flattened residual network architecture (FRnet). We also introduce a new boundary loss function to highlight ambiguous and low contrast regions between brain and non-brain regions. To make the whole framework more robust to MR images with different imaging quality, we further introduce an artifact simulator for data augmentation. We have trained and tested our proposed framework on a large dataset (N=343), covering newborns to 48-month-olds, and obtained performance better than the state-of-the-art methods in all age groups.

Enhancement of Perivascular Spaces Using Densely Connected Deep Convolutional Neural Network.

Perivascular spaces (PVS) in the human brain are related to various brain diseases. However, it is difficult to quantify them due to their thin and blurry appearance. In this paper, we introduce a deep-learning-based method, which can enhance a magnetic resonance (MR) image to better visualize the PVS. To accurately predict the enhanced image, we propose a very deep 3D convolutional neural network that contains densely connected networks with skip connections. The proposed networks can utilize rich contextual information derived from low-level to high-level features and effectively alleviate the gradient vanishing problem caused by the deep layers. The proposed method is evaluated on 17 7T MR images by a twofold cross-validation. The experiments show that our proposed network is much more effective to enhance the PVS than the previous PVS enhancement methods.

Quantitative phase contrast MRI of penetrating arteries in centrum semiovale at 7T.

Pathological changes of penetrating arteries (PA) within the centrum semiovale is an important contributing factor of cerebral small vessel disease (SVD). However, quantitative characterization of the PAs remains challenging due to their sub-voxel sizes. Here, we proposed a Model-based Analysis of Complex Difference images (MACD) of phase contrast MRI capable of measuring the mean velocities (vmean), diameters (D), and volume flow rates (VFR) of PAs without contamination from neighboring static tissues at 7 T. Simulation, phantom and in vivo studies were performed to evaluate the reproducibility and errors of the proposed method. For comparison, a Model-based Analysis of Phase difference images (MAP) was also carried out in the simulation. The proposed MACD analysis approach was applied in vivo to study the age dependence of PA properties in healthy subjects between 21 and 55 years old. Simulation showed that our proposed MACD approach yielded smaller errors than MAP, with errors increasing at lower velocities and diameters for both methods. In the phantom study, errors of the MACD-derived vmean, D, and VFR were ≤20% of their true values when vmean≥1cm/s and similar at different spatial resolutions. On the other hand, errors of the uncorrected apparent velocities were 24-60% and depended strongly on voxel size. The MACD errors linearly increased with the angle (α) between the vessel and slice normal direction at α ≤ 2° but remained almost constant at larger α. Results of the in vivo studies showed that the coefficients of repeatability for vmean, D, and VFR for PAs with α = 0° were 0.67 cm/s, 0.060 mm, and 0.067 mm3/s, respectively. No significant age dependence was found for the number, vmean, D, and VFR of PAs. The mean vmean, D, and VFR over all PAs with α = 0° were 1.79 ±â€¯0.62 cm/s, 0.17 ±â€¯0.05 mm, and 0.36 ±â€¯0.18 mm3/s, respectively. Quantitative measurements of PAs with the MACD method may serve as a useful tool for illuminating the vascular pathology in cerebral SVD.

Multi-channel multi-scale fully convolutional network for 3D perivascular spaces segmentation in 7T MR images.

Accurate segmentation of perivascular spaces (PVSs) is an important step for quantitative study of PVS morphology. However, since PVSs are the thin tubular structures with relatively low contrast and also the number of PVSs is often large, it is challenging and time-consuming for manual delineation of PVSs. Although several automatic/semi-automatic methods, especially the traditional learning-based approaches, have been proposed for segmentation of 3D PVSs, their performance often depends on the hand-crafted image features, as well as sophisticated preprocessing operations prior to segmentation (e.g., specially defined regions-of-interest (ROIs)). In this paper, a novel fully convolutional neural network (FCN) with no requirement of any specified hand-crafted features and ROIs is proposed for efficient segmentation of PVSs. Particularly, the original T2-weighted 7T magnetic resonance (MR) images are first filtered via a non-local Haar-transform-based line singularity representation method to enhance the thin tubular structures. Both the original and enhanced MR images are used as multi-channel inputs to complementarily provide detailed image information and enhanced tubular structural information for the localization of PVSs. Multi-scale features are then automatically learned to characterize the spatial associations between PVSs and adjacent brain tissues. Finally, the produced PVS probability maps are recursively loaded into the network as an additional channel of inputs to provide the auxiliary contextual information for further refining the segmentation results. The proposed multi-channel multi-scale FCN has been evaluated on the 7T brain MR images scanned from 20 subjects. The experimental results show its superior performance compared with several state-of-the-art methods.