High-resolution diffusion MRI can provide the ability to resolve small brain structures, enabling investigations of detailed white matter architecture. A major challenge for in vivo high-resolution diffusion MRI is the low signal-to-noise ratio. In this work, we combine two highly compatible methods, ultra-high field and three-dimensional multi-slab acquisition to improve the SNR of high-resolution diffusion MRI. As each kz plane is encoded using a single-shot echo planar readout, scan speeds of the proposed technique are similar to the commonly used two-dimensional diffusion MRI. In-plane parallel acceleration is applied to reduce image distortions. To reduce the sensitivity of auto-calibration signal data to subject motion and respiration, several new adaptions of the fast low angle excitation echo-planar technique (FLEET) that are suitable for 3D multi-slab echo planar imaging are proposed and evaluated. A modified reconstruction scheme is proposed for auto-calibration with the most robust method, Slice-FLEET acquisition, to make it compatible with navigator correction of motion induced phase errors. Slab boundary artefacts are corrected using the nonlinear slab profile encoding method recently proposed by our group. In vivo results demonstrate that using 7T and three-dimensional multi-slab acquisition with improved auto-calibration signal acquisition and nonlinear slab boundary artefacts correction, high-quality diffusion MRI data with ~1 mm isotropic resolution can be achieved.
•High-resolution (1 mm) in vivo diffusion MRI with high SNR.•3D multi-slab acquisition at 7T for high SNR efficiency.•Motion-robust parallel imaging kernel for 3D multi-slab acquisition.•Reducing slab boundary artefacts using nonlinear slab profile encoding.•High-quality fibre tractography with excellent anatomical fidelity.
Diffusion; High resolution; 7T; Tractography; 3D; Multi-slab
Permutation tests are increasingly being used as a reliable method for inference in neuroimaging analysis. However, they are computationally intensive. For small, non-imaging datasets, recomputing a model thousands of times is seldom a problem, but for large, complex models this can be prohibitively slow, even with the availability of inexpensive computing power. Here we exploit properties of statistics used with the general linear model (GLM) and their distributions to obtain accelerations irrespective of generic software or hardware improvements. We compare the following approaches: (i) performing a small number of permutations; (ii) estimating the p-value as a parameter of a negative binomial distribution; (iii) fitting a generalised Pareto distribution to the tail of the permutation distribution; (iv) computing p-values based on the expected moments of the permutation distribution, approximated from a gamma distribution; (v) direct fitting of a gamma distribution to the empirical permutation distribution; and (vi) permuting a reduced number of voxels, with completion of the remainder using low rank matrix theory. Using synthetic data we assessed the different methods in terms of their error rates, power, agreement with a reference result, and the risk of taking a different decision regarding the rejection of the null hypotheses (known as the resampling risk). We also conducted a re-analysis of a voxel-based morphometry study as a real-data example. All methods yielded exact error rates. Likewise, power was similar across methods. Resampling risk was higher for methods (i), (iii) and (v). For comparable resampling risks, the method in which no permutations are done (iv) was the absolute fastest. All methods produced visually similar maps for the real data, with stronger effects being detected in the family-wise error rate corrected maps by (iii) and (v), and generally similar to the results seen in the reference set. Overall, for uncorrected p-values, method (iv) was found the best as long as symmetric errors can be assumed. In all other settings, including for familywise error corrected p-values, we recommend the tail approximation (iii). The methods considered are freely available in the tool PALM — Permutation Analysis of Linear Models.
•Permutation methods can be accelerated through additional statistical approaches.•Six approaches are described and assessed.•Methods can be 100 times faster than in the non-accelerated case.•Recommendations are provided for various common scenarios.
Permutation tests; Negative binomial distribution; Tail approximation; Gamma distribution; Generalised Pareto distribution; Low rank matrix completion; Pearson type III distribution
Readout-segmented echo-planar imaging (rs-EPI) can provide high quality diffusion data because it is less prone to distortion and blurring artifacts than single-shot echo-planar imaging (ss-EPI), particularly at higher resolution and higher field. Readout segmentation allows shorter echo-spacing and echo train duration, resulting in reduced image distortion and blurring, respectively, in the phase-encoding direction. However, these benefits come at the expense of longer scan times because the segments are acquired in multiple repetitions times (TRs). This study shortened rs-EPI scan times by reducing the TR duration with simultaneous multislice acceleration.
The blipped-CAIPI method for slice acceleration with reduced g-factor SNR loss was incorporated into the diffusion-weighted rs-EPI sequence. The rs- and ss-EPI sequences were compared at a range of resolutions at both 3 and 7 Tesla in terms of image fidelity and diffusion postprocessing results.
Slice-accelerated clinically useful trace-weighted images and tractography results are presented. Tractography analysis showed that the reduced artifacts in rs-EPI allowed better discrimination of tracts than ss-EPI.
Slice acceleration reduces rs-EPI scan times providing a practical alternative to diffusion-weighted ss-EPI with reduced distortion and high resolution.
diffusion MRI; readout-segmented EPI; simultaneous multislice; blipped-CAIPI; distortion; tractography
Accurate segmentation of the subcortical structures is frequently required in neuroimaging studies. Most existing methods use only a T1-weighted MRI volume to segment all supported structures and usually rely on a database of training data. We propose a new method that can use multiple image modalities simultaneously and a single reference segmentation for initialisation, without the need for a manually labelled training set. The method models intensity profiles in multiple images around the boundaries of the structure after nonlinear registration. It is trained using a set of unlabelled training data, which may be the same images that are to be segmented, and it can automatically infer the location of the physical boundary using user-specified priors. We show that the method produces high-quality segmentations of the striatum, which is clearly visible on T1-weighted scans, and the globus pallidus, which has poor contrast on such scans. The method compares favourably to existing methods, showing greater overlap with manual segmentations and better consistency.
•We describe a new multimodal method for subcortical segmentation and apply it to the striatum and globus pallidus.•The method does not require multiple manual segmentations for training.•The method improves upon exiting methods that only use a T1‐weighted image (FIRST and FreeSurfer).•Multimodal segmentation is particularly advantageous for the globus pallidus, which has poor contrast on T1‐weighted scans.
Segmentation; Multimodal; Striatum; Globus pallidus; Brain; Huntington
The identification of resting state networks (RSNs) and the quantification of their functional connectivity in resting-state fMRI (rfMRI) are seriously hindered by the presence of artefacts, many of which overlap spatially or spectrally with RSNs. Moreover, recent developments in fMRI acquisition yield data with higher spatial and temporal resolutions, but may increase artefacts both spatially and/or temporally. Hence the correct identification and removal of non-neural fluctuations is crucial, especially in accelerated acquisitions. In this paper we investigate the effectiveness of three data-driven cleaning procedures, compare standard against higher (spatial and temporal) resolution accelerated fMRI acquisitions, and investigate the combined effect of different acquisitions and different cleanup approaches. We applied single-subject independent component analysis (ICA), followed by automatic component classification with FMRIB’s ICA-based X-noiseifier (FIX) to identify artefactual components. We then compared two first-level (within-subject) cleaning approaches for removing those artefacts and motion-related fluctuations from the data. The effectiveness of the cleaning procedures were assessed using timeseries (amplitude and spectra), network matrix and spatial map analyses. For timeseries and network analyses we also tested the effect of a second-level cleaning (informed by group-level analysis). Comparing these approaches, the preferable balance between noise removal and signal loss was achieved by regressing out of the data the full space of motion-related fluctuations and only the unique variance of the artefactual ICA components. Using similar analyses, we also investigated the effects of different cleaning approaches on data from different acquisition sequences. With the optimal cleaning procedures, functional connectivity results from accelerated data were statistically comparable or significantly better than the standard (unaccelerated) acquisition, and, crucially, with higher spatial and temporal resolution. Moreover, we were able to perform higher dimensionality ICA decompositions with the accelerated data, which is very valuable for detailed network analyses.
functional magnetic resonance imaging (fMRI); resting-state; artefact removal; functional connectivity; multiband acceleration
Many sources of fluctuation contribute to the fMRI signal, and this makes identifying the effects that are truly related to the underlying neuronal activity difficult. Independent component analysis (ICA) - one of the most widely used techniques for the exploratory analysis of fMRI data - has shown to be a powerful technique in identifying various sources of neuronally-related and artefactual fluctuation in fMRI data (both with the application of external stimuli and with the subject “at rest”). ICA decomposes fMRI data into patterns of activity (a set of spatial maps and their corresponding time series) that are statistically independent and add linearly to explain voxel-wise time series. Given the set of ICA components, if the components representing “signal” (brain activity) can be distinguished form the “noise” components (effects of motion, non-neuronal physiology, scanner artefacts and other nuisance sources), the latter can then be removed from the data, providing an effective cleanup of structured noise. Manual classification of components is labour intensive and requires expertise; hence, a fully automatic noise detection algorithm that can reliably detect various types of noise sources (in both task and resting fMRI) is desirable. In this paper, we introduce FIX (“FMRIB’s ICA-based X-noiseifier”), which provides an automatic solution for denoising fMRI data via accurate classification of ICA components. For each ICA component FIX generates a large number of distinct spatial and temporal features, each describing a different aspect of the data (e.g., what proportion of temporal fluctuations are at high frequencies). The set of features is then fed into a multi-level classifier (built around several different Classifiers). Once trained through the hand-classification of a sufficient number of training datasets, the classifier can then automatically classify new datasets. The noise components can then be subtracted from (or regressed out of) the original data, to provide automated cleanup. On conventional resting-state fMRI (rfMRI) single-run datasets, FIX achieved about 95% overall accuracy. On high-quality rfMRI data from the Human Connectome Project, FIX achieves over 99% classification accuracy, and as a result is being used in the default rfMRI processing pipeline for generating HCP connectomes. FIX is publicly available as a plugin for FSL.
Resting-state fMRI (R-fMRI) has shown considerable promise in providing potential biomarkers for diagnosis, prognosis and drug response across a range of diseases. Incorporating R-fMRI into multi-center studies is becoming increasingly popular, imposing technical challenges on data acquisition and analysis, as fMRI data is particularly sensitive to structured noise resulting from hardware, software, and environmental differences. Here, we investigated whether a novel clean up tool for structured noise was capable of reducing center-related R-fMRI differences between healthy subjects. We analyzed three Tesla R-fMRI data from 72 subjects, half of whom were scanned with eyes closed in a Philips Achieva system in The Netherlands, and half of whom were scanned with eyes open in a Siemens Trio system in the UK. After pre-statistical processing and individual Independent Component Analysis (ICA), FMRIB's ICA-based X-noiseifier (FIX) was used to remove noise components from the data. GICA and dual regression were run and non-parametric statistics were used to compare spatial maps between groups before and after applying FIX. Large significant differences were found in all resting-state networks between study sites before using FIX, most of which were reduced to non-significant after applying FIX. The between-center difference in the medial/primary visual network, presumably reflecting a between-center difference in protocol, remained statistically significant. FIX helps facilitate multi-center R-fMRI research by diminishing structured noise from R-fMRI data. In doing so, it improves combination of existing data from different centers in new settings and comparison of rare diseases and risk genes for which adequate sample size remains a challenge.
resting-state functional MRI; multi-center analysis; independent component analysis; dual regression; structured noise reduction
Resting-state functional magnetic resonance imaging (rfMRI) allows one to study functional connectivity in the brain by acquiring fMRI data while subjects lie inactive in the MRI scanner, and taking advantage of the fact that functionally related brain regions spontaneously co-activate. rfMRI is one of the two primary data modalities being acquired for the Human Connectome Project (the other being diffusion MRI). A key objective is to generate a detailed in vivo mapping of functional connectivity in a large cohort of healthy adults (over 1,000 subjects), and to make these datasets freely available for use by the neuroimaging community. In each subject we acquire a total of one hour of whole-brain rfMRI data at 3 Tesla, with a spatial resolution of 2×2×2mm and a temporal resolution of 0.7s, capitalizing on recent developments in slice-accelerated echo-planar imaging. We will also scan a subset of the cohort at higher field strength and resolution. In this paper we outline the work behind, and rationale for, decisions taken regarding the rfMRI data acquisition protocol and pre-processing pipelines, and present some initial results showing data quality and example functional connectivity analyses.
The ability to predict learning performance from brain imaging data has implications for selecting individuals for training or rehabilitation interventions. Here, we used structural MRI to test whether baseline variations in gray matter (GM) volume correlated with subsequent performance after a long-term training of a complex whole-body task. 44 naïve participants were scanned before undertaking daily juggling practice for 6 weeks, following either a high intensity or a low intensity training regime. To assess performance across the training period participants' practice sessions were filmed. Greater GM volume in medial occipito-parietal areas at baseline correlated with steeper learning slopes. We also tested whether practice time or performance outcomes modulated the degree of structural brain change detected between the baseline scan and additional scans performed immediately after training and following a further 4 weeks without training. Participants with better performance had higher increases in GM volume during the period following training (i.e., between scans 2 and 3) in dorsal parietal cortex and M1. When contrasting brain changes between the practice intensity groups, we did not find any straightforward effects of practice time though practice modulated the relationship between performance and GM volume change in dorsolateral prefrontal cortex. These results suggest that practice time and performance modulate the degree of structural brain change evoked by long-term training regimes.
•Inter-individual differences in brain structure correlate with subsequent performance outcome.•Performance outcome plays an important role in positive structural brain change.•Performance outcome and amount of practice modulate structural brain change.
Structural plasticity; Skill learning; MRI
Menke/Koerner et al. use structural MRI to explore the extent of longitudinal changes in cerebral pathology in amyotrophic lateral sclerosis, and their relationship to clinical features. A characteristic white matter tract pathological signature is seen cross-sectionally, while cortical involvement dominates longitudinally. This has implications for the development of biomarkers for diagnosis versus therapeutic monitoring.
Diagnosis, stratification and monitoring of disease progression in amyotrophic lateral sclerosis currently rely on clinical history and examination. The phenotypic heterogeneity of amyotrophic lateral sclerosis, including extramotor cognitive impairments is now well recognized. Candidate biomarkers have shown variable sensitivity and specificity, and studies have been mainly undertaken only cross-sectionally. Sixty patients with sporadic amyotrophic lateral sclerosis (without a family history of amyotrophic lateral sclerosis or dementia) underwent baseline multimodal magnetic resonance imaging at 3 T. Grey matter pathology was identified through analysis of T1-weighted images using voxel-based morphometry. White matter pathology was assessed using tract-based spatial statistics analysis of indices derived from diffusion tensor imaging. Cross-sectional analyses included group comparison with a group of healthy controls (n = 36) and correlations with clinical features, including regional disability, clinical upper motor neuron signs and cognitive impairment. Patients were offered 6-monthly follow-up MRI, and the last available scan was used for a separate longitudinal analysis (n = 27). In cross-sectional study, the core signature of white matter pathology was confirmed within the corticospinal tract and callosal body, and linked strongly to clinical upper motor neuron burden, but also to limb disability subscore and progression rate. Localized grey matter abnormalities were detected in a topographically appropriate region of the left motor cortex in relation to bulbar disability, and in Broca’s area and its homologue in relation to verbal fluency. Longitudinal analysis revealed progressive and widespread changes in the grey matter, notably including the basal ganglia. In contrast there was limited white matter pathology progression, in keeping with a previously unrecognized limited change in individual clinical upper motor neuron scores, despite advancing disability. Although a consistent core white matter pathology was found cross-sectionally, grey matter pathology was dominant longitudinally, and included progression in clinically silent areas such as the basal ganglia, believed to reflect their wider cortical connectivity. Such changes were significant across a range of apparently sporadic patients rather than being a genotype-specific effect. It is also suggested that the upper motor neuron lesion in amyotrophic lateral sclerosis may be relatively constant during the established symptomatic period. These findings have implications for the development of effective diagnostic versus therapeutic monitoring magnetic resonance imaging biomarkers. Amyotrophic lateral sclerosis may be characterized initially by a predominantly white matter tract pathological signature, evolving as a widespread cortical network degeneration.
motor neuron disease; biomarker; magnetic resonance imaging; voxel-based morphometry; diffusion tensor imaging
Diffusion imaging of post-mortem brains could provide valuable data for validation of diffusion tractography of white matter pathways. Long scans (e.g., overnight) may also enable high-resolution diffusion images for visualization of fine structures. However, alterations to post-mortem tissue (T2 and diffusion coefficient) present significant challenges to diffusion imaging with conventional diffusion-weighted spin echo (DW-SE) acquisitions, particularly for imaging human brains on clinical scanners. Diffusion-weighted steady-state free precession (DW-SSFP) has been proposed as an alternative acquisition technique to ameliorate this tradeoff in large-bore clinical scanners. In this study, both DWSE and DW-SSFP are optimized for use in fixed white matter on a clinical 3-Tesla scanner. Signal calculations predict superior performance from DW-SSFP across a broad range of protocols and conditions. DW-SE and DW-SSFP data in a whole, post-mortem human brain are compared for 6- and 12-hour scan durations. Tractography is performed in major projection, commissural and association tracts (corticospinal tract, corpus callosum, superior longitudinal fasciculus and cingulum bundle). The results demonstrate superior tract-tracing from DW-SSFP data, with 6-hour DW-SSFP data performing as well as or better than 12-hour DW-SE scans. These results suggest that DW-SSFP may be a preferred method for diffusion imaging of post-mortem human brains. The ability to estimate multiple fibers in imaging voxels is also demonstrated, again with greater success in DW-SSFP data.
► Comparison of DW-SE and DW-SSFP for post-mortem imaging on clinical scanners. ► Optimization of protocols predicts 50-130% higher SNR efficiency in DW-SSFP. ► Comparison of tractography 6- and 12-hour DW-SE and DW-SSFP scans. ► Lower uncertainty on fibre direction in DW-SSFP produces superior tractography. ► Crossing fibres can be estimated from 12-hour DW-SSFP data.
Diffusion; Tractography; Post mortem; Steady-state free precession; DTI
The brain regions functionally engaged in motor sequence performance are well-established, but the structural characteristics of these regions and the fiber pathways involved have been less well studied. In addition, relatively few studies have combined multiple magnetic resonance imaging (MRI) and behavioral performance measures in the same sample. Therefore, the current study used diffusion tensor imaging (DTI), probabilistic tractography, and voxel-based morphometry (VBM) to determine the structural correlates of skilled motor performance. Further, we compared these findings with fMRI results in the same sample. We correlated final performance and rate of improvement measures on a temporal motor sequence task (TMST) with skeletonized fractional anisotropy (FA) and whole brain gray matter (GM) volume. Final synchronization performance was negatively correlated with FA in white matter (WM) underlying bilateral sensorimotor cortex—an effect that was mediated by a positive correlation with radial diffusivity. Multi-fiber tractography indicated that this region contained crossing fibers from the corticospinal tract (CST) and superior longitudinal fasciculus (SLF). The identified SLF pathway linked parietal and auditory cortical regions that have been shown to be functionally engaged in this task. Thus, we hypothesize that enhanced synchronization performance on this task may be related to greater fiber integrity of the SLF. Rate of improvement on synchronization was positively correlated with GM volume in cerebellar lobules HVI and V—regions that showed training-related decreases in activity in the same sample. Taken together, our results link individual differences in brain structure and function to motor sequence performance on the same task. Further, our study illustrates the utility of using multiple MR measures and analysis techniques to specify the interpretation of structural findings.
superior longitudinal fasciculus; individual differences; motor sequence performance; fractional anisotropy; diffusion tensor imaging; gray matter volume
Diffusion imaging of post mortem brains has great potential both as a reference for brain specimens that undergo sectioning, and as a link between in vivo diffusion studies and “gold standard” histology/dissection. While there is a relatively mature literature on post mortem diffusion imaging of animals, human brains have proven more challenging due to their incompatibility with high-performance scanners. This study presents a method for post mortem diffusion imaging of whole, human brains using a clinical 3-Tesla scanner with a 3D segmented EPI spin-echo sequence. Results in eleven brains at 0.94 × 0.94 × 0.94 mm resolution are presented, and in a single brain at 0.73 × 0.73 × 0.73 mm resolution. Region-of-interest analysis of diffusion tensor parameters indicate that these properties are altered compared to in vivo (reduced diffusivity and anisotropy), with significant dependence on post mortem interval (time from death to fixation). Despite these alterations, diffusion tractography of several major tracts is successfully demonstrated at both resolutions. We also report novel findings of cortical anisotropy and partial volume effects.
► Acquisition and processing protocols for diffusion MRI of post-mortem human brains. ► Effect of post-mortem and scan intervals on diffusion indices. ► Tractography in post-mortem human brains. ► Radial diffusion anisotropy in cortical gray matter.
Diffusion tensor imaging; Tractography; Post mortem; Human; Brain