Author name cluster

Fernando Calamante

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

35 papers

1 author row

JBHI Journal 2025 Journal Article

Voxel-Level Brain States Prediction Using Swin Transformer

Yifei Sun
Daniel Chahine
Qinghao Wen
Tianming Liu
Xiang Li
Yixuan Yuan
Fernando Calamante
Jinglei Lv

Understanding brain dynamics is important for neuroscience and mental health. Functional magnetic resonance imaging (fMRI) enables the measurement of neural activities through blood-oxygen-level-dependent (BOLD) signals, which represent brain states. In this study, we aim to predict future human resting brain states with fMRI. Due to the 3D voxel-wise spatial organization and temporal dependencies of the fMRI data, we propose a novel architecture which employs a 4D Shifted Window (Swin) Transformer as encoder to efficiently learn spatio-temporal information and a convolutional decoder to enable brain state prediction at the same spatial and temporal resolution as the input fMRI data. We used 100 unrelated subjects from the Human Connectome Project (HCP) for model training and testing. Our novel model has shown high accuracy when predicting 7. 2s resting-state brain activities based on the prior 23. 04s fMRI time series. The predicted brain states highly resemble BOLD contrast and dynamics. This work shows promising evidence that the spatiotemporal organization of the human brain can be learned by a Swin Transformer model, at high resolution, which provides a potential for reducing the fMRI scan time and the development of brain-computer interfaces in the future.

AIIM Journal 2024 Journal Article

Improving multiple sclerosis lesion segmentation across clinical sites: A federated learning approach with noise-resilient training

Lei Bai
Dongang Wang
Hengrui Wang
Michael Barnett
Mariano Cabezas
Weidong Cai
Fernando Calamante
Kain Kyle

Accurately measuring the evolution of Multiple Sclerosis (MS) with magnetic resonance imaging (MRI) critically informs understanding of disease progression and helps to direct therapeutic strategy. Deep learning models have shown promise for automatically segmenting MS lesions, but the scarcity of accurately annotated data hinders progress in this area. Obtaining sufficient data from a single clinical site is challenging and does not address the heterogeneous need for model robustness. Conversely, the collection of data from multiple sites introduces data privacy concerns and potential label noise due to varying annotation standards. To address this dilemma, we explore the use of the federated learning framework while considering label noise. Our approach enables collaboration among multiple clinical sites without compromising data privacy under a federated learning paradigm that incorporates a noise-robust training strategy based on label correction. Specifically, we introduce a Decoupled Hard Label Correction (DHLC) strategy that considers the imbalanced distribution and fuzzy boundaries of MS lesions, enabling the correction of false annotations based on prediction confidence. We also introduce a Centrally Enhanced Label Correction (CELC) strategy, which leverages the aggregated central model as a correction teacher for all sites, enhancing the reliability of the correction process. Extensive experiments conducted on two multi-site datasets demonstrate the effectiveness and robustness of our proposed methods, indicating their potential for clinical applications in multi-site collaborations to train better deep learning models with lower cost in data collection and annotation.

YNIMG Journal 2022 Journal Article

Network communication models narrow the gap between the modular organization of structural and functional brain networks

Caio Seguin
Sina Mansour L
Olaf Sporns
Andrew Zalesky
Fernando Calamante

Structural and functional brain networks are modular. Canonical functional systems, such as the default mode network, are well-known modules of the human brain and have been implicated in a large number of cognitive, behavioral and clinical processes. However, modules delineated in structural brain networks inferred from tractography generally do not recapitulate canonical functional systems. Neuroimaging evidence suggests that functional connectivity between regions in the same systems is not always underpinned by anatomical connections. As such, direct structural connectivity alone would be insufficient to characterize the functional modular organization of the brain. Here, we demonstrate that augmenting structural brain networks with models of indirect (polysynaptic) communication unveils a modular network architecture that more closely resembles the brain's established functional systems. We find that diffusion models of polysynaptic connectivity, particularly communicability, narrow the gap between the modular organization of structural and functional brain networks by 20-60%, whereas routing models based on single efficient paths do not improve mesoscopic structure-function correspondence. This suggests that functional modules emerge from the constraints imposed by local network structure that facilitates diffusive neural communication. Our work establishes the importance of modeling polysynaptic communication to understand the structural basis of functional systems.

YNIMG Journal 2021 Journal Article

Characterisation of white matter asymmetries in the healthy human brain using diffusion MRI fixel-based analysis

Arush Honnedevasthana Arun
Alan Connelly
Robert E. Smith
Fernando Calamante

The diffusion tensor model for diffusion MRI has been used extensively to study asymmetry in the human brain white matter. However, given the limitations of the tensor model, the nature of any underlying asymmetries remains uncertain, particularly in crossing fibre regions. Here, we provide a more robust characterisation of human brain white matter asymmetries based on fibre-specific diffusion MRI metrics and a whole-brain data-driven approach. We used high-quality diffusion MRI data (n = 100) from the Human Connectome Project, the spherical deconvolution model for fibre orientation distribution estimation, and the Fixel-Based Analysis framework to utilise crossing fibre information in registration, data smoothing and statistical inference. We found many significant asymmetries, widespread throughout the brain white matter, with both left>right and right>left dominances observed in different pathways. No influences of sex, age, or handedness on asymmetry were found. We also report on the relative contributions of microstructural and morphological white matter properties toward the asymmetry findings. Our findings should provide important information to future studies focussing on how these asymmetries are affected by disease, development/ageing, or how they correlate to functional/cognitive measures.

YNIMG Journal 2021 Journal Article

TractLearn: A geodesic learning framework for quantitative analysis of brain bundles

Arnaud Attyé
Félix Renard
Monica Baciu
Elise Roger
Laurent Lamalle
Patrick Dehail
Hélène Cassoudesalle
Fernando Calamante

Deep learning-based convolutional neural networks have recently proved their efficiency in providing fast segmentation of major brain fascicles structures, based on diffusion-weighted imaging. The quantitative analysis of brain fascicles then relies on metrics either coming from the tractography process itself or from each voxel along the bundle. Statistical detection of abnormal voxels in the context of disease usually relies on univariate and multivariate statistics models, such as the General Linear Model (GLM). Yet in the case of high-dimensional low sample size data, the GLM often implies high standard deviation range in controls due to anatomical variability, despite the commonly used smoothing process. This can lead to difficulties to detect subtle quantitative alterations from a brain bundle at the voxel scale. Here we introduce TractLearn, a unified framework for brain fascicles quantitative analyses by using geodesic learning as a data-driven learning task. TractLearn allows a mapping between the image high-dimensional domain and the reduced latent space of brain fascicles using a Riemannian approach. We illustrate the robustness of this method on a healthy population with test-retest acquisition of multi-shell diffusion MRI data, demonstrating that it is possible to separately study the global effect due to different MRI sessions from the effect of local bundle alterations. We have then tested the efficiency of our algorithm on a sample of 5 age-matched subjects referred with mild traumatic brain injury. Our contributions are to propose: 1/ A manifold approach to capture controls variability as standard reference instead of an atlas approach based on a Euclidean mean. 2/ A tool to detect global variation of voxels' quantitative values, which accounts for voxels' interactions in a structure rather than analyzing voxels independently. 3/ A ready-to-plug algorithm to highlight nonlinear variation of diffusion MRI metrics. With this regard, TractLearn is a ready-to-use algorithm for precision medicine.

YNIMG Journal 2021 Journal Article

Tractography dissection variability: What happens when 42 groups dissect 14 white matter bundles on the same dataset?

Kurt G. Schilling
François Rheault
Laurent Petit
Colin B. Hansen
Vishwesh Nath
Fang-Cheng Yeh
Gabriel Girard
Muhamed Barakovic

White matter bundle segmentation using diffusion MRI fiber tractography has become the method of choice to identify white matter fiber pathways in vivo in human brains. However, like other analyses of complex data, there is considerable variability in segmentation protocols and techniques. This can result in different reconstructions of the same intended white matter pathways, which directly affects tractography results, quantification, and interpretation. In this study, we aim to evaluate and quantify the variability that arises from different protocols for bundle segmentation. Through an open call to users of fiber tractography, including anatomists, clinicians, and algorithm developers, 42 independent teams were given processed sets of human whole-brain streamlines and asked to segment 14 white matter fascicles on six subjects. In total, we received 57 different bundle segmentation protocols, which enabled detailed volume-based and streamline-based analyses of agreement and disagreement among protocols for each fiber pathway. Results show that even when given the exact same sets of underlying streamlines, the variability across protocols for bundle segmentation is greater than all other sources of variability in the virtual dissection process, including variability within protocols and variability across subjects. In order to foster the use of tractography bundle dissection in routine clinical settings, and as a fundamental analytical tool, future endeavors must aim to resolve and reduce this heterogeneity. Although external validation is needed to verify the anatomical accuracy of bundle dissections, reducing heterogeneity is a step towards reproducible research and may be achieved through the use of standard nomenclature and definitions of white matter bundles and well-chosen constraints and decisions in the dissection process.

YNIMG Journal 2019 Journal Article

Connectomes from streamlines tractography: Assigning streamlines to brain parcellations is not trivial but highly consequential

Chun-Hung Yeh
Robert E. Smith
Thijs Dhollander
Fernando Calamante
Alan Connelly

When using diffusion MRI streamlines tractograms to construct structural connectomes, ideally, each streamline should connect exactly 2 regions-of-interest (i. e. network nodes) as defined by a given brain parcellation scheme. However, the ill-posed nature of termination criteria in many tractography algorithms can cause streamlines apparently being associated with zero, one, or more than two grey matter (GM) nodes; streamlines that terminate in white matter or cerebrospinal fluid may even end up being assigned to nodes if the definitions of these nodes are not strictly constrained to genuine GM areas, resulting in a misleading connectome in non-trivial ways. Based on both in-house MRI data and state-of-the-art data provided by the Human Connectome Project, this study investigates the actual influence of streamline-to-node assignment methods, and their interactions with fibre-tracking terminations and brain parcellations, on the construction of pairwise regional connectivity and subsequent connectomic measures. Our results show that the frequency of generating successful pairwise connectivity is heavily affected by the convoluted interactions between the applied strategies for connectome construction, and that minor changes in the mechanism can cause significant variations in the within- and between-module connectivity strengths as well as in the commonly-used graph theory metrics. Our data suggest that these fundamental processes should not be overlooked in structural connectomics research, and that improved data quality is not in itself sufficient to solve the underlying problems associated with assigning streamlines to brain nodes. We demonstrate that the application of advanced fibre-tracking techniques that are designed to correct for inaccuracies of track terminations with respect to anatomical information at the fibre-tracking stage is advantageous to the subsequent connectome construction process, in which pairs of parcellation nodes can be more robustly identified from streamline terminations via a suitable assignment mechanism.

YNIMG Journal 2019 Journal Article

Is removal of weak connections necessary for graph-theoretical analysis of dense weighted structural connectomes from diffusion MRI?

Oren Civier
Robert Elton Smith
Chun-Hung Yeh
Alan Connelly
Fernando Calamante

Recent advances in diffusion MRI tractography permit the generation of dense weighted structural connectomes that offer greater insight into brain organization. However, these efforts are hampered by the lack of consensus on how to extract topological measures from the resulting graphs. Here we evaluate the common practice of removing the graphs’ weak connections, which is primarily intended to eliminate spurious connections and emphasize strong connections. Because this processing step requires arbitrary or heuristic-based choices (e. g. , setting a threshold level below which connections are removed), and such choices might complicate statistical analysis and inter-study comparisons, in this work we test whether removing weak connections is indeed necessary. To this end, we systematically evaluated the effect of removing weak connections on a range of popular graph-theoretical metrics. Specifically, we investigated if (and at what extent) removal of weak connections introduces a statistically significant difference between two otherwise equal groups of healthy subjects when only applied to one of the groups. Using data from the Human Connectome Project, we found that removal of weak connections had no statistical effect even when removing the weakest ∼70–90% connections. Removing yet a larger extent of weak connections, thus reducing connectivity density even further, did produce a predictably significant effect. However, metric values became sensitive to the exact connectivity density, which has ramifications regarding the stability of the statistical analysis. This pattern persisted whether connections were removed by connection strength threshold or connectivity density, and for connectomes generated using parcellations at different resolutions. Finally, we showed that the same pattern also applies for data from a clinical-grade MRI scanner. In conclusion, our analysis revealed that removing weak connections is not necessary for graph-theoretical analysis of dense weighted connectomes. Because removal of weak connections provides no practical utility to offset the undesirable requirement for arbitrary or heuristic-based choices, we recommend that this step is avoided in future studies.

YNIMG Journal 2016 Journal Article

Correction for diffusion MRI fibre tracking biases: The consequences for structural connectomic metrics

Chun-Hung Yeh
Robert E. Smith
Xiaoyun Liang
Fernando Calamante
Alan Connelly

Diffusion MRI streamlines tractography has become a major technique for inferring structural networks through reconstruction of brain connectome. However, quantification of structural connectivity based on the number of streamlines interconnecting brain grey matter regions is known to be problematic in a number of aspects, such as the ill-posed nature of streamlines terminations and the non-quantitative nature of streamline counts. This study investigates the effects of state-of-the-art connectome construction methods on the subsequent analyses of structural brain networks using graph theoretical approaches. Our results demonstrate that the characteristics of structural connectivity, including connectome variability, global network metrics, small-world attributes and network hubs, alter significantly following the improvement in biological accuracy of streamlines tractograms provided by anatomically-constrained tractography (ACT) and spherical-deconvolution informed filtering of tractograms (SIFT). Importantly, the commonly-used correction for connection density based on scaling the contribution of each streamline to the connectome by its inverse length is shown to provide incomplete correction, highlighting the necessity for the use of advanced tractogram reconstruction techniques in structural connectomics research.

YNIMG Journal 2016 Journal Article

The contribution of geometry to the human connectome

James A. Roberts
Alistair Perry
Anton R. Lord
Gloria Roberts
Philip B. Mitchell
Robert E. Smith
Fernando Calamante
Michael Breakspear

The human connectome is a topologically complex, spatially embedded network. While its topological properties have been richly characterized, the constraints imposed by its spatial embedding are poorly understood. By applying a novel resampling method to tractography data, we show that the brain's spatial embedding makes a major, but not definitive, contribution to the topology of the human connectome. We first identify where the brain's structural hubs would likely be located if geometry was the sole determinant of brain topology. Empirical networks show a widespread shift away from this geometric center toward more peripheral interconnected skeletons in each hemisphere, with discrete clusters around the anterior insula, and the anterior and posterior midline regions of the cortex. A relatively small number of strong inter-hemispheric connections assimilate these intra-hemispheric structures into a rich club, whose connections are locally more clustered but globally longer than predicted by geometry. We also quantify the extent to which the segregation, integration, and modularity of the human brain are passively inherited from its geometry. These analyses reveal novel insights into the influence of spatial geometry on the human connectome, highlighting specific topological features that likely confer functional advantages but carry an additional metabolic cost.

YNIMG Journal 2015 Journal Article

Fourier Tract Sampling (FouTS): A framework for improved inference of white matter tracts from diffusion MRI by explicitly modelling tract volume

Thomas G. Close
Jacques-Donald Tournier
Leigh A. Johnston
Fernando Calamante
Iven Mareels
Alan Connelly

Diffusion MRI tractography algorithm development is increasingly moving towards global techniques to incorporate “downstream” information and conditional probabilities between neighbouring tracts. Such approaches also enable white matter to be represented more tangibly than the abstract lines generated by the most common approaches to fibre tracking. However, previously proposed algorithms still use fibre-like models of white matter corresponding to thin strands of white matter tracts rather than the tracts themselves, and therefore require many components for accurate representations, which leads to poorly constrained inverse problems. We propose a novel tract-based model of white matter, the ‘Fourier tract’, which is able to represent rich tract shapes with a relatively low number of parameters, and explicitly decouples the spatial extent of the modelled tract from its ‘Apparent Connection Strength (ACS)’. The Fourier tract model is placed within a novel Bayesian framework, which relates the tract parameters directly to the observed signal, enabling a wide range of acquisition schemes to be used. The posterior distribution of the Bayesian framework is characterised via Markov-chain Monte-Carlo sampling to infer probable values of the ACS and spatial extent of the imaged white matter tracts, providing measures that can be directly applied to many research and clinical studies. The robustness of the proposed tractography algorithm is demonstrated on simulated basic tract configurations, such as curving, twisting, crossing and kissing tracts, and sections of more complex numerical phantoms. As an illustration of the approach in vivo, fibre tracking is performed on a central section of the brain in three subjects from 60 direction HARDI datasets.

YNIMG Journal 2015 Journal Article

Quantification of voxel-wise total fibre density: Investigating the problems associated with track-count mapping

Fernando Calamante
Robert E. Smith
Jacques-Donald Tournier
David Raffelt
Alan Connelly

A biological parameter that would be valuable to be able to extract from diffusion MRI data is the local white matter axonal density. Track-density imaging (TDI) has been used as if it could provide such a measure; however, this has been the subject of controversy, primarily due to the fact that track-count quantitation is highly sensitive to tracking biases and errors. The spherical-deconvolution informed filtering of tractograms (SIFT) post-processing method was recently introduced to minimise tractography biases, and thus provides a more biologically meaningful measure that could be used in track-count mapping (i. e. TDI following SIFT). The TDI intensity following SIFT ideally corresponds to the orientational average of the fibre orientation distribution (FOD), which corresponds to the total Apparent Fibre Density (AFDtotal) within the AFD framework; in fact, AFDtotal provides a direct measure of local fibre density at native resolution that does not rely on fibre-tracking. In this study, we demonstrate problems associated with quantitative TDI investigations, which can be avoided by using SIFT processing or directly by using AFDtotal maps. We also characterise the intra- and inter-subject reproducibility of TDI maps (with and without SIFT pre-processing) and AFDtotal maps. It is shown that SIFT improves the quantitative characteristics of TDI, but is still vastly inferior to the properties of the AFDtotal parameter itself, because the latter does not require tracking. While standard TDI might be preferable in applications when high anatomical contrast is required, particularly when combined with super-resolution, for voxel-wise quantitation of total tract density (i. e. without tract orientation information) at native resolution, the total AFD maps are preferable to TDI or other related track-count maps. Regardless of the track-count measure, it should be noted that all of these voxel-averaged approaches discard important information that is retained in fibre-specific approaches such as AFD.

YNIMG Journal 2015 Journal Article

Reproducibility of multiphase pseudo-continuous arterial spin labeling and the effect of post-processing analysis methods

Amir Fazlollahi
Pierrick Bourgeat
Xiaoyun Liang
Fabrice Meriaudeau
Alan Connelly
Olivier Salvado
Fernando Calamante

Arterial spin labeling (ASL) is an emerging MRI technique for non-invasive measurement of cerebral blood flow (CBF). Compared to invasive perfusion imaging modalities, ASL suffers from low sensitivity due to poor signal-to-noise ratio (SNR), susceptibility to motion artifacts and low spatial resolution, all of which limit its reliability. In this work, the effects of various state of the art image processing techniques for addressing these ASL limitations are investigated. A processing pipeline consisting of motion correction, ASL motion correction imprecision removal, temporal and spatial filtering, partial volume effect correction, and CBF quantification was developed and assessed. To further improve the SNR for pseudo-continuous ASL (PCASL) by accounting for errors in tagging efficiency, the data from multiphase (MP) acquisitions were analyzed using a novel weighted-averaging scheme. The performances of each step in terms of SNR and reproducibility were evaluated using test–retest ASL data acquired from 12 young healthy subjects. The proposed processing pipeline was shown to improve the within-subject coefficient of variation and regional reproducibility by 17% and 16%, respectively, compared to CBF maps computed following motion correction but without the other processing steps. The CBF measurements of MP-PCASL compared to PCASL had on average 23% and 10% higher SNR and reproducibility, respectively.

YNIMG Journal 2015 Journal Article

SIFT2: Enabling dense quantitative assessment of brain white matter connectivity using streamlines tractography

Robert E. Smith
Jacques-Donald Tournier
Fernando Calamante
Alan Connelly

Diffusion MRI streamlines tractography allows for the investigation of the brain white matter pathways non-invasively. However a fundamental limitation of this technology is its non-quantitative nature, i. e. the density of reconstructed connections is not reflective of the density of underlying white matter fibres. As a solution to this problem, we have previously published the “spherical-deconvolution informed filtering of tractograms (SIFT)” method, which determines a subset of the streamlines reconstruction such that the streamlines densities throughout the white matter are as close as possible to fibre densities estimated using the spherical deconvolution diffusion model; this permits the use of streamline count as a valid biological marker of connection density. Particular aspects of its performance may have however limited its uptake in the diffusion MRI research community. Here we present an alternative to this method, entitled SIFT2, which provides a more logically direct and computationally efficient solution to the streamlines connectivity quantification problem: by determining an appropriate cross-sectional area multiplier for each streamline rather than removing streamlines altogether, biologically accurate measures of fibre connectivity are obtained whilst making use of the complete streamlines reconstruction.

YNIMG Journal 2015 Journal Article

The effects of SIFT on the reproducibility and biological accuracy of the structural connectome

Robert E. Smith
Jacques-Donald Tournier
Fernando Calamante
Alan Connelly

Diffusion MRI streamlines tractography is increasingly being used to characterise and assess the structural connectome of the human brain. However, issues pertaining to quantification of structural connectivity using streamlines reconstructions are well-established in the field, and therefore the validity of any conclusions that may be drawn from these analyses remains ambiguous. We recently proposed a post-processing method entitled “SIFT: Spherical-deconvolution Informed Filtering of Tractograms” as a mechanism for reducing the biases in quantitative measures of connectivity introduced by the streamlines reconstruction method. Here, we demonstrate the advantage of this approach in the context of connectomics in three steps. Firstly, we carefully consider the model imposed by the SIFT method, and the implications this has for connectivity quantification. Secondly, we investigate the effects of SIFT on the reproducibility of structural connectome construction. Thirdly, we compare quantitative measures extracted from structural connectomes derived from streamlines tractography, with and without the application of SIFT, to published estimates drawn from post-mortem brain dissection. The combination of these sources of evidence demonstrates the important role the SIFT methodology has for the robust quantification of structural connectivity of the brain using diffusion MRI.

YNIMG Journal 2014 Journal Article

Graph analysis of resting-state ASL perfusion MRI data: Nonlinear correlations among CBF and network metrics

Xiaoyun Liang
Alan Connelly
Fernando Calamante

Human connectome mapping is important to understand both normal brain function and disease-related dysfunction. Although blood-oxygen-level-dependent (BOLD) fMRI has been the most commonly used method for human connectome mapping, arterial spin labeling (ASL) is an fMRI technique to measure cerebral blood flow (CBF) directly and noninvasively, and thus provides a more direct quantitative correlate of neural activity. In this study, investigations on properties of CBF networks using ASL perfusion data have been conducted on 10 healthy subjects. As with BOLD fMRI studies, the extracted networks exhibited small-world network properties. In addition, highly connected brain regions are shown to overlap mostly with hub regions detected from BOLD fMRI studies. Taken together, this demonstrates the capability of ASL fMRI for mapping the brain connectome. Furthermore, a sigmoid model was then employed to fit the extracted network metrics vs. CBF measurements. Interestingly, the relationships between 4 specific network metrics and region-wise CBF demonstrate that consistently nonlinear patterns exist across all subjects. In contrast to the positive nonlinear pattern of other network metrics (degree, vulnerability, and eigenvector centrality), the characteristic path length shows a negative nonlinear pattern, reflecting the mechanism underlying the small-world properties. To our knowledge, this is the first study to unravel the intrinsic relationships between specific network metrics and CBF estimates. This should have diagnostic and therapeutic implications for those studies focusing on patients who suffer from abnormal functional connectivity.

YNIMG Journal 2014 Journal Article

Mapping somatosensory connectivity in adult mice using diffusion MRI tractography and super-resolution track density imaging

Kay Richards
Fernando Calamante
Jacques-Donald Tournier
Nyoman D. Kurniawan
Farnoosh Sadeghian
Alexander R. Retchford
Gabriel Davis Jones
Christopher A. Reid

In this study we combined ultra-high field diffusion MRI fiber tracking and super-resolution track density imaging (TDI) to map the relay locations and connectivity of the somatosensory pathway in paraformaldehyde fixed, C57Bl/6J mouse brains. Super-resolution TDI was used to achieve 20μm isotropic resolution to inform the 3D topography of the relay locations including thalamic barreloids and brainstem barrelettes, not described previously using MRI methodology. TDI-guided mapping results for thalamo-cortical connectivity were consistent with thalamo-cortical projections labeled using virus mediated fluorescent protein expression. Trigemino-thalamic TDI connectivity maps were concordant with results obtained using anterograde dye tracing from brainstem to thalamus. Importantly, TDI mapping overcame the constraint of tissue distortion observed in mechanically sectioned tissue, enabling 3D reconstruction and long-range connectivity data. In conclusion, our results showed that diffusion micro-imaging at ultra-high field MRI revealed the stereotypical pattern of somatosensory connectivity and is a valuable tool to complement histologic methods, achieving 3D spatial preservation of whole brain networks for characterization in mouse models of human disease.

YNIMG Journal 2014 Journal Article

Quantification of track-weighted imaging (TWI): Characterisation of within-subject reproducibility and between-subject variability

Lisa Willats
David Raffelt
Robert E. Smith
J.-Donald Tournier
Alan Connelly
Fernando Calamante

Recently several novel image contrasts derived from whole-brain fibre tracking-data (tractograms) have been introduced. The novel contrasts of these track-weighted imaging (TWI) methods may provide important information for clinical neuroscience studies. However, before they can be used reliably to generate quantitative measures, it is important to characterise their within-subject reproducibility, and between-subject variability. In this work we compute the within-subject reproducibility (intra-scan, intra-session and inter-session), and between-subject variability of TWI for a number of different TWI contrasts across multiple subjects. The results are used in simple voxel-wise power calculations within illustrative regions of interest to provide guidelines for required sample sizes and observable effect sizes for individual subjects and between groups. It was found that the required sample sizes and observable effect sizes varied considerably between different TWI maps and for different ROIs. For some TWI contrast and ROI combinations, the power calculations yielded clinically practical values. These results provide important information concerning the potential usefulness and sensitivity of TWI maps for individual diagnosis, longitudinal studies and group comparisons, as well as for study designs.

YNIMG Journal 2014 Journal Article

Visualization of mouse barrel cortex using ex-vivo track density imaging

Nyoman D. Kurniawan
Kay L. Richards
Zhengyi Yang
David She
Jeremy F.P. Ullmann
Randal X. Moldrich
Sha Liu
Javier Urriola Yaksic

We describe the visualization of the barrel cortex of the primary somatosensory area (S1) of ex vivo adult mouse brain with short-tracks track density imaging (stTDI). stTDI produced much higher definition of barrel structures than conventional fractional anisotropy (FA), directionally-encoded color FA maps, spin-echo T 1- and T 2-weighted imaging and gradient echo T 1/T 2*-weighted imaging. 3D high angular resolution diffusion imaging (HARDI) data were acquired at 48micron isotropic resolution for a (3mm)3 block of cortex containing the barrel field and reconstructed using stTDI at 10micron isotropic resolution. HARDI data were also acquired at 100micron isotropic resolution to image the whole brain and reconstructed using stTDI at 20micron isotropic resolution. The 10micron resolution stTDI maps showed exceptionally clear delineation of barrel structures. Individual barrels could also be distinguished in the 20micron stTDI maps but the septa separating the individual barrels appeared thicker compared to the 10micron maps, indicating that the ability of stTDI to produce high quality structural delineation is dependent upon acquisition resolution. Close homology was observed between the barrel structure delineated using stTDI and reconstructed histological data from the same samples. stTDI also detects barrel deletions in the posterior medial barrel sub-field in mice with infraorbital nerve cuts. The results demonstrate that stTDI is a novel imaging technique that enables three-dimensional characterization of complex structures such as the barrels in S1 and provides an important complementary non-invasive imaging tool for studying synaptic connectivity, development and plasticity of the sensory system.

YNIMG Journal 2013 Journal Article

SIFT: Spherical-deconvolution informed filtering of tractograms

Robert E. Smith
Jacques-Donald Tournier
Fernando Calamante
Alan Connelly

Diffusion MRI allows the structural connectivity of the whole brain (the ‘tractogram’) to be estimated in vivo non-invasively using streamline tractography. The biological accuracy of these data sets is however limited by the inherent biases associated with the reconstruction method. Here we propose a method to retrospectively improve the accuracy of these reconstructions, by selectively filtering out streamlines from the tractogram in a manner that improves the fit between the streamline reconstruction and the underlying diffusion images. This filtering is guided by the results of spherical deconvolution of the diffusion signal, hence the acronym SIFT: spherical-deconvolution informed filtering of tractograms. Data sets processed by this algorithm show a marked reduction in known reconstruction biases, and improved biological plausibility. Emerging methods in diffusion MRI, particularly those that aim to characterise and compare the structural connectivity of the brain, should benefit from the improved accuracy of the reconstruction.

YNIMG Journal 2013 Journal Article

Track-weighted functional connectivity (TW-FC): A tool for characterizing the structural–functional connections in the brain

Fernando Calamante
Richard A.J. Masterton
Jacques-Donald Tournier
Robert E. Smith
Lisa Willats
David Raffelt
Alan Connelly

MRI provides a powerful tool for studying the functional and structural connections in the brain non-invasively. The technique of functional connectivity (FC) exploits the intrinsic temporal correlations of slow spontaneous signal fluctuations to characterise brain functional networks. In addition, diffusion MRI fibre-tracking can be used to study the white matter structural connections. In recent years, there has been considerable interest in combining these two techniques to provide an overall structural–functional description of the brain. In this work we applied the recently proposed super-resolution track-weighted imaging (TWI) methodology to demonstrate how whole-brain fibre-tracking data can be combined with FC data to generate a track-weighted (TW) FC map of FC networks. The method was applied to data from 8 healthy volunteers, and illustrated with (i) FC networks obtained using a seeded connectivity-based analysis (seeding in the precuneus/posterior cingulate cortex, PCC, known to be part of the default mode network), and (ii) with FC networks generated using independent component analysis (in particular, the default mode, attention, visual, and sensory-motor networks). TW-FC maps showed high intensity in white matter structures connecting the nodes of the FC networks. For example, the cingulum bundles show the strongest TW-FC values in the PCC seeded-based analysis, due to their major role in the connection between medial frontal cortex and precuneus/posterior cingulate cortex; similarly the superior longitudinal fasciculus was well represented in the attention network, the optic radiations in the visual network, and the corticospinal tract and corpus callosum in the sensory-motor network. The TW-FC maps highlight the white matter connections associated with a given FC network, and their intensity in a given voxel reflects the functional connectivity of the part of the nodes of the network linked by the structural connections traversing that voxel. They therefore contain a different (and novel) image contrast from that of the images used to generate them. The results shown in this study illustrate the potential of the TW-FC approach for the fusion of structural and functional data into a single quantitative image. This technique could therefore have important applications in neuroscience and neurology, such as for voxel-based comparison studies.

YNIMG Journal 2012 Journal Article

A generalised framework for super-resolution track-weighted imaging

Fernando Calamante
Jacques-Donald Tournier
Robert E. Smith
Alan Connelly

Track-density imaging (TDI) was recently introduced as a method to achieve super-resolution imaging using whole-brain fibre-tracking data (the so called tractogram). A similar approach to achieve super-resolution was later applied for average pathlength mapping (APM). These two methods have in common that the tractogram information is used to create an image with novel contrast and super-resolution properties. In this study, we present a generalised framework for creating super-resolution track-weighted imaging (TWI), where the intensity of the map can be made dependent on any specific property of the streamlines or their set of spatial coordinates. Furthermore, each contrast can be determined by a number of characteristics that are under user control. It is shown that TDI and APM represent specific cases of this generalised framework, and that this framework opens up the possibility of generating a large range of images with novel image contrasts. Finally, it is shown that the same super-resolution principles as those introduced in the original TDI method are also applicable to any of these new images.

YNIMG Journal 2012 Journal Article

Anatomically-constrained tractography: Improved diffusion MRI streamlines tractography through effective use of anatomical information

Robert E. Smith
Jacques-Donald Tournier
Fernando Calamante
Alan Connelly

Diffusion MRI streamlines tractography suffers from a number of inherent limitations, one of which is the accurate determination of when streamlines should be terminated. Use of an accurate streamlines propagation mask from segmentation of an anatomical image confines the streamlines to the volume of the brain white matter, but does not take full advantage of all of the information available from such an image. We present a modular addition to streamlines tractography, which makes more effective use of the information available from anatomical image segmentation, and the known properties of the neuronal axons being reconstructed, to apply biologically realistic priors to the streamlines generated; we refer to this as “Anatomically-Constrained Tractography”. Results indicate that some of the known false positives associated with tractography algorithms are prevented, such that the biological accuracy of the reconstructions should be improved, provided that state-of-the-art streamlines tractography methods are used.

YNIMG Journal 2012 Journal Article

Comment on time-varying eddy currents effects on diffusion-weighting echo-planar imaging

Fernando Calamante
Alan Connelly

This commentary addresses the recent paper entitled “Dynamic correction of artifacts due to susceptibility effects and time-varying eddy currents in diffusion tensor imaging” (Truong et al. , 2011) and discusses their findings in relation to previous studies from the late 1990s that showed equivalent findings regarding the complex time-varying nature of eddy currents during the imaging readout. The similarities between the findings from these studies are highlighted, and their possible implications for present commonly used eddy current reduction methods are discussed.

YNIMG Journal 2012 Journal Article

Super-resolution track-density imaging studies of mouse brain: Comparison to histology

Fernando Calamante
Jacques-Donald Tournier
Nyoman D. Kurniawan
Zhengyi Yang
Erika Gyengesi
Graham J. Galloway
David C. Reutens
Alan Connelly

The recently proposed track-density imaging (TDI) technique was introduced as a means to achieve super-resolution using diffusion MRI. This technique is able to increase the spatial resolution of the reconstructed images beyond the acquired MRI resolution by incorporating information from whole-brain fibre-tracking results. It not only achieves super-resolution, but also provides very high anatomical contrast with a new MRI contrast mechanism. However, the anatomical information-content of this novel contrast mechanism has not yet been assessed. In this work, we perform such a study using diffusion MRI of ex vivo mouse brains acquired at 16. 4T, to compare the results of the super-resolution TDI technique with histological staining (myelin and Nissl stains) in the same brains. Furthermore, a modified version of the directionally-encoded colour TDI map using short-tracks is introduced, which reduces the TDI intensity dynamic range, and therefore enhances the directionality colour-contrast. Good agreement was observed between structures visualised in the super-resolution TDI maps and in the histological sections, supporting the anatomical information-content of the images generated using the TDI technique. The results therefore show that the TDI methodology does provide meaningful and rich anatomical contrast, in addition to achieving super-resolution. Furthermore, this study is the first to show the application of TDI to mouse brain imaging: the high-resolution, high-quality images demonstrate the useful complementary information that can be achieved using super-resolution TDI.

YNIMG Journal 2011 Journal Article

Track density imaging (TDI): Validation of super resolution property

Fernando Calamante
Jacques-Donald Tournier
Robin M. Heidemann
Alfred Anwander
Graeme D. Jackson
Alan Connelly

We have recently introduced a novel MRI methodology, so-called super resolution track-density imaging (TDI), which produces high-quality white matter images, with high spatial resolution and exquisite anatomical contrast not available from other MRI modalities. This method achieves super resolution by utilising the long-range information contained in the diffusion MRI fibre tracks. In this study, we validate the super resolution property of the TDI method by using in vivo diffusion MRI data acquired at ultra-high magnetic field strength (7T), and in silico diffusion MRI data from a well-characterised numerical phantom. Furthermore, an alternative version of the TDI technique is described, which mitigates the track length weighting of the TDI map intensity. For the in vivo data, high-resolution diffusion images were down-sampled to simulate low-resolution data, for which the high-resolution images serve as a gold standard. For the in silico data, the gold standard is given by the known simulated structures of the numerical phantom. Both the in vivo and in silico data show that the structures that could be identified in the TDI maps only after using super resolution were consistent with the corresponding structures identified in the reference maps. This supports the claim that the structures identified by the super resolution step are accurate, thus providing further evidence for the important potential role of the super resolution TDI methodology in neuroscience.

YNIMG Journal 2010 Journal Article

The effect of finite diffusion gradient pulse duration on fibre orientation estimation in diffusion MRI

Chun-Hung Yeh
J-Donald Tournier
Kuan-Hung Cho
Ching-Po Lin
Fernando Calamante
Alan Connelly

An essential step for fibre-tracking is the accurate estimation of neuronal fibre orientations within each imaging voxel, and a number of methods have been proposed to reconstruct the orientation distribution function based on sampling three-dimensional q-space. In the q-space formalism, very short (infinitesimal) gradient pulses are the basic requirement to obtain the true spin displacement probability density function. On current clinical MR systems however, the diffusion gradient pulse duration (δ) is inevitably finite due to the limit on the achievable gradient intensity. The failure to satisfy the short gradient pulse (SGP) requirement has been a recurrent criticism for fibre orientation estimation based on the q-space approach. In this study, the influence of a finite δ on the DW signal measured as a function of gradient direction is described theoretically and demonstrated through simulations and experimental models. Our results suggest that the current practice of using long δ for DW imaging on human clinical MR scanners, which is enforced by hardware limitations, might in fact be beneficial for estimating fibre orientations. For a given b-value, the prolongation of δ is advantageous for estimating fibre orientations for two reasons: first, it leads to a boost in DW signal in the transverse plane of the fibre. Second, it stretches out the shape of the measured diffusion profile, which improves the contrast between DW orientations. This is especially beneficial for resolving crossing fibres, as this contrast is essential to discriminate between different fibre directions.

YNIMG Journal 2010 Journal Article

Track-density imaging (TDI): Super-resolution white matter imaging using whole-brain track-density mapping

Fernando Calamante
Jacques-Donald Tournier
Graeme D. Jackson
Alan Connelly

Neuroimaging advances have given rise to major progress in neurosciences and neurology, as ever more subtle and specific imaging methods reveal new aspects of the brain. One major limitation of current methods is the spatial scale of the information available. We present an approach to gain spatial resolution using post-processing methods based on diffusion MRI fiber-tracking, to reveal structures beyond the resolution of the acquired imaging voxel; we term such a method as super-resolution track-density imaging (TDI). A major unmet challenge in imaging is the identification of abnormalities in white matter as a cause of illness; super-resolution TDI is shown to produce high-quality white matter images, with high spatial resolution and outstanding anatomical contrast. A unique property of these maps is demonstrated: their spatial resolution and signal-to-noise ratio can be tailored depending on the chosen image resolution and total number of fiber-tracks generated. Super-resolution TDI should greatly enhance the study of white matter in disorders of the brain and mind.

YNIMG Journal 2009 Journal Article

A software tool to generate simulated white matter structures for the assessment of fibre-tracking algorithms

Thomas G. Close
Jacques-Donald Tournier
Fernando Calamante
Leigh A. Johnston
Iven Mareels
Alan Connelly

The assessment of Diffusion-Weighted MRI (DW-MRI) fibre-tracking algorithms has been limited by the lack of an appropriate ‘gold standard’. Practical limitations of alternative methods and physical models have meant that numerical simulations have become the method of choice in practice. However, previous numerical phantoms have consisted of separate fibres embedded in homogeneous backgrounds, which do not capture the true nature of white matter. In this paper we describe a method that is able to randomly generate numerical structures consisting of densely packed bundles of fibres, which are much more representative of human white matter, and simulate the DW-MR images that would arise from them under many imaging conditions. User-defined parameters may be adjusted to produce structures with a range of complexities that spans the levels we would expect to find in vivo. These structures are shown to contain many different features that occur in human white matter and which could confound fibre-tracking algorithms, such as tract kissing and crossing. Furthermore, combinations of such features can be sampled by the random generation of many different structures with consistent levels of complexity. The proposed software provides means for quantitative assessment via direct comparison between tracking results and the exact location of the generated fibres. This should greatly improve our understanding of algorithm performance and therefore prove an important tool for fibre tracking development.

YNIMG Journal 2008 Journal Article

Resolving crossing fibres using constrained spherical deconvolution: Validation using diffusion-weighted imaging phantom data

J.-Donald Tournier
Chun-Hung Yeh
Fernando Calamante
Kuan-Hung Cho
Alan Connelly
Ching-Po Lin

Diffusion-weighted imaging can potentially be used to infer the connectivity of the human brain in vivo using fibre-tracking techniques, and is therefore of great interest to neuroscientists and clinicians. A key requirement for fibre tracking is the accurate estimation of white matter fibre orientations within each imaging voxel. The diffusion tensor model, which is widely used for this purpose, has been shown to be inadequate in crossing fibre regions. A number of approaches have recently been proposed to address this issue, based on high angular resolution diffusion-weighted imaging (HARDI) data. In this study, an experimental model of crossing fibres, consisting of water-filled plastic capillaries, is used to thoroughly assess three such techniques: constrained spherical deconvolution (CSD), super-resolved CSD (super-CSD) and Q-ball imaging (QBI). HARDI data were acquired over a range of crossing angles and b-values, from which fibre orientations were computed using each technique. All techniques were capable of resolving the two fibre populations down to a crossing angle of 45°, and down to 30° for super-CSD. A bias was observed in the fibre orientations estimated by QBI for crossing angles other than 90°, consistent with previous simulation results. Finally, for a 45° crossing, the minimum b-value required to resolve the fibre orientations was 4000s/mm2 for QBI, 2000s/mm2 for CSD, and 1000s/mm2 for super-CSD. The quality of estimation of fibre orientations may profoundly affect fibre tracking attempts, and the results presented provide important additional information regarding performance characteristics of well-known methods.

YNIMG Journal 2007 Journal Article

Cortical abnormalities and language function in young patients with basal ganglia stroke

Alison Rowan
Faraneh Vargha-Khadem
Fernando Calamante
Jacques-Donald Tournier
Fenella J. Kirkham
Wui K. Chong
Torsten Baldeweg
Alan Connelly

We examined MRI abnormalities and language function in young patients with infarctions apparently confined to the basal ganglia to establish whether impaired performance was attributable to basal ganglia damage per se or to additional cerebral abnormalities. Seventeen stroke patients (10 with left- and 7 with right-hemispheric damage) and seventeen controls participated. MRI included perfusion imaging and voxel-based morphometry analyses of T1-weighted and diffusion data sets. Language was assessed using the CELF-III test. Analysis of CELF-III scores showed a main effect of presence or absence of stroke, with patients performing more poorly than controls. There was no evidence of differences between the left- and right-hemisphere groups. However individual patients with left-hemisphere lesions showed large variations in performance. In the patients with left-hemisphere damage, voxel-based morphometry showed significant relationships between language function and gray matter density in cortical language areas. The white matter analyses also showed correlations with language function, and in addition there were hemodynamic abnormalities in cortical language areas in the three patients with poorest language function. We suggest that language impairments following basal ganglia damage may be attributable primarily to abnormalities in cortical language areas that are too subtle to detect on conventional structural MRI.

YNIMG Journal 2007 Journal Article

Robust determination of the fibre orientation distribution in diffusion MRI: Non-negativity constrained super-resolved spherical deconvolution

J-Donald Tournier
Fernando Calamante
Alan Connelly

Diffusion-weighted (DW) MR images contain information about the orientation of brain white matter fibres that potentially can be used to study human brain connectivity in vivo using tractography techniques. Currently, the diffusion tensor model is widely used to extract fibre directions from DW-MRI data, but fails in regions containing multiple fibre orientations. The spherical deconvolution technique has recently been proposed to address this limitation. It provides an estimate of the fibre orientation distribution (FOD) by assuming the DW signal measured from any fibre bundle is adequately described by a single response function. However, the deconvolution is ill-conditioned and susceptible to noise contamination. This tends to introduce artefactual negative regions in the FOD, which are clearly physically impossible. In this study, the introduction of a constraint on such negative regions is proposed to improve the conditioning of the spherical deconvolution. This approach is shown to provide FOD estimates that are robust to noise whilst preserving angular resolution. The approach also permits the use of super-resolution, whereby more FOD parameters are estimated than were actually measured, improving the angular resolution of the results. The method provides much better defined fibre orientation estimates, and allows orientations to be resolved that are separated by smaller angles than previously possible. This should allow tractography algorithms to be designed that are able to track reliably through crossing fibre regions.

YNIMG Journal 2004 Journal Article

Direct estimation of the fiber orientation density function from diffusion-weighted MRI data using spherical deconvolution

J.-Donald Tournier
Fernando Calamante
David G. Gadian
Alan Connelly

Diffusion-weighted magnetic resonance imaging can provide information related to the arrangement of white matter fibers. The diffusion tensor is the model most commonly used to derive the orientation of the fibers within a voxel. However, this model has been shown to fail in regions containing several fiber populations with distinct orientations. A number of alternative models have been suggested, such as multiple tensor fitting, q-space, and Q-ball imaging. However, each of these has inherent limitations. In this study, we propose a novel method for estimating the fiber orientation distribution directly from high angular resolution diffusion-weighted MR data without the need for prior assumptions regarding the number of fiber populations present. We assume that all white matter fiber bundles in the brain share identical diffusion characteristics, thus implicitly assigning any differences in diffusion anisotropy to partial volume effects. The diffusion-weighted signal attenuation measured over the surface of a sphere can then be expressed as the convolution over the sphere of a response function (the diffusion-weighted attenuation profile for a typical fiber bundle) with the fiber orientation density function (ODF). The fiber ODF (the distribution of fiber orientations within the voxel) can therefore be obtained using spherical deconvolution. The properties of the technique are demonstrated using simulations and on data acquired from a volunteer using a standard 1. 5-T clinical scanner. The technique can recover the fiber ODF in regions of multiple fiber crossing and holds promise for applications such as tractography.

YNIMG Journal 2003 Journal Article

Diffusion-weighted magnetic resonance imaging fibre tracking using a front evolution algorithm

J.-Donald Tournier
Fernando Calamante
David G Gadian
Alan Connelly

A novel technique is presented for estimating white matter connectivity in vivo using diffusion-weighted magnetic resonance imaging. The concept of a fibre orientation density function (ODF) is described, which characterises the uncertainty in the orientation of the underlying white matter fibres, given the set of diffusion-weighted signal intensities at the point of interest. The proposed algorithm is based on the evolution of a front from a seed region, using the information provided by the fibre ODF. Each point reached by the front is assigned an index of connectivity with the seed region. The algorithm was used to track various major white matter fibre pathways in two data sets acquired on the same healthy adult volunteer over separate occasions. Example tracks are shown to illustrate some of the properties of the algorithm, such as robustness to noise and branching capability. Finally, the dependence of the algorithm on the model used to derive the fibre ODF is discussed.

YNIMG Journal 2003 Journal Article

Estimation of bolus dispersion effects in perfusion MRI using image-based computational fluid dynamics

Fernando Calamante
Peter J. Yim
Juan R. Cebral

Bolus tracking magnetic resonance imaging (MRI) is a powerful technique for measuring perfusion, and is playing an increasing role in the investigation of acute stroke. However, limitations have been reported when assessing patients with steno-occlusive disease. The presence of a steno-occlusive disease in the artery may cause bolus dispersion, which has been shown to introduce significant errors in cerebral blood flow (CBF) quantification. Bolus dispersion is commonly described by a vascular transport function, but the function that properly characterizes the dispersion is unknown. A novel method to quantify bolus dispersion errors on perfusion measurements is presented. A realistic patient-specific model is constructed from anatomical and physiologic MR data, and the arterial blood flow pattern and the transport of the bolus of contrast agent are computed using finite element analysis. The methodology presented was used also to evaluate the accuracy of three simple vascular models. The methodology was tested on MR data from two normal subjects and two subjects with mild carotid artery stenosis. The estimated CBF errors were of the order of 15% to 20%. However, the presence of stenosis did not necessarily introduce larger dispersion (not only the geometrical model but also the particular physiologic conditions influence the degree of bolus dispersion). The method described will contribute to a better understanding of errors introduced by dispersion effects, to the assessment and validation of vascular models, and to the development of new methods for the correction of dispersion errors in CBF quantification.