DBB - A Distorted Brain Benchmark for Automatic Tissue Segmentation in Paediatric Patients.

T1-weighted magnetic resonance images provide a comprehensive view of the morphology of the human brain at the macro scale. These images are usually the input of a segmentation process that aims detecting the anatomical structures labeling them according to a predefined set of target tissues. Automated methods for brain tissue segmentation rely on anatomical priors of the human brain structures. This is the reason why their performance is quite accurate on healthy individuals. Nevertheless model-based tools become less accurate in clinical practice, specifically in the cases of severe lesions or highly distorted cerebral anatomy. More recently there are empirical evidences that a data-driven approach can be more robust in presence of alterations of brain structures, even though the learning model is trained on healthy brains. Our contribution is a benchmark to support an open investigation on how the tissue segmentation of distorted brains can be improved by adopting a supervised learning approach. We formulate a precise definition of the task and propose an evaluation metric for a fair and quantitative comparison. The training sample is composed of almost one thousand healthy individuals. Data include both T1-weighted MR images and their labeling of brain tissues. The test sample is a collection of several tens of individuals with severe brain distortions. Data and code are openly published on BrainLife, an open science platform for reproducible neuroscience data analysis.

Classifyber, a robust streamline-based linear classifier for white matter bundle segmentation.

Virtual delineation of white matter bundles in the human brain is of paramount importance for multiple applications, such as pre-surgical planning and connectomics. A substantial body of literature is related to methods that automatically segment bundles from diffusion Magnetic Resonance Imaging (dMRI) data indirectly, by exploiting either the idea of connectivity between regions or the geometry of fiber paths obtained with tractography techniques, or, directly, through the information in volumetric data. Despite the remarkable improvement in automatic segmentation methods over the years, their segmentation quality is not yet satisfactory, especially when dealing with datasets with very diverse characteristics, such as different tracking methods, bundle sizes or data quality. In this work, we propose a novel, supervised streamline-based segmentation method, called Classifyber, which combines information from atlases, connectivity patterns, and the geometry of fiber paths into a simple linear model. With a wide range of experiments on multiple datasets that span from research to clinical domains, we show that Classifyber substantially improves the quality of segmentation as compared to other state-of-the-art methods and, more importantly, that it is robust across very diverse settings. We provide an implementation of the proposed method as open source code, as well as web service.

Spontaneous unilateral spatial neglect recovery after brain tumour resection: A multimodal diffusion and rs-fMRI case report.

Patients with unilateral spatial neglect (USN) are unable to explore or to report stimuli presented in the left personal and extra-personal space. USN is usually caused by lesion of the right parietal lobe: nowadays, it is also clear the key role of structural connections (the second and the third branch of the right Superior Longitudinal Fasciculus, respectively, SLF II and III) and functional networks (Dorsal and Ventral Attention Network, respectively, DAN and VAN) in USN. In this multimodal case report, we have merged those structural and functional information derived from a patient with a right parietal lobe tumour and USN before surgery. Functional, structural and neuropsychological data were also collected 6 months after surgery, when the USN was spontaneously recovered. Diffusion metrics and Functional Connectivity (FC) of the right SLF and DAN, before and after surgery, were compared with the same data of a patient with a tumour in a similar location, but without USN, and with a control sample. Results indicate an impairment in the right SLF III and a reduction of FC of the right DAN in patients with USN before surgery compared to controls; after surgery, when USN was recovered, patient's diffusion metrics and FC showed no differences compared to the controls. This single case and its multimodal approach reinforce the crucial role of the right SLF III and DAN in the development and recovery of egocentric and allocentric extra-personal USN, highlighting the need to preserve these structural and functional areas during brain surgery.

Accurate localization and coactivation profiles of the frontal eye field and inferior frontal junction: an ALE and MACM fMRI meta-analysis.

The frontal eye field (FEF) and the inferior frontal junction (IFJ) are prefrontal structures involved in mediating multiple aspects of goal-driven behavior. Despite being recognized as prominent nodes of the networks underlying spatial attention and oculomotor control, and working memory and cognitive control, respectively, the limited quantitative evidence on their precise localization has considerably impeded the detailed understanding of their structure and connectivity. In this study, we performed an activation likelihood estimation (ALE) fMRI meta-analysis by selecting studies that employed standard paradigms to accurately infer the localization of these regions in stereotaxic space. For the FEF, we found the highest spatial convergence of activations for prosaccade and antisaccade paradigms at the junction of the precentral sulcus and superior frontal sulcus. For the IFJ, we found consistent activations across oddball/attention, working memory, task-switching and Stroop paradigms at the junction of the inferior precentral sulcus and inferior frontal sulcus. We related these clusters to previous meta-analyses, sulcal/gyral neuroanatomy, and a comprehensive brain parcellation, highlighting important differences compared to their results and taxonomy. Finally, we leveraged the ALE peak coordinates as seeds to perform a meta-analytic connectivity modeling (MACM) analysis, which revealed systematic coactivation patterns spanning the frontal, parietal, and temporal cortices. We decoded the behavioral domains associated with these coactivations, suggesting that these may allow FEF and IFJ to support their specialized roles in flexible behavior. Our study provides the meta-analytic groundwork for investigating the relationship between functional specialization and connectivity of two crucial control structures of the prefrontal cortex.

BundleWarp, streamline-based nonlinear registration of white matter tracts.

Nonlinear registration plays a central role in most neuroimage analysis methods and pipelines, such as in tractography-based individual and group-level analysis methods. However, nonlinear registration is a non-trivial task, especially when dealing with tractography data that digitally represent the underlying anatomy of the brain's white matter. Furthermore, such process often changes the structure of the data, causing artifacts that can suppress the underlying anatomical and structural details. In this paper, we introduce BundleWarp, a novel and robust streamline-based nonlinear registration method for the registration of white matter tracts. BundleWarp intelligently warps two bundles while preserving the bundles' crucial topological features. BundleWarp has two main steps. The first step involves the solution of an assignment problem that matches corresponding streamlines from the two bundles (iterLAP step). The second step introduces streamline-specific point-based deformations while keeping the topology of the bundle intact (mlCPD step). We provide comparisons against streamline-based linear registration and image-based nonlinear registration methods. BundleWarp quantitatively and qualitatively outperforms both, and we show that BundleWarp can deform and, at the same time, preserve important characteristics of the original anatomical shape of the bundles. Results are shown on 1,728 pairs of bundle registrations across 27 different bundle types. In addition, we present an application of BundleWarp for quantifying bundle shape differences using the generated deformation fields.

Supervised tractogram filtering using Geometric Deep Learning.

A tractogram is a virtual representation of the brain white matter. It is composed of millions of virtual fibers, encoded as 3D polylines, which approximate the white matter axonal pathways. To date, tractograms are the most accurate white matter representation and thus are used for tasks like presurgical planning and investigations of neuroplasticity, brain disorders, or brain networks. However, it is a well-known issue that a large portion of tractogram fibers is not anatomically plausible and can be considered artifacts of the tracking procedure. With Verifyber, we tackle the problem of filtering out such non-plausible fibers using a novel fully-supervised learning approach. Differently from other approaches based on signal reconstruction and/or brain topology regularization, we guide our method with the existing anatomical knowledge of the white matter. Using tractograms annotated according to anatomical principles, we train our model, Verifyber, to classify fibers as either anatomically plausible or non-plausible. The proposed Verifyber model is an original Geometric Deep Learning method that can deal with variable size fibers, while being invariant to fiber orientation. Our model considers each fiber as a graph of points, and by learning features of the edges between consecutive points via the proposed sequence Edge Convolution, it can capture the underlying anatomical properties. The output filtering results highly accurate and robust across an extensive set of experiments, and fast; with a 12GB GPU, filtering a tractogram of 1M fibers requires less than a minute.

Planning Brain Tumor Resection Using a Probabilistic Atlas of Cortical and Subcortical Structures Critical for Functional Processing: A Proof of Concept.

BACKGROUND: Functional preoperative planning for resection of intrinsic brain tumors in eloquent areas is still a challenge. Predicting subcortical functional framework is especially difficult. Direct electrical stimulation (DES) is the recommended technique for resection of these lesions. A reliable probabilistic atlas of the critical cortical epicenters and subcortical framework based on DES data was recently published. OBJECTIVE: To propose a pipeline for the automated alignment of the corticosubcortical maps of this atlas with T1-weighted MRI. METHODS: To test the alignment, we selected 10 patients who underwent resection of brain lesions by using DES. We aligned different cortical and subcortical functional maps to preoperative volumetric T1 MRIs (with/without gadolinium). For each patient we quantified the quality of the alignment, and we calculated the match between the location of the functional sites found at DES and the functional maps of the atlas. RESULTS: We found an accurate brain extraction and alignment of the functional maps with both the T1 MRIs of each patient. The matching analysis between functional maps and functional responses collected during surgeries was 88% at cortical and, importantly, 100% at subcortical level, providing a further proof of the correct alignment. CONCLUSION: We demonstrated quantitatively and qualitatively the reliability of this tool that may be used for presurgical planning, providing further functional information at the cortical level and a unique probabilistic prevision of distribution of the critical subcortical structures. Finally, this tool offers the chance for multimodal planning through integrating this functional information with other neuroradiological and neurophysiological techniques.

The open diffusion data derivatives, brain data upcycling via integrated publishing of derivatives and reproducible open cloud services.

We describe the Open Diffusion Data Derivatives (O3D) repository: an integrated collection of preserved brain data derivatives and processing pipelines, published together using a single digital-object-identifier. The data derivatives were generated using modern diffusion-weighted magnetic resonance imaging data (dMRI) with diverse properties of resolution and signal-to-noise ratio. In addition to the data, we publish all processing pipelines (also referred to as open cloud services). The pipelines utilize modern methods for neuroimaging data processing (diffusion-signal modelling, fiber tracking, tractography evaluation, white matter segmentation, and structural connectome construction). The O3D open services can allow cognitive and clinical neuroscientists to run the connectome mapping algorithms on new, user-uploaded, data. Open source code implementing all O3D services is also provided to allow computational and computer scientists to reuse and extend the processing methods. Publishing both data-derivatives and integrated processing pipeline promotes practices for scientific reproducibility and data upcycling by providing open access to the research assets for utilization by multiple scientific communities.

Classification-Based Prediction of Effective Connectivity Between Timeseries With a Realistic Cortical Network Model.

Effective connectivity measures the pattern of causal interactions between brain regions. Traditionally, these patterns of causality are inferred from brain recordings using either non-parametric, i.e., model-free, or parametric, i.e., model-based, approaches. The latter approaches, when based on biophysically plausible models, have the advantage that they may facilitate the interpretation of causality in terms of underlying neural mechanisms. Recent biophysically plausible neural network models of recurrent microcircuits have shown the ability to reproduce well the characteristics of real neural activity and can be applied to model interacting cortical circuits. Unfortunately, however, it is challenging to invert these models in order to estimate effective connectivity from observed data. Here, we propose to use a classification-based method to approximate the result of such complex model inversion. The classifier predicts the pattern of causal interactions given a multivariate timeseries as input. The classifier is trained on a large number of pairs of multivariate timeseries and the respective pattern of causal interactions, which are generated by simulation from the neural network model. In simulated experiments, we show that the proposed method is much more accurate in detecting the causal structure of timeseries than current best practice methods. Additionally, we present further results to characterize the validity of the neural network model and the ability of the classifier to adapt to the generative model of the data.

Photogrammetry of the Human Brain: A Novel Method for Three-Dimensional Quantitative Exploration of the Structural Connectivity in Neurosurgery and Neurosciences.

BACKGROUND: Anatomic awareness of the structural connectivity of the brain is mandatory for neurosurgeons, to select the most effective approaches for brain resections. Although standard microdissection is a validated technique to investigate the different white matter (WM) pathways and to verify the results of tractography, the possibility of interactive exploration of the specimens and reliable acquisition of quantitative information has not been described. Photogrammetry is a well-established technique allowing an accurate metrology on highly defined three-dimensional (3D) models. The aim of this work is to propose the application of the photogrammetric technique for supporting the 3D exploration and the quantitative analysis on the cerebral WM connectivity. METHODS: The main perisylvian pathways, including the superior longitudinal fascicle and the arcuate fascicle were exposed using the Klingler technique. The photogrammetric acquisition followed each dissection step. The point clouds were registered to a reference magnetic resonance image of the specimen. All the acquisitions were coregistered into an open-source model. RESULTS: We analyzed 5 steps, including the cortical surface, the short intergyral fibers, the indirect posterior and anterior superior longitudinal fascicle, and the arcuate fascicle. The coregistration between the magnetic resonance imaging mesh and the point clouds models was highly accurate. Multiple measures of distances between specific cortical landmarks and WM tracts were collected on the photogrammetric model. CONCLUSIONS: Photogrammetry allows an accurate 3D reproduction of WM anatomy and the acquisition of unlimited quantitative data directly on the real specimen during the postdissection analysis. These results open many new promising neuroscientific and educational perspectives and also optimize the quality of neurosurgical treatments.

White Matter Tract Segmentation as Multiple Linear Assignment Problems.

Diffusion magnetic resonance imaging (dMRI) allows to reconstruct the main pathways of axons within the white matter of the brain as a set of polylines, called streamlines. The set of streamlines of the whole brain is called the tractogram. Organizing tractograms into anatomically meaningful structures, called tracts, is known as the tract segmentation problem, with important applications to neurosurgical planning and tractometry. Automatic tract segmentation techniques can be unsupervised or supervised. A common criticism of unsupervised methods, like clustering, is that there is no guarantee to obtain anatomically meaningful tracts. In this work, we focus on supervised tract segmentation, which is driven by prior knowledge from anatomical atlases or from examples, i.e., segmented tracts from different subjects. We present a supervised tract segmentation method that segments a given tract of interest in the tractogram of a new subject using multiple examples as prior information. Our proposed tract segmentation method is based on the idea of streamline correspondence i.e., on finding corresponding streamlines across different tractograms. In the literature, streamline correspondence has been addressed with the nearest neighbor (NN) strategy. Differently, here we formulate the problem of streamline correspondence as a linear assignment problem (LAP), which is a cornerstone of combinatorial optimization. With respect to the NN, the LAP introduces a constraint of one-to-one correspondence between streamlines, that forces the correspondences to follow the local anatomical differences between the example and the target tract, neglected by the NN. In the proposed solution, we combined the Jonker-Volgenant algorithm (LAPJV) for solving the LAP together with an efficient way of computing the nearest neighbors of a streamline, which massively reduces the total amount of computations needed to segment a tract. Moreover, we propose a ranking strategy to merge correspondences coming from different examples. We validate the proposed method on tractograms generated from the human connectome project (HCP) dataset and compare the segmentations with the NN method and the ROI-based method. The results show that LAP-based segmentation is vastly more accurate than ROI-based segmentation and substantially more accurate than the NN strategy. We provide a Free/OpenSource implementation of the proposed method.

Supervised Estimation of Granger-Based Causality between Time Series.

Brain effective connectivity aims to detect causal interactions between distinct brain units and it is typically studied through the analysis of direct measurements of the neural activity, e.g., magneto/electroencephalography (M/EEG) signals. The literature on methods for causal inference is vast. It includes model-based methods in which a generative model of the data is assumed and model-free methods that directly infer causality from the probability distribution of the underlying stochastic process. Here, we firstly focus on the model-based methods developed from the Granger criterion of causality, which assumes the autoregressive model of the data. Secondly, we introduce a new perspective, that looks at the problem in a way that is typical of the machine learning literature. Then, we formulate the problem of causality detection as a supervised learning task, by proposing a classification-based approach. A classifier is trained to identify causal interactions between time series for the chosen model and by means of a proposed feature space. In this paper, we are interested in comparing this classification-based approach with the standard Geweke measure of causality in the time domain, through simulation study. Thus, we customized our approach to the case of a MAR model and designed a feature space which contains causality measures based on the idea of precedence and predictability in time. Two variations of the supervised method are proposed and compared to a standard Granger causal analysis method. The results of the simulations show that the supervised method outperforms the standard approach, in particular it is more robust to noise. As evidence of the efficacy of the proposed method, we report the details of our submission to the causality detection competition of Biomag2014, where the proposed method reached the 2nd place. Moreover, as empirical application, we applied the supervised approach on a dataset of neural recordings of rats obtaining an important reduction in the false positive rate.

Bayesian estimation of directed functional coupling from brain recordings.

In many fields of science, there is the need of assessing the causal influences among time series. Especially in neuroscience, understanding the causal interactions between brain regions is of primary importance. A family of measures have been developed from the parametric implementation of the Granger criteria of causality based on the linear autoregressive modelling of the signals. We propose a new Bayesian method for linear model identification with a structured prior (GMEP) aiming to apply it as linear regression method in the context of the parametric Granger causal inference. GMEP assumes a Gaussian scale mixture distribution for the group sparsity prior and it enables flexible definition of the coefficient groups. Approximate posterior inference is achieved using Expectation Propagation for both the linear coefficients and the hyperparameters. GMEP is investigated both on simulated data and on empirical fMRI data in which we show how adding information on the sparsity structure of the coefficients positively improves the inference process. In the same simulation framework, GMEP is compared with others standard linear regression methods. Moreover, the causal inferences derived from GMEP estimates and from a standard Granger method are compared across simulated datasets of different dimensionality, density connection and level of noise. GMEP allows a better model identification and consequent causal inference when prior knowledge on the sparsity structure are integrated in the structured prior.

Alignment of Tractograms As Graph Matching.

The white matter pathways of the brain can be reconstructed as 3D polylines, called streamlines, through the analysis of diffusion magnetic resonance imaging (dMRI) data. The whole set of streamlines is called tractogram and represents the structural connectome of the brain. In multiple applications, like group-analysis, segmentation, or atlasing, tractograms of different subjects need to be aligned. Typically, this is done with registration methods, that transform the tractograms in order to increase their similarity. In contrast with transformation-based registration methods, in this work we propose the concept of tractogram correspondence, whose aim is to find which streamline of one tractogram corresponds to which streamline in another tractogram, i.e., a map from one tractogram to another. As a further contribution, we propose to use the relational information of each streamline, i.e., its distances from the other streamlines in its own tractogram, as the building block to define the optimal correspondence. We provide an operational procedure to find the optimal correspondence through a combinatorial optimization problem and we discuss its similarity to the graph matching problem. In this work, we propose to represent tractograms as graphs and we adopt a recent inexact sub-graph matching algorithm to approximate the solution of the tractogram correspondence problem. On tractograms generated from the Human Connectome Project dataset, we report experimental evidence that tractogram correspondence, implemented as graph matching, provides much better alignment than affine registration and comparable if not better results than non-linear registration of volumes.

Differential Effects of Brain Disorders on Structural and Functional Connectivity.

Different measures of brain connectivity can be defined based on neuroimaging read-outs, including structural and functional connectivity. Neurological and psychiatric conditions are often associated with abnormal connectivity, but comparing the effects of the disease on different types of connectivity remains a challenge. In this paper, we address the problem of quantifying the relative effects of brain disease on structural and functional connectivity at a group level. Within the framework of a graph representation of connectivity, we introduce a kernel two-sample test as an effective method to assess the difference between the patients and control group. Moreover, we propose a common representation space for structural and functional connectivity networks, and a novel test statistics to quantitatively assess differential effects of the disease on different types of connectivity. We apply this approach to a dataset from BTBR mice, a murine model of Agenesis of the Corpus Callosum (ACC), a congenital disorder characterized by the absence of the main bundle of fibers connecting the two hemispheres. We used normo-callosal mice (B6) as a comparator. The application of the proposed methods to this data-set shows that the two types of connectivity can be successfully used to discriminate between BTBR and B6, meaning that both types of connectivity are affected by ACC. However, our novel test statistics shows that structural connectivity is significantly more affected than functional connectivity, consistent with the idea that functional connectivity has a robust topology that can tolerate substantial alterations in its structural connectivity substrate.

Statistical independence for the evaluation of classifier-based diagnosis.

Machine learning techniques are increasingly adopted in computer-aided diagnosis. Evaluation methods for classification results that are based on the study of one or more metrics can be unable to distinguish between cases in which the classifier is discriminating the classes from cases in which it is not. In the binary setting, such circumstances can be encountered when data are unbalanced with respect to the diagnostic groups. Having more healthy controls than pathological subjects, datasets meant for diagnosis frequently show a certain degree of unbalancedness. In this work, we propose to recast the evaluation of classification results as a test of statistical independence between the predicted and the actual diagnostic groups. We address the problem within the Bayesian hypothesis testing framework. Different from the standard metrics, the proposed method is able to handle unbalanced data and takes into account the size of the available data. We show experimental evidence of the efficacy of the approach both on simulated data and on real data about the diagnosis of the Attention Deficit Hyperactivity Disorder (ADHD).

ADHD diagnosis from multiple data sources with batch effects.

The Attention Deficit Hyperactivity Disorder (ADHD) affects the school-age population and has large social costs. The scientific community is still lacking a pathophysiological model of the disorder and there are no objective biomarkers to support the diagnosis. In 2011 the ADHD-200 Consortium provided a rich, heterogeneous neuroimaging dataset aimed at studying neural correlates of ADHD and to promote the development of systems for automated diagnosis. Concurrently a competition was set up with the goal of addressing the wide range of different types of data for the accurate prediction of the presence of ADHD. Phenotypic information, structural magnetic resonance imaging (MRI) scans and resting state fMRI recordings were provided for nearly 1000 typical and non-typical young individuals. Data were collected by eight different research centers in the consortium. This work is not concerned with the main task of the contest, i.e., achieving a high prediction accuracy on the competition dataset, but we rather address the proper handling of such a heterogeneous dataset when performing classification-based analysis. Our interest lies in the clustered structure of the data causing the so-called batch effects which have strong impact when assessing the performance of classifiers built on the ADHD-200 dataset. We propose a method to eliminate the biases introduced by such batch effects. Its application on the ADHD-200 dataset generates such a significant drop in prediction accuracy that most of the conclusions from a standard analysis had to be revised. In addition we propose to adopt the dissimilarity representation to set up effective representation spaces for the heterogeneous ADHD-200 dataset. Moreover we propose to evaluate the quality of predictions through a recently proposed test of independence in order to cope with the unbalancedness of the dataset.

Intercepting the First Pass: Rapid Categorization is Suppressed for Unseen Stimuli.

The operations and processes that the human brain employs to achieve fast visual categorization remain a matter of debate. A first issue concerns the timing and place of rapid visual categorization and to what extent it can be performed with an early feed-forward pass of information through the visual system. A second issue involves the categorization of stimuli that do not reach visual awareness. There is disagreement over the degree to which these stimuli activate the same early mechanisms as stimuli that are consciously perceived. We employed continuous flash suppression (CFS), EEG recordings, and machine learning techniques to study visual categorization of seen and unseen stimuli. Our classifiers were able to predict from the EEG recordings the category of stimuli on seen trials but not on unseen trials. Rapid categorization of conscious images could be detected around 100 ms on the occipital electrodes, consistent with a fast, feed-forward mechanism of target detection. For the invisible stimuli, however, CFS eliminated all traces of early processing. Our results support the idea of a fast mechanism of categorization and suggest that this early categorization process plays an important role in later, more subtle categorizations, and perceptual processes.

PyMVPA: A Unifying Approach to the Analysis of Neuroscientific Data.

The Python programming language is steadily increasing in popularity as the language of choice for scientific computing. The ability of this scripting environment to access a huge code base in various languages, combined with its syntactical simplicity, make it the ideal tool for implementing and sharing ideas among scientists from numerous fields and with heterogeneous methodological backgrounds. The recent rise of reciprocal interest between the machine learning (ML) and neuroscience communities is an example of the desire for an inter-disciplinary transfer of computational methods that can benefit from a Python-based framework. For many years, a large fraction of both research communities have addressed, almost independently, very high-dimensional problems with almost completely non-overlapping methods. However, a number of recently published studies that applied ML methods to neuroscience research questions attracted a lot of attention from researchers from both fields, as well as the general public, and showed that this approach can provide novel and fruitful insights into the functioning of the brain. In this article we show how PyMVPA, a specialized Python framework for machine learning based data analysis, can help to facilitate this inter-disciplinary technology transfer by providing a single interface to a wide array of machine learning libraries and neural data-processing methods. We demonstrate the general applicability and power of PyMVPA via analyses of a number of neural data modalities, including fMRI, EEG, MEG, and extracellular recordings.