Atlases of cognition with large-scale human brain mapping.

To map the neural substrate of mental function, cognitive neuroimaging relies on controlled psychological manipulations that engage brain systems associated with specific cognitive processes. In order to build comprehensive atlases of cognitive function in the brain, it must assemble maps for many different cognitive processes, which often evoke overlapping patterns of activation. Such data aggregation faces contrasting goals: on the one hand finding correspondences across vastly different cognitive experiments, while on the other hand precisely describing the function of any given brain region. Here we introduce a new analysis framework that tackles these difficulties and thereby enables the generation of brain atlases for cognitive function. The approach leverages ontologies of cognitive concepts and multi-label brain decoding to map the neural substrate of these concepts. We demonstrate the approach by building an atlas of functional brain organization based on 30 diverse functional neuroimaging studies, totaling 196 different experimental conditions. Unlike conventional brain mapping, this functional atlas supports robust reverse inference: predicting the mental processes from brain activity in the regions delineated by the atlas. To establish that this reverse inference is indeed governed by the corresponding concepts, and not idiosyncrasies of experimental designs, we show that it can accurately decode the cognitive concepts recruited in new tasks. These results demonstrate that aggregating independent task-fMRI studies can provide a more precise global atlas of selective associations between brain and cognition.

FReM - Scalable and stable decoding with fast regularized ensemble of models.

Brain decoding relates behavior to brain activity through predictive models. These are also used to identify brain regions involved in the cognitive operations related to the observed behavior. Training such multivariate models is a high-dimensional statistical problem that calls for suitable priors. State of the art priors -eg small total-variation- enforce spatial structure on the maps to stabilize them and improve prediction. However, they come with a hefty computational cost. We build upon very fast dimension reduction with spatial structure and model ensembling to achieve decoders that are fast on large datasets and increase the stability of the predictions and the maps. Our approach, fast regularized ensemble of models (FReM), includes an implicit spatial regularization by using a voxel grouping with a fast clustering algorithm. In addition, it aggregates different estimators obtained across splits of a cross-validation loop, each time keeping the best possible model. Experiments on a large number of brain imaging datasets show that our combination of voxel clustering and model ensembling improves decoding maps stability and reduces the variance of prediction accuracy. Importantly, our method requires less samples than state-of-the-art methods to achieve a given level of prediction accuracy. Finally, FreM is much faster than other spatially-regularized methods and, in addition, it can better exploit parallel computing resources.

The Brainomics/Localizer database.

The Brainomics/Localizer database exposes part of the data collected by the in-house Localizer project, which planned to acquire four types of data from volunteer research subjects: anatomical MRI scans, functional MRI data, behavioral and demographic data, and DNA sampling. Over the years, this local project has been collecting such data from hundreds of subjects. We had selected 94 of these subjects for their complete datasets, including all four types of data, as the basis for a prior publication; the Brainomics/Localizer database publishes the data associated with these 94 subjects. Since regulatory rules prevent us from making genetic data available for download, the database serves only anatomical MRI scans, functional MRI data, behavioral and demographic data. To publish this set of heterogeneous data, we use dedicated software based on the open-source CubicWeb semantic web framework. Through genericity in the data model and flexibility in the display of data (web pages, CSV, JSON, XML), CubicWeb helps us expose these complex datasets in original and efficient ways.

Assessing and tuning brain decoders: Cross-validation, caveats, and guidelines.

Decoding, i.e. prediction from brain images or signals, calls for empirical evaluation of its predictive power. Such evaluation is achieved via cross-validation, a method also used to tune decoders' hyper-parameters. This paper is a review on cross-validation procedures for decoding in neuroimaging. It includes a didactic overview of the relevant theoretical considerations. Practical aspects are highlighted with an extensive empirical study of the common decoders in within- and across-subject predictions, on multiple datasets -anatomical and functional MRI and MEG- and simulations. Theory and experiments outline that the popular "leave-one-out" strategy leads to unstable and biased estimates, and a repeated random splits method should be preferred. Experiments outline the large error bars of cross-validation in neuroimaging settings: typical confidence intervals of 10%. Nested cross-validation can tune decoders' parameters while avoiding circularity bias. However we find that it can be favorable to use sane defaults, in particular for non-sparse decoders.

NeuroVault.org: A repository for sharing unthresholded statistical maps, parcellations, and atlases of the human brain.

NeuroVault.org is dedicated to storing outputs of analyses in the form of statistical maps, parcellations and atlases, a unique strategy that contrasts with most neuroimaging repositories that store raw acquisition data or stereotaxic coordinates. Such maps are indispensable for performing meta-analyses, validating novel methodology, and deciding on precise outlines for regions of interest (ROIs). NeuroVault is open to maps derived from both healthy and clinical populations, as well as from various imaging modalities (sMRI, fMRI, EEG, MEG, PET, etc.). The repository uses modern web technologies such as interactive web-based visualization, cognitive decoding, and comparison with other maps to provide researchers with efficient, intuitive tools to improve the understanding of their results. Each dataset and map is assigned a permanent Universal Resource Locator (URL), and all of the data is accessible through a REST Application Programming Interface (API). Additionally, the repository supports the NIDM-Results standard and has the ability to parse outputs from popular FSL and SPM software packages to automatically extract relevant metadata. This ease of use, modern web-integration, and pioneering functionality holds promise to improve the workflow for making inferences about and sharing whole-brain statistical maps.

Electronic versus paper-based assessment of health-related quality of life specific to HIV disease: reliability study of the PROQOL-HIV questionnaire.

BACKGROUND: Electronic patient-reported outcomes (PRO) provide quick and usually reliable assessments of patients' health-related quality of life (HRQL). OBJECTIVE: An electronic version of the Patient-Reported Outcomes Quality of Life-human immunodeficiency virus (PROQOL-HIV) questionnaire was developed, and its face validity and reliability were assessed using standard psychometric methods. METHODS: A sample of 80 French outpatients (66% male, 52/79; mean age 46.7 years, SD 10.9) were recruited. Paper-based and electronic questionnaires were completed in a randomized crossover design (2-7 day interval). Biomedical data were collected. Questionnaire version and order effects were tested on full-scale scores in a 2-way ANOVA with patients as random effects. Test-retest reliability was evaluated using Pearson and intraclass correlation coefficients (ICC, with 95% confidence interval) for each dimension. Usability testing was carried out from patients' survey reports, specifically, general satisfaction, ease of completion, quality and clarity of user interface, and motivation to participate in follow-up PROQOL-HIV electronic assessments. RESULTS: Questionnaire version and administration order effects (N=59 complete cases) were not significant at the 5% level, and no interaction was found between these 2 factors (P=.94). Reliability indexes were acceptable, with Pearson correlations greater than .7 and ICCs ranging from .708 to .939; scores were not statistically different between the two versions. A total of 63 (79%) complete patients' survey reports were available, and 55% of patients (30/55) reported being satisfied and interested in electronic assessment of their HRQL in clinical follow-up. Individual ratings of PROQOL-HIV user interface (85%-100% of positive responses) confirmed user interface clarity and usability. CONCLUSIONS: The electronic PROQOL-HIV introduces minor modifications to the original paper-based version, following International Society for Pharmacoeconomics and Outcomes Research (ISPOR) ePRO Task Force guidelines, and shows good reliability and face validity. Patients can complete the computerized PROQOL-HIV questionnaire and the scores from the paper or electronic versions share comparable accuracy and interpretation.

FTO, obesity and the adolescent brain.

Genetic variations in fat mass- and obesity (FTO)-associated gene, a well-replicated gene locus of obesity, appear to be associated also with reduced regional brain volumes in elderly. Here, we examined whether FTO is associated with total brain volume in adolescence, thus exploring possible developmental effects of FTO. We studied a population-based sample of 598 adolescents recruited from the French Canadian founder population in whom we measured brain volume by magnetic resonance imaging. Total fat mass was assessed with bioimpedance and body mass index was determined with anthropometry. Genotype-phenotype associations were tested with Merlin under an additive model. We found that the G allele of FTO (rs9930333) was associated with higher total body fat [TBF (P = 0.002) and lower brain volume (P = 0.005)]. The same allele was also associated with higher lean body mass (P = 0.03) and no difference in height (P = 0.99). Principal component analysis identified a shared inverse variance between the brain volume and TBF, which was associated with FTO at P = 5.5 × 10(-6). These results were replicated in two independent samples of 413 and 718 adolescents, and in a meta-analysis of all three samples (n = 1729 adolescents), FTO was associated with this shared inverse variance at P = 1.3 × 10(-9). Co-expression networks analysis supported the possibility that the underlying FTO effects may occur during embryogenesis. In conclusion, FTO is associated with shared inverse variance between body adiposity and brain volume, suggesting that this gene may exert inverse effects on adipose and brain tissues. Given the completion of the overall brain growth in early childhood, these effects may have their origins during early development.

Cohort-level brain mapping: learning cognitive atoms to single out specialized regions.

Functional Magnetic Resonance Imaging (fMRI) studies map the human brain by testing the response of groups of individuals to carefully-crafted and contrasted tasks in order to delineate specialized brain regions and networks. The number of functional networks extracted is limited by the number of subject-level contrasts and does not grow with the cohort. Here, we introduce a new group-level brain mapping strategy to differentiate many regions reflecting the variety of brain network configurations observed in the population. Based on the principle of functional segregation, our approach singles out functionally-specialized brain regions by learning group-level functional profiles on which the response of brain regions can be represented sparsely. We use a dictionary-learning formulation that can be solved efficiently with on-line algorithms, scaling to arbitrary large datasets. Importantly, we model inter-subject correspondence as structure imposed in the estimated functional profiles, integrating a structure-inducing regularization with no additional computational cost. On a large multi-subject study, our approach extracts a large number of brain networks with meaningful functional profiles.

PyXNAT: XNAT in Python.

As neuroimaging databases grow in size and complexity, the time researchers spend investigating and managing the data increases to the expense of data analysis. As a result, investigators rely more and more heavily on scripting using high-level languages to automate data management and processing tasks. For this, a structured and programmatic access to the data store is necessary. Web services are a first step toward this goal. They however lack in functionality and ease of use because they provide only low-level interfaces to databases. We introduce here PyXNAT, a Python module that interacts with The Extensible Neuroimaging Archive Toolkit (XNAT) through native Python calls across multiple operating systems. The choice of Python enables PyXNAT to expose the XNAT Web Services and unify their features with a higher level and more expressive language. PyXNAT provides XNAT users direct access to all the scientific packages in Python. Finally PyXNAT aims to be efficient and easy to use, both as a back-end library to build XNAT clients and as an alternative front-end from the command line.

Data sharing in neuroimaging research.

Significant resources around the world have been invested in neuroimaging studies of brain function and disease. Easier access to this large body of work should have profound impact on research in cognitive neuroscience and psychiatry, leading to advances in the diagnosis and treatment of psychiatric and neurological disease. A trend toward increased sharing of neuroimaging data has emerged in recent years. Nevertheless, a number of barriers continue to impede momentum. Many researchers and institutions remain uncertain about how to share data or lack the tools and expertise to participate in data sharing. The use of electronic data capture (EDC) methods for neuroimaging greatly simplifies the task of data collection and has the potential to help standardize many aspects of data sharing. We review here the motivations for sharing neuroimaging data, the current data sharing landscape, and the sociological or technical barriers that still need to be addressed. The INCF Task Force on Neuroimaging Datasharing, in conjunction with several collaborative groups around the world, has started work on several tools to ease and eventually automate the practice of data sharing. It is hoped that such tools will allow researchers to easily share raw, processed, and derived neuroimaging data, with appropriate metadata and provenance records, and will improve the reproducibility of neuroimaging studies. By providing seamless integration of data sharing and analysis tools within a commodity research environment, the Task Force seeks to identify and minimize barriers to data sharing in the field of neuroimaging.

Very large fMRI study using the IMAGEN database: sensitivity-specificity and population effect modeling in relation to the underlying anatomy.

In this paper we investigate the use of classical fMRI Random Effect (RFX) group statistics when analyzing a very large cohort and the possible improvement brought from anatomical information. Using 1326 subjects from the IMAGEN study, we first give a global picture of the evolution of the group effect t-value from a simple face-watching contrast with increasing cohort size. We obtain a wide activated pattern, far from being limited to the reasonably expected brain areas, illustrating the difference between statistical significance and practical significance. This motivates us to inject tissue-probability information into the group estimation, we model the BOLD contrast using a matter-weighted mixture of Gaussians and compare it to the common, single-Gaussian model. In both cases, the model parameters are estimated per-voxel for one subgroup, and the likelihood of both models is computed on a second, separate subgroup to reflect model generalization capacity. Various group sizes are tested, and significance is asserted using a 10-fold cross-validation scheme. We conclude that adding matter information consistently improves the quantitative analysis of BOLD responses in some areas of the brain, particularly those where accurate inter-subject registration remains challenging.

Improving accuracy and power with transfer learning using a meta-analytic database.

Typical cohorts in brain imaging studies are not large enough for systematic testing of all the information contained in the images. To build testable working hypotheses, investigators thus rely on analysis of previous work, sometimes formalized in a so-called meta-analysis. In brain imaging, this approach underlies the specification of regions of interest (ROIs) that are usually selected on the basis of the coordinates of previously detected effects. In this paper, we propose to use a database of images, rather than coordinates, and frame the problem as transfer learning: learning a discriminant model on a reference task to apply it to a different but related new task. To facilitate statistical analysis of small cohorts, we use a sparse discriminant model that selects predictive voxels on the reference task and thus provides a principled procedure to define ROIs. The benefits of our approach are twofold. First it uses the reference database for prediction, i.e., to provide potential biomarkers in a clinical setting. Second it increases statistical power on the new task. We demonstrate on a set of 18 pairs of functional MRI experimental conditions that our approach gives good prediction. In addition, on a specific transfer situation involving different scanners at different locations, we show that voxel selection based on transfer learning leads to higher detection power on small cohorts.