Bertrand Thirion

• National Institute for Research in Computer Science and Control

Conditional Permutation Importance (CPI) has been recently introduced for Variable Importance analysis with good empirical results. In this work, we first provide theoretical guarantees for CPI. We establish a double robustness property to detect null covariates, making it a suitable model for variable selection. We then present a modified and stil...

"Number sense", the ability to quickly estimate quantities of objects in a visual scene, is present in humans and many other animals, and has recently been demonstrated in biologically inspired vision models, even before training. However, real-world number perception requires abstraction from the properties of individual objects and their contexts...

Causal machine learning (ML) promises to provide powerful tools for estimating individual treatment effects. Although causal ML methods are now well established, they still face the significant challenge of interpretability, which is crucial for medical applications. In this work, we propose a new algorithm based on the Conditional Permutation Impo...

Knockoffs are a popular statistical framework that addresses the challenging problem of conditional variable selection in high-dimensional settings with statistical control. Such statistical control is essential for the reliability of inference. However, knockoff guarantees rely on an exchangeability assumption that is difficult to test in practice...

Decoding cognitive states from functional magnetic resonance imaging is central to understanding the functional organization of the brain. Within-subject decoding avoids between-subject correspondence problems but requires large sample sizes to make accurate predictions; obtaining such large sample sizes is both challenging and expensive. Here, we...

The Individual Brain Charting (IBC) is a multi-task functional Magnetic Resonance Imaging dataset acquired at high spatial-resolution and dedicated to the cognitive mapping of the human brain. It consists in the deep phenotyping of twelve individuals, covering a broad range of psychological domains suitable for functional-atlasing applications. Her...

The quest for higher spatial and/or temporal resolution in functional MRI (fMRI) while preserving a sufficient temporal signal-to-noise ratio (tSNR) has generated a tremendous amount of methodological contributions in the last decade ranging from Cartesian vs. non-Cartesian readouts, 2D vs. 3D acquisition strategies, parallel imaging and/or compres...

Explaining the decision process of machine learning algorithms is nowadays crucial for both model’s performance enhancement and human comprehension. This can be achieved by assessing the variable importance of single variables, even for high-capacity non-linear methods, e.g. Deep Neural Networks (DNNs). While only removal-based approaches, such as...

Reducing contributions from non-neuronal sources is a crucial step in functional magnetic resonance imaging (fMRI) connectivity analyses. Many viable strategies for denoising fMRI are used in the literature, and practitioners rely on denoising benchmarks for guidance in the selection of an appropriate choice for their study. However, fMRI denoising...

Purpose Static and dynamic B0$$ {\mathrm{B}}_0 $$ field imperfections are detrimental to functional MRI (fMRI) applications, especially at ultra‐high magnetic fields (UHF). In this work, a field camera is used to assess the benefits of retrospectively correcting B0$$ {\mathrm{B}}_0 $$ field perturbations on Blood Oxygen Level Dependent (BOLD) sensi...

The analysis and understanding of brain characteristics often require considering region-level information rather than voxel-sampled data. Subject-specific parcellations have been put forward in recent years, as they can adapt to individual brain organization and thus offer more accurate individual summaries than standard atlases. However, the pric...

Videogames are emerging as a promising experimental paradigm in neuroimaging. Acquiring gameplay in a scanner remains challenging due to the lack of a scanner-compatible videogame controller that provides a similar experience to standard, commercial devices. In this paper, we introduce a videogame controller designed for use in the functional magne...

The shared response model (SRM) provides a simple but effective framework to analyze fMRI data of subjects exposed to naturalistic stimuli. However, when the number of subjects or runs is large, fitting the model requires a large amount of memory and computational power, which limits its use in practice. Furthermore, SRM is not identifiable, which...

A fundamental question in neurolinguistics concerns the brain regions involved in syntactic and semantic processing during speech comprehension, both at the lexical (word processing) and supra-lexical levels (sentence and discourse processing).To what extent are these regions separated or intertwined? To address this question, we introduce a novel...

Two fundamental questions in neurolinguistics concerns the brain regions that integrate information beyond the lexical level, and the size of their window of integration. To address these questions we introduce a new approach named masked-attention generation. It uses GPT-2 transformers to generate word embeddings that capture a fixed amount of con...

Reducing contributions from non-neuronal sources is a crucial step in functional magnetic resonance imaging (fMRI) analyses. Many viable strategies for denoising fMRI are used in the literature, and practitioners rely on denoising benchmarks for guidance in the selection of an appropriate choice for their study. However, fMRI denoising software is...

A fundamental question in neurolinguistics concerns the brain regions involved in syntactic and semantic processing during speech comprehension, both at the lexical (word processing) and supra-lexical levels (sentence and discourse processing). To what extent are these regions separated or intertwined? To address this question, we trained a lexical...

We consider the inference problem for high-dimensional linear models, when covariates have an underlying spatial organization reflected in their correlation. A typical example of such a setting is high-resolution imaging, in which neighboring pixels are usually very similar. Accurate point and confidence intervals estimation is not possible in this...

In this article we develop a method for performing post hoc inference of the False Discovery Proportion (FDP) over multiple contrasts of interest in the multivariate linear model. To do so we use the bootstrap to simulate from the distribution of the null contrasts. We combine the bootstrap with the post hoc inference bounds of Blanchard (2020) and...

The study of associations between inter-individual differences in brain structure and behaviour has a long history in psychology and neuroscience. Many associations between psychometric data, particularly intelligence and personality measures and local variations of brain structure have been reported. While the impact of such reported associations...

Neural Language Models (NLMs) have made tremendous advances during the last years, achieving impressive performance on various linguistic tasks. Capitalizing on this, studies in neuroscience have started to use NLMs to study neural activity in the human brain during language processing. However, many questions remain unanswered regarding which fact...

Cluster-level inference procedures are widely used for brain mapping. These methods compare the size of clusters obtained by thresholding brain maps to an upper bound under the global null hypothesis, computed using Random Field Theory or permutations. However, the guarantees obtained by this type of inference - i.e. at least one voxel is truly act...

Individual brains vary in both anatomy and functional organization, even within a given species. Inter-individual variability is a major impediment when trying to draw generalizable conclusions from neuroimaging data collected on groups of subjects. Current co-registration procedures rely on limited data, and thus lead to very coarse inter-subject...

Identifying the relevant variables for a classification model with correct confidence levels is a central but difficult task in high-dimension. Despite the core role of sparse logistic regression in statistics and machine learning, it still lacks a good solution for accurate inference in the regime where the number of features $p$ is as large as or...

Associating brain systems with mental processes requires statistical analysis of brain activity across many cognitive processes. These analyses typically face a difficult compromise between scope—from domain-specific to system-level analysis—and accuracy. Using all the functional Magnetic Resonance Imaging (fMRI) statistical maps of the largest dat...

Cluster-level inference procedures are widely used for brain mapping. These methods compare the size of clusters obtained by thresholding brain maps to an upper bound under the global null hypothesis, computed using Random Field Theory or permutations. However, the guarantees obtained by this type of inference - i.e. at least one voxel is truly act...

Background With increasing data sizes and more easily available computational methods, neurosciences rely more and more on predictive modeling with machine learning, e.g., to extract disease biomarkers. Yet, a successful prediction may capture a confounding effect correlated with the outcome instead of brain features specific to the outcome of inte...

Background: Magnetic resonance imaging (MRI) is currently considered a safe imaging technique because, unlike computed tomography, MRI does not expose patients to ionising radiation. However, conflicting literature reports possible genotoxic effects of MRI. We herein examine the chromosomal effects of repeated MRI scans by performing a longitudina...

We consider shared response modeling, a multi-view learning problem where one wants to identify common components from multiple datasets or views. We introduce Shared Independent Component Analysis (ShICA) that models each view as a linear transform of shared independent components contaminated by additive Gaussian noise. We show that this model is...

Background Biological aging is revealed by physical measures, e.g., DNA probes or brain scans. In contrast, individual differences in mental function are explained by psychological constructs, e.g., intelligence or neuroticism. These constructs are typically assessed by tailored neuropsychological tests that build on expert judgement and require ca...

Inter-individual variability in the functional organization of the brain presents a major obstacle to identifying generalizable neural coding principles. Functional alignment—a class of methods that matches subjects’ neural signals based on their functional similarity—is a promising strategy for addressing this variability. To date, however, a rang...

Advances in computational cognitive neuroimaging research are related to the availability of large amounts of labeled brain imaging data, but such data are scarce and expensive to generate. While powerful data generation mechanisms, such as Generative Adversarial Networks (GANs), have been designed in the last decade for computer vision, such impro...

How can neuroimaging inform us about the function of brain structures? This simple question immediately brings out two pertinent issues: Firstly, an inference problem, namely the fact that the function of a region can only be asserted after observing a large array of experimental conditions or contrasts; and second, the fact that the identity of a...

Advances in computational cognitive neuroimaging research are related to the availability of large amounts of labeled brain imaging data, but such data are scarce and expensive to generate. While powerful data generation mechanisms, such as Generative Adversarial Networks (GANs), have been designed in the last decade for computer vision, such impro...

We consider the inference problem for high-dimensional linear models, when covariates have an underlying spatial organization reflected in their correlation. A typical example of such a setting is high-resolution imaging, in which neighboring pixels are usually very similar. Accurate point and confidence intervals estimation is not possible in this...

Cognitive brain imaging is accumulating datasets about the neural substrate of many different mental processes. Yet, most studies are based on few subjects and have low statistical power. Analyzing data across studies could bring more statistical power; yet the current brain-imaging analytic framework cannot be used at scale as it requires casting...

Lesion-behaviour mapping aims at predicting individual behavioural deficits, given a certain pattern of brain lesions. It also brings fundamental insights on brain organization, as lesions can be understood as interventions on normal brain function. We focus here on the case of stroke. The most standard approach to lesion-behaviour mapping is mass-...

In brain imaging, decoding is widely used to infer relationships between brain and cognition, or to craft brain-imaging biomarkers of pathologies. Yet, standard decoding procedures do not come with statistical guarantees, and thus do not give confidence bounds to interpret the pattern maps that they produce. Indeed, in whole-brain decoding settings...

We consider a multi-view learning problem known as group independent component analysis (group ICA), where the goal is to recover shared independent sources from many views. The statistical modeling of this problem requires to take noise into account. When the model includes additive noise on the observations, the likelihood is intractable. By cont...

A key goal in neuroscience is to understand brain mechanisms of cognitive functions. An emerging approach is “brain decoding”, which consists of inferring a set of experimental conditions performed by a participant, using pattern classification of brain activity. Few works so far have attempted to train a brain decoding model that would generalize...

Functional neuroimaging provides the unique opportunity to characterize brain regions based on their response to tasks or ongoing activity. As such, it holds the premise to capture brain spatial organization. Yet, the conceptual framework to describe this organization has remained elusive: on the one hand, parcellations build implicitly on a piecew...

Functional magnetic resonance imaging (fMRI) has opened the possibility to investigate how brain activity is modulated by behavior. Most studies so far are bound to one single task, in which functional responses to a handful of contrasts are analyzed and reported as a group average brain map. Contrariwise, recent data‐collection efforts have starte...

Inter-individual variability in the functional organization of the brain presents a major obstacle to identifying generalizable neural coding principles. Functional alignment—a class of methods that matches subjects’ neural signals based on their functional similarity—is a promising strategy for addressing this variability. At present, however, a r...

We present an extension of the Individual Brain Charting dataset –a high spatial-resolution, multi-task, functional Magnetic Resonance Imaging dataset, intended to support the investigation on the functional principles governing cognition in the human brain. The concomitant data acquisition from the same 12 participants, in the same environment, al...

In the 20th century, many advances in biological knowledge and evidence-based medicine were supported by p values and accompanying methods. In the early 21st century, ambitions toward precision medicine place a premium on detailed predictions for single individuals. The shift causes tension between traditional regression methods used to infer stati...

Detecting where and when brain regions activate in a cognitive task or in a given clinical condition is the promise of non-invasive techniques like magnetoencephalography (MEG) or electroencephalography (EEG). This problem, referred to as source localization, or source imaging, poses however a high-dimensional statistical inference challenge. While...

Background Biological aging is revealed by physical measures, e . g ., DNA probes or brain scans. Instead, individual differences in mental function are explained by psychological constructs, e.g., intelligence or neuroticism. These constructs are typically assessed by tailored neuropsychological tests that build on expert judgement and require car...

We simultaneously revisited the ADI-R and ADOS with a comprehensive data-analytics strategy. Here, the combination of pattern analysis algorithms and an extensive data resources (n=266 patients aged 7 to 49 years) allowed identifying coherent clinical constellations in and across ADI-R and ADOS assessments widespread in clinical practice. Our clust...

Population imaging markedly increased the size of functional-imaging datasets, shedding new light on the neural basis of inter-individual differences. Analyzing these large data entails new scalability challenges, computational and statistical. For this reason, brain images are typically summarized in a few signals, for instance reducing voxel-leve...

Group studies involving large cohorts of subjects are important to draw general conclusions about brain functional organization. However, the aggregation of data coming from multiple subjects is challenging, since it requires accounting for large variability in anatomy, functional topography and stimulus response across individuals. Data modeling i...

Magnetoencephalography and electroencephalography (M/EEG) are non-invasive modalities that measure the weak electromagnetic fields generated by neural activity. Estimating the location and magnitude of the current sources that generated these electromagnetic fields is an inverse problem. Although it can be cast as a linear regression, this problem...

A key goal in neuroscience is to understand brain mechanisms of cognitive functions. An emerging approach is brain decoding, which consists of inferring a set of experimental conditions performed by a participant, using pattern classification of brain activity. Few works so far have attempted to train a brain decoding model that would generalize ac...

Brain networks are increasingly characterized at different scales, including summary statistics, community connectivity, and individual edges. While research relating brain networks to behavioral measurements has yielded many insights into brain‐phenotype relationships, common analytical approaches only consider network information at a single scal...

Population imaging markedly increased the size of functional-imaging datasets, shedding new light on the neural basis of inter-individual differences. Analyzing these large data entails new scalability challenges, computational and statistical. For this reason, brain images are typically summarized in a few signals, for instance reducing voxel-leve...

Reaching a global view of brain organization requires assembling evidence on widely different mental processes and mechanisms. The variety of human neuroscience concepts and terminology poses a fundamental challenge to relating brain imaging results across the scientific literature. Existing meta-analysis methods perform statistical tests on sets o...

Reaching a global view of brain organization requires assembling evidence on widely different mental processes and mechanisms. The variety of human neuroscience concepts and terminology poses a fundamental challenge to relating brain imaging results across the scientific literature. Existing meta-analysis methods perform statistical tests on sets o...

Reaching a global view of brain organization requires assembling evidence on widely different mental processes and mechanisms. The variety of human neuroscience concepts and terminology poses a fundamental challenge to relating brain imaging results across the scientific literature. Existing meta-analysis methods perform statistical tests on sets o...

We develop an extension of the Knockoff Inference procedure, introduced by Barber and Candes (2015). This new method, called Aggregation of Multiple Knockoffs (AKO), addresses the instability inherent to the random nature of Knockoff-based inference. Specifically, AKO improves both the stability and power compared with the original Knockoff algorit...

Reaching a global view of brain organization requires assembling evidence on widely different mental processes and mechanisms. The variety of human neuroscience concepts and terminology poses a fundamental challenge to relating brain imaging results across the scientific literature. Existing meta-analysis methods perform statistical tests on sets o...

Magnetoencephalography and electroencephalography (M/EEG) are non-invasive modalities that measure the weak electromagnetic fields generated by neural activity. Estimating the location and magnitude of the current sources that generated these electromagnetic fields is a challenging ill-posed regression problem known as \emph{source imaging}. When c...

We simultaneously revisited the ADI-R and ADOS with a comprehensive data-analytics strategy. Here, the combination of pattern analysis algorithms and an extensive data resources (n=266 patients aged 7 to 49 years) allowed identifying coherent clinical constellations in and across ADI-R and ADOS assessments widespread in clinical practice. The colle...

The shared response model provides a simple but effective framework toanalyse fMRI data of subjects exposed to naturalistic stimuli. However whenthe number of subjects or runs is large, fitting the model requires a large amountof memory and computational power, which limits its use in practice. In thiswork, we introduce the FastSRM algorithm that r...

Schizophrenia is a devastating brain disorder that disturbs sensory perception, motor action, and abstract thought. Its clinical phenotype implies dysfunction of various mental domains, which has motivated a series of theories regarding the underlying pathophysiology. Aiming at a predictive benchmark of a catalog of cognitive functions, we develope...

Purpose A calibration‐free pulse design method is introduced to alleviate B1+ artifacts in clinical routine with parallel transmission at high field, dealing with significant inter‐subject variability, found for instance in the abdomen. Theory and Methods From a dual‐transmit 3T scanner, a database of B1+ and off‐resonance abdominal maps from 50 s...

An important goal of cognitive brain imaging studies is to model the functional organization of the brain; yet there exists currently no functional brain atlas built from existing data. One of the main roadblocks to the creation of such an atlas is the functional variability that is observed in subjects performing the same task; this variability go...

Continuous improvement in medical imaging techniques allows the acquisition of higher-resolution images. When these are used in a predictive setting, a greater number of explanatory variables are potentially related to the dependent variable (the response). Meanwhile, the number of acquisitions per experiment remains limited. In such high dimension...

More than two decades of functional magnetic resonance imaging (fMRI) of the human brain have succeeded to identify, with a growing level of precision, the neural basis of multiple cognitive skills within various domains (perception, sensorimotor processes, language, emotion and social cognition …). Progress has been made in the comprehension of th...

Magnetoencephalography (MEG) and electroencephalography (EEG) are non-invasive modalities that measure the weak electromagnetic fields generated by neural activity. Inferring the location of the current sources that generated these magnetic fields is an ill-posed inverse problem known as source imaging. When considering a group study, a baseline ap...

Functional connectomes reveal biomarkers of individual psychological or clinical traits. However, there is great variability in the analytic pipelines typically used to derive them from rest-fMRI cohorts. Here, we consider a specific type of studies, using predictive models on the edge weights of functional connectomes, for which we highlight the b...

Estimating covariances from functional Magnetic Resonance Imaging at rest (r-fMRI) can quantify interactions between brain regions. Also known as brain functional connectivity, it reflects inter-subject variations in behavior and cognition, and characterizes neuropathologies. Yet, with noisy and short time-series, as in r-fMRI, covariance estimatio...

Magnetoencephalography (MEG) and electroencephalogra-phy (EEG) are non-invasive modalities that measure the weak electromagnetic fields generated by neural activity. Inferring the location of the current sources that generated these magnetic fields is an ill-posed inverse problem known as source imaging. When considering a group study, a baseline a...

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 ov...

Distribution of terms in our database. (PDF)

Maps for consensus between forward and reverse. Left: maps for the different inferences on the “place” concept. Right: the overlaid inferences for this concept. The consensus singles out the PPA for the “place” concept. (TIFF)

Prediction scores for different methods. AUC (area under the curve) of the ROC. OD: ontology decoding, LOG: logistic, NB: Naive Bayes. Left: leave-subject-out cross-validation, Right: leave-study-out cross-validation. (TIFF)

Prediction scores for different methods. AUC (area under the curve) of the ROC curve. OD: ontology decoding, LOG: logistic regression, NB: Naive Bayes, NS: NeuroSynth. The OD (ontology decoding) method performs very well (chance is at .5), including when predicting to new studies. Leave-subject-out cross-validation scheme tend to display a higher p...

Forward analysis: Ontology-based design across studies. (PDF)

Reverse inference: Decoding with cognitive ontologies. (PDF)

Terms correlations. Correlation matrix between terms across images. (TIFF)

First-level classifiers used. We train three types of classifier to learn the hierarchy of terms: category classifiers (with a OvA approach), and terms classifiers (both with OvA and OvO approaches). The classifiers’ decision functions span an intermediate feature space tailored to our ontology, upon which we perform a standard OvA approach to pred...

Terms and categories we use to characterize tasks associated with images in our database. We used CogPO categories for task-related description, and add necessary terms from Cognitive Atlas to describe higher-level cognitive aspect. Here we report only terms that were present in more than one study—aside from the “left foot”, which maps in the anal...