Jean-Baptiste Poline
Research Interests
Make your next move.
Prepare your application
Review drafts before you send them.
Explore related searches
Articles (15)
Numerical variability is rarely quantified in neuroimaging despite many measures relying on subtle morphometric differences across individuals. We instrumented FreeSurfer, a widely used neuroimaging pipeline, to simulate numerical differences across computational environments, and used it to measure numerical variability in MRI analyses of Parkinson’s disease patients and controls. In multiple cortical and subcortical regions, numerical variation reached nearly one-third of the population variability, altering statistical conclusions about group differences and clinical associations. To assess the impact of numerical noise in existing studies, we developed a practical tool that estimates the Numerical-Population Variability Ratio (NPVR) in a study, and propagates the resulting numerical uncertainty to common statistics and associated p-values. By applying this framework to thirteen previously published studies reporting MRI measures in Parkinson’s disease, we quantified the probability of numerically induced false positives and false negatives in the literature, highlighting a substantial impact of numerical variability on MRI measures of Parkinson’s disease. These results underscore the importance of systematically evaluating numerical stability in neuroimaging and provides a practical framework to do so.
Year:
2026
Genetic studies have advanced our understanding of Parkinson's disease (PD) pathogenesis, establishing a role for autophagy and lysosomal dysfunction. However, how genetic risk translates into neuronal vulnerability remains mostly unknown. Using recent genome-wide association studies and neuroimaging data from UK Biobank, we show that higher polygenic risk score of PD correlates with greater cortical surface area, white matter fractional anisotropy and subcortical volumes. Mendelian randomization supports a causal relation from increased brain size to PD, and cortical regions showing the greatest polygenic expansion in surface area show the greatest atrophy in PD. Lifespan gene expression and pathway-specific analyses identify autophagy-lysosomal and neurodevelopmental pathways as separate mechanisms of vulnerability. We show that a portion of PD susceptibility originates from neurodevelopmental processes that regulate neuronal proliferation. Genetically-determined increases in brain size confer vulnerability to PD in later life, supporting the existence of shared neurobiological pathways between brain development and neurodegeneration.
Year:
2025
Recent work suggests that thousands of individuals are required in multivariate brain-behaviour analyses to obtain consistently replicable results. Some believe, however, that smaller sample sizes may be sufficient if specific subpopulations are targeted. We investigated how sample size and cohort composition influence the replicability of Canonical Correlation Analysis (CCA) results using the UK Biobank (N=40,514). We applied CCA to diffusion-weighted magnetic resonance imaging (dMRI) phenotypes and cognitive assessment test scores. We defined four participant cohorts based on clinical profile and found that, across all cohorts, sample sizes of around 500 were needed to obtain replicable canonical correlations and variable loadings. The most targeted cohort required much fewer samples to achieve similar or greater correlations than the other cohorts. Variable loadings were consistent between sample sizes of ∼500 to thousands, suggesting that sample sizes in the order of hundreds may be sufficient for obtaining reliable CCA results.
Year:
2025
The adoption of open science principles for data has been instrumental in enabling the creation and reuse of large datasets and efficient collaboration, but requires a dedicated open data sharing infrastructure. Although clinical neuroscience research increasingly involves large-scale collaboration on controlled, sensitive data, open science principles can bring key benefits in this setting. The unique privacy and access requirements of clinical research create challenges that require a specialized data infrastructure, focused on local governance. Here we present a narrative account of a pilot project to establish such an infrastructure for the data of the ENIGMA Parkinson’s Disease working group by adopting the decentralized Nipoppy and Neurobagel tool stack. Our collaboration has led to the adoption of community data standards among international sites of the working group and has launched an open portal that enables the cross-site discovery of cohorts based on harmonized metadata. We discuss our approach to this collaboration, the social and technical challenges we have encountered and how we addressed them in the hope that groups in similar situations may benefit from our experiences.
Year:
2025
Researchers in brain imaging have access to a multitude of analysis tools, many of which carry out the same or similar tasks but yield different results when applied to the same data. This analytical flexibility often undermines reproducibility and raises concerns about the robustness of neuroimaging studies. However, the array of software packages to investigate and address analytical flexibility is decentralized, scattered, and not well documented. Consequently, researchers often lack the necessary information and protocols to buttress the reliability of their findings across analytical tools. This review catalogs and describes software platforms (i.e., software or computational libraries) that can be used to address result variability arising from computational pipelines and environments and explores the use of computing platforms and neuroimaging pipeline frameworks in addressing this issue. This study offers guidance to the research community on accessing, understanding, and utilizing these platforms to address brain imaging analytical flexibility. Additionally, the article provides specific recommendations tailored to different user groups, considering the tools they intend to use with these platforms and their computational constraints.
Year:
2025
Predicting cognitive processes from brain activation maps has remained an open question within the neuroscience community for many years. Meta-analytic functional decoding methods aim to tackle this issue by providing a quantitative estimation of behavioral profiles associated with specific brain regions. Existing methods face intrinsic challenges in neuroimaging meta-analysis, particularly in consolidating textual information from publications, as they rely on limited metrics that do not capture the semantic context of the text. The combination of large language models (LLMs) with advanced deep contrastive learning models (e.g., CLIP) for aligning text with images has revolutionized neuroimaging meta-analysis, potentially offering solutions to functional decoding challenges. In this work, we present NiCLIP, a contrastive language-image pretrained model that predicts cognitive tasks, concepts, and domains from brain activation patterns. We leveraged over 23,000 neuroscientific articles to train a CLIP model for text-to-brain association. We demonstrated that fine-tuned LLMs (e.g., BrainGPT models) outperform their base LLM counterparts. Evaluation of NiCLIP predictions revealed that performance is optimized when using full-text articles instead of abstracts, as well as a curated cognitive ontology with precise task-concept-domain mappings. Our results indicated that NiCLIP accurately predicts cognitive tasks from group-level activation maps provided by the Human Connectome Project across multiple domains (e.g., emotion, language, motor) and precisely characterizes the functional roles of specific brain regions, including the amygdala, hippocampus, and temporoparietal junction. However, NiCLIP showed limitations with noisy subject-level activation maps. NiCLIP represents a significant advancement in quantitative functional decoding for neuroimaging, offering researchers a powerful tool for hypothesis generation and scientific discovery.
Year:
2025
Researchers in brain imaging have access to a multitude of analysis tools, many of which carry out the same or similar tasks but yield different results when applied to the same data. This analytical flexibility often undermines reproducibility and raises concerns about the robustness of neuroimaging studies. However, the array of software packages to investigate and address analytical flexibility is decentralized, scattered, and not well documented. Consequently, researchers often lack the necessary information and protocols to buttress the reliability of their findings across analytical tools. This review catalogs and describes software platforms (i.e., software or computational libraries) that can be used to address result variability arising from computational pipelines and environments and explores the use of computing platforms and neuroimaging pipeline frameworks in addressing this issue. This study offers guidance to the research community on accessing, understanding, and utilizing these platforms to address brain imaging analytical flexibility. Additionally, the article provides specific recommendations tailored to different user groups, considering the tools they intend to use with these platforms and their computational constraints.
Year:
2025
Image-based meta-analysis (IBMA) is a powerful method for synthesizing results from various fMRI studies. However, challenges related to data accessibility and the lack of available tools and methods have limited its widespread use. This study examined the current state of the NeuroVault repository and developed a comprehensive framework for selecting and analyzing neuroimaging statistical maps within it. By systematically assessing the quality of NeuroVault’s data and implementing novel selection and meta-analysis techniques, we demonstrated the repository’s potential for IBMA. We created a multi-stage selection framework that includes preliminary, heuristic, and manual image selection methods. We conducted meta-analyses for three distinct domains: working memory, motor, and emotion processing. The results from the three manual IBMA approaches closely resembled reference maps from the Human Connectome Project. Importantly, we found that while manual selection provides the most precise results, heuristic methods can serve as robust alternatives, especially for domains that include a heterogeneous set of fMRI tasks and contrasts, such as emotion processing. Additionally, we evaluated five different meta-analytic estimator methods to assess their effectiveness in handling spurious images. For domains characterized by heterogeneous tasks, employing a robust estimator (e.g., median) is essential. This study is the first to present a systematic approach for implementing IBMA using the NeuroVault repository. We introduce an accessible and reproducible methodology that allows researchers to make the most of NeuroVault’s extensive neuroimaging resources, potentially fostering greater interest in data sharing and future meta-analyses utilizing NeuroVault data.
Year:
2025
Meta-analyses are invaluable tools for navigating the rapidly expanding scientific literature. Given their high value, ensuring the quality of meta-analyses is paramount. We conducted a multifaceted overview, examining each step in a manual neuroimaging meta-analysis on a large scale. We used four novel datasets comprising over 14,000 papers, including fMRI meta-analyses, fMRI studies, studies included in meta-analyses, and studies associated with image data on NeuroVault. Regarding successes, two-thirds of meta-analyses stated that they followed PRISMA guidelines, and 65% included a flowchart describing their inclusion process. We point out several areas for improvement. Pre-registration was fairly rare (20%), and only half listed their exact search strategy. There could be a location bias in which papers are included, and many did not include enough studies to be robust against publication bias (68% of meta analyses have less than 30 studies included). We also offer ideas for future directions. As image based meta-analysis is the gold standard, we have indicated which topics have the most image data available. The potential redundancy of topics can be visualized in our paper, and we recommend future meta-analyses be in conversation with past ones by citing and discussing previous similar work. By addressing these findings, the neuroimaging community can collectively improve the field of neuroimaging meta-analyses.
Year:
2025
Automated analysis of the biomedical literature ( literature mining ) offers a rich source of insights. However, such analysis requires collecting a large number of articles and extracting and processing their content. This task is often prohibitively difficult and time-consuming. Here, we provide tools to easily collect, process, and annotate the biomedical literature. In particular, https://neuroquery.github.io/pubget/pubget.html is an efficient and reliable command-line tool for downloading articles in bulk from PubMed Central, extracting their contents and metadata into convenient formats, and extracting and analyzing information such as stereotactic brain coordinates. https://jeromedockes.github.io/labelbuddy/labelbuddy/current/ is a lightweight local application for annotating text, which facilitates the extraction of complex information or the creation of ground-truth labels to validate automated information extraction methods. Further, we describe repositories where researchers can share their analysis code and their manual annotations in a format that facilitates reuse. These resources can help streamline text mining and meta-science projects and make text mining of the biomedical literature more accessible, effective, and reproducible. We describe a typical workflow based on these tools and illustrate it with several example projects.
Year:
2025
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Collaborators (18)
Georgios D. Mitsis
McGill University
Yashar Zeighami
Assistant Professor
McGill University
Nikhil Bhagwat
McGill University
Ziv Gan-Or
Montreal Neurological Institute and Hospital
Thomas E. Nichols
University of Oxford
Ysbrand van der Werf
Professor of Functional Neuroanatomy
Amsterdam UMC
Celia Greenwood
McGill University
Neda Jahanshad
University of Southern California
Bratislav Misic
Associate Professor
McGill University
Chris Vriend
assistant professor
VU University Medical Center
Maryann Martone
Health Science Specialist
San Francisco VA Medical Center
Filip Morys
Research Associate
McGill University
Tim D van Balkom
Health care psychologist
Amsterdam University Medical Centers
Julio A Peraza Goicolea
Florida International University
Alejandro de la Vega
Research Assistant Professor
University of Texas at Austin
Angela Laird
Distinguished University Professor
Florida International University
Franco Pestilli
Professor
University of Texas at Austin
Alain Dagher
Professor
McGill University
Make your next move.
Prepare your application
Review drafts before you send them.