In Brief

I am a researcher at NeuroSpin CEA-Saclay, France. I have a Ph.D. in the field of medical image processing awarded by University of Strasbourg in 2013. Since 2014 I’m in charge of the Neurospin analysis platform. My interests include computer science, ultra-high field MRI and its application to neuroscience using various imaging techniques such as diffusion MRI for probing structural brain connectivity. I have skilled experience in the installation and configuration of operating systems and applications software related to High Performance Computing (HPC) and parallel architectures. More recently I have been actively involved in challenging software research projects focused on applying Deep Learning technique to neuroimaging and genetic data. I am also in charge of developping cloud services to collect, process and share heterogeneous, and complex data as part of the IMAGEN, cVEDA, STRATIFY, EU-AIMS, or R-LINK european or international projects.

The NeuroSpin Analysis Platform

Health research strategies using neuroimaging have shifted in recent years: the focus has moved from patient care only, to a combination of patient care and prevention. In the case of neurodegenerative and psychiatric diseases, this drives the creation of increasingly numerous massive imaging studies also known as Population Imaging (PI) surveys. It should be noticed that PI studies no longer consist of image data only. The recent wide availability of high-throughput genomics has augmented the subject data with genetics, epigenetics, and functional genomics. In this context, the challenge for NeuroSpin is to be able to process individual, high-quality datasets with advanced algorithms as well as large heterogenous, possibly multi-center datasets.

The NeuroSpin data Analysis Platform (NSAp) was born in 2014 with the objective to develop imaging genetic offering and services with high added support for the research conducted in NeuroSpin. The aim of this innovative platform is to propose tools, methods, and services to the researchers with focuses on the concept of reproducible research. The need for reproducibility is increasing dramatically as data analyses become more complex, involving larger datasets and more sophisticated computations. The NeuroSpin data Analysis Platform has leveraged softwares to process neuroimaging gentic datasets in a reproducible fashion, to simplify the access to available high-performance computing (HPC) resources, to manage the incoming platform requests with a web service while ensuring more transparency and better management, and to handle the data generated by a project using web semantic technologies and emerging ontologies for the data collection, moderation, and publication. The platform regularly evaluates its range of services and changes them to suit current research needs by integrating new state of the art processing pipelines or data managment tools. Since its creation, I am in charge of this platform.

Currently the NeuroSpin data Analysis Platform is involved in three major European projects, IMAGEN (, EU-AIMS (, and more recently R-link, where its principal contributions are data processing, and data collection, moderation, and sharing using web services. The NeuroSpin data Analysis Platform also supports the local research by processing international cohorts (i.e., HCP, UKB, HBN, …) or data acquired within the laboratory using all the plaftorm expertise.