«Changing landscape for neuroimaging data sharing and data processing: towards distributed solutions»
27.10.2025 15:00 – 16:00
In this presentation, I will review some of the challenges that neuroscience and neuroimaging are facing, and in particular, the challenges of reproducible and generalizable studies in the context of data distributed and under variable legal and ethical constraints. The field of neuroscience is facing new questions on the impact of variable processing and the difficulty of generalizing results. These challenges have both technical and sociological roots, and the solutions will likely also be both technological and sociological. I will provide examples of neuroinformatics projects developed in my laboratory or in collaboration to transform how we share and analyze data, with a specific example being the development of imaging biomarkers for Parkinson’s Disease. I will show preliminary results showing that distributed infrastructures can be leveraged for federated learning. I will open the discussion on some key aspects to consider, for instance, how these infrastructures can be maintained, how their governance structure should be thought out, and how this offers opportunities to build a community of practices.
Lieu
Bâtiment: CMU
Auditoire Alex‐F. Müller
Possibilité de suivre l'événement par Zoom:
https://epfl.zoom.us/meeting/register/7XogAcTDQWaniJWKdxqnSg
(Meeting ID: 699 7585 9085)
Organisé par
Département de radiologie et informatique médicaleCenter for Biomedical Imaging (CIBM)
Intervenant-e-s
Jean-Baptiste POLINE, Professor, Montreal Neurological Institute, McGill University, Faculty of Medicine and Health Sciences, Department of Neurology and Neurosurgery and School of Computer Sciencesentrée libre
Classement
Catégorie: Séminaire
Mots clés: neuroimagerie, Neuroscience, reproductibilité, Data Science, parkinson

haut