Visitors Talk: Jean-Baptiste Poline, McGill University, Canada

On Monday, 27 October, CIBM was delighted to host Professor Jean-Baptiste Poline from the Montreal Neurological Institute, McGill University, for the Visitors Talk entitled “Changing landscape for neuroimaging data sharing and data processing: towards distributed solutions.”

The session, hosted by Associate Professor Oscar Esteban (HES-SO University of Applied Sciences and Arts Western Switzerland), brought together around 50 researchers both in person at the University of Geneva and online to discuss the current challenges and opportunities in data sharing, reproducibility, and federated approaches in neuroimaging research.

Rethinking How We Share and Process Data

In his talk, Prof. Poline addressed a central issue in modern neuroscience: how to ensure that data analysis pipelines and findings remain reproducible, generalizable, and ethically compliant, even when data are distributed across different institutions and subject to diverse legal constraints. Drawing from his experience as Director of the ORIGAMI Neuro-Data-Science Laboratory at McGill, Prof. Poline showcased several neuroinformatics projects developed to support open science, including initiatives such as Neurobagel and Nipoppy, which promote distributed analysis and community-based data standards.

One of the key themes of the presentation was the transition from centralized data repositories to distributed infrastructures that enable collaboration without compromising data privacy. Prof. Poline highlighted that such systems, when governed transparently and sustainably, can facilitate federated learning approaches, particularly in sensitive domains such as the development of imaging biomarkers for Parkinson’s Disease.

Take-Home Messages: Building Sustainable and Inclusive Communities

The final part of the talk summarized the lessons learned from decades of neuroinformatics collaboration:

• Build standards and communities of practice to ensure interoperability and knowledge sharing.
• Be sustainable – scalable, modular, portable, and user-oriented systems are key.
• Avoid lock-in and ensure freedom to evolve tools and workflows.
• Foster transparent and distributed governance for long-term trust and efficiency.
• Build communities of practice and consider all incentives: love, fame, money, and rules, recognizing the human and social dimensions behind technological innovation.

This engaging reflection resonated with the audience, underscoring the need to combine technical excellence with community building to achieve reproducible and impactful science.