
Francis Jeanson
Head of the Centre for Analytics (CfA), Ontario Brain Institute

Azadeh Kushki
Associate Chief of Data Science, Bloorview Research Institute

Jean-Baptiste Poline
Professor & Associate Professor, McGill University

Erin Dickie
Staff Scientist, Krembil Centre for Neuroinformatics, CAMH
The Ontario Brain Institute is enabling privacy-enhancing AI collaboration in brain health research, positioning Canada as a global leader.
Brain research is complex and requires massive amounts of data, diverse patient populations, and collaboration across institutions. Federated learning — a technique that sends AI models directly to data rather than trying to move data to a central repository — addresses the need for data privacy and sovereignty and reduces the friction that differing privacy and governance rules across jurisdictions can cause.
Instead of having one-size-fits-all approaches based on broad diagnostic labels, we want to be able to tailor things to each individual’s biology.
The Ontario Brain Institute (OBI) is leading the way with its NeuroFL federated learning platform. “Federated learning is particularly valuable for brain health research because you can increase your model’s precision and decrease its bias with access to a greater set of data,” says Francis Jeanson, Head of the Centre for Analytics (CfA) at OBI.
Developing personalized, proactive approaches
“Instead of having one-size-fits-all approaches based on broad diagnostic labels, we want to be able to tailor things to each individual’s biology,” says Azadeh Kushki, Associate Chief of Data Science at the Bloorview Research Institute.
Canada is well-positioned to lead in secure and privacy-conscious AI innovation in health care. “We have a strong distributed computing infrastructure and the capacity to experiment locally, with our different provincial legislative environments,” says Jean-Baptiste Poline, Professor in the Department of Neurology and Neurosurgery and Associate Professor at the School of Computer Science at McGill University.
“Collaborations across Canadian hospitals could bring a lot of movement forward in this space,” says Erin Dickie, Staff Scientist at the Krembil Centre for Neuroinformatics at CAMH.
Learn more about the Centre for Analytics at www.braininstitute.ca/research-data-sharing/centre-for-analytics
Shortly after this article went to print, the Honourable Evan Solomon (Federal Minister of AI) announced up to $100 million in funding to unlock Canada’s healthcare data through a federated learning network. This initiative, called Vital, is led by St. Michael’s Hospital.