Artificial intelligence
Imagia partners with top Canadian and US hospitals
December 18, 2019
MONTREAL – Imagia, a leading healthcare AI company that unites healthcare expertise and advanced AI technology to improve patient outcomes, has announced EVIDENS partnerships with the following major North American hospitals: Center hospitalier universitaire de Québec – Université Laval, Centre intégré universitaire de santé et de services sociaux de l’Estrie – Centre hospitalier universitaire de Sherbrooke, Jewish General Hospital (JGH), the McGill University Health Centre (MUHC) and its Research Institute (RI-MUHC), Penn Medicine, and the University of Montreal Hospital Centre (CHUM).
“We are building the largest international AI-driven clinical collaboration ecosystem to discover new paths to treatment,” said Imagia CEO Frédéric Francis (pictured). “Our latest hospital partners across North America will transform once static, siloed raw data into valuable sources of insight. Research collaborators joining our ecosystem can now access the ingredients needed to unleash the next wave of healthcare innovation that will lead to truly personalized healthcare.”
The partnerships empower clinicians to create cohorted and structured data, live from hospital systems, to provide a truly data-driven approach to AI-powered clinical research. EVIDENS members unlock institution-wide patient data access, empowering clinicians to make use of data from live hospital systems. This further enables the potential for AI-driven discoveries in personalized care pathways and treatment plans.
The Imagia EVIDENS AI platform ingests and indexes real-world data, such as labs, reports, and images, to design evidence-based clinical studies. Clinicians can now build patient cohorts associated with diagnostic and treatment responses and annotate them to achieve AI-friendly data.
Ultimately, the platform enables clinical researchers to learn from distributed patient data and scale up traditional discovery processes through federated learning capabilities while preserving privacy.
“We are the first to combine automated AI model discovery and federated learning in a hospital setting for researchers to undertake different discovery processes and validate clinical intuitions, without requiring a team of AI experts by their side,” said Imagia CTO Florent Chandelier.
“These functionalities will be available in the EVIDENS platform, and are designed to generalize to a broad range of diseases and data sources. As a use case of reproducible science, we demonstrated that by federating automatic model discovery, and we achieved 90% accuracy for the complex task of detecting lung cancer in 3D CT scans from a public dataset.”
By federating the knowledge extracted by AI algorithms from within the Imagia ecosystem, the EVIDENS platform trains AI models on the decentralized data sets. The growing and continually updated inputs from hospital partners significantly improve model discovery and robustness with each new EVIDENS onboarding.
EVIDENS member healthcare organizations CHUM, Center hospitalier universitaire de Québec, CIUSSS de l’Estrie – CHUS, JGH, MUHC, and Penn Medicine represent potential access to over 4 million patients treated annually, representing a solid foundation for understanding how evidence-based research may transition and be adopted into routine clinical practice.
“Federated Learning is one of the key drivers for multi-site collaborative machine learning and enables discoveries at scale in the healthcare industry,” said Yoshua Bengio, founder and scientific director of MILA and Imagia Advisor.
“Eliminating the need to share raw data beyond each site yet being able to build models derived from all the sites will translate into new discoveries,” Dr. Bengio added. “While further research is necessary to ensure patient privacy in edge cases of physical server breach, appropriate safeguards currently exist through encryption. This ensures that clinicians around the world can benefit from shared insights that will foster open innovation and save lives.”
“Leveraging these kinds of platforms is important for enhancing patient outcomes and investigating new patient management strategies,” said Dr. Benoit Desjardins, MD, PhD, FAHA, FACR, Associate Professor of Radiology in the Perelman School of Medicine at the University of Pennsylvania. “By cohorting the radiology data and images, the research collaboration enables us to find unique insights derived not only in our own datasets, but across other hospitals.”
About Imagia
Imagia brings together healthcare expertise and advanced artificial intelligence technology to achieve better patient outcomes. Through its ecosystem, Imagia enables hospitals, medical device manufacturers, pharmaceutical companies, and diagnostic companies around the world to access and utilize healthcare data. Headquartered in Montreal, Imagia’s mission is to unleash the potential of all data across all organizations to achieve collaborative, medical breakthroughs. For more information, visit www.imagia.com.