UHN partners on data platform with Mayo Clinic

TORONTO – UHN has become a founding member of Mayo Clinic’s Platform_Connect, a global distributed data network designed to enable the creation of new healthcare solutions driven by data science and artificial intelligence.

“Teamwork, technology and bold thinking are essential to transforming the ways that healthcare is delivered and putting patients first – goals UHN shares with Mayo Clinic,” Dr. Kevin Smith (pictured), UHN’s president & CEO, said after the agreement was signed at the Mayo campus in Rochester, Minn.

Alongside UHN, Hospital Israelita Albert Einstein in Brazil and Sheba Medical Center in Israel on May 4 joined with Mercy and the Mayo Clinic in the United States as founding members. It is a first-of-its-kind global alliance among large healthcare systems that transcends language and other barriers.

Mayo Clinic’s Platform_Connect provides a solution to create secure, cloud-based access to de-identified clinical data generated by network members across three continents. The platform will enable federated learning that keeps “Data Behind Glass” in member institutions, while unlocking discovery, validation and implementation of new data driven algorithms informed by AI.

This approach creates new pathways for collaboration and at the same time ensures that each healthcare system maintains control over its de-identified data.

“We describe the data needed for fair, equitable AI as having depth (types of information), breadth (number of patients) and spread (heterogeneity),” says Dr. John Halamka, president of Mayo Clinic Platform. “If we are to transform healthcare globally, we must expand our distributed data networks for machine learning to every continent.”

“We must protect privacy, adhere to international laws/regulations, and incorporate knowledge from every language. Today, three premier medical centres in South America, Canada, and the Middle East are joining our network.”

The alliance will initially focus on discovery of new solutions to improve patient outcomes:

• Information collaboration – Secure cloud-based use of “Data Behind Glass” allows each collaborator to base decisions on a wider range of clinical outcomes gathered over time. The information will help scientists analyze patterns of effective disease treatment taking into account the diversity amongst patient populations and care team approaches.
• Solution and algorithm development, validation and deployment – The resulting AI algorithms will provide proven treatment paths based on years of patient outcomes, representing the next generation of proactive and predictive medicine that can be used by care providers around the world to access best practices in medical care.

Until recently, complex types of medical information such as integrated electronic health records data, radiology and pathology images, videos, and clinical notes were too unstructured and complex to share across institutions.

With the combination of privacy-protected, cloud-based storage and the growth of AI and machine learning, Mayo Clinic Platform and the members of Connect will use aggregated clinical de-identified data to generate patterns to find diseases earlier and start patients on the path to be health faster.

“We are thrilled to be part of a collaboration of this unprecedented scale and breadth,” says Dr. Brad Wouters, UHN’s executive vice president, Science and Research. “This alliance is the result of UHN and Mayo Clinic’s shared vision of driving excellence in discovery and innovation to improve the lives of patients.

“By unlocking AI-enabled healthcare solutions, while protecting patient data, this alliance will be key to accelerating the translation of discovery to practice.”