Synthetic data project attracts multinational corporation to Alberta

A project aiming to demonstrate how synthetic data can be useful for the healthcare industry has gained a new partner in Merck Canada, a Quebec-based pharmaceutical company.

Health City and its partners have been working on the Synthetic Data project since last fall — which it says is the first of its kind in Canada. Merck Canada's engagement also marks the first time a multinational corporation has invested in a synthetic data project in Alberta.

Synthetic data is information that is artificially generated using algorithms. Synthetic databases don't use real patient health information, which Health City CEO Reg Joseph said is important because these databases can be shared freely among innovators and researchers without raising patient privacy concerns.

"Before we engaged the project with Merck Canada, we actually launched an academic version of the project. We were able to demonstrate that we could create a synthetic database, prove that it's useful and that the data was safe," Joseph told Taproot.

"Positioning Edmonton as sort of the first mover on this is interesting because it lends itself now towards more industrial relationships in the health sector if we can be one of the first jurisdictions to demonstrate how to use data safely in this kind of environment."

The project is a collaboration between Health City, the School of Public Health at the University of Alberta, the Insitute of Health Economics (IHE), Alberta Innovates, and Replica Analytics.

Multinational corporation Merck Canada has joined the Alberta-based Synthetic Data project. (Courtesy of Unsplash/Carlos Muza) Multinational corporation Merck Canada has joined the Alberta-based Synthetic Data project. (Courtesy of Unsplash/Carlos Muza)

"We looked at multiple databases and then we took a number of aggregate characteristics of the real data. In those aggregate characteristics, there's no personalized information at all," said Joseph. "Then we literally synthesized fake data to match those aggregate characteristics. As a result, we found that there's actually utility in the synthetic data."

Joseph told Taproot that one of the great strengths of synthetic data is that it enables artificial intelligence and machine learning.

"Most health research projects are hypothesis-driven. We have a world-renowned AI facility, talent, and capability, but because it doesn't do hypothesis-driven research, it actually can't get access to health data at all. So, the reason why we went down the track of synthetic data was to enable AI."

Joseph said this project allows Health City and its partners to help small and medium-sized enterprises in Alberta, so they can get their innovations to market sooner.

"What Merck Canada is helping us to do is bring a real-life industry case to the table, and see if synthetic data can work. If it does work, that will give us a lot more confidence to be able to show our (enterprises) how synthetic data works."

According to Health City, validating synthetic datasets and establishing opportunities for their usage creates further opportunities for innovators to work alongside the health system while preserving patient privacy.

"Working together through innovative collaborations such as this one will help facilitate access to health information critical to scientific advancement, ultimately helping improve Canadian health outcomes," said Heidi Waser, the executive director of patient access at Merck Canada.