Nov 26 2018

Accelerating health research in the Data Safe Haven

A new partnership will help researchers and doctors apply cutting-edge technologies such as machine learning and high-performance computing to gain valuable insights from Ontario’s population-wide data sets.

The Institute for Clinical Evaluative Services (ICES), HPC4Health and Compute Ontario have launched a pilot initiative called the Ontario Data Safe Haven (ODSH), a secure private cloud that will provide access to a wide range of data sets dating back to 1991 and include as many as 13 million linkable health records.

“It’s the first time we’ve been able to access data held by ICES in a secure, high-performance computing environment,” says Michael Brudno, who is leading the ODHS pilot project. Brudno is a professor in the Department of Computer Science at the University of Toronto. He is also the director of the Centre of Computational Medicine at the Hospital for Sick Children and scientific director of the HPC4Health Consortium. Brudno’s focus is the use of high-performance computing to analyze the data sets of large numbers of electronic medical records (EMRs).

“The goal is to conduct analysis on the data and learn. For example, we can look at anonymized data of patients who have had frequent hospital visits and investigate what makes them similar and different from those in the general population. This can help us better prepare and plan future health care processes and policies.

“Data Safe Haven is a proof of concept that you can bring sensitive, protected data sets into an environment where you can link them with other sensitive data and perform computations on them,” he says.

The ICES Data Repository consists of publicly funded administrative health services records for the Ontario population. The data captured include patient-level administrative data from both health and non-health government ministries and agencies, and clinical data captured from hospitals. Other data include research studies and biometric and biologic data that broadly represent the Ontario population. This pilot project will empower researchers to post, access and analyze data and build linkages that could lead to insights around more efficient, economical and appropriate health care for Ontarians.

“Health care resources are precious and need to be properly managed across the province. By identifying the heaviest users of the system, we can make sure their needs are satisfied adequately and free up resources. Hopefully, this will result in shorter wait times at the doctor’s office and in hospital waiting rooms and emergency departments,” explains Brudno.

“Compute Ontario, one of the executive sponsors of this project, wants to better understand how computing can be applied to sensitive data. Compute Ontario’s CEO and President, Nizar Ladak has been phenomenal in supporting us, both financially and as a member of our steering committee.”

The collaborative team includes Michael Schull, Alison Paprica, Carl Virtanen and Rob Naccarato. The project is funded by Compute Ontario, the Province of Ontario, the Canadian Foundation for Innovation, the SickKids Foundation, the University Health Network and Princess Margaret Cancer Foundation, and ICES.