Alan Bernstein is the president and CEO of the Canadian Institute of Advanced Research. Steven J. Hoffman is the scientific director of the Canadian Institutes of Health Research's Institute of Population & Public Health and a professor of Global Health, Law and Politics at York University.
Chances are this winter you've already had the aches, fever and exhaustion that typically accompany the flu, or at least you know people who have. As this becomes one of the worst flu seasons on record, public-health agencies around the world are working to monitor and track the spread of the virus. But it's a difficult process.
Artificial intelligence will soon help us undertake this task better, faster and cheaper. Canadian researchers are developing AI systems that monitor social media, calls to health lines, emergency-room visits and other real-time data sources, so that computer algorithms can identify flu outbreaks much earlier. The goal is to help public-health agencies accurately identify the start and spread of infectious diseases so that we can all react more effectively. For example, real-time analysis of huge quantities of data – enabled by AI – could allow authorities to create systems of mobile notifications, arm clinicians with relevant information, and rapidly evaluate the effectiveness of their efforts.
This is just one way in which AI is already starting to improve public health. However, solving problems like these is easier said than done.
With the amount of available data increasing exponentially, combined with increasing computer power and artificial intelligence techniques, it is now possible to make much more accurate predictions about everything – from the weather to traffic collisions to the next disease outbreak. The applications of AI are endless.
AI is particularly helpful for medicine, with applications ranging from the detection of genetic mutations in our DNA to improvements in medical diagnostics, image recognition, pharmaceuticals and personalized medicine, among others. The great promise of AI for medicine explains why the Senate's standing committee on social affairs, science and technology is working on recommendations to further integrate AI into Canada's health-care system.
But so far AI approaches have not been widely used to tackle public-health challenges, such as infectious disease outbreaks or efforts to promote healthy behaviours and prevent chronic disease. This is a missed opportunity – one that makes us all less healthy than we could be.
Public health poses an interesting challenge: It primarily deals with disease prevention. It is hard to demonstrate the value of public-health interventions when the desired outcome is the absence of disease. Consider the polio vaccine. Before it was introduced in 1955, thousands of Canadian children were affected by the disease. Today, we have largely forgotten about this victory.
Public health also deals with the underlying determinants of health – poverty, relationships, education and environments – as well as the growing health inequities among different populations. These are all social challenges that, when addressed, have great benefits for both society and individuals.
AI could be a powerful tool for public health because it can predict future events based on past data. This makes it possible to answer many questions of great social significance. Where will the next pandemic occur? How should we design our cities to reduce chronic diseases? At what scale should screening programs be implemented? Answering these questions will help to improve everyone's health.
The challenges and opportunities for applying AI to public health requires a concerted effort to make it happen. To enable researchers to explore these issues, the Canadian Institutes of Health Research's Institute of Population & Public Health (CIHR-IPPH) and the Canadian Institute for Advanced Research (CIFAR) are working collaboratively under the Government of Canada's broader Pan-Canadian Artificial Intelligence Strategy. For the most part, researchers in these fields – public health and AI – have worked independently of each other. We need these communities to discuss common issues and learn from each other to spark the innovation necessary to address social challenges related to public health. Last week CIHR-IPPH and CIFAR hosted a workshop in Toronto to get those sparks flying.
By bringing these two fields together, we hope Canada will be positioned to take a leadership role in utilizing AI tools to solve public-health challenges that affect us all. Our goal is to ensure that AI benefits all Canadians. The workshop is important because public health affects us all, and with the advancements AI can provide, we'll all be a lot healthier.
In a few years, our collective efforts could even be the reason you don't get the flu.