How AI can support people’s health at home and ease pressure on acute and long-term care ―if we act now

By Dr. Alex Mihailidis

There’s understandably a lot of focus these days on the role that artificial intelligence (AI) can play in the health-care system. With its ability to analyze vast amounts of data, AI holds great promise as a tool that health professionals can use when making diagnoses and treatment decisions.

What’s often missing in the conversation about AI is how it can support people’s health and well-being at home and in the community – keeping us out of hospital and long-term care. This has to change and soon. It’s no secret that our population is aging, placing strain on the health-care system and overburdened long-term care facilities. And most of us want to remain in our own homes as long as we can.

AI can make this possible. Smart-home systems can help people stay safe, independent and healthy. Sensors, for example, have the potential to detect changes in a person’s health and provide warnings before things deteriorate. Imagine a bathroom with flooring that can monitor someone’s blood pressure and heart rate. In my own work, we have created an intelligent emergency-response system that uses AI and computer vision to detect falls in the home. It ‘learns’ a person’s habits, knows when something has gone wrong and can interact with the person or call for help, if necessary.

Other AI-based systems now in development can prompt older adults with cognitive impairment to do daily activities such as taking medications and making meals.

A generation ago, none of this would have been imaginable. Welcome to the future. Today, we have widely-recognized expertise in AI in Canada, but we need to focus more attention – and more people – on developing AI applications to help older adults stay longer in their homes. This area has tremendous potential but also challenges. Training AI algorithms requires high-quality data. Hospitals have electronic health records and other sources – imperfect though these may be. Home is a different story. Data sets there are small and ‘messy.’

Case in point: my team set out to develop a data set that could be used to predict dementia. We installed non-invasive motion sensors in the homes of more than 300 people. But there were so many problems – the system would pick up multiple people moving around in the environment, for instance – that we could only use the data set from approximately 60 people.

Despite these challenges, research teams at AGE-WELL, Canada’s Technology and Aging Network, are making significant strides in applying AI to home settings. We are developing socially-assistive robots, in-home therapies and remote health-monitoring tools. There are systems to detect night-time wandering, or to show if someone is unstable when rising in the morning, indicating a decline in mobility. AGE-WELL also studies ethical, privacy and security factors associated with new technologies.

But for several reasons, work in this area is in a precarious situation. First, the federal Networks of Centres of Excellence (NCE) program, which has funded AGE-WELL since 2015, is being phased out over the next few years. AGE-WELL can only apply for three further years of NCE funding – and nothing after that. Alternative funding must be found.

Second, the federal government announced a $125 million Pan-Canadian Artificial Intelligence Strategy in 2017, led by the Canadian Institute for Advanced Research (CIFAR). It’s a laudable initiative but it needs to pay more attention to AI applications that can support people’s health at home and in the community.

Third, Canada’s new AI institutes in Edmonton, Montreal and Toronto must, in addition to hospital-based data, place a greater emphasis on developing repositories of community-based data that researchers can use to build AI systems for the home and community. Partnering with organizations that look after people in the community can help to generate this much-needed data. By creating rich data sets, we will attract more AI researchers to this crucial area.

AGE-WELL, which was established to bring together everyone in the field of technology and aging, is ready to be the leader in pushing this forward – because the future of aging depends on it.

Dr. Alex Mihailidis is scientific co-director and CEO of AGE-WELL (agewell-nce.ca), a federally-funded Network of Centres of Excellence. Dr. Mihailidis is also a professor at the University of Toronto and a senior scientist at KITE Toronto Rehab-University Health Network.