A chance conversation was all it took to give Alex Mihailidis's burgeoning career a new focus.
As a graduate student in 1996, the now 41-year-old met a man whose wife had developed dementia in her early 50s. Her bizarre behaviour, such as taking soiled toilet paper, folding it up and hiding it around the house, had her husband at wit's end.
But in describing ways to coach his wife through basic behaviours that most of us take for granted, such as getting dressed in the morning, brushing our teeth or going to the bathroom, it lit a fire in Dr. Mihailidis's mind.
"He said, 'Wouldn't it be great if computers could do all this for us,' and for my first thesis and my masters I developed a computer-based prompting system for people with Alzheimer's," Dr. Mihailidis says.
Now the scientific director of Age-Well Network of Centres of Excellence Inc., based at the Toronto Rehabilitation Institute and hosted by the University Health Network, Dr. Mihailidis's work is focused on improving the daily lives of elderly patients.
Robotics plays a key role in his work, but not the sort that pick people up off a bed or the toilet. Instead he works with the kind that provides reminders and prompts to guide people through the steps involved in everyday tasks.
"It's more meant for older adults who have some kind of cognitive impairment, so that's kind of a project that's taken off and is garnering a lot of interest and potential down the road for these robots to become commonplace in people's homes," he says.
Besides, as Geoff Fernie, the institute director for research at Toronto Rehab, says, robots aren't really suited for that in their field of work.
"Robots actually aren't very strong," says Dr. Fernie, who has worked with Dr. Mihailidis for more than 20 years. "Unlike the robots you see in Transformers, they are generally pretty weak, and if you have a strong robot you usually want to be a long way away from it."
Alongside robotics, big data and predictive algorithms also form an important facet of Dr. Mihailidis's work. One of his main goals is to ensure that people can continue to live and function in their home as they get older, and using sensors to reveal and interpret patterns of behaviour are some of the tools that he and his team use to capture that.
Age-Well collaborated with a team based at the Oregon Health and Science University in Portland, Ore., which put sensors in the homes of about 300 seniors in the Portland area, monitoring things such as movement in the home. On looking at the collected data, Dr. Mihailidis and his team realized that comparing real-time data with historical data enabled them to tell if there had been a change in a person's behaviour.
That kind of big data can prove revealing when it comes to understanding the well-being of a patient.
"For example, we have some early results that show that, just from three months' worth of data collected in a person's home, it's [with] about 90 per cent accuracy [that] we can predict that that person is going to develop dementia down the road," he says.
Telltale signs that Dr. Mihailidis would be looking for in this case are diminished activity around the house, as well as a decrease in walking speed. But to do that it is essential that those habits are monitored on a daily basis. As he explains, part of the problem with the current assessment tools used by doctors to check in on patients is that the patients only see a doctor every six months or every year and the evaluation is done on how they're feeling at that moment in time. There's no long-term assessment.
For instance, older patients with congestive heart failure or diabetes are required to monitor their vital signs on a regular basis, but cognitive impairment can affect their ability to comply with this. Consequently, Dr. Mihailidis and his team have been trying to combine sensors into daily routines so that the patients don't even have to think about them.
They have been working on embedding sensors in the cushions of a sofa to take an electrocardiogram while someone sits there, or a floor tile that can measure blood pressure while someone is standing on it, even in shoes and socks.
"It's really taking the approach of smart home systems. As you go through your morning routine, right there the house is already sensing your heart rate, blood pressure, weight, etc., on several occasions to get that longitudinal data," he says.
But while the field is growing, he says, in Canada it is growing slower than in other countries. The federal government, partially based on the recognition that large numbers of baby boomers are moving into their senior stages, granted $36.6-million to Age-Well last year.
However, one of the problems that Dr. Mihailidis and his team face is in establishing partnerships with companies that can help realize some of their ideas.
"There are not a lot of companies that are knocking on our door saying, 'Hey our core business is technology to support seniors, let's take your work and commercialize it.'"
He tells of families who buy cameras and sensors from their local electronics store to jerry-rig a home-monitoring system, allowing offspring to remotely keep an eye on elderly parents at home while they are at work
Dr. Mihailidis is mindful, though, of the human touch, and the important role that caregivers play in the process, acknowledging that the tools that he is developing are designed to aid them, not replace them.
"That's always been the double-edged sword in the technology area. We're talking about technology that can provide care and monitoring and robots, etc., but is it going to decrease the amount of social connectedness that a person has as well?" he asks.
It's an area that he says needs more study to ensure that as much as technology can help a person in their home, it must also enable them to maintain their connections with their families and communities.
"We don't really say that these technologies are replacement for care or for human contact," he says. "These are tools that families can use to provide better care."