Leading Edge / Summer 2015
A Robot’s Helping Hand

A U of T lab is developing new technologies to assist older people with daily tasks


"Robot & Frank," Sony Pictures Entertainment

“Robot & Frank,” Sony Pictures Entertainment


According to recent statistics, about a quarter of Canadians aged 65 and up are unable to perform simple activities, such as making a meal. This takes a toll not just on seniors themselves, but also on their caregivers, many of whom take time off work to look after family members. One Canadian study puts the cost in lost productivity as equivalent to 157,000 full-time employees annually. As the number of seniors soars in coming years due to an aging population, the burden on caregivers will intensify – unless new ways are found to assist aging Canadians.

Enter Alex Mihailidis, a professor of occupational science and occupational therapy, whose lab at the Toronto Rehabilitation Institute is developing, among other things, a robot that can coach people to wash their hands and make a cup of tea. It can propel itself (like those robotic vacuum cleaners) and will eventually assist with other straightforward tasks, such as grooming, bathing, getting dressed and preparing simple meals. Studies of caregivers suggest these activities are where people need the most help.

The robot won’t sympathize with your woes but it can alter its visual and verbal prompts to suit individual needs. Some people might need only verbal prompts, for example, while others might require visual ones, or to be prompted differently, depending on the time of day. “The system can recognize this and adapt its prompts automatically,” says Mihailidis.

The technology appears to work even for people with moderate and severe dementia, but it’s not designed to replace personal caregivers or the human touch. “That’s always going to be needed,” says Mihailidis. But the innovative device does aim to reduce the burden on family members.

Mihailidis also hopes to alleviate caregivers’ concerns about injuries. Some 20 to 30 per cent of seniors fall every year and the effects can have a lasting impact. Falls cause 95 per cent of all hip fractures and 20 per cent of these cases result in the patient’s death within a year. A fall-detection system developed by Mihailidis’s lab aims to bring help quickly.

Mounted on the ceiling like a smoke alarm, the device uses a vision sensor to monitor the room. If it detects that someone has fallen, it uses speech recognition technology to ask a series of simple yes and no questions and then, depending on what it detects, either stops speaking or calls for help; 911 is automatically included in the system and other contacts, such as friends or neighbours, can be programmed in.

The fall-detection system is closer to the marketplace than the robot, but both will be affordable (under $100 for the fall-detection system) and Mihailidis is hoping that a new research network will speed their commercialization.

Launched earlier this year, AGE-WELL (Aging Gracefully across Environments to Ensure Wellbeing, Engagement and Long Life) is Canada’s first research network for technology and aging. It links academic, business, government and community partners and aims to help older adults maintain their health and independence through technology. This January, the federal government announced $36.6 million for the network over five years. As one of the network’s two scientific directors, Mihailidis can’t recommend his own lab for funding, but he’s optimistic that the network will help get his lab’s products to the people who need them.


Reader Comments

# 1
Posted by Evelyn Stefov on June 21st, 2015 @ 7:44 am

Brilliant work and congratulations, Alex. I am proud of the work and the commitment to help the elderly lead a better and safer life. Wishing you success.

# 2
Posted by Betty Navratil on June 25th, 2015 @ 11:52 am

I like the robot to help the elderly. Since our daily tasks are routine, they match the mechanical movements of the robot. However, not everyone can afford to buy one.

Add a Comment

required, use real name
required, Not for Publication
optional, eg: BSc 2008

Previous story in this issue: «