Imagine you get to school and realize you forgot your lunch at home. You write an exam and you can’t remember half the answers. Or worse: imagine you’ve got Alzheimer’s and don’t remember how to wash your hands.

Alzheimer’s is the most common type of neurodegenerative dementia. Sufferers have difficulty remembering events in their proper sequence, which seriously impairs their ability to carry out daily activities. Often, a caregiver must continually provide verbal prompts to the patient. Assisting patients with memory loss or other cognitive problems, however, is no easy task for caregiver or patient, requiring invasions of privacy.

For the past three years, a research team led by Dr. Alex Mihailidis, a mechanical and biomedical engineer at the Intelligent Assistive Technology and Systems Lab at the University of Toronto, have been developing a home-based computer system that uses artificial intelligence to promote independence and ensure the safety of older people living with cognitive disabilities like dementia.

“We are using artificial intelligence to support aging-in-place so that people can remain in their homes for as long as possible,” said Mihailidis.

The group’s latest design is a talking bathroom equipped with a computer screen that gives video and verbal cues to assist hand-washing.

The team has also developed a personal emergency response system that detects when a person has fallen and calls for help. It uses ceiling-mounted cameras that feed pictures into a computer system which analyses the images, determining the position of the occupant and whether or not an accident has occurred. If it decides there has been an accident, the speech recognition system asks the occupant if they need help.

Mihailidis explained the difference between this system’s “computer vision” and traditional video cameras.

“The computer takes the images, processes them, gets the required data but does not store anything that we can visually identify the person from.”

The system considers probabilities and statistics when evaluating a situation and devising a course of action for the person to take. This distinguishes an “intelligent home” from a “smart home,” as a smart home, according to Mihailidis, responds to preprogrammed requests but cannot learn and adapt.

“Our systems use computer algorithms that act more like a human in terms of rational thought and decision-making. They actually learn and adapt to a person’s needs,” he said.

Studies of the technology have found that Alzheimer’s sufferers being assisted by the technology were about 25 per cent more capable of washing their hands without further help from a caregiver. The computer also detected 77 per cent of falls staged in the lab.

As the Canadian population ages, the number of people affected by Alzheimer’s disease is expected to increase dramatically. Today, an estimated 300,000 Canadians over 65 have Alzheimer’s disease, and by 2031, more than 750,000 Canadians may have Alzheimer’s or a related dementia.

Mihailidis and his research team hope that their intelligent home will be ready seven to 10 years from now.