University of Waterloo researchers working on a virtual assistant to help people with Alzheimer’s disease

ACT@Home

Computer scientists at the University of Waterloo are creating a virtual assistant that would help people living with Alzheimer’s disease. Called ACT@Home, it combines artificial intelligence (AI) with social psychological models to support people with the neurodegenerative disease.

The system is designed to prompt patients to complete day-to-day tasks, such as handwashing, in a way that aligns with how they feel and what they are thinking at a specific moment.

ACT@Home will ensure that tasks align with a person’s feelings or thoughts by picking up emotional cues from a person with Alzheimer’s disease, such as facial expressions, posture, their way of moving, and tone of voice. The aim is to help people cope with day-to-day challenges, such as forgetting when to take their medicine or whether they need to dry their hands after washing them.

“Convincing a person with Alzheimer’s disease to do something is very difficult because it depends on the person’s interpretation of the situation, what they think is going on, what cues they are getting and who they think they are in that moment,” said Jesse Hoey, a professor at the David R. Cheriton School of Computer Science at Waterloo and lead researcher on the project. “This prototype will work by building a model of what’s going on emotionally in the mind of someone with the cognitive difficulty and then prompting them to complete an activity of daily living in a way that makes sense to them in that moment.”

Hoey added that even before building the prototype device, the research team interviewed people with Alzheimer’s disease and their caregivers to better understand their needs. The team is also receiving two awards for a paper in the American Sociological Review that presents some of the core ideas of the project.

He believes that in the future, virtual assistants like ACT@Home will also be able to help people with other conditions, including Down Syndrome, dementia, and traumatic brain injuries. Hoey hopes the tool will be available within the next four years.