Kim Adams

Chance tour of Glenrose Hospital leads to new career direction for assistive technology researcher

Kim Adams applies her love of math to help kids with physical disabilities engage more fully with their world.

When Kim Adams was in her third year of an electrical engineering degree, her roommate — a speech-language pathology student — took a tour of the Glenrose Hospital with their assistive technology service. When she got home, she told Adams she should check out the work the facility was doing for people with physical and communication impairments. Her curiosity piqued, a tour led to a stint as a volunteer, an eventual job and a new career direction for Adams, associate professor in the Faculty of Rehabilitation Medicine.

How did you get into your area of research?

After I went to check out the Glenrose, I ended up volunteering there, then working there. Eventually I picked that as a career, applying engineering to technologies for access and communication. After many years, I did my PhD in rehabilitation science, focusing on strategies for children with physical disabilities to control assistive robots to do play and learning activities. There have since been so many interesting directions to take with that.

How has the field of assistive technology changed since you began?

Assistive technology was actually quite innovative back then, with some creative access methods such as using your voice to control the computer, or switches to do scanning, or eye gaze to select items.

In recent years there has been another surge of innovation in assistive technology. The brain-computer interfaces that I've been working with lately, the virtual reality systems, gesture recognition, wearable computers — all are now becoming feasible to use.

Where do you see that technology heading in the next five to 10 years?

I think there will be a lot of work in applying machine learning to improve the technologies. We've used machine learning to teach a robot a task, so we don't have to program every possible task a child might do with the robot. But machine learning can also do things like learn to take on more of the robot control so the children can just focus on playing or learning, rather than on how to control the robot to do the play or learning they want to do.

Machine learning could also help in personalizing the access methods that kids will use to control communication systems, robots and games. Sometimes traditional eye-gaze or switch-access methods don't quite work for an individual, so we're looking at brain-computer interfaces. Perhaps machine learning can determine an optimal hybrid combination of access methods for each unique person and task.

What’s the most rewarding aspect of your work?

I find working with people the most rewarding, like when the kids we work with get to do something they hadn't done before. One example early on was with the LEGO robot. Children could control a robot that had a pen attached to it and they could make a picture. It was something they made themselves, that they could put on the fridge at home. The kids and parents were so proud of that. More recently, with brain-computer interfaces, kids are able to play games on the same computer as their playmate, rather than just watching them.

What’s the most challenging aspect of your work?

When the technology’s just not there yet and you can't find a way to make it work with the kids.

If you hadn't become a researcher, what do you think you'd be doing instead?

When I was younger, the two pathways for those who loved math were engineering or teaching math. Being a teacher or working with kids somehow would have been fun. But I get to do some of that with the LEGO robot programming outreach that we do for the nine- to 14-year-olds. It's not math so much as it is programming and building robots, but I love working with the kids.