Kids with autism spark this scientist's inventions
By Louise Kinross
Azadeh Kushki (centre) is a scientist at Holland Bloorview who analyzes patterns in children’s brain and behaviour data to guide treatments for autism. She also develops technologies that make life easier for kids with autism, like this Google Glass software that coaches kids through social interactions. Azadeh, who trained as an electrical and computer engineer, works in the hospital’s Autism Research Centre and at the Institute of Biomaterials and Biomedical Engineering at the University of Toronto.
BLOOM: How did you get into the field?
Azadeh Kushki: I did my PhD in electrical and computer engineering, unrelated to disabilities. Then I was trying to figure out what I wanted to do with my life. I had volunteered in high school as a classroom aid for other students with disabilities. I thought I might want to do something with children with disabilities, so I started volunteering at Holland Bloorview.
BLOOM: Where did you volunteer?
Azadeh Kushki: In the therapeutic playroom. We would do music with the kids or go to the Snoezelen room. The staff provide amazing guidance and mentorship, and the kids are awesome, so it was a lot of fun. I came once a week.
BLOOM: How did that lead you into research?
Azadeh Kushki: I started thinking that technology could have a lot of uses here, and people told me Tom Chau does this work. So I got in touch with him. He was supportive of my writing an application for a post-doctoral fellowship on access technologies. I got the funding and was exposed to a range of technologies. Then I met Evdokia Anagnostou, who got me into autism research. We did a joint project with Tom to understand motor function in children with autism spectrum disorder.
BLOOM: What interested you about autism?
Azadeh Kushki: There are so many differences in how autism affects [individual] children. That intrigued me as an engineer. Engineers build things to solve problems. Evdokia and Tom were the best mentors I could have, and were tremendous in helping me find my way as a scientist.
BLOOM: When did you become a staff scientist here?
Azadeh Kushki: In 2012, I applied for the Mary and James W. Davie scientist position. When you’re a post-doc, you have a lot of limitations with the grants you can apply for and the students you can supervise. Tom Chau made an argument that we needed to create these funded scientist positions, to help post-docs transition into being independent researchers.
BLOOM: So your position is donor-funded?
Azadeh Kushki: Yes, thanks to the generous donation from Jim and Mary Davie.
BLOOM: How would you describe the research you do now?
Azadeh Kushki: It’s very multidisciplinary. The unique thing about it is that I have the opportunity to combine my engineering perspective with a clinical perspective. I get to collaborate with everyone else at Holland Bloorview, so I have access to tremendous clinical [knowledge]. It's [unusual] for an engineer in my field to be so integrated into clinical practice. Most of my engineering classmates work for tech companies now.
BLOOM: How does the clinical side inform your work?
Azadeh Kushki: It’s direct contact with families, kids and clinicians who can give input on everything from what the problems are, to what impactful solutions may look like. I'm very grateful for these partnerships, and for the time families spend with us to inform our research.
BLOOM: Can you describe the two areas of your research?
Azadeh Kushki: One area is to find patterns in brain and behavioural data across children with neurodevelopmental disabilities.
Every child you meet with autism is unique, so, how do you characterize this uniqueness in a way that can inform treatment approaches? That's the question I try to answer.
The idea, based on work by Evdokia and the POND network, is that there are things shared by kids with autism, obsessive compulsive disorder and ADHD, in terms of their biology or behaviour. If we can find commonalities, these groups may benefit from similar treatment approaches.
The second part of my research involves taking these findings and translating them into technology-based interventions.
BLOOM: Can you talk about some of these technologies?
Azadeh Kushki: Sure. The anxiety meter is an app for helping children with autism recognize their anxiety. It's on its way to being commercialized.
Up to 84 per cent of children with autism experience anxiety symptoms that are impairing. The difference between kids with autism and other kids with anxiety is that many children with autism have difficulty recognizing the signs in themselves. We can teach relaxation strategies, like deep breathing, but if a child doesn’t know when they’re beginning to get anxious, they can’t apply the strategies in time.
The anxiety meter measures their heart rate and identifies when they need to start deep breathing—or doing whatever their personal calming strategy is. It’s software that can run on a tablet, a phone or a smart watch. If the child is using a smart watch, the watch vibrates to tell them when they need to start deep breathing.
BLOOM: How accurate is it?
Azadeh Kushki: In a lab setting it can detect anxiety with 95 per cent accuracy. We now have a randomized control trial going with 30 children with autism. Everyone learns deep breathing as their relaxation strategy. Then they’re put in a stressful situation—giving a speech in front of three people—and half get feedback from the anxiety meter to say they’re getting anxious, and the others don’t. Based on preliminary analysis, almost 100 per cent of the kids in the group using the meter start deep breathing.
BLOOM: And in the group that have to identify the stress on their own?
Azadeh Kushki: Around 30 per cent.
BLOOM: When will it be available?
Azadeh Kushki: With appropriate funding, in about six months. The cost of the software download is minimal. We’ll also offer value-added services, like a cloud-based system to track your progress over time.
BLOOM: What is the software you developed for Google Glass?
Azadeh Kushki: It’s called Holli. We wanted a technology-based coach for children with autism that goes with them to help with social interactions. Google Glass has speech recognition and it can listen to a conversation, then prompt the child with appropriate responses.
In our study, we tested how Holli would work in a restaurant where a child would order fast food. For example, the child would go to the counter and the staff person says ‘Hi, how are you?’ The child looks at the little TV mounted on the top of one of the glass lenses, and sees options like ‘I’m good,’ ‘I’m well,’ ‘Thanks.’ They say the words they choose. Then the software waits for the next part of the conversation, which is usually ‘What would you like to order?’ And they see options like hamburger, cheeseburger, fries.
We did a usability study and all of the kids enjoyed using it and were able to complete the conversation. We’re looking for more funding to move that forward—which is the case for most of our work.
BLOOM: What do you love about what you do?
Azadeh Kushki: Everything. I love the fact that I have the opportunity to make an impact. Part of that opportunity is interacting closely with families and kids to hear what they say is needed. The solution is usually not the hard part. The tough part is identifying the need.
BLOOM: What have you learned from families?
Azadeh Kushki: That a diagnosis doesn’t define a person. Every child has his or her own strengths, likes and dislikes, and is unique. The uniqueness of every child is the thing that I appreciate now.
BLOOM: What’s the greatest challenge of your work?
Azadeh Kushki: Waiting to get these products out to families. It takes time to get from research and development into the users’ hands, and that’s the frustrating part. It comes down to time, money and resources.
BLOOM: What’s the work culture like here?
Azadeh Kushki: It’s amazing. Everybody is so passionate and committed to improving the quality of life for, and care of, children with disabilities.
BLOOM: What emotions come with the job?
Azadeh Kushki: There’s a sense of privilege to have the opportunity to be able to make an impact. I’m very thankful for that. I’m so humbled by the kids and families I [work] with. There’s also the frustration that I mentioned.
BLOOM: How do you cope with that frustration?
Azadeh Kushki: I haven’t figured it out, to be honest with you. I don’t knit, like Melanie Penner does. The best I can do is to write grant applications.
The frustration motivates me to write a grant for the fifth or sixth time, after the first versions have been rejected. If I know there are people who can use it, and there’s a need, it gives me the drive to keep going.
BLOOM: What are your hopes for the future?
Azadeh Kushki: I think technology has a lot of potential to help the field of childhood disability and transform the way we think about delivering services.
Kids spend so much time with technology these days—for better or worse—so how can we use that time and turn it into productive time? The key is to make technologies that are so interesting and motivating that a child uses them. If it’s boring or cumbersome, they won’t. That’s why we spend so much time consulting with kids and families.
