"We hypothesize there's a pattern we can find between those who have had a stroke and haven't in their eye scans." - Tanha Kabir
Q: What is the overarching goal of your research?
A: The goal of our research is to see how we can improve stroke prevention by aiming to predict the likelihood of one getting a stroke in the future. We hypothesize that there's a pattern that we can find between those who have had a stroke and haven't in their eye scans.
Q: How do you go about testing this hypothesis?
A: My goal is to see if we can see if there's a correlation of certain features of someone's eyes that we can say are statistically significant to predict a stroke. This involves taking thousands of eye scans, processing them through some computer vision algorithms to:
a) cut out the parts in interest, this is called segmentation, and then measuring those parts and putting them in a mathematical model to see if there is any strong pattern.
b) see if there's the presence of certain parts in some eyes but not others, and then pass this into another mathematical model to see if there's any pattern of some individuals having a certain feature versus not.
Q: So what's next for this research?
A: Eventually, if it's shown that there is a correlation, I will take this process and make it an application that anyone can use for all eye scans to predict the likelihood of an individual having a stroke without touching any of the code. The prediction would work off of the mathematical model we found that showed a pattern.
Tanha Kabir is a 4th-year Computer Science at the University of British Columbia. She is a Next36 alum and has done several PM internships at Microsoft throughout her degree.
We can't wait to hear how the rest of Tanha's research pans out! If you are currently a student and also doing research in the field of AI, reach out to me at email@example.com and let’s chat! We would love to feature your work.
This blog post was written to accompany our All About AI newsletter, our newsletter containing news from the world of artificial intelligence, as well as research papers to help you learn more about AI. Subscribe here for more content like this!