By Mareska Chettiar, Co-T/E Life Editor
During the summer before his junior year, senior Faizaan Siddique began working on an AI tool that detects early signs of glaucoma. Siddique’s research paper “Opening the Black Box: A Novel, Cost-Effective Pipeline for Automated, Explainable Glaucoma Diagnosis and Blindness Prevention” won him the title of Regeneron Science Talent Search (STS) Scholar, one of 300 recipients internationally. The Society for Science hosts the annual Regeneron STS competition, which recognizes and awards high school students who present new discoveries and research in science and math. Both the school and the student receive a $2,000 grant for the accomplishment.
After seeing a family friend diagnosed with glaucoma, Siddique decided to create a tool that analyzes fundus images, images of the back of the eye, to output the percent likelihood of someone having glaucoma. He did this as an independent project in his free time, relying on prior research experience.
“I had worked on a research paper on a separate topic that gave me the foundational knowledge that I needed for this independent project,” Siddique said. “That machine learning expertise that I gained in addition to the coding side of things and the medical side of knowledge that I (gained) really helped me conduct this project.”
Siddique designed the AI tool to operate in a user-friendly and transparent manner, displaying what methods it uses to compute its results. Siddique’s goal is to make it more accessible and cost-effective in order to enable widespread use.
“That is the ‘black box’ part of my research title, where most AI tools out there are a black box in that it’s taking an input, giving you an output, without you, the user, ever knowing how it got to that point,” Siddique said. “The part of opening the black box and making this AI tool explainable, in addition to being cost-effective (so) that users like clinicians and patients are able to see why the AI tool arrived at the diagnosis.”
Siddique presented his research to a panel of science and math teachers before he submitted to Regeneron STS. The panel included teachers Jacqueline Gontarek, Dr. Scott Best, Janet Wolfe and Travis Hartley.
“It was fascinating to see the math behind (the tool) and the computer science behind it. Those are two things that I don’t know much about, so it was a huge learning opportunity for me,” Gontarek said. “He did so much work, and it was so impressive to see how he put it all together.”
In the future, Siddique looks forward to joining the academic space in medical research and AI application. He aims to work in the field of medical research, combining technology and healthcare.
“I’m still thinking about what I want to do,” Siddique said. “But I do want to make sure that I’m being a young innovator who makes the world a better place using technology and science.”
Mareska Chettiar can be reached at [email protected].
