The winners of York University’s Science Fair in April 2022 — a group of students from St. Robert’s Catholic High School in Markham — has developed an artificial neural network to detect cancer from just a blood sample at 95 per cent accuracy, which, according to Alex Gan, a student within and representative of their group, FourSight, is over 15 per cent more accurate than the traditional diagnosis methods.
According to FourSight’s website, “using genetic test data retrieved from the Gene Expression Omnibus Database, over 27,000 sample tests were used to train our multilayer perceptron neural network to recognize 12 different types of cancers.”
The project premiered at the York University Science Fair in April 2022 where the team won a gold medal and a Challenge Award in Disease and Illness, and have gone on to represent Canada at the EU Contest for Young Scientists in the Netherlands, winning second place. They were also sponsored by Student Life Network at The Remarkable Students Showcase where they won $10,000 prize money, among many other recognitions.
FourSight is a cancer diagnosis program that uses AI to read genetic data that is extracted from genetic sequences in the blood. The program functions by taking a blood test from a patient from which the genetic sequence is extracted and inserted into a microarray chip. The AI then creates a diagnosis when it is plugged into the chip. The goal is to make cancer detection as non-invasive as possible.
Hanze Wu, a member of FourSight is optimistic of the technology’s development, stating that “FourSight is hoping to get sample data to make the AI more accurate — some actual practical usage of our program tested in the medical field, such as going to hospitals and requesting blood samples. Actual clinical data extracted from patients is necessary to see if our AI can help them.”
The group is now looking to expand FourSight to create ground-level applications such as using it in hospitals to aid in cancer detection. Gan says, “We believe we have what it takes to truly revolutionize early cancer detection and can help save countless lives.”
The group researched previous cancer detection tools and connected with various medical professionals and lab workers cognizant of the type of technology being used in hospitals. By understanding market demand and shortcomings of the current cancer diagnostic methods, the team decided on developing the AI.
“One thing that’s interesting about our project is a lot of people, despite medical professionals not always being accurate, trust a human more than an AI. The emotional connection is valued more than AI,” adds Wu. The group hopes to achieve more attention from professionals that have the power to publicize methods used in AI in order to gain public trust.
The team members’ biology teacher, Christina Gueli comments that, “they were armed with research information and eager to find a solution to whatever problem was at hand. They are remarkable students with exceptional potential. Their commitment to their project was extraordinary and their resilience when faced with challenges praiseworthy. “With students like these developing novel solutions to old problems, the future of cancer detection is bright!”
Other faculty at St. Robert’s Catholic High School that aided the group of students were Aimee Rusli, Jennifer Thome and Giulia Provenzano, their chemistry, biology and mathematics teachers respectively.
