A 17-year-old from Kitchener, Ontario, has taken on a problem in blood oxygen monitoring that judges say has been around for 35 years.
Gurnoor Kaur, a Grade 11 student at Cameron Height Collegiate Institute, won the best project award for innovation at the Canada-Wide Science Fair in Edmonton.
Her project was titled Eigenpulse: Eliminating Demographic Bias in Pulse Oximetry and Remote PPG from First Principles.
Judges said the work fixes a 35-year-old problem in blood oxygen sensors, which has led to higher mortality in Black patients.
Before heading to Edmonton, Kaur spoke to CBC K-W’s The Morning Edition about another device she created to detect hospital-induced delirium, which can affect patients’ cognitive state.
She said nurses are often busy with other work, so many cases of delirium go undetected.
“It can detect emotions and micro expressions to understand patients’ emotional state and it also can detect heart rate and respiratory rate through non-contact, camera-based monitoring, eliminating the need for bulky sensors in hospitals as well,” she said.
“I’ve integrated a chatbot to be able to continuously converse with patients and run reorientation techniques, which have been shown to decrease risk by up to 50 percent.”
Kaur said her work on blood oxygen sensors grew out of that project.
She said she noticed a demographic bias in systems that monitored vital signs and detected oxygen.
“So on lighter skin patients, the error is lower than it is on darker skin patients,” she said.
Kaur said the field currently treats the issue as a data problem.
“Currently, the field assumes that it’s an issue with the data that the models are being trained on, not enough diverse data, and that they don’t have enough videos from darker skinned patients,” she said.
“While that does contribute to the issue, I also found out that there is a mathematical instability in current cardiac models and to be able to resolve that, you need to add a missing term and that’s what my project focused on. So this is an aspect of the hospital-induced delirium project. But what I did was I solved the mathematical instability in the cardiac model and using that, I was able to start to remove this demographic bias.”
Kaur told CBC K-W she is interested in a career in medicine and said her current interest is computational biophysics.
“I want to use math and physics to be able to model our biological systems and understand how light interacts with them, to be able to make better diagnosis and treatment tools that can remove the biases and inequities currently found in health care,” she said.
Read more from CBC.
