The Researcher
University of Sherbrooke researcher Thierry Judge
The Challenge
Automate a time-consuming, necessary step in diagnosing heart disease
The Solution
A new technology that identifies when results generated by emerging artificial intelligence (AI) systems are incorrect or uncertain
The Outcome
Automated diagnostics are now one step closer to widespread clinical use, which will reduce wait times and free up doctors to perform other important tasks
Thierry Judge has automated the time-consuming and redundant task of contouring ultrasound images of the heart, a necessary step in diagnosing heart disease that up until now has been conducted manually
Novel technology allows for the automation of a crucial task needed to diagnose heart disease
Thierry Judge, a master’s student in the Department of Computer Science at the University of Sherbrooke, has developed a technology that identifies when results generated by emerging artificial intelligence (AI) systems — which speed up analysis of ultrasound images to detect heart disease — are incorrect or uncertain. The software, called CRISP, is currently being tested by Oxford, UK-based, Ultromics Ltd, a leader in AI echocardiography.
His innovation automates the time-consuming and redundant task of contouring ultrasound images of the heart, a necessary step typically conducted manually, in diagnosing heart disease. In simple terms, AI-assisted heart contouring automates the process of extracting key clinical metrics from ultrasound images, such as the volume of blood being pumped by the left heart ventricle, that are routinely used to identify and diagnose heart disease. Without it, doctors need to manually set contours on images in order to see specific parts of the heart and obtain cardiac output data required to make a diagnosis. 
“AI methods are usually as accurate as clinicians at understanding images, but there are still times when very obvious mistakes are made, which is slowing down the integration of AI-based diagnostics into clinical practice,” explained Judge.
Because emerging and time-saving AI-based approaches are still prone to error, doctors must validate each result for accuracy before proceeding to a diagnosis. With Judge’s advancement — software that accurately flags indefinite contouring metrics generated by the automated system — only unclear results need to be validated, resulting in additional time savings and greater confidence when making a diagnosis.
“The ultimate goal is to build trust in AI systems so that we can free up time for clinicians to see more patients and spend more time on patient care instead of manual tasks,” says Judge. “We know these AI systems will sometimes fail. The goal of my research is to accurately identify when they do.”
The innovative research has earned Judge, a Mitacs intern working under the supervision of Professor Pierre-Marc Jodoin, the Mitacs Award for Outstanding Innovation—Master’s. The award, presented at a ceremony at the National Arts Centre in Ottawa on November 22, 2022, recognizes significant achievement in research and development innovation during Mitacs-funded research. 
Judge’s research was recently published, and the software he developed is currently in the process of being integrated into Ultromics’s AI-assisted contouring product. He credits Mitacs for helping him on the path towards commercialization.
“Before I partnered with Ultromics, I had a more academic view of my research,” said Judge. “Now I see how the software development pipeline works in practice and I have a better understanding of what’s needed to make my methods usable for clinicians. I’m also able to tap into valuable input from people who work across various parts of the development pipeline and apply their feedback in my work.”
Mitacs’s programs receive funding from valued partners across Canada. We thank the Government of Canada, the Government of Alberta, the Government of British Columbia, Research Manitoba, the Government of New Brunswick, the Government of Newfoundland and Labrador, the Government of Nova Scotia, the Government of Ontario, Innovation PEI, the Government of Quebec, the Government of Saskatchewan, and the Government of Yukon for supporting us to foster innovation and economic growth throughout the country.
Do you have a business challenge that could benefit from a research solution? If so, contact Mitacs today to discuss partnership opportunities: BD@mitacs.ca.
Mitacs Team
Mitacs’s website content is created by people throughout our organization, united in their passion for innovation and eager to share their perspectives with others in the innovation ecosystem.
