Incidental extracardiac findings (ECFs) occur when doctors perform medical imaging scans of the heart and discover unexpected health issues in other organs in the body.
Almost 30% of heart scans find unexpected issues, and about half of these issues require follow up. Unfortunately, cardiologists can miss up to 95% of non-heart issues and getting a second review from a radiologist is very costly, rare and time-consuming.
To give patients a quick diagnosis and the treatment they need, Telfer’s Christopher Sun, a professor and scientist at the University of Ottawa Heart Institute, has received funding from the Canadian Institutes of Health Research Project Grant Program for his project titled “Actionable Incidental Extracardiac Findings Detection via Interactive Artificial Intelligence.”
Quick analysis and clear explanations
Sun and his team aim to develop the first interactive AI model for automated and explainable ECF detection and characterization in heart scans, generating insights to improve patient outcomes. Their approach will allow for quick analysis of images and provide clear explanations of its findings, helping cardiologists and radiologists understand and act on unexpected health issues.
Sun and his team will validate their system by comparing its results to those of expert radiologists and cardiologists across multiple Canadian hospitals partnered with the study, ensuring the technology works well in different real-world applications.
Their study will be the first AI model of its kind made for detecting incidental findings in heart imaging.
A new tool could enhance overall patient safety, reduce strain on the health-care system and save lives, securing Canada’s leadership in AI health-care innovation.
