Medtech group VentriPoint Diagnostics Ltd (CVE:VPT) updated on the project to advance the artificial intelligence (AI) capabilities already embedded in its heart monitoring VMS+ products.
In March this year, the firm reported on its collaboration with Ryerson University after Ventripoint revealed it was taking part in a Natural Sciences and Engineering Research Council (NSERC) Engage grant.
The university researchers have now completed an initial prototype of the AI engine aimed at analysing all four heart chambers, Ventripoint said Monday.
"This software can be used to detect the end systolic and end diastolic frames in cardiac video clips in a number of different views.
"In initial testing with images provided by the company, results were impressive with an up to 90% accuracy rate in identifying cardiac ultrasound views," the firm said.
Now more images are being provided to the Ryerson team to further the development of their deep-learning algorithm to enable automatic, real-time cardiac-feature detection using AI on ultrasound imaging.
Ventripointb shares shed around 4% to C$0.235.