Meeting Banner
Abstract #0795

Fully automated aortic pulse wave velocity calculation from 4D flow MRI with AI-based segmentation

Ethan M I Johnson1, Haben Berhane1, Michael Scott1, Kelly Jarvis1, Alex Barker2, and Michael Markl1
1Northwestern University, Chicago, IL, United States, 2University of Colorado Anschutz Medical Campus, Aurora, CO, United States


Pulse wave velocity (PWV) is an important measure of vessel stiffness and can be quantified from 4D flow MRI. Accurate assessment of PWV requires placement of multiple analysis planes in an imaged volume, which typically is performed manually to some degree, such as through use of a vessel segmentation made by an experienced human observer. Here we present evaluation of a fully-automated method for calculating PWV in the thoracic aorta, incorporating an artificial intelligence (AI) convolutional neural network to segment the aorta. Estimates of PWV using AI segmentations were compared those from human-observer segmentations and found to be similarly reliable.

This abstract and the presentation materials are available to members only; a login is required.

Join Here