Is automatic analysis of multicontrast MRI ready for clinical studies on plaque tissue composition?

Automatic plaque segmentation algorithms have been proposed to be helpful for large-scale and/or multicenter studies. Yet few data are their performance in real-world applications. In this study, by using a prospective cohort study with baseline and follow-up imaging, quantitative measurements on plaque tissue composition including calcification and lipid-rich necrotic core as analyzed by an automatic classifier were compared to expert human review. Results showed excellent correlation between MEPPS and manual review in analyzing compositional volumes with images of a single time point, and moderate to good correlation in analyzing compositional changes with images of multiple time points.

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

Join Here