Characterization of blood flow through the left ventricle (LV) using 4D Flow MRI requires accurate segmentation of the underlying anatomy. The combination of poor resolution and low contrast in 4D flow MR images, however, makes the LV segmentation challenging and prone to observer bias. We propose an image registration based approach that allows reliable and automatic correction of LV segmentations in 4D Flow images using the high contrast in routinely acquired 2D cine-MR images. The proposed approach is shown to achieve high in plane spatial accuracy and comparable values of blood flow parameters when evaluated against manually obtained expert segmentations of 4D flow MR images.
This abstract and the presentation materials are available to members only; a login is required.