Intimai flap segmentation on CTA aortic dissection images based on mathematical morphology

Abstract

Aortic dissections are caused by tissue perforations: it consists of one or several intimal tears, two separate blood flow channels and a layer between them, called intimal flap. As far as we know, only a single study has been proposed to segment intimal flaps of aortic dissections images, it is based on geometrical assumptions. In this study, we use one Mathematical Morphology operator to extract the intimal flap from Computed Tomography Angiography images acquired after an aortic dissection. This image of flap directly allowed us to use a holes closing algorithm for 3D binary images to materialize intimal tears, by considering them as 3D holes. As far as we know, there is no other study which leads to intimal tears localization. That work is included into the preoperative part of a whole process consisting in registering aortic dissection features, extracted from a Computed Tomography Angiography image, onto intraoperative X-ray angiographic images, to assist clinicians during their intervention.