Université d'Auvergne Clermont1 | CNRS


Axis 2: Interventional planning and assistance for ascending aorta dissections


Aortic dissections is a real problem of Public Health, it is a medical emergency that may quickly lead to death. Aortic dissections are characterized by the presence of tears inside lumen (or holes in the intimal tissue). These tears are difficult to detect because they do not correspond to a filled organ to segment.

Our first purpose is to detect these intimal tears to help cardiac surgeons either during a pre-operative step: the visualization and location of intimal tears could lead to take decision more easily about the kind of intervention, and could facilitate the choice of the endoprothesis size; or during a per-operative step: with the registration of these tears on angiographic images, the surgeon’s gestures could be more accurate and safe, what would enhance patients' care.

Initial grayscale image 3d reconstruction Curve skeleton Intimal flap Closing of intimal flap holes

Preliminary study about intimal tears in aortic dissection images (digital topology, C. Lohou)




Aortic lumens segmentation results (fast-marching, N. Fetnaci)




3D/2D Registration of lumens static image [True Lumen (red), False lumen (blue), retrieved from a Ct-scan,

merged onto an angiographic view]

(ITD - image transformation descriptors, ITK, P. Lubniewski  )




3D/2D Registration of lumens animation onto the angiographic sequence [P. Lubniewski]




Initial DRR simulation of an aortic dissection CT-scan Augmented view: with the DRR rendering of the interior of a tubular envelope

Angiography simulation (DRR-base, H. Feuillâtre, now Ph.D. student at LTSI Rennes)




Tubular envelope (V. Sauvage, P. Lubniewski) DICOM Volume Rendering (Qt, ITK, VTK, P. Lubniewski)

Graphical User Interfaces.