Meeting Banner
Abstract #3935

Automated Model-Based Assessment of the Aortic Root for Optimal Planning of TAVI Procedures Using CMRI a Comparison with Manual Measurements

Sabine Mollus1, Maximilian Pietsch1, Irina Waechter-Stehle2, Yang-Chul Boering3, Meike Schmidt3, Jason Foerst3, Sebastian Gruenig3, Mirja Neizel3, Burkhard Sievers<sup

1Philips Research, Aachen, Germany; 2Philips Research, Hamburg, Germany; 3Clinic for Cardiology, Pneumology and Angiology, University Hospital Duesseldorf, Germany; 4Department of Diagnostic and Interventional Radiology, University Hospital Duesseldorf, Germany

TAVI is a catheter-based intervention for high-risk patients with severe aortic valve stenosis. It is based on the replacement of the non-functioning valve by a stent-based prosthesis. Comprehensive, (MR) imaging-based planning is essential for TAVI device selection. Methods for automatic aortic valve planimetry can improve manual evaluations. For this purpose, we developed a MR-based model of the relevant anatomy. In this study, we compare model-based with manual measurements, evaluate the inter-operator variability and perform inter-modality comparisons between MR and CT. The results show that model-based automated aortic root measurements in MR correlate well with manual evaluations and with CT-based measurements. Thus, geometrical modelling proves to be an accurate and innovative technique for automatic aortic root planimetry.

Keywords

able according accuracy accurate acquisition addition agents aging aligned amounts anatomical annulus another aortic array assessed assessment automated automatic automatically balanced balloon benefits best bland bull cardiac cardiologists cardiology clinic coefficient cohort coils complement comprehensive computed concerned constrained context contrast correlate correlation correlations dependency derived developed device diameter diameters dimension disease diseases early enrolled essential evaluation evaluations even faster flexibility furthermore gated geometrical geometry growing hamburg hardware health heart highly hospital implantation inter intervention intraclass invasive ionizing junction lancet late literature manual medical medicine minimally minimize model models navigator necessity operator optimal partner patient patients planning play plot population precise prerequisite prevalent previous problem procedure procedures prosthesis protocol protocols proven proves radiation radiology randomly rater recent reduce reformat reliability reliable replace root scanning screening selected selection selectively sets shape sievers statistician structures studies surrounding system systematically tomography toxic training treatment trial tubular turbo valve valvular variability views volume whole width yang