3D Geometric Analysis of the Pediatric Aorta in 3D MRA Follow-Up Images with Application to Aortic Coarctation

Summary

Background: Coarctation of the aorta is one of the most common congenital heart diseases. Despite different treatment opportunities, long-term outcome after surgical or interventional therapy is diverse. Serial morphologic follow-up of vessel growth is necessary, because vessel growth cannot be predicted by primer morphology or a therapeutic option. Objectives: For the analysis of the long-term outcome after therapy of congenital diseases such as aortic coarctation, accurate 3D geometric analysis of the aorta from follow-up 3D medical image data such as magnetic resonance angiography (MRA) is important. However, for an objective, fast, and accurate 3D geometric analysis, an automatic approach for 3D segmentation and quantification of the aorta from pediatric images is required. Methods: We introduce a new model-based approach for the segmentation of the thoracic aorta and its main branches from follow-up pediatric 3D MRA image data. For robust segmentation of vessels even in difficult cases (e.g., neighboring structures), we propose a new extended parametric cylinder model that requires only relatively few model parameters. Moreover, we include a novel adaptive background-masking scheme used for least-squares model fitting, we use a spatial normalization scheme to align the segmentation results from follow-up examinations, and we determine relevant 3D geometric parameters of the aortic arch. Results: We have evaluated our proposed approach using different 3D synthetic images. Moreover, we have successfully applied the approach to follow-up pediatric 3D MRA image data, we have normalized the 3D segmentation results of follow-up images of individual patients, and we have combined the results of all patients. We also present a quantitative evaluation of our approach for four follow-up 3D MRA images of a patient, which confirms that our approach yields accurate 3D segmentation results. An experimental comparison with two previous approaches demonstrates that our approach yields superior results. Conclusions: From the results, we found that our approach is well suited for the quantification of the 3D geometry of the aortic arch from follow-up pediatric 3D MRA image data. In future work, this will enable to investigate the long-term outcome of different surgical and interventional therapies for aortic coarctation.

Keywords

3D vessel segmentation - 3D model-based segmentation - cylindrical intensity model - aortic coarctation - pediatric image data - follow-up MRA - thoracic aorta

^* These authors contributed equally to this work

• References

• 1 Eichhorn J, Fink C, Delorme S, Ulmer H. Rings, Slings and other Vascular Abnormalities – Ultra-fast Computed Tomography and Magnetic Resonance Angiography in Pediatric Cardiology. Zeitschrift für Kardiologie 2004; 93 (Suppl. 03) 201-208.
• 2 Abbruzzese P, Aidala E. Aortic Coarctation: An Overview. J Cardiovasc Med 2007; 8: 123-128.
• 3 Waldhausen J, Nahrwold D. Repair of coarctation of the aorta with a subclavian flap. J Thorac Cardiovasc Surg 1966; 51: 532-533.
• 4 Vossschulte K. Surgical Correction of Coarctation of the Aorta by an “Isthmusplastic” Operation. Thorax 1961; 16 (Suppl. 04) 338-345.
• 5 von Kodolitsch Y, Aydin MA, Koschyk DH, Loose R, Schalwat I, Karck M, Cremer J, Haverich A, Berger J, Meinertz T, Nienaber CA. Predictors of Aneurysmal Formation after Surgical Correction of Aortic Coarctation. J Am Coll Cardiol 2002; 39 (Suppl. 04) 617-24.
• 6 Behrens T, Rohr K, Stiehl H. Robust Segmentation of Tubular Structures in 3D Medical Images by Parametric Object Detection and Tracking. IEEE Transactions on Systems, Man and Cybernetics 2003; 33 (Suppl. 04) 554-561.
• 7 Taeprasartsit P, Higgins W. Method for Extracting the Aorta from 3D CT Images. In: Proc. SPIE Medical Imaging 2007: Image Processing (MI’07). San Diego, CA/USA: 2007
• 8 Isgum I, Staring M, Rutten A, Prokop M, Viergever M, van Ginneken B. Multi-Atlas-Based Segmentation With Local Decision Fusion – Application to Cardiac and Aortic Segmentation in CT Scans. IEEE Transactions on Medical Imaging, 2009; 28 (Suppl. 07) 1000-1010.
• 9 Zhao F, Zhang H, Wahle A, Matthew T, Stolpen A, Scholz T, Sonka M. Congenital Aortic Disease: 4D Magnetic Resonance Segmentation and Quantitative Analysis. Medical Image Analysis 2009; 13 (Suppl. 03) 483-493.
• 10 de Bruijne M, van Ginneken B, Viergever M, Niessen W. Interactive Segmentation of Abdominal Aortic Aneurysms in CTA Images. Medical Image Analysis 2004; 8 (Suppl. 02) 127-138.
• 11 Zheng Y, John M, Liao R, Nöttling A, Boese J, Kempfert J, Walther T, Brockmann G, Comaniciu D. Automatic Aorta Segmentation and Valve Landmark Detection in C-Arm CT for Transcatheter Aortic Valve Implantation. IEEE Transactions on Medical Imaging 2012; 31 (Suppl. 12) 2307-2321.
• 12 Vitanovski D, Ralovich K, Ionasec R, Zheng Y, Suehling M, Krawtschuk W, Hornegger J, Comaniciu D. Personalized Learning-Based Segmentation of Thoracic Aorta and Main Branches for Diagnosis and Treatment Planning. In: Proc. 9th IEEE Internat. Symposium on Biomedical Imaging: From Nano to Macro (ISBI’12). Barcelona, Spain: 2012
• 13 Wörz S, von Tengg-Kobligk H, Henninger V, Rengier F, Schumacher H, Böckler D, Kauczor HU, Rohr K. 3D Quantification of the Aortic Arch Morphology in 3D CTA Data for Endovascular Aortic Repair. IEEE Transactions on Biomedical Engineering 2010; 57 (Suppl. 10) 2359-2368.
• 14 Biesdorf A, Rohr K, Feng D, von Tengg-Kobligk H, Rengier F, Böckler D, Kauczor HU, Wörz S. Segmentation and Quantification of the Aortic Arch using Joint 3D Model-Based Segmentation and Elastic Image Registration. Medical Image Analysis 2012; 16: 1187-1201.
• 15 Wörz S, Alrajab A, Arnold R, Eichhorn J, von Tengg-Kobligk H, Schenk JP, Rohr K. 3D Geometric Analysis of the Aorta in 3D MRA Follow-Up Pediatric Image Data. In: Proc. SPIE Medical Imaging 2014: Image Processing (MI’14). San Diego, CA/USA: 2014
• 16 Noordmans H, Smeulders A. High Accuracy Tracking of 2D/3D Curved Line Structures by Consecutive Cross-Section Matching. Pattern Recognition Letters 1998; 19 (Suppl. 01) 97-111.
• 17 Wörz S, Rohr K. Segmentation and Quantification of Human Vessels using a 3-D Cylindrical Intensity Model. IEEE Transactions on Image Processing 2007; 16 (Suppl. 08) 1994-2004.
• 18 Friman O, Hindennach M, Kühnel C, Peitgen HO. Multiple Hypothesis Template Tracking of Small 3D Vessel Structures. Medical Image Analysis 2010; 14 (Suppl. 02) 160-171.
• 19 Press W, Flannery B, Teukolsky S, Vetterling W. Numerical Recipes in C. Cambridge: Cambridge University Press; 1992
• 20 Caselles V, Kimmel R, Sapiro G. Geodesic Active Contours. International Journal of Computer Vision 1997; 22 (Suppl. 01) 61-79.