Registration with Adjacent Anatomical Structures for Cardiac Resynchronization Therapy Guidance
The clinical applications and benefits of multi-modal image registration are wide-ranging and well established. Current image based approaches exploit cross-modality information, such as landmarks or anatomical structures, which is visible in both modalities. A lack of cross-modality information can prohibit accurate automatic registration. This paper proposes a novel approach for MR to X-ray image registration which uses prior knowledge of adjacent anatomical structures to enable registration without cross-modality image information. The registration of adjacent structures formulated as a partial surface registration problem which is solved using a globally optimal ICP method. The practical clinical application of the approach is demonstrated on an image guided cardiac resynchronization therapy procedure. The left ventricle (segmented from pre-operative MR) is registered to the coronary vessel tree (extracted from intra-operative fluoroscopic images). The proposed approach is validated on synthetic and phantom data, where the results show a good comparison with the ground truth registrations. The vertex-to-vertex MAE was \(3.28\pm 1.18\) mm for 10 X-ray image pairs of the phantom.
KeywordsCardiac Resynchronization Therapy Iterative Close Point Vessel Tree Phantom Data Adjacent Anatomical Structure
Acknowledgements and Disclaimer
The authors are grateful for the support from the Innovate UK grant 32684-234174. The research was supported by the National Institute for Health Research (NIHR) Biomedical Research Centre based at Guys and St Thomas NHS Foundation Trust and Kings College London. The views expressed are those of the authors and not necessarily those of the NHS, NIHR or the Department of Health. Concepts and information presented are based on research and are not commercially available.
- 3.Truong, M.V.N., Aslam, A., Rinaldi, C.A., Razavi, R., Penney, G.P., Rhode, K.: Preliminary investigation: 2D–3D registration of MR and X-ray cardiac images using catheter constraints. In: MICCAI Workshop on Cardiovascular Interventional Imaging and Biophysical Modelling, London, pp. 1–9 (2009)Google Scholar
- 4.Bourier, F., Brost, A., Yatziv, L., Hornegger, J., Strobel, N., Kurzidim, K.: Coronary sinus extraction for multimodality registration to guide transseptal puncture. In: 8th Interventional MRI Symposium, Leipzig, pp. 311–313 (2010)Google Scholar
- 5.Faber, T.L., Santana, C.A., Garcia, E.V., Candell-Riera, J., Folks, R.D., Peifer, J.W., Hopper, A., Aguade, S., Angel, J., Klein, J.L.: Three-dimensional fusion of coronary arteries with myocardial perfusion distributions: clinical validation. J. Nuclear Med. 45(5), 745–753 (2004)Google Scholar
- 6.Jolly, M.-P., Guetter, C., Lu, X., Xue, H., Guehring, J.: Automatic segmentation of the myocardium in cine MR images using deformable registration. In: Camara, O., Konukoglu, E., Pop, M., Rhode, K., Sermesant, M., Young, A. (eds.) STACOM 2011. LNCS, vol. 7085, pp. 98–108. Springer, Heidelberg (2012). doi: 10.1007/978-3-642-28326-0_10 CrossRefGoogle Scholar
- 9.Yang, J., Li, H., Jia, Y.: Go-ICP: solving 3D registration efficiently and globally optimally. In: 2013 IEEE International Conference on Computer Vision, pp. 1457–1464 (2013)Google Scholar
- 10.Yang, J., Li, H., Campbell, D., Jia, Y.: Go-ICP: a globally optimal solution to 3D ICP point-set registration. IEEE Trans. Pattern Anal. Mach. Intell. PP(99), 1–14 (2015)Google Scholar