Lung Registration Using Automatically Detected Landmarks

Summary

Objectives: Accurate registration of lung CT images is inevitable for numerous clinical applications. Usually, nonlinear intensity-based methods are used. Their accuracy is typically evaluated using corresponding anatomical points (landmarks; e.g. bifurcations of bronchial and vessel trees) annotated by medical experts in the images to register. As image registration can be interpreted as correspond ence finding problem, these corresponding landmarks can also be used in feature-based registration techniques. Recently, approaches for automated identification of such landmark correspondences in lung CT images have been presented. In this work, a novel combination of variational nonlinear intensity-based registration with an approach for automated landmark correspond ence detection in lung CT pairs is presented and evaluated.

Methods: The main blocks of the proposed hybrid intensity- and feature-based registration scheme are a two-step landmark correspondence detection and the so-called CoLD (Combining Landmarks and Distance Measures) framework. The landmark correspondence identification starts with feature detection in one image followed by a blockmatching-based transfer of the features to the other image. The established correspond ences are used to compute a thin-plate spline (TPS) transformation. Within CoLD, the TPS transformation is improved by minimization of an objective function consisting of a Normalized Gradient Field distance measure and a curvature regularizer; the landmark correspondences are guaranteed to be preserved by optimization on the kernel of the discretized landmark constraints.

Results: Based on ten publicly available end-inspiration/expiration CT scan pairs with anatomical landmark sets annotated by medical experts from the DIR-Lab database, it is shown that the hybrid registration approach is superior in terms of accuracy: The mean distance of expert landmarks is decreased from 8.46 mm before to 1.15 mm after registration, outperforming both the TPS transformation (1.68 mm) and a nonlinear registration without usage of automatically detected landmarks (2.44 mm). The improvement is statistically significant in eight of ten datasets in comparison to TPS and in nine of ten datasets in comparison to the intensity-based registration. Furthermore, CoLD globally estimates the breathing-induced lung volume change well and results in smooth and physiologically plausible motion fields of the lungs.

Conclusions: We demonstrated that our novel landmark-based registration pipeline outperforms both TPS and the underlying nonlinear intensity-based registration without landmark usage. This highlights the potential of automatic landmark correspondence detection for improvement of lung CT registration accuracy.

Keywords

Lung - image registration - landmark detection - pulmonary diseases - computer-assisted image processing - anatomic landmarks - computed X ray tomography - mathematical concepts

• References

• 1 Wang H, Dong L, O’Daniel J, Mohan R, Garden AS, Ang KK. et al Validation of an Accelerated ‘Demons’ Algorithm for Deformable Image Registration in Radiation Therapy. Phys Med Biol 2005; 50: 2887-2905.
• 2 Murphy K, van Ginneken B, Reinhardt JM, Kabus S, Ding K, Deng X. et al Evaluation of Registration Methods on Thoracic CT: the EMPIRE10 Challenge. IEEE T Med Imaging 2011; 30: 1901-1920.
• 3 Galbán CJ, Han MK, Boes JL, Chughtai KA, Meyer CR, Johnson TD. et al Computed Tomography-based Biomarker Provides Unique Signature for Diagnosis of COPD Phenotypes and Disease Progression. Nat Med 2012; 18: 1711-1715.
• 4 Rühaak J, Heldmann S, Kipshagen T, Fischer B. Highly Accurate Fast Lung CT Registration. In. Ourselin S, Haynor DR. editors SPIE Medical Imaging 2013 Image Processing. Proceedings of SPIE Volume 8669. Lake Buena Vista, Florida, USA: 2013. Feb 9-14. 86690Y-1-9.
• 5 Castillo E, Castillo R, Martinez J, Shenoy M, Guerrero T. Four-dimensional Deformable Image Registration using Trajectory Modeling. Phys Med Biol 2010; 55: 305-327.
• 6 Castillo E, Castillo R, White B, Rojo J, Guerrero T. Least Median of Squares Filtering of Locally Optimal Point Matches for Compressible Flow Image Registration. Phys Med Biol 2012; 57: 4827-4833.
• 7 Castillo R, Castillo E, Guerra R, Johnson VE, McPhail T, Garg AK. et al A Framework for Evaluation of Deformable Image Registration Spatial Accuracy using Large Landmark Point Sets. Phys Med Biol 2009; 54: 1849-1870.
• 8 Kabus S, Klinder T, Murphy K, van Ginneken B, Lorenz C, Pluim JPW. Evaluation of 4D-CT Lung Registration. In. Yang GZ, Hawkes D, Rueckert D, Noble A, Taylor C. editors MICCAI. 2009. Proceedings of the 12th International Conference on Medical Image Computing and Computer-Assisted Intervention London, UK. Berlin: Springer LNCS; 2009. Sep 20-24 vol5761 747-754.
• 9 Schmidt-Richberg A, Werner R, Ehrhardt J, Wolf JC, Handels H. Landmark-driven Parameter Optimization for Non-linear Image Registration. In. Dawant BM, Haynor DR. editors SPIE Medical Imaging 2011 Image Processing. Proceedings of SPIE Volume 7962. Lake Buena Vista, Florida, USA: 2011. Feb 12-17 OT1-8.
• 10 Riyahi-Alam S, Peroni M, Baroni G, Riboldi M. Regularization in Deformable Registration of Biomedical Images Based on Divergence and Curl Operators. Methods Inf Med 2014; 53: 21-28.
• 11 Likar B, Pernuš F. Automatic Extraction of Corresponding Points for the Registration of Medical Images. Med Phys. 1999; 26: 1678-1686.
• 12 Werner R, Duscha C, Schmidt-Richberg A, Ehrhardt J, Handels H. Assessing Accuracy of Non-linear Registration in 4D Image Data using Automatically Detected Landmark Correspondences. In Ourselin S, Haynor DR. editors SPIE Medical Imaging 2013 Image Processing. Proceedings of SPIE Volume 8669. Lake Buena Vista, Florida, USA: 2013. Feb 9-1 86690Z-1-9.
• 13 Murphy K, van Ginneken B, Klein S, Staring M, de Hoop BJ. Viergever MA. et al Semi-automatic Construction of Reference Standards for Evaluation of Image Registration. Med Image Anal 2011; 15: 71-84.
• 14 Olesch J, Papenberg N, Lange T, Conrad M, Fischer B. Matching CT and Ultrasound Data of the Liver by Landmark Constrained Image Registration. In. Miga MI, Wong KH. editors SPIE Medical Imaging 2009: Visualization, Image-Guided Procedures, and Modeling. Proceedings of SPIE Volume 7261. Lake Buena Vista, Florida, USA: 2009. Feb 8-12 72610G-1-7.
• 15 Papenberg N, Olesch J, Lange J, Schlag PM, Fischer B. Landmark Constrained Non-parametric Image Registration with Isotropic Tolerances. In. Meinzer HP, Deserno TM, Handels H, Tolxdorff T. editors BVM 2009; Proceedings of the 12th Workshop on Bildverarbeitung für die Medizin. 2009 Mar 22-25. Heidelberg, Germany. Berlin: Springer; 2009: 122-126.
• 16 Lange T, Papenberg N, Olesch J, Fischer B, Schlag PM. Landmark Constrained Non-rigid Image Registration with Anisotropic Tolerances. In. Dössel O, Schlegel WC. editors World Congress on Medical Physics and Biomedical Engineering. Sep 7-12 2007. Munich, Germany. Berlin: Springer IFMBE Proceedings; vol. 25/4 2238-2241.
• 17 Hellier P, Barillot C. Coupling Dense and Landmark-Based Approaches for Nonrigid Registration. IEEE T Med Imaging. 2003; 22: 217-227.
• 18 Johnson HJ, Christensen GE. Consistent Landmark and Intensity-based Image Registration. IEEE T Med Imaging 2002; 21: 450-461.
• 19 Kybic J, Unser M. Fast Parametric Elastic Image Registration. IEEE T Image Process 2003; 12: 1427-1442.
• 20 Papademetris X, Jackowski AP, Schultz RT, Staib LH, Duncan JS. Integrated Intensity and Point-Feature Nonrigid Registration. In. Barillot C, Haynor D, Hellier P. editors MICCAI Proceedings of the 7th International Conference on Medical Image Computing and Computer-Assisted Intervention. 2004. Saint-Malo, France. Berlin: Springer LNCS; 2004. Sep 26-29 vol. 3216 763-770.
• 21 Fischer B, Modersitzki J. Combining Landmark and Intensity Driven Registrations. Proc Appl Math Mech 2003; 3: 32-35.
• 22 Haber E, Heldmann S, Modersitzki J. A Scale-Space Approach to Landmark Constrained Image Registration. In. Tai XC, Morken K, Lysaker M, Lie KA. editors SSVM. 2009. Proceedings of the 2nd International Conference on Scale Space Methods and Variational Methods in Computer Vision Voss, Norway: Springer LNCS; 2009. Jun 1-5 vol.5567 612-623.
• 23 Modersitzki J. FAIR: Flexible Algorithms for Image Registration. Philadelphia: SIAM; 2009
• 24 Rohr K. Landmark-based Image Analysis. Norwell, MA, USA: Kluwer Academic Publishers; 2001
• 25 Haber E, Modersitzki J. Intensity Gradient Based Registration and Fusion of Multi-modal Images. Methods Inf Med 2007; 46: 292-299.
• 26 Fischer B, Modersitzki J. Curvature Based Image Registration. J Math Imaging Vis 2003; 18: 81-85.
• 27 Lassen B, Kuhnigk JM, Schmidt M, Krass S, Peitgen HO. Lung and Lung Lobe Segmentation Methods at Fraunhofer MEVIS. In. Beichel RR, de Bruijne M, van Ginneken B, Kabus S, Kiraly AP, Kuhnigk JM. et al. editors Proceedings of the 4th International Workshop on Pulmonary Image Analysis. Toronto, Canada: 2011. Sep 18 185-199.
• 28 Lola11.com [Internet]. LObe and Lung Analysis 2011 [cited 2014 Mar 7>. Available from. http://lola11.com/Results/Overview
• 29 Nocedal J, Wright SJ. Numerical Optimization. 2nd ed. New York: Springer; 2006
• 30 Modersitzki J. Numerical Methods for Image Registration. New York: Oxford University Press; 2004
• 31 Ciarlet PG. Mathematical Elasticity. New York: North-Holland; 1988
• 32 Polzin T, Rühaak J, Werner R, Strehlow J, Heldmann S, Handels H, Modersitzki J. Combining Automatic Landmark Detection and Variational Methods for Lung CT Registration. In. Beichel RR, de Bruijne M, Kabus S, Kiraly AP, Kuhnigk JM. et al. editors Proceedings of the 5th International Workshop on Pulmonary Image Analysis. Nagoya, Japan: 2013. Sep 26 85-96.