Efficient and generalizable statistical models of shape and appearance for analysis of cardiac MRI

We present a framework for the analysis of short axis cardiac MRI, using statistical models of shape and appearance. The framework integrates temporal and structural constraints and avoids common optimization problems inherent in such high dimensional models. The first contribution is the introduction of an algorithm for fitting 3D active appearance models (AAMs) on short axis cardiac MRI. We observe a 44-fold increase in fitting speed and a segmentation accuracy that is on par with Gauss-Newton optimization, one of the most widely used optimization algorithms for such problems. The second contribution involves an investigation on hierarchical 2D+time active shape models (ASMs), that integrate temporal constraints and simultaneously improve the 3D AAM based segmentation. We obtain encouraging results (endocardial/epicardial error 1.43+/-0.49 mm/1.51+/-0.48 mm) on 7980 short axis cardiac MR images acquired from 33 subjects. We have placed our dataset online, for the community to use and build upon.

[1]  Milan Sonka,et al.  Multistage hybrid active appearance model matching: segmentation of left and right ventricles in cardiac MR images , 2001, IEEE Transactions on Medical Imaging.

[2]  S. Mallat.  A wavelet tour of signal processing , 1998 .

[3]  Timothy F. Cootes,et al.  Statistical models of appearance for computer vision , 1999 .

[4]  Simon Baker,et al.  Lucas-Kanade 20 Years On: A Unifying Framework , 2004, International Journal of Computer Vision.

[5]  Jinah Park,et al.  Deformable models with parameter functions for cardiac motion analysis from tagged MRI data , 1996, IEEE Trans. Medical Imaging.

[6]  David T. Gering,et al.  Automatic Segmentation of Cardiac MRI , 2003, MICCAI.

[7]  James S. Duncan,et al.  Estimation of 3-D left ventricular deformation from medical images using biomechanical models , 2002, IEEE Transactions on Medical Imaging.

[8]  Demetri Terzopoulos,et al.  A dynamic finite element surface model for segmentation and tracking in multidimensional medical images with application to cardiac 4D image analysis. , 1995, Computerized medical imaging and graphics : the official journal of the Computerized Medical Imaging Society.

[9]  Dinggang Shen,et al.  An Adaptive-Focus Deformable Model Using Statistical and Geometric Information , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[10]  Hervé Delingette,et al.  Deformable biomechanical models: Application to 4D cardiac image analysis , 2003, Medical Image Anal..

[11]  Alejandro F. Frangi,et al.  Three-dimensional modeling for functional analysis of cardiac images, a review , 2001, IEEE Transactions on Medical Imaging.

[12]  Ghassan Hamarneh,et al.  Deformable Spatio-Temporal Shape Models: Extending ASM to 2D+Time , 2001, BMVC.

[13]  Dinggang Shen,et al.  Hierarchical active shape models, using the wavelet transform , 2003, IEEE Transactions on Medical Imaging.

[14]  Timothy F. Cootes,et al.  Combining point distribution models with shape models based on finite element analysis , 1994, Image Vis. Comput..

[15]  Timothy F. Cootes,et al.  Active Appearance Models , 1998, ECCV.

[16]  Juha Koikkalainen,et al.  Artificially Enlarged Training Set in Image Segmentation , 2006, MICCAI.

[17]  Timothy F. Cootes,et al.  Face recognition using the active appearance model. , 1998, European Conference on Computer Vision.

[18]  James S. Duncan,et al.  Medical Image Analysis , 1999, IEEE Pulse.

[19]  Cornelius Rosse,et al.  Hollinshead's Textbook of Anatomy , 1997 .

[20]  Simon Baker,et al.  Active Appearance Models Revisited , 2004, International Journal of Computer Vision.

[21]  Aaron F. Bobick,et al.  Shape-Driven 3D Segmentation Using Spherical Wavelets , 2006, MICCAI.

[22]  James S. Duncan,et al.  Bending and stretching models for LV wall motion analysis from curves and surfaces , 1992, Image Vis. Comput..

[23]  Jan Flusser,et al.  Image registration methods: a survey , 2003, Image Vis. Comput..

[24]  James S. Duncan,et al.  A Boundary Element-Based Approach to Analysis of LV Deformation , 2005, MICCAI.

[25]  Juha Koikkalainen,et al.  Artificial Enlargement of a Training Set for Statistical Shape Models: Application to Cardiac Images , 2005, FIMH.

[26]  Simon Baker,et al.  Equivalence and efficiency of image alignment algorithms , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[27]  Rachid Deriche,et al.  Implicit Active Shape Models for 3D Segmentation in MR Imaging , 2004, MICCAI.

[28]  I.E. Magnin,et al.  A dynamic 3-D cardiac surface model from MR images , 2005, Computers in Cardiology, 2005.

[29]  Hildur Ólafsdóttir,et al.  Unsupervised motion-compensation of multi-slice cardiac perfusion MRI , 2005, Medical Image Anal..

[30]  Milan Sonka,et al.  Automatic segmentation of echocardiographic sequences by active appearance motion models , 2002, IEEE Transactions on Medical Imaging.

[31]  Juha Koikkalainen,et al.  Statistical shape model of atria, ventricles and epicardium from short- and long-axis MR images , 2004, Medical Image Anal..

[32]  Alex Pentland,et al.  Recovery of Nonrigid Motion and Structure , 1991, IEEE Trans. Pattern Anal. Mach. Intell..

[33]  Laurent D. Cohen,et al.  Tracking and motion analysis of the left ventricle with deformable superquadrics , 1996, Medical Image Anal..

[34]  Arun V. Holden,et al.  Computational biology of the heart , 1998, The Mathematical Gazette.

[35]  Nicholas Ayache,et al.  Dense Non-Rigid Motion Estimation in Sequences of Medical Images Using Differential Constraints , 2004, International Journal of Computer Vision.

[36]  Jens von Berg,et al.  Automated Segmentation of the Left Ventricle in Cardiac MRI , 2003, MICCAI.

[37]  Laurent D. Cohen,et al.  Introducing new deformable surfaces to segment 3D images , 1991, Proceedings. 1991 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[38]  Milan Sonka,et al.  3-D active appearance models: segmentation of cardiac MR and ultrasound images , 2002, IEEE Transactions on Medical Imaging.

[39]  Laurent D. Cohen,et al.  Deformable models for 3-D medical images using finite elements and balloons , 1992, Proceedings 1992 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[40]  Marleen de Bruijne,et al.  Adapting Active Shape Models for 3D Segmentation of Tubular Structures in Medical Images , 2003, IPMI.

[41]  Bjarne Kjær Ersbøll,et al.  Generative Interpretation of Medical Images , 2004 .

[42]  Ghassan Hamarneh,et al.  Deformable spatio-temporal shape models: extending active shape models to 2D+time , 2004, Image Vis. Comput..

[43]  Alejandro F. Frangi,et al.  Active shape model segmentation with optimal features , 2002, IEEE Transactions on Medical Imaging.

[44]  Daniel Rueckert,et al.  Segmentation of 4D cardiac MR images using a probabilistic atlas and the EM algorithm , 2004, Medical Image Anal..

[45]  Dinggang Shen,et al.  An adaptive-focus statistical shape model for segmentation and shape modeling of 3-D brain structures , 2001, IEEE Transactions on Medical Imaging.