Segmentation of the Skull in MRI Volumes Using Deformable Model. Segmentation par Modèle Déformable des Régions Osseuses de la tête dans les Volumes IRM

Segmentation of the Skull in MRI Volumes Using Deformable Model

Segmentation par Modèle Déformable des Régions Osseuses de la tête dans les Volumes IRM

Hilmi Rifai Isabelle Bloch  Seth Hutchinson 

Ecole Nationale Supérieure des Télécommunications, départementTSI, CNRS URA 820, 46 rue Barrault, 75013 Paris, France

University of Illinois at Urbana-Champaign, 2017 Beckman Institute, 405 North Matthews Avenue, Urbana, IL 61801

20 July 1998
31 August 1999
| Citation



In this paper, we present a new approach for segmenting regions of bone in MRI volumes using a deformable model. Our method takes into account the partial volume effects that occur with MRI data, thus permitting a precise segmentation of these bone regions . Partial volume is estimated, in a narrow band around the deformable model, at each iteration of the propagation of the model . Segmentation of the skull in medical imagery is an important stage in applications that require the construction of realistic models of the head. Such models are used, for example, to simulate the behavior of electro- magnetic fields in the head and to model the electrical activity of the cortex in EEG andMEG data. Our segmentation method begins with a pre-segmentation stage, in which a preliminary segmentation of the skull is constructed using a region-growing method . The surface which bounds the pre-segmented skull region offers an automatic 3D initialization of the deformable model . This surface is propagated (in 3D) in the direction of its normal. This propagation is achieved using level set method, thus permitting changes to occur in the topology of the surface as it evolves, an essential capability for our problem. The speed at which the surface evolves is a function of the estimated partial volume. This provides a sub-voxel accuracy in the resulting segmentation .


Dans ce papier, nous présentons une méthode de segmentation par modèle déformable des régions osseuses de la tête à partir de données IRM 3D. Cette segmentation prend en compte l'effet du volume partiel présent en IRM permettant ainsi une segmentation précise de l'os. La segmentation du crâne est une étape importante dans les applications nécessitant la construction d'un modèle réaliste de la tête . Ce type de modèle est utilisé, entre autres, pour la simulation du comportement d'un champ électromagnétique dans les tissus de la tête, ainsi que pour la modélisation de l'activité électrique du cortex en EEG et MEG. La méthode de segmentation proposée commence par une pré-segmentation du crâne avec une technique de croissance de région. Le résultat de la pré-segmentation est ensuite raffiné par la propagation, en 3D, de la surface de la région pré-segmentée dans le sens de la normale à cette surface. La propagation est réalisée par la détection des courbes de niveau d'une hypersurface, permettant ainsi des changements de topologie avantageux dans notre cas . L'effet du volume partiel est pris en considération lors de la formulation du terme de vitesse de la surface propagée, ce qui permet de réaliser une segmentation sub-voxelique du crâne.


Segmentation of the skull, MRI volumes, 3D deformable model, level sets, partial volume estimation .

Mots clés 

Segmentation du crâne, données IRM, modèle déformable 3D, courbes de niveau, estimation du volume partiel .

1. Introduction
2. Segmentation du Crâne par Modèle Déformable[
3. Propagation de Fronts par la Détection des Courbes de Niveaux
4. Estimation du Volume Partiel et Calcul de la Vitesse du Modèle
5. Détermination des Paramètres du Modèle
6. Résultats
7. Conclusions

[1] D. Adalsteinsson and J.A. Sethian. "A Fast Level Set Method for Propagating Interfaces", J. Comp. Phys, 118(2) :269-277, 1995. 

[2] L.D. Cohen, E. Bardinet and N. Ayache. "Surface Reconstruction Using Active Contour Models", In Conference on Geometric Methods in Computer Vision, volume 2031, pages 38-50.SPIE, 1993. 

[3] C. Jaggi, S. Ruan, J. Fadili and D. BLOYET. "ApprocheMarkovienne pour la Segmentation 3D des Tissus Cérébraux en IRM", In GRETSI'97, pages 327-330,1997.

[4] C. Studholme, D.L.G. Hill and D.J. Hawkes. "Automated 3-D Registration of MR and CT Images of the Head", Medical Image Analysis, 1(2) :163-175, 1996. 

[5] I. Cohen and L.D. Cohen. "A Hybrid Hyperquadric Model for 2-D and 3-D Data Fitting", In12th IEEE International Conference on Pattern Recognition (ICPR'94),volume B, pages 403-405, 1994. 

[6] L. D. Cohen. "On Active Contour Models and Balloons", CVGIP : 1U, 53(2) :211-218,1991. 

[7] D. Rueckert, P. Burger, S.M. Forbat, R.D. Mohiaddin and G.Z. Yang. "Automatic Tracking of the Aorta in Cardiovascular MR Images Using Deformable Models", IEEETranson Medical Imaging, 16(5):581-590,1997. 

[8] D. S. Yoo, L. Lemieux and P. Tofts. "Validation of an Automated Method to Detect Skull Contours in MR Images", In Oxford, editor, MIUA'97, March 1997. 

[9] C. Davatzikos and R.N. Bryan. "Using a Deformable Model to Obtain a Shape Representation of the Cortex", IEEE trans on Medical Imaging, 15(6) :785795, December 1996. 

[10] F.W. Wehli, D. Shaw and J.B. Kneeland. Biomedical Magnetic Resonance Imaging : Principles, Methodology and Applications. VCH, 1988. 

[11] H. Rifai, I. Bloch, J. Wiart, L. Garnero and D. Dormont. "Segmentation, Suivi, Modélisation3D et Recalage del'Oreille Interne APartirdeDonnées IRM", InRFIA,volume 3, pages 71-80, 1998. 

[12] H.S. Choi, D.R. Haynor and Y Kim. "Partial Volume Tissue Classification of Multichannel Magnetic Resonnance Images-a Mixel Model", IEEETrans on Medical Imaging, 10(3) :395-407, 1991. 

[13] J-F.Mangin. Mise en Correspondance d'Images Médicales 3D MultiModalités Multi-individus pour la Corrélation Anatomofonctionnelle Cérébrale. PhD thesis, ENST-E010, 1995. 

[14] J. Gee, M. Reivich and R. Bajcsy. "Elastically Deforming 3D Atlas to Match Anatomical Brain Images", Journal of Computer Assisted Tomography, 17(2) :225-236, 1993. 

[15] J.W.H. Meijs, F.G.C. Bosch,M.J. Peters and F.H. LopesDa Silva. "On the Magnetic Field Distribution Generated by a Dipolar Current Source Situated in a Realistically Shaped Compartment Model of the Head", Electroenceph. andClin. Neurophysiol, 66 :286-296, 1987. 

[16] K. Held, E.R. Kops, B.J. Krause, W.M. Wells, R. Kikinis and H.W. Mullergartner. "Markov Random Field Segmentation of Brain MR Images", IEEE Transon Medical Imaging, 16(6) :878-886, December 1997.

[17] K. Vincken, A. Koster and M. Viergever. "Probabilistic Segmentation of Partial Volume Voxels", Pattern Recognition Letters, pages 477-484, May 1994. 

[18] S. Lobregt and M.A. Viergever. "A Discrete Dynamic Contour Model",IEEE TransonMedicalImaging, 14(l) :12-24, 1995.

[19] M. Kass, A. Witkin and D. Terzopoulos. "Snakes : Active Contour Models", IJCV, 1(4) :321-331, 1988. 

[20] M.A.Jensenand Y. Rahmat-Samii. "EM Interaction of Handset Antennas and a Human in Personal Communications",Proceedings ofIEEE, 83(1) :7-17, 1995. 

[21] G. Marin. Utilisationde laMéthode des Eléments Finispour le Calcul des ChampsElectromagnétiquesA l'Aide d'un Modèle Réaliste de Tite en MEG et EEG. PhD thesis,Université PARIS XI, 1997. 

[22] T. McInerney and D. Terzopoulos. "Deformable Models in Medical Image Analysis : a Survey", MedicalImage Analysis, 1(2) :91-108, 1996. 

[23] S. Osher and J. A. Sethian. "Fronts Propagating with Curvature Dependent Speed : Algorithms Based on Hamilton-Jacobi Formulation",J.ofComputational Physics, 79 :12-49, 1988. 

[24] P. Clarysse, D. Friboulet and I.E. Magnin. "Tracking Geometrical Descriptors on 3-D Deformable Surfaces :Application to the Left-ventricular Surface of the Heart",IEEE trans MedicalImaging, 16(4) :392-404, August 1997. 

[25] R. Malladi, J. A. Sethian and B. C. Vemuri. "Shape Modeling with Front Propagation : A Level Set Approach", IEEE-PA MI, 17(2):158-175, February 1995. 

[26] P. Santago and H. Gage. "Quatification of MR Brain Images by Mixture Density and Partial Volume Modeling", IEEE Trans on Medical Imaging, 12(3) :566--574. 1993. 

[27] JA. Sethian. LevelSetMethods. Cambridge University Press, 1996. 

[28] J.A. Sethian and J. Strain. "Crystal Growth and Dendritic Solidification", Journal ofComputational Physics, 98 :231-253, 1992. 

[29] G. Subsol. Construction Automatique d'Atlas Anatomiques Morphométriques A Partir d'Images Médicales Tridimentionnelles. PhD thesis, Ecole centralede Paris, 1995. 

[30] T. Geraud, J.-F. Mangin, I. Bloch and H. Maitre. "Segmenting Internal Structures in 3D MR Images of the Brain by Markovian Relaxation on a Watershed Based Adjacency Graph", InIEEE International Conference on Image Processing ICIP-95, pages 548-551, 1995.

[31] T. Geraud, L. Aurdal, H. Maitre, I. Blochand C. Adamsbaum. "Estimation of Partial Volume Effect Using Spatial Context . Application to Morphometry in Cerebral Imaging", In IEEE Medical Imaging Conference, 1995. 

[32] T. Heinonen, H. Eskola, P. Dastidar, P. Laarne and J. Malmivuo. "Segmentation of TI MR Scans for Reconstruction of Resistive Head Models", Comput MethodsProgramsBiomed, 3(54) :173-181, 1997. 

[33] V. Casselles, R. Kimmel and G. Sapiro. "Geodesic Active Contours", IJCV, 22(l) :61-79, 1997. 

[34] J. Wiart and R. Mittra. "Calculation of the Power Absorbed by Tissues in Case of Hand Set Mobile Antenna Close to Biological Tissue", IEEE APS, Baltimore, 1996. 

[35] D.J. Williams and M. Shah. "A Fast Algorithm for Active Contours and Curvature Estimation", CVGIP : Image Understanding, 55(l) :14-26, 1992. 

[36] H. Soltanian-Zadeh and J.P. Windham. "A Multiresolution Approach for Contour Extraction from Brain Images", MedPhys,12(24):1844-1853,1997. 

[37] Y.L. Fok, J.C.K. Chan and R.T. Chin. "Automated Analysis of Nervecell Images Using Active Contour Models", IEEE trans Medical Imaging, 15(3):353-368, June 1996.