Active Contour and Segmentation Models using Geometric PDE’s for Medical Imaging
This paper is devoted to the analysis and the extraction of information from bio-medical images. The proposed technique is based on object and contour detection, curve evolution and segmentation. We present a particular active contour model for 2D and 3D images, formulated using the level set method, and based on a 2-phase piecewise-constant segmentation. We then show how this model can be generalized to segmentation of images with more than two segments. The techniques used are based on the Mumford-Shah  model. By the proposed models, we can extract in addition measurements of the detected objects, such as average intensity, perimeter, area, or volume. Such informations are useful when in particular a time evolution of the subject is known, or when we need to make comparisons between different subjects, for instance between a normal subject and an abnormal one. Finally, all these will give more informations about the dynamic of a disease, or about how the human body growths. We illustrate the efficiency of the proposed models by calculations on two-dimensional and three-dimensional bio-medical images.
Unable to display preview. Download preview PDF.
- 8.Chan, T., Vese, L.: Image segmentation using level sets and the piecewiseconstant Mumford-Shah model. UCLA CAM Report 00-14 (2000).Google Scholar
- 10.Kichenassamy, S., Kumar, A., Olver, P., Tannenbaum, A., Yezzi, A.: Gradient flows and geometric active contour models. Proceedings of ICCV, Cambridge, (1995) 810–815.Google Scholar
- 13.Malladi, R., Kimmel, R., Adalsteinsson, D., Caselles, V., Sapiro, G., Sethian, J.A.: A Geometric Approach to Segmentation and Analysis of 3D Medical Images. Proc. of IEEE/SIAM Workshop on Biomedical Image Analysis, San-Francisco, California, (1996).Google Scholar
- 14.Malladi, R., Sethian, J.A.: A Real-Time Algorithm for Medical Shape Recovery. Proc. of International Conf. on Computer Vision. Mumbai, India (1998) 304–310.Google Scholar
- 16.Malladi, R., Sethian, J.A., Vemuri, B.C.: A Topology Independent Shape Modeling Scheme. Proc. SPIE Conf. on Geometric Methods in Computer Vision II 2031 (1993) 246–258, San Diego.Google Scholar
- 17.Malladi, R., Sethian, J.A., Vemuri, B.C.: Evolutionary Fronts for TopologyIndependent Shape Modeling and Recovery. Proc. of the Third European Conference on Computer Vision, LNCS 800 (1994) 3–13, Stockholm, Sweden.Google Scholar
- 20.Morel J.M., Solimini, S.: Variational Methods in Image Segmentation. Birkhäuser, PNLDE 14 (1994).Google Scholar
- 23.Samson, C., Blanc-Féraud, L., Aubert, G., Zerubia, J.: A Level Set Model for Image Classification. M. Nilsen et al. (Eds.): Scale-Space’99, LNCS 1682 (1999) 306–317, Springer-Verlag Berlin Heidelberg.Google Scholar
- 24.Sapiro, G., Kimmel, R., Caselles, V.: Measurements in medical images via geodesic deformable contours. Proc. SPIE-Vision Geometry IV, Vol. 2573 (1995), San Diego, California.Google Scholar
- 25.Shah, J.: A Common Framework for Curve Evolution, Segmentation and Anisotropic Diffusion. IEEE Conference on Computer Vision and Pattern Recognition (1996).Google Scholar
- 26.Shah, J.: Riemannian Drums, Anisotropic Curve Evolution and Segmentation. M. Nilsen et al. (Eds.): Scale-Space’99, LNCS 1682 (1999) 129–140, Springer-Verlag Berlin Heidelberg.Google Scholar
- 28.Yezzi, A., Tsai, A., Willsky, A.: A statistical approach to snakes for bimodal and trimodal imagery. Int. Conf. on Computer Vision (1999).Google Scholar
- 30.Zhu, S.C., Lee, T.S., Yuille, A.L.: Region competition: Unifying snakes, region growing, energy/Bayes/MDL for multi-band image segmentation. Proceedings of the IEEE 5th ICCV, Cambridge (1995) 416–423.Google Scholar