Document Type : Full Research Paper

Authors

1 Electrical and Computer Engineering Department, Faculty of Engineering, University of Tehran School of Intelligent Systems, Institute for Studies in Theoretical Physics and Mathematics

2 Electrical and Computer Engineering Department, Faculty of Engineering, University of Tehran School of Intelligent Systems, Institute for Studies in Theoretical Physics and Mathematics Radiology Research Image Analysis Laboratory, Henry Ford Health System - Detroit, USA

3 Radiology Department, Faculty of Medicine, University of Tehran

10.22041/ijbme.2005.13578

Abstract

Based on a discrete dynamic contour model, a method for segmentation of brain structures like thalamus and red nucleus from magnetic resonance images (MRI) is developed. A new method for solving common problems in extracting the discontinuous boundary of a structure from a low contrast image is presented. External and internal forces deform the dynamic contour model. Internal forces are obtained from local geometry of the contour, which consist of vertices and edges, connecting adjacent vertices. The image data and desired image features such as image energy are utilized to obtain external forces. The problem of low contrast image data and unclear edges in the image energy is overcome by the proposed algorithm that uses several methods like thresholding, unsupervised clustering methods such as fuzzy C-means (FCM), edge-finding filters like Prewitt, and morphological operations. We also present a method for generating an initial contour for the model from the image data automatically. Evaluation and validation of the methods are conducted by comparing radiologist and automatic segmentation results. The average of the similarity between segmentation results is 0.8 for the left and right thalami indicating excellent performance of the new method. Additional noise and intensity inhomogeneity changed the evaluation results slightly illustrating the robustness of the proposed method to the image noise and intensity inhomogeneity. 

Keywords

Main Subjects

[1]     Tamraz JC, Comair YG; Atlas of Regional Anatomy of the Brain using MRI with Functional Correlations; Springer-Verlag, Berlin Heidelberg; 2000.
[2]     Barra V, Boire J; Automatic segmentation of subcortical brain structures in MR images using information fusion; Med Imag IEEE Trans 2001; 20: 549-558.
[3]     Hendelman WJ; Atlas of Functional Neuroanatomy; CRC Press LLC; 2000.
[4]     Amini L, Soltanian-Zadeh H, Locas C; Segmentation of specific brain structures from MRI; Proc of the Int Conf on Diag Imag and Anal; Shanghai, China, August 2002: 113-118.
[5]     Sonka M, Tadikonda SK and Collins L; Knowledgebased interpretation of MR brain images; Med Imag IEEE Trans 1996; 15: 443-452.
[6]     Xue JH, Ruan S, Morehi B, Revenu M, Bloyet D, Philips W; Fuzzy modeling of knowledge for MRI brain structure segmentation; Med Imag IEEE Trans 2000; 1: 617-620.
[7]     Saha PK, Udupa JK; Optimum image thresholding via class uncertainty and region homogeneity; Pattern Analysis & Machine Intelligence, IEEE Trans 2001; 23: 689-706.
[8]     Kass M, Witkin A, Terzopoulos D; Snakes: Active contour models; Int Comput Vision 1988; 1: 321-331.
[9]     McInerney T, Terzopoulos D; Deformable models in medical image analysis: A survey; Med Image Anal, IEEE Trans 1996; 1: 91-108.
[10] Sethian JA; Level Set Methods and Fast Marching Methods (2ed); Cambridge, UK, Cambridge Univ Press; 1999.
[11] Davatzikos C, Prine JL; Brain image registration based on curve mapping; Biomed Imag Los Alamitos CA USA, Proc IEEE Workshop 1994: 245-254.
[12] Ghanei A, Soltanian Zadeh H, Windham JP; Segmentation of the hippocampus from brain MRI using deformable contours; Computerized Med Imag and Graphics 1998; 22: 203-216.
[13] Ghanei A, Soltanian Zadeh H, Windham JP; A deformable model for hippocampus segmentation: Improvements and extension to 3D; Computer in Biology and Medicine 1998; 28: 239-258.
[14] Shen D, Moffat S, Resnick SM, Davatzikos C; Measuring size and shape of the hippocampus in MR images using a deformable shape model; Neuroimag in Press.
[15] Lobregt S, Viergever MA; A discrete dynamic contour model; Med Imag, IEEE Trans 1995; 14: 12-24.
[16] Faugeras O; Three dimensional computer vision: A Geometric Viewpoint; Cambridge MA, MIT Press; 1993.
[17] Teodiridis S, Kourtoumbas K; Pattern Recognition; Academic Press; 1998.
[18] Jain AK, Duin RPW, Mao J; Statistical pattern recognition: A review; Pat Ana & Mach Intell, IEEE Trans 2000; 22(1): 4 37.
[19] Parker JR; Algorithms for Image Processing and Computer Vision; John Wiley & Sons Inc, New York; 1997: 23-29.
[20] امینی لادن؛ تحلیل ساختارهای خاص مغزی از روی تصاویر MRI؛ پایان‌نامه کارشناسی ارشد، دانشگاه تهران، 1381.
[21] Barra V, Boire JY; Tissue segmentation on MR images by a possibilistic clustering on a 3D wavelet representation; J Magn Reson Imag 2000; 11: 267-278.
[22] Krishnapuram R, Keller J; The possibilistic c-means algorithm: Insights and recommendations; Fuzzy Syst, IEEE Trans 1996; 4: 385-393.
[23] Gonzalez RC, Woods RE; Digital Image Processing; Addison-Wesley Publishing Company; 1992.
[24] Pitiot A, Toga AW, Thompson PM; Adaptive elastic segmentation of brain MRI via shape-model-guided evolutionary programming; Med Imag, IEEE Trans 2002; 21: 910-923.
[25] Chalana V, Kim Y; A methodology for evaluation of boundary detection algorithms on medical images; Med Imag, IEEE Trans 1997; 16: 642-652.
[26] Brejl M, Sonka M; Object localization and border detection criteria design in edge-based image segmentation: Automated learning from examples; Med Imag, IEEE Trans 2000; 19: 973-985.
[27] Duta N, Sonka M; Segmentation and interpretation of MR brain images: An improved active shape model; Med Imag, IEEE Trans 1998; 17: 1049-1062.