Rough-Fuzzy Clustering Algorithm for Segmentation of Brain MR Images - Rough Fuzzy Image Analysis: Foundations and Methodologies - page 16

2

Rough-Fuzzy Clustering Algorithm
for Segmentation of Brain MR
Images

2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1

2.2 Fuzzy C-Means and Rough Sets 2-3

. . . . . . . . . . . . . . . . . . . .

Fuzzy C-Means • Rough Sets

2.3 Rough-Fuzzy C-Means Algorithm 2-5

. . . . . . . . . . . . . . . . . .

Objective Function • Cluster Prototypes • Details of
the Algorithm

2.4 Pixel Classification of Brain MR Images . . . . . . . . . . . 2-7

2.5 Segmentation of Brain MR Images 2-9 . . . . . . . . . . . . . . . . .

Feature Extraction • Selection of Initial Centroids

2.6 Experimental Results and Discussion 2-13

. . . . . . . . . . . . . .

Pradipta Maji
Machine Intelligence Unit, Indian Statistical
Institute, Kolkata, 700 108, India

Haralick's Features Versus Proposed Features •
Random Versus Discriminant Analysis Based
Initialization • Comparative Performance Analysis

2.7 Conclusion 2-18

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Sankar K. Pal
Machine Intelligence Unit, Indian Statistical
Institute, Kolkata, 700 108, India

Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-18

Bibliography 2-19

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2.1 Introduction

Segmentation is a process of partitioning an image space into some non-overlapping mean-
ingful homogeneous regions. The success of an image analysis system depends on the quality
of segmentation (Rosenfeld and Kak, 1982). A segmentation method is supposed to find
those sets that correspond to distinct anatomical structures or regions of interest in the
image. In the analysis of medical images for computer-aided diagnosis and therapy, seg-
mentation is often required as a preliminary stage. However, medical image segmentation is
a complex and challenging task due to intrinsic nature of the images. The brain has a par-
ticularly complicated structure and its precise segmentation is very important for detecting
tumors, edema, and necrotic tissues, in order to prescribe appropriate therapy (Suetens,
2002).
In medical imaging technology, a number of complementary diagnostic tools such as x-
ray computer tomography (CT), magnetic resonance imaging (MRI), and position emission
tomography (PET) are available. MRI is an important diagnostic imaging technique for
the early detection of abnormal changes in tissues and organs. Its unique advantage over
other modalities is that it can provide multispectral images of tissues with a variety of

2-1