MR brain image segmentation using fuzzy clustering

• Published in: FUZZ-IEEE'99. 1999 IEEE International Fuzzy Systems. Conference Proceedings (Cat. No.99CH36315) Vol. 2; pp. 853 - 857 vol.2
• Main Authors: Ock-Kyung Yoon, Dong-Min Kwak, Dong-Whee Kim, Kil-Houm Park
• Format: Conference Proceeding
• Language: English
• Published: IEEE 1999
• Subjects: Biomedical imaging; Brain; Detection algorithms; Image segmentation; Lesions; Magnetic resonance; Magnetic resonance imaging; Optical imaging; Ultrasonic imaging; X-ray imaging
• ISBN: 9780780354067; 0780354060
• ISSN: 1098-7584
• DOI: 10.1109/FUZZY.1999.793060

In anatomical aspects, magnetic resonance (MR) imaging offers more accurate information for medical examination than other medical images such as X-ray, ultrasonic and CT images. In this paper, an automated segmentation and lesion detection algorithm are proposed for axial MR brain images. The proposed segmentation algorithm consists of two steps in order to reduce computation time for classifying tissues. In the first step, the cerebrum region is extracted by using thresholding, morphological operation, and labeling algorithm. In the second step, white matter, gray matter, and cerebrospinal fluid in the cerebrum are detected using fuzzy c-means (FCM) algorithm. The new lesion detection algorithm uses anatomical knowledge and local symmetry. A symmetric measure is defined to quantify the normality of MRI slice, which is based on the number of pixels, moment invariants, and Fourier descriptors. The proposed method has been applied to forty normal and abnormal slices. The experimental results show that the proposed segmentation algorithm is appropriate for classifying a large amount of axial brain MR data, and also show that the proposed lesion detection algorithm is successful.