User-Guided Segmentation of the Frontal Lobe Using Fuzzy Rule-Based Active Contour Model
Yuji Fujiki, Syoji Kobashi, Katsuya Kondo, Yutaka Hata, Mieko Matsui*
Brain Function Research Institute, Inc.
Proc. of 1st Int. Conf. on Soft Computing and Intelligent Systems, CD-ROM, (2002)
This paper describes a computer-aided system for extracting the frontal lobe from 3-D human brain SPGR MR images using fuzzy rule-based active contour model (ACM). The volumetry and surface rendering of the frontal lobe are effective for diagnosing the cerebral diseases with local atrophy. The proposed system uses criterial curves consisting the central sulcus and the sylvian fissure drawn by a user using proposed graphical user interface (GUI). The criterial curves are optimized by fuzzy rule-based ACM. The fuzzy rule-based ACM can represent physiciansf knowledge with fuzzy if-then rules. Using the optimized curves and the anterior and posterior commissure (AC and PC), the frontal lobe is segmented automatically. The experimental results showed that the proposed system could segment the frontal lobe with high accuracy.