The geometric Model of Collagen Fiber

• Main Authors: Hsien-Sheng Lee, 李献生
• Other Authors: Ching-Shoei Chiang
• Format: Others
• Language: zh-TW
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/55212531765823239921

碩士 === 東吳大學 === 資訊科學系 === 91 === Given the advanced medical technology nowadays, a lot of medical research is conducted and completed by computer-based automation, for example, by means of Magnetic Resonance Imaging (MRI), Computerized Tomography (CT) and Ultrasonic Scanning. This paper is intended to explore the presently popular topic about further collagen fiber images processing. Collagen fiber is a polymer constituent of extracellular matrix. In the event that it is injured and thus its features are changed, diseases may arise. Hence, it is important to understand the structural characteristics of collagen fibers. For this reason, the focus of this paper is to uncover the structural characteristics of collagen fibers. Since the structure of a fiber has complicated network topology, we analyze the geometric information embodied in the structure of the fiber so as to understand the characteristics of collagen fibers. Besides, we create a geometric model image automated analysis system for collagen fibers. With a series of z-axis-oriented planar image collected by means of a confocal microscope, we build a three-dimensional imaging space wherein a central axis of collagen fibers is then sought for. The information about the central axis is analyzed. Separate any branches that aggregate around intersection points. We determine the full length of a fiber, we retrieve the geometric information about every single fiber, and transform all the fibers into 3D images of the collagen fibers positioned in the three-directional geometric space. Reveal the geometric topological information about the collagen fibers, for example, length, caliber, quantity, number of times of mutual contact, density, direction and curvature of the collagen fibers. With this system, medical professionals only need to input a few parameters in order to retrieve definite related geometrical characteristics rapidly, so that they are able to offer medical service more efficiently.