Automatic Stem Cell Agarose Gel Analysis

• Main Authors: Zi-Xuan Yan, 嚴志軒
• Other Authors: Yen-Ping Chu
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/27854201899976310047

碩士 === 國立中興大學 === 資訊科學研究所 === 93 === 　　In medical research, the stem cell can be used to diagnose diseases by the biological experimental technique of the Agarose Gel Electrophoresis. Its experimental result can be stored into the image, calling it Agarose Gel Electrophoresis Image; hence one can use the image processing software effectively and efficiently to analyze the image. However, the majority of the stem cell images are still interpreted and corresponded to the possible diseases manually. It is not only a laborious and time-consuming job, but also may likely cause various erroneous diagnoses. The purpose of this thesis is to develop a software system automatically analyzing the Agarose Gel Electrophoresis Image. A crucial step before interpreting an Agarose Gel Electrophoresis Image is to segment all the objects from the image and extracts the underlying information from them. This thesis hence proposes an automatic stem cell segmentation method (ASCS method) to segment the objects from the Agarose Gel automatically and analyze them to diagnose diseases. 　　The ASCS method uses the K-mean method to classify all pixels in the image first; carries out object segmenting; extracts the expected objects according to the characteristics of each different kind of objects, and then diagnoses disease by checking up the expected objects. Its main advantage is no parameter needed for this method. The experimental results show that the proposed method provides an accuracy of 85~95% in automatically objects extracting, and it also marks the possible expected objects which is convenient for the user manually to carry on his final judgment. The ASCS method cannot only minimize human interaction, but also reduce the time consuming problem.