An Automatic Liver Extraction Method Using Moving Points and Neural Networks

摘要

Automatic extraction of tumors is needed in a medical field Usually, a 3-dimensional image is composed of the sequence of 2-dimentional CT images. The extraction of a liver field in those images is difficult to be done, because they include various internal organs with similar gray scale levels. Almost of all studies for these problems are not very good for real systems, in this paper, the method using moving points is proposed to analyze the information of CT images and to extract livers. It is shown this method has ability as a preprocessing system of a liver extraction system for extraction of liver tumors. Finally, in order to show the effectiveness of the proposed method, it is applied for real systems.