Design And Realization Of Lung Tumor Segmentation System Based On CT Image

It is well known that cancer has become one of the most important causes of human death,and lung cancer is the leading cause of cancer death.Lung tumor segmentation is an important part of computer assisted diagnosis.At present,it is mainly through the way of doctors manual delineation of lung tumor.But this method is not only inefficient,but also has certain subjectivity.In recent years,there has been a great deal of research on the segmentation of lung tumors.Because of the diversity of the tumors,it is still not possible to use an algorithm to accurately and automatically segment all types of lung tumors.In this paper,we design and implement a complete lung tumor segmentation system by using VTK and Qt toolkit.The main contents of this paper are as follows:(1)The boundary region growing interactive segmentation algorithm was used to achieve a segmentation result of two-dimensional adhesion lung tumors,which solves the problem of over segmentation of region growing algorithm.The accuracy of the segmentation results is verified by the comparison between a large number of segmentation experiments of system and the manual segmentation results.(2)According to the actual needs of the lung tumor segmentation system,the system is divided into the reading and saving module,the display module,the interactive module and the segmentation module.The reading and saving module realized the unified management of image data through the VTK interface.The display module realized the images two-dimensional and three-dimensional visualization,it is also realized the interactive display function module of the system at the same time.The interactive module realized interaction of external events and internal events.The segmentation module realized 2D segmentation of lung tumors.And it is also realized 3D segmentation and 3D reconstruction of pulmonary tumors through human-computer interaction.(3)In order to solve the issue of time-consuming of region growing algorithm in large amount of CT images data,the VOI extraction and selection of multi-seed points was used in the system.By using isosurface threshold adjustment tool,which provide users a best display effection of 3D segmentation and reconstruction.At last,the function of the system is tested.And the system is used to the experiment of segmention.The results show that the design of system can achieve the desired design goals in this paper.It is not only has a good segmentation efficient,but also has a certain application value.