The Method Of Detecting Pulmonary Nodules By Level Set Model

Lung cancer occurred on the mucosa of bronchus is a kind of mutation in cells,when cells are in an extreme condition,such as the stimulation of toxin.With the development of industry in china,the quality of air is getting worse and worse which causes the high morbidity of lung cancer and makes the patients younger.Although the time of early-stage cancer may last ten or fifteen years,there are mild symptoms in incubation period of cancer which leads most of patients missing the best time to treat.The manifestation of early-stage cancer is lung nodules,so the key to treat lung cancer is detecting nodules fast and accurately.The article uses Level Set Model to extract the contours of nodules.Mainly includes the following aspects of the content:1)The segmentation of pulmonary parenchymaThe article gives an efficient way combined with morphological filter,threshold segmentation,area filter,morphological opening and backfill to segment pulmonary parenchyma.2)Improved Distance Regularized Level Set EvolutionThe basic idea of the Level Set Method is to represent a contour as the zero level set of a higher dimensional function,called a Level Set Function?LSF?,and formulate the motion of the contour as the evolution of the Level Set Function.The article introduces the Distance Regularized Level Set Evolution?DRLSE?.Firstly,to improve DRLSE,Nonlinear isotropic Diffusion Filter and Edge-enhancing Diffusion Filter are introduced into optimizing the edge indicator function g.Then the article optimizes the parameter a of area energy term,but it is not perfect.For overcoming the drawback of the optimized parameter a,the article raises a new binary matrix named Y,which can be got by region growing,to add on the area energy term.Finally,the article attempts a new potential-well function called p₃?s? in the level setregularization term to overcome the drawbacks of single-well function and double-well function which invented by Li.In addition,to compare the new method with the old one in a same segmented condition,the article gives a curve-changed rate as a reference.3)The way of segmenting nodules by coarse segmentation and resegmentationThe article optimizes the initial curve position combining with the way of morphological method and threshold segmentation which is called coarse segmentation.After that,the article uses the improved Level Set Method to extract boundaries of objects.This process is called resegmentation.Through large experiments between improved Level set method and original Level set method,the result can be concluded that the improved method can raise the speed of segmentation and improve the adaptability to different images.When the improved Level set method is applied to the division of pulmonary nodules,the method also can make out the contours of targets accurately and quickly.The data are shown in table 4-1 and table5-1.The way of segmenting nodules by coarse segmentation and resegmentation has the ability of ruling out interference.