Li B.,South China University of Technology |
Zhang J.,South China University of Technology |
Tian L.,South China University of Technology |
Tan L.,South China University of Technology |
And 2 more authors.
International Journal of Computational Intelligence Systems | Year: 2012
Computer-aided detection(CAD) system for lung nodules plays the important role in the diagnosis of lung cancer. In this paper, an improved intelligent recognition method of lung nodule in HRCT combing rule-based and costsensitive support vector machine(C-SVM) classifiers is proposed for detecting both solid nodules and ground-glass opacity(GGO) nodules(part solid and nonsolid). This method consists of several steps. Firstly, segmentation of regions of interest(ROIs), including pulmonary parenchyma and lung nodule candidates, is a difficult task. On one side, the presence of noise lowers the visibility of low-contrast objects. On the other side, different types of nodules, including small nodules, nodules connecting to vasculature or other structures, part-solid or nonsolid nodules, are complex, noisy, weak edge or difficult to define the boundary. In order to overcome the difficulties of obvious boundary-leak and slow evolvement speed problem in segmentatioin of weak edge, an overall segmentation method is proposed, they are: the lung parenchyma is extracted based on threshold and morphologic segmentation method; the image denoising and enhancing is realized by nonlinear anisotropic diffusion filtering(NADF) method; candidate pulmonary nodules are segmented by the improved C-V level set method, in which the segmentation result of EM-based fuzzy threshold method is used as the initial contour of active contour model and a constrained energy term is added into the PDE of level set function. Then, lung nodules are classified by using the intelligent classifiers combining rules and C-SVM. Rule-based classification is first used to remove easily dismissible nonnodule objects, then C-SVM classification are used to further classify nodule candidates and reduce the number of false positive(FP) objects. In order to increase the efficiency of SVM, an improved training method is used to train SVM, which uses the grid search method to search the optimal parameters of C-SVM and uses second order information to achieve fast convergence to select the Sequential Minimal Optimization(SMO) working set. Experimental results of recognition for lung nodules show desirable performances of the proposed method.