GACM Based Segmentation Method for Lung Nodule Detection and Classification of Stages Using CT Images

This paper process the GACM (Gradient based Active Contour Model) for segmenting the lung region from digital CT (Computed Tomography) images. In GACM, the gradient information present in the image at lung boundaries are extracted using active contour model. Then shape and GLCM feature sets are extracted from normalized images. 

The PCA is used for optimal feature set evaluation based on the consistency of the feature attributes. The lung nodules detection and stage classifications are carried out using multi class Support Vector Machine which makes the decision boundary between the various classes using hyper planes. 

The experiments result shows that the stage classification accuracy and sensitivity achieved by the proposed method is 95.3% and 92.17% respectively. The results of proposed system are improved when compared with existing system in terms of accuracy and sensitivity.