Detection of Lung Cancer by Modified Irregular Tree Structure Bayesian Network Model Based Image Segmentation

Abstract

Identification of lung cancer at an initial stage is an important and crucial aspect of image processing. Different methods have been used to detect lung cancer at early stage. Medical image segmentation has got an important role in the medical research field. The Computed Tomography (CT) lung images of the patient can be classified into normal or abnormal depending upon the presence and absence of a tumor. Here a method has been presented which will diagnose lung cancer at an early stage using CT scan images. Automatic image segmentation can also be used for interactive segmentation. Initially, the watershed segmentation is used to identify the different regions based on the intensity variation. The relation between the different regions is found by Irregular Tree Structure Bayesian Network (ITSBN) model. Extracted features from different regions are fed into Radial Basis Function Neural Network (RBFNN), which identifies a tumor.