Lung cancer image segmentation using rough set theory

Background/Objectives: Lung cancer seems to be the common cause of death among people through the world. Early detection of lung cancer can increase the chance of survival among people. An attempt is made to segment the CT image of lung cancer using Rough K-Means clustering, which is one of the most important unsupervised learning methods in machine learning. Methods/Statistical analysis: The necessary CT images are collected from Mitra Scan centre, Salem for this study. The proposed method is compared with the bench mark K-means algorithm in order to achieve the efficiency. Findings: the performance of proposed Rough Set technique is compared with existing Clustering (k means) work which shows its efficiency level of segmented image portion and the prediction rate is better than its counterpart. Improvements/Applications: The proposed technique predicts the early symptoms of the disease with segmented region of image matched to the similar patterns of diseased portions of trained patient images. Keywords Computed Tomography (CT) image, Segmentation, Rough K-Means, Clustering.