Lung Nodule Segmentation in Computed Tomography Images for Preoperative Planning and Diagnosis Purposes

Abstract

Abstract-The precise segmentation of lung nodule is of great significance to identify tumors for the diagnosis of the lung cancer. Computed tomography (CT) is powerful evaluation technique for producing 2D cross sectional images of the organs under test by providing characteristics of the internal structure of an organ. Here, Segmentation concept is used to analyze the nodule in the CT lung image. In this proposed work, the lung CT image is segmented from which nodule is isolated and then classified as either benign or malignant. Segmentation is done using optimal threshold technique which is followed by Region of Interest (ROI) extraction. Then texture features of this ROI have been extracted using Gray Level Co-occurrence Matrix (GLCM) and classification is performed on these features. Experiments have been conducted to classify nodule as benign or malignant. and obtained above 95% of accuracy. CT images required for the work are collected from the diagnostic center.-CT images; Segmentation; lung nodules (key words)