Efficient Lung CT Image Segmentation using Mathematical Morphology and the Region Growing algorithm

Abstract

Computer Aided Diagnosis (CAD) systems are often used during Today's medical practicality. It helps the physician to perform an accurate detection and diagnosis of Lung Pathologies. Lung CT image segmentation is a prerequisite in lung CT analysis. In this paper we proposed an automated hybrid method for lung segmentation based on both mathematical morphology and the region growing algorithm. the seed points are selected automatically without any user interaction. Also, the structuring element used in mathematical morphology operation is dynamic and it changes its shape and parameters according to the input 2D lung CT slices.