Automatic Lung Segmentation in CT Images Using Anisotropic Diffusion and Morphology Operation

The preprocessing step of most computer-aided diagnosis (CAD) systems for identifying the lung diseases is lung segmentation. We present a novel lung segmentation technique based on anisotropic diffusion and morphological operation which is performed fast and accurately. The proposed method consists of three steps. At first step, gray image is produced by the input image. And then anisotropic diffusion is preformed to blur the gray image. The second step is that morphological operation is performed to remove the airway and mediastinum and get the right and left lung area. At the third step, the binary image of the right and left lung area obtained in the second step is generated and is matched to the original to segment the lung part. The proposed method eliminates the tasks of finding an optimal threshold and separating the attached left and right lungs. We have applied our new approach on several pulmonary CT images and the results reveal the speed, robustness and accuracy of this method.