Abstract
Lung segmentation is the initial step for detection and diagnosis for lung-related abnormalities and disease. In CAD system for lung cancer, this step traces the boundary for the pulmonary region from thorax in CT images. It decreases the overhead for a further step in CAD system by reducing the space for finding the ROIs. The major issue and challenging task for the segmentation is the inclusion of juxtapleural nodules in the segmented lungs. The chapter attempts 3D lung segmentation of CT images using active contour and morphological operations. The major steps in the proposed approach contain: preprocessing through various techniques, Otsu's thresholding for the binarizing the image; morphological operations are applied for elimination of undesired region and, finally, active contour for the segmentation of the lungs in 3D. For experiment, 10 subjects are taken from the public dataset of LIDC-IDRI. The proposed method achieved accuracies 0.979 Jaccard's similarity index value, 0.989 Dice similarity coefficient, and 0.073 volume overlap error when compared to ground truth.
TopLung segmentation has been a keen research area for various authors in current decades. The advancement of latest modality like CT, PET (Positron Emission Tomography), MRI (Magnetic Resonance Imaging), etc. facilitate the 3D segmentation for lungs for better visualization and accuracies in further diagnosis. Silva et al. (Nery et al, 2012) had given method based on 3D Gaussian filter to reduce some noise. In this technique, the initial lung lobes are extracted through thresholding followed by sequential erosion for separation of two lung regions. The inferior and superior border of lungs is obtained and morphological operation is applied for border correction subsequently subtracting with original generates the final region. A hybrid method using automatic 3D region growing and morphological operation and multiatlas segmentation had been proposed by Rikxoort et al. (Van et al, 2009). The procedure for lung segmentation given by Da Nobrega et al. (Da et al, 2017) follows: image addition, locating the trachea, lungs (with respiratory track) segmentation, trachea segmentation and volume subtraction. Yim et al. (Yim et al, 2005) proposed a procedure using inverse seeded region growing and connected component labeling followed by the extraction of respiratory track and lungs region. Then, trachea and large airways are delineated from the lungs using 3D region growing. At final step, the borders of lung region are obtained accurately by the subtraction of the second step result from the result of first step. 3D region growing segmentation had shown the comparable performance or better results with respect to the other methods of automatic segmentation.
Key Terms in this Chapter
Computer-Aided Diagnosis System (CADx): It is the system for the detection of nodules and further diagnosis of nodules into malignant and benign candidates.
Lung Segmentation: The process of extracting the lungs ROI or contour from thorax region.
Thresholding: It is method of finding the threshold (s) to segment the image into region (s).
Juxtapleural Nodules: The nodules present in wall region of lungs.
Computer-Aided Detection System (CADe): It is the system for the detection of various types of nodule candidates in lung region.
Active Contour: It is the method of image segmentation-based snake/curvature of the image.