Segmentation of lung images for tumor detection using color based kmeans clustering algorithm

Abstract

Lung tumor is a complex disease that occurs due to the abnormal growth of lung cells. For efficient treatment planning,
earlier detection of tumor is necessary. Magnetic Resonance Imaging (MRI) is now recognized as an important tool for the
detection of Lung tumor. The proposed Lung Tumor detection algorithm is composed of three stages: Image acquisition,
Preprocessing and Segmentation. Major steps in preprocessing are noise filtering, contrast enhancement. A novel algorithm
for lungimage segmentation using modified color based K-means clustering is proposed in this work. Modified Histogram
Equalization is used to remove noise and high frequency components from database without affecting its edges and
bandwidth. It is used to enhance lungimages. Modified color based K-means clustering based segmentation process is
implemented to detect Lung Tumor. Here, the performance of segmentation algorithm in terms of accuracy, specificity and
sensitivity are computed using manually segmented images. In this study Region Growing was implemented to authenticate
the performance of proposed method. The parameters used to evaluate the performance of segmentation algorithm are
Mean Square Error (MSE) and Peak Signal to Noise Ratio (PSNR), accuracy. Finally all these image processing steps are
integrated and get the results