MEDICAL IMAGE SEGMENTATION AND DETECTION OF LUNG TUMOR ON CT SCAN IMAGES

Abstract

In order to improve the survival rate of lung cancer patient, early detection has a significantly more hopeful
prognosis and is the key treatment. CT scanning presents great opportunities for lung cancer diagnosis. In this paper
median filter is used for image preprocessing. For image segmentation, thresholding and marker controlled Watershed
segmentation approach is used to segment the lung of CT image. In feature extraction step, Gray-level co-occurrence
matrix is applied. Depending on the lung feature extraction, decision is made whether the lung has nodule or not.
Diagnosis is mostly based on CT (computed tomography) images. CT scanning presents great opportunities for lung
cancer diagnosis. The main objective of this paper is to implement lung nodule segmentation and feature extraction using
digital image processing for the classification of the disease stages to avoid serious stages early and to reduce lung
cancer percentage distribution. These images are more efficient and detailed than X-ray or other conventional methods.
MATLAB is one of the most widely used computer program for the examination and study of CT scanned images. This
prototype work proposes a convenient and low-cost procedure to detect the cancerous cells accurately from the captured
lung CT scanned images.