Fast and cost efficient method of lung cancer detection using watershed segmentation.

The detection and treatment of lung cancer at an early stage is very much important when it bothers human mortality rate. Techniques such as Positron Emission Tomography (PET) and Computed Tomography (CT) are not capable enough to detect the cancer at an early stage. Symptoms usually show up when the disease is in advanced stage and the survival rate of the patients is 17.8%. This method is about diagnosing the disease at early and crucial stages with intelligent computational techniques with various distortion removals by segmentation techniques and algorithms, which are the root concepts of image processing. The CT images are obtained from Wayne State University, School of Engineering and Medicine. These images are analyzed using MATLAB. Also this method is tested on a public dataset containing 983 lung cancer images and this method is able to achieve 98% of the success rate in tumor detection.