Lung Cancer Research Paper

EARLY DETECTION OF LUNG CANCER IN X-RAY IMAGES USING CELLULAR NEURAL NETWORK AND IMAGE INTERPOLATION

1 K.Lalitha, 2 K.Meenakshi,

1Assistant Professor, 2Assistant Professor, Department of ECE.

1, 2Panimalar Engineering College,Chennai – 600 123, India.

Email: 1 [e-mail] , 2 [e-mail].

Abstract— The interpolation method is to increase the resolution of the lung cancer image based on the bicubic interpolation and the 2D interpolation filtering. Lung cancer is one of the leading causes of death and one of the most lethal cancers. Therefore one of the solutions to this problem is the early diagnosis. The algorithm is mainly developed for the use of general practitioners in helping them to do accurate early diagnosis of

Today Smoking causes nearly nine out of 10 lung cancer deaths while radon gas, pollution and other chemical exposures plays a smaller role. Cancer occurs when normal cells undergo a transformation that causes them to grow and multiply without control. The cells form a mass or tumor that differs from the surrounding tissues from which it arises. Tumors are dangerous because they take oxygen, nutrients, and space from healthy cells and because they invade and destroy or reduce the ability of normal tissues to function. Doctors aren't sure whether people with no signs or symptoms of lung cancer should undergo screening for the disease. Some studies show that lung cancer screening may save lives by finding cancer earlier, when it may be treated more successfully. But other studies find that lung cancer screening often reveals more benign conditions that require invasive testing and expose people to unnecessary risks. Screening for lung cancer is controversial among doctors. Studies are ongoing to determine what types of tests may be helpful and who would benefit from lung cancer screening. X-rays are a form of radiant energy that offer a wide variety of medical uses. In addition to their applications as both curative and palliative lung cancer treatments, X-rays are also used in several imaging tests that can lead to an early diagnosis of the disease. The most common X-ray imaging tests can help doctors

SMOOTHING TEMPLATE Average threshold or average is the old names for smoothing template. This template will smooth with binary output by the corresponding CNN template given by

(7)

H. EDGE DETECTION TEMPLATE The edge detection template is used to detect the lung cancer region boundary on the image. The corresponding CNN template is given by

(8)

The cellular neural network templates are used to detect the boundary of the lung cancer region.

2. INTERPOLATION METHOD

The non linear interpolation techniques are better than the linear interpolation technique. In the method bicubic interpolation is used to increase the resolution of the low resolution X-ray images which is a linear interpolation technique. Bicubic interpolation is known for its interpolating pixels by collecting the information from its sixteen neighbours but it performs poorly on