A Computer Based Feature Extraction of Lung Nodule in Chest X-Ray Image

Abstract

Feature extraction is one of the most important step in CAD (Computer Assisted Diagnosis) system. It helps CAD system to take correct decision and increase its accuracy by providing distinguish feature of malignant and benign tumor. Computer based system is proposed in this paper for feature extraction of lung nodule from the X-ray image. In recent years, the image processing mechanisms are widely used in several medical areas for early detection and in deciding treatment stages, where the time and cost factor is very important to discover the disease in the patient. Among the cancer, lung cancer is one of the most common causes of death worldwide. Therefore, early detection using diagnostic tests promises to reduce mortality from lung cancer. Present paper deals with the problem of developing a computer based system for the extraction of maximum features from the segmented suspicious area from the lung X-ray image. Further, these properties can be used to classify lung tumor as benign or malignant from the X-ray image directly. Calculated features will help the CAD system to take correct decision.

Early Lung Cancer Detection Using Artificial Neural Network

“…Of the 25 lung nodule samples, 20 samples were used for training and 5 samples were used to test the neural network. In each lung nodule, the perimeter value, the area and the size of the shape of the object were calculated using Matlab program [2,4]. The calculation results are presented in Table 1.…”

“…There are several studies about early lung cancer detection system. Lingayat and Tarambale [2], Patil and Udupi [3], also Ramaraju and Praveen [4] have identified and characterized lung tumor objects that measure the area, perimeter and shape or irregularity of tumor objects. Kumar and Saini have detected lung cancer using artificial neural networks, however the characterization of cancer object has not been identified [5].…”

Advances in Optimal Detection of Cancer by Image Processing; Experience with Lung and Breast Cancers

“…The developed system with this algorithm, acts as a second opinion to the physicians and radiologists. The results show malignant and benign lung tumor's distinguishes has been improved (Lingayat and Tarambale, 2013).…”

An Enhanced Approach for Detection and Classification of Computed Tomography Lung Cancer

“…In [9], many filters were applied in the preprocessing step, such as low pass filters, contrast stretching histogram equalization, negativity and power law transformation. For segmentation modified thresholding, labeling algorithm and edge detection were taken off.…”