期刊名称：International Journal of Computer Science and Information Technologies
电子版ISSN：0975-9646
出版年度：2017
卷号：8
期号：2
页码：159-162
出版社：TechScience Publications
摘要：Lung cancer is one of the main causes of death inhumans worldwide. The death rate due to lung cancer ishighest among all other types of cancer. Early detection andcorrect diagnosis of the growing cancer cells can increase thesurvival rate. The present techniques like X-ray, MRI, CTscan and PET images are used for the diagnosis of the disease.For early detection and treatment stages, image processingtechniques are widely used. With the help of expert physicians,images are examined and the stage of cancer is detected. Timefactor plays very important role in the diagnosis of theabnormal cells as it is directly related to the survival rate. Inthis research work, we are using significant pattern tool forprediction of lung cancer. The proposed system will useHistogram Equalization for pre-processing followed bysegmentation principles under adaptive segmentationalgorithm and feature extraction processes. An enhancedregion of the object of interest is obtained. On the basis offeatures obtained a normality comparison is made to check thestate of the patient. If the detection of lung cancer is predictedin its early stages it will reduce the number of painfultreatments and also reduce the surgery risk which willincrease the survival rate. The effectiveness of the proposedsystem is validated through MATLAB environment whichclearly verifies the pre-processing techniques of early detectionand prediction of lung cancer.
关键词：Image Processing; Early Detection; Prediction;Lung Cancer; Segmentation; Feature Extraction.
