Effective Classification of Lung Cancer Using Hybrid Neuro Fuzzy Binary Cuckoo Search Algorithm

K. Vasanthi and N. Balakumar

Lung malignant growth is one among the significant reasons for malignancy related passings. Luckily, a beginning period conclusion can increment the endurance paces of the patients. Early location of the lung malignant growth is conceivable utilizing Computed Tomography (CT) Images. Right now, CT examines images process through lung images preprocessing, highlight extraction and grouping. In this proposed, the CT images are processed by extraction and classification of the characteristics. The pulmonary images are pretreated by an anisotropic diffusion mechanism, and then the characteristics are extracted using the histogram of the oriented gradient (HOG) technique, which is established in the selective characteristics Neuro Fuzzy Classifier with Binary Cuckoo Search (NFCBCS). In this classification process, we have improved the weight of the NFCBCS algorithm to classify the cancer growth whether it is a typical or unusual. This proposed shows that images can be identified using image processing methods and that this can have some significance in detection early lung cancer.

Volume 12 | 04-Special Issue

Pages: 73-84

DOI: 10.5373/JARDCS/V12SP4/20201468