Lung cancer is the most deadly disease in the world and it should be identified early. Early detection helps to increase the survival rate of patients. But, the early recognition of lung cancer is a complex process because of the size and shape of nodules. In lung cancer, useful information is ensured by CT (computed tomography) scan. The major aim of this research work is classify the nodules as normal, benign and malignant. This work undergoes automatic CAD (computer aided design) for automatic diagnosis. Initially, the input CT image is pre-processed by a Gaussian filter to eliminate the noise that exist in the image. Then, the segmentation process is carried out by FTEC (fuzzy Tsallis entropy) clustering. Then, the texture features are extracted by GLCM (Gray-Level Co-Occurrence Matrix) and Gabor filter (GF). Finally, to classify the nodules as normal, benign and malignant the ML (machine learning) classifier Multi-Support Vector Machine (M-SVM) is used. Further, optimizing the weights of M-SVM is optimized by the optimization EBOA (enhanced butterfly optimization algorithm). Theexperimentation is carried out Python platform on the benchmark dataset Image Database Resource Initiative (LIDC-IDRI). The simulation results proved that the proposed M-SVM-EBOA achieved better accuracy and precision of 0.998 and 0.993 respectively.

Lung Cancer, Computed Tomography, Computer Aided Design, Machine Learning, Classification