MACHINE LEARNING TECHNIQUES AND HYBRID FEATURE EXTRACTION BASED CT LUNG CANCER CLASSIFICATION

Abstract

Finding malignant nodules in the lungs using as medical system of computed tomography (CT) is a difficult and time-consuming task for a radiologist. A computer-aided diagnostic (CAD) system is proposed to alleviate this burden. In recent years, the thorough training approach has proven effective. Better results than classic methods in various areas. Currently, researchers are using various machine learning methods to improve the efficiency of the CAD system in detecting lung cancer using computed tomography. In this work, To improve the CAD based CT lung cancer detection, Hybrid feature extraction techniques is introduced. The hybrid feature extraction is based on three different features such as Local Binary Pattern (LBP), Gray-Level Co-Occurrence Matrix (GLCM) and histogram of oriented gradients (HOG). In order to test the classification accuracy by using three different kind of machine learning methods are used such as Random forest, Decision tree and Artificial neural network. The proposed system has evaluated with the help of online available LIDC-IDRI Dataset CT images. The proposed system performances are validated in the terms of sensitivity (SE), Recall (R), Precision (P), specificity (SP), and accuracy (AC).