Neuro Fuzzy Inference approaches for early detection of lung cancer

Abstract

Prediction and diagnosis of cancer is very important nowadays and has become a newlinecrucial task. Lung cancer is one of the leading causes of death globally. Early detection of lung cancer seems to be the only factor which can increase the survival rate of the patients which is a challenging task due to structure of cancer cell, where most of the cells are newlineoverlapped each other. Image processing techniques are widely used for prediction and early detection of lung cancer. A typical Computer Aided Diagnosis (CAD) system for lung cancer diagnosis is composed of six processing steps: preprocessing of lung CT image, newlinesegmentation of the lung fields, detection of nodules inside the lung fields, segmentation of the detected nodules, feature extraction and selection and classification of the nodules as newlinebenign or malignant. As per the available literature, there exist several challenges and aspects that CAD systems have been facing for lung cancer. Some of them are: elimination of noises in lung CT images, segment the challenging types of nodules and providing accurate segmentation of lung fields for effective reduction in the search space for lung nodules and selection of appropriate classification method. Although considerable amount of newlineresearch work has been done by various researchers for early detection of lung cancer, still there is a scope for improving the detection accuracy for better decision making through newlinemachine learning techniques. Neural Networks and Fuzzy Logic have become one of the most successful technologies in machine learning. This research work proposes Neuro-Fuzzy inference approaches for early detection of lung cancer from raw chest CT scan images with newlineimproved accuracy so as to help the radiologists in effective decision making. Enhancing the image quality, improving the accuracy in predicting the presence of lung nodules,feature selection and classifying the nodules into one of the four stages of lung cancer are the core factors of this research work,this thesis has made five significant contributions.