Prediction of lung tumors from lung CT images using combination of an enhanced image processing and data mining techniques

Abstract

The research work titled and#8213;Prediction of Lung Tumors from Lung CT newlineImages using Combination of an Enhanced Image Processing and Data newlineMining Techniquesand#8214; is yet another effort to bring out more proficient and an newlineefficient research work to predict and detect lung tumors from lung CT newlineimages using computer-based lung tumor detection system. Lung cancer is a newlinemajor cause of mortality throughout the world. Detection and prediction of newlinelung cancer stages in its prodromal stages have to be facilitated with newlineintelligent methodologies and techniques. With the advancement in medical newlineindustry, automated techniques simplify the process of detection and accurate newlineclassification of lung cancers. The chances of survival are immensely newlineenhanced owing to earlier detection and diagnosis. Classification of lung newlinemalignancy stages will assist in better treatment and survival rate. Modalities newlineinvolved in imaging lungs serve as a major contributor in detection. newlineThe proposed model is a Computer Aided Detection (CADe) system newlinewhich automates the detection and classification process, with ease and newlineimproved accuracy. This system is developed using MATLAB code. The newlinealgorithms used in this research are selected based on extensive surveys and newlinereviews conducted by many researchers on this field. The combination of newlinehigh-performance segmentation algorithm, content-based image retrieval and newlineinclusion of data mining technique yields better and desired results. This newlineresearch work uses an amalgamation of two major areas such as Medical newlineImage Processing and Data Mining. In medical image processing, effective newlineimage segmentation algorithm contributes a major proportion of the proposed newlineimplementation. Novel approach in the thesis analyses irregular boundaries newlineand delivered accurate results with respect to detection of nodules in CT newlineimages. Another observation in Computer Aided Detection (CADe) systems newlineis the absence of detection techniques for classification grounds newline