期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2018
卷号：6
期号：1
页码：267
DOI：10.15680/IJIRCCE.2017.0601046
出版社：S&S Publications
摘要：In this paper a new classification algorithm is proposed for the Efficient Classification of Lung Tumor.In order to develop algorithm 150 CT scan images of patients have been considered consisting of Benign andMalignant Tumor Computed tomography (CT) Scan image. With a view to extract features from the CT scanimages after image processing, an algorithm proposes WHT Transform domain coefficients. The Efficient classifiersbased on Multilayer Perceptron (MLP) Neural Network. A separate Cross-Validation dataset is used for properevaluation of the proposed classification algorithm with respect to important performance measures, such as MSEand classification accuracy. The Average Classification Accuracy of MLP Neural Network comprising of one hiddenlayers with 37 PE’s organized in a typical topology is found to be superior (95.54 %) for Training . Finally, optimalalgorithm has been developed on the basis of the best classifier performance. The algorithm will provide aneffective alternative to traditional method of Lung Computed tomography (CT) scan image analysis for deciding thetumor in lung is Benign or Malignant.
