Prediction therapy outcomes of HCV patients treated with interferon/ribavirin. (August 2018)

• Record Type: Journal Article
• Authors: Ebrahimi, Zeinab; Gharesi, Niloofar; Arefi, Mohammad Mehdi; Safavi, Ali Akbar; Zadeh, Mehrdad Hosseini
• Is Part Of: Biomedical signal processing and control. Volume 45(2018)
• Page Start: 128
• Page End: 135
• Publication Date: 2018-08
• Subjects: Hepatitis C virus (HCV) -- Discrete wavelet transform (DWT) -- Wavelet neural network (Wave-Net) -- Support vector machine (SVM) -- Linear discriminant analysis (LDA) -- Interferon and ribavirin therapy; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2018.05.030
• Languages: English
• ISSNs: 1746-8094

Abstract

Abstract: Hepatitis C is a kind of an infectious disease that mainly has an impact on the liver and also disrupts its activities. As an approximation, 130∼170 millions of people around the world have been suffering from hepatitis C virus. Until now, a combination of interferon-alpha (IFN-Alpha) and ribavirin (RBV) is employed as a therapy to those who infected with hepatitis C virus (HCV). This paper presents powerful and novel methods to predict and classify therapy outcomes based on two techniques and two classifiers. Here, discrete wavelet transform (DWT) is invoked for decomposing the initial datasets up several levels. The datasets that used in the procedure of prediction and classification are the full-length nucleotide sequences of HCV subtypes 1a and 1b. Next, the reduction of data dimension as well as correlation amongst the datasets are carried out by exerting linear discriminant analysis (LDA). After acquiring the most significant and vital features from the full-length nucleotide sequences of HCV subtypes 1a and 1b, two effective and powerful methods are presented for classifying and identifying genetic determinatives of treatment consequence. Thus, wavelet neural network (Wave-Net) and support vector machine (SVM) with various parameters and wavelets are used to classify and predict the therapy outcome. The experimental results indicate the efficiency and accuracy of the proposed techniques compared to other classification and prediction methods.