Improving Ischemic Beat Classification Using Fuzzy-Genetic Based PCA And ICA

In this paper, an improved version of Principal Component Analysis (PCA) and Independent Component Analysis (ICA) is proposed for feature extraction to classify the ischemic beats from electrocardiogram (ECG) signal. The Fuzzy C-Means (FCM) and Genetic Algorithm (GA) is combined with PCA and ICA to extract more relevant features; the proposed methods are named as Fuzzy-Genetic based PCA (FGPCA) and Fuzzy-Genetic based ICA (FGICA). Least Square Support Vector Machine (LSSVM) is used to classify the beats into ischemic or non-ischemic, with the features from the FGPCA and FGICA. The ECG beats used in this paper are collected from European ST-T database. There is totally 2040 beats extracted from 17 different patients. The performance of our proposed method is compared with the linear PCA and ICA, shown that the proposed methods improve the sensitivity of ischemic classification. Keywords-Principal Component Analysis, Independent Component Analysis, Fuzzy Logic, Genetic Algorithm.