Sparse representation of ECG signals for automated recognition of cardiac arrhythmias
作者：
Highlights：
• A new method i.e. sparse decomposition over a gabor dictionary is proposed.
• Four different features are extracted and concatenated from the atoms of dictionary.
• An optimal least-square twin SVM classifier model is developed using ABC technique.
• The experiments are evaluated under category and personalized schemes.
• A higher accuracy of proposed method outperforms the results from the literature.
摘要
•A new method i.e. sparse decomposition over a gabor dictionary is proposed.•Four different features are extracted and concatenated from the atoms of dictionary.•An optimal least-square twin SVM classifier model is developed using ABC technique.•The experiments are evaluated under category and personalized schemes.•A higher accuracy of proposed method outperforms the results from the literature.
论文关键词：Electrocardiogram signal,Sparse representation,Overcomplete dictionary,Least-square twin support vector machines,Artificial bee colony,Particle swarm optimization
论文评审过程：Received 12 November 2017, Revised 20 March 2018, Accepted 21 March 2018, Available online 23 March 2018, Version of Record 4 April 2018.
论文官网地址：https://doi.org/10.1016/j.eswa.2018.03.038
