Dual-convolutional neural network for electrocardiogram type recognition
作者： Wang Chaoyang , Yan Tianhong , Zhu Junjiang , Yao Weijing
DOI:
关键词: Fragment (computer graphics) 、 Artificial intelligence 、 Matrix (mathematics) 、 Value (computer science) 、 Type (model theory) 、 Dual (category theory) 、 Signal 、 Artificial neural network 、 Pattern recognition 、 Convolutional neural network 、 Computer science
摘要: The application of the invention relates to dual-convolutional neural network for electrocardiogram type recognition. comprises a first CNN model and second model, which number leads is same as that multi-lead electrocardiosignals, input value signal fragment obtained by cutting off electrocardiosignals each lead using window length (a+b) seconds pace 0.01-0.05 seconds, output result successive [X1,X2]; matrix formed through values respectively inputting fragments different electrocardiosignal into trained [Y1,Y2]. According disclosed invention, are considered, mutual relationship an can be final has advantage being high in accuracy.
lens.org 参考文章(5)
Jia Li, Jérémy Rapin, Mathurin Massias, An automatic method to delineate or categorize an electrocardiogram ,(2016)
Gao Yan, Yu Zhilou, Automatic diagnosis method for electrocardiographic abnormality ,(2017)
Hu Chuanyan, Liu Tao, Zhang Xue, Tian Liang, Cao Jun, Liu Chang, AUTOMATIC RECOGNITION AND CLASSIFICATION METHOD FOR ELECTROCARDIOGRAM HEARTBEAT BASED ON ARTIFICIAL INTELLIGENCE ,(2020)
Zhou Shusen, Jia Shixiang, Liu Chanjuan, Zou Hailin, Wei Yimin, Liu Tong, Zang Mujun, Automatic arrhythmia analysis method based on lead deep neural network ,(2018)
Guo Wei, Li Bin, Chen Bin, Wang Hui, Method and system for classifying multi-lead ECG signals ,(2018)