A kernel-cnn based ECG signal recognition and classification method

Abstract

The invention discloses a method for recognizing and classifying electrocardiographic signals based on Kernel-CNN, which introduces kernel transformation into the convolution process to form a kernel transformation convolution operation, further enhances the ability of model feature extraction; and passes it through the Massachusetts Institute of Technology The data provided in the provided MIT-BIH database is verified, and the results show that the model of the present invention has a lower LOSS value than the convolutional neural network on the same prediction accuracy. The convolutional neural network of the present invention has excellent feature extraction ability, and the nonlinear mapping of data is realized by introducing the kernel transformation into the convolution operation, which further enhances the feature extraction ability of the convolution process. After the ECG signal is input into the trained network, the probability values ​​of five categories can be obtained, and the one with the highest probability value is selected as the type of the data. It does not require signal prior knowledge or expert input, and can extract effective features from ECG signals, which can be applied to the identification and classification of ECG by medical devices.

Description

technical field [0001] The invention belongs to the technical field of medical devices, relates to intelligent recognition of electrocardiographic signals in medical devices, and in particular to a method for recognizing and classifying electrocardiographic signals based on Kernel-CNN. Background technique [0002] According to the latest report of the World Health Organization (WHO) in 2019, cardiovascular disease (CVD) is one of the main diseases that cause human death. The high mortality rate of cardiovascular disease makes it continue to affect our normal life. The prevention, diagnosis and treatment of cardiovascular diseases have become important issues that society needs to solve. [0003] The automatic classification technology of electrocardiosignal (ECG) can be summarized as: signal acquisition, preprocessing, feature extraction and classification. Throughout the process, feature extraction plays a vital role and can directly affect the final classification result...

AI Technical Summary

Problems solved by technology

These methods require certain prior knowledge of the signal and often require expert input, which limits the application of the method

Images