Electrocardiosignal quality discrimination method based on neural network model

Abstract

The invention discloses an electrocardiosignal quality discrimination method based on a neural network model. The method comprises the steps that an independent and single-channel electrocardiosignal is converted into three feature values, namely, a QRS energy specific value, signal kurtosis and a base line energy specific value before learning through a technological means of solving an integral and a kurtosis coefficient through power spectral density, and then a discrimination model is accurately set up in an optimizing mode of gradient descent through an artificial neural network learning algorithm according to the feature values. The electrocardiosignal quality discrimination method is achieved by restoring a system model, and then whether the electrocardiosignal can be used for diagnosis or not is effectively discriminated.

Description

technical field [0001] The invention belongs to the technical field of medical devices, and in particular relates to a method for discriminating the quality of electrocardiographic signals based on a neural network model. Background technique [0002] ECG signals are often subject to severe noise and artifact interference, and filtering algorithms often cannot remove these interferences well, especially because interference signals and ECG signals often have similar frequency components and similar shapes. Therefore, the interference will reduce the quality of the ECG signal, and affect the automatic disease diagnosis based on the ECG, thus causing more false alarms (false positives). For example, if the ECG signal is too poor, it will cause a large number of false alarms in the ICU, and the false alarms in the ICU may even be as high as 86%. [0003] With the gradual increase in human life expectancy, healthy aging in modern society will become a global focus. The World H...

AI Technical Summary

Problems solved by technology

For a specific signal, it may contain many factors that affect the quality, how to classify it into a category that is more similar to it, the previous processing method is difficult to do this

Images