A Deep Regression Heart Rate Estimation Method for Ballistocardiogram Signals

Abstract

The present invention proposes a deep regression heart rate estimation method for a shock cardiogram signal, which is used to solve the technical problem of large estimation error existing in the prior art. The implementation steps are: collecting a shock cardiogram signal and a heart pulse signal; Filter the shockogram signal; use the periodic prior knowledge of the shockcardiogram signal to obtain the training sample set and test sample set; build a heart rate regression estimation network model based on the periodicity and amplitude characteristics of the shockogram signal; the heart rate regression estimation network The model is trained; heart rate estimates are obtained from the cardiocardiogram signal. The present invention obtains the periodic characteristics and amplitude characteristics of the heartbeat signal through a supervised learning method, and uses the bidirectional cyclic neural network to obtain the periodic characteristics and amplitude characteristics of the heartbeat signal, and simultaneously uses the periodic characteristics and amplitude characteristics of the shock cardiogram signal to estimate the heart rate value through a regression network, thereby simplifying the heart rate. The step of estimating the heart rate of the shock cardiogram signal effectively reduces the estimation error of the heart rate of the shock cardiogram signal.

Description

technical field [0001] The invention belongs to the technical field of biomedical information processing, and relates to a method for estimating a heart rate of a ballistocardiogram signal, in particular to a method for estimating a depth regression heart rate of a ballocardiogram signal. Background technique [0002] In recent years, with the improvement of technology and economic level, people pay more and more attention to their own health problems. Changes in heart rhythm beyond the normal range usually indicate the occurrence of a certain disease, such as sudden cardiac death, asphyxia, arrhythmia, etc. Therefore, heart rate monitoring in daily life is of great significance for the early detection and treatment of people's own diseases. [0003] Currently, electrocardiogram (ECG) is widely used in heart rate monitoring clinically, but this requires close contact of electrodes or heart probes with the human body, which brings great inconvenience and psychological pressu...

AI Technical Summary

Problems solved by technology

[0006] The disadvantage of the above two methods is that the estimation error is relatively large. The reasons are as follows: First, although the periodicity of the ballistocardiogram signal is used to overcome the interference of the non-periodic noise signal, it cannot solve the periodicity in the ballistocardiogram signal. The impact of sexual noise on heart rate estimation, and the above two methods do not use more additional information to guide the estimation of heart rate, so the estimation error is large

Second, the method of calculating the spectral components first and then estimating the heart rate is cumbersome, and the rounding operation when obtaining the frequency of the heartbeat signal will increase the estimation error

Images