A method and device for extracting respiration rate

Abstract

The invention discloses a respiration rate extraction method. The method comprises the following steps: obtaining original electrocardio signals and carrying out power frequency notch on the original electrocardio signals to obtain electrocardio signals to be processed; extracting the electrocardio signals to be processed via a pre-trained neural network model to obtain first respiration signals, and calculating according to the first respiration signals to obtain a first respiration rate at the moment, wherein an input sample space formed by each lead RR interval and an R peak amplitude value of collected multi-lead electrocardio signals is processed by using a principal component analysis method to obtain a principal component score matrix; and the principal component score matrix and target respiration signals are trained to obtain the neural network model. The invention also discloses a respiration rate extraction device which is capable of rapidly and effectively extracting respiration signals from the electrocardio signals so as to calculate and obtain an accurate and stable respiration rate.

Description

technical field [0001] The invention relates to the field of respiration detection, in particular to a method and device for extracting respiration rate. Background technique [0002] Respiration is an important physiological process of the human body, and the monitoring and detection of human respiration is also an important part of modern medical monitoring technology. Whether the disease of the patient's respiratory system itself or other important organs develops to a certain extent, the respiratory center will be affected. The failure of multiple organ systems often involves the failure of respiratory function, and the failure of respiratory function leads to the failure of other organ functions, which is a mutual cause and effect. [0003] The prior art mainly uses the following methods to detect respiratory motion: Impedance volume method: measure the change of chest impedance with a high-frequency constant current source to extract respiratory information; sensor me...

AI Technical Summary

Problems solved by technology

For ECG signals, multi-lead ECG signals can be used as input samples, but the standard multi-lead ECG signals are 12 leads. After calculating the RR interval and R peak amplitude of each lead separately, it is necessary to The number of eigenvalues ​​input to the neural network reaches 24, and the leads are correlated with each other. Therefore, the input sample has a large dimension and contains linearly related input items, which leads to poor training efficiency and training effect of the neural network model, which in turn affects Accuracy of signal extraction and accuracy of detected respiration rate

Images