CRNN-BP based intelligent blood pressure value measurement method

Abstract

Disclosed is a CRNN-BP based intelligent blood pressure value measurement method. The method mainly includes the following steps: a) preprocessing real data; b) performing data standardization on preprocessed data and establishing a non-linear mathematical mapping relationship between pulse wave sequence signals and blood pressure values based on a CRNN-BP model; and c) performing intelligent blood pressure value measurement on real pulse wave characteristics by using the trained CRNN-BP model. In the step a), the preprocessing operation of the real data concretely includes baseline drift removal, cardiac cycle division and noise sample removal. The step a) includes the following steps: (1) using real human physiological data in a MIMIC database as an experimental sample, using a frequencyof 125 HZ, performing the baseline drift removal on a collected PPG signal by wavelet decomposition, and then removing a low frequency part of the signal to remove a baseline; (2) performing the cardiac cycle division on the data processed in the step (1) and collected ABP pulse wave signals, and combining the PPG and ABP signals belonging to the same cardiac cycle according to time relationships; and (3) removing the cardiac cycles with obvious abnormal waveform shape by using a rule-based method. The CRNN-BP based intelligent blood pressure value measurement method has the advantages of simper measurement and high detection precision, is more suitable for tracking and detecting cardiovascular diseases, and can also play a certain monitoring role for normal people without cardiovasculardiseases.

Description

technical field [0001] The invention relates to a CRNN-BP-based intelligent blood pressure measurement method, which belongs to the field of intelligent measurement of blood pressure (diastolic and systolic blood pressure). Background technique [0002] Blood pressure refers to the pressure on the vessel wall when blood flows through the vessel. It is an important physiological parameter that can reflect the functional state of the human cardiovascular system and is an important basis for clinical diagnosis of related diseases. Arterial blood pressure fluctuates periodically. In a cardiac cycle, the maximum value is called systolic blood pressure (SBP), and the minimum value is called diastolic blood pressure (DBP). [0003] So far, many literatures have proposed the method of continuous monitoring of blood pressure using the pulse wave waveform, which is of great significance for the prevention and diagnosis of hypertension, and it is helpful to understand the changes of bl...

AI Technical Summary

Problems solved by technology

However, there are problems in the above two methods. For the first method, because there are large differences in the shape of the pulse wave between individuals, and as the age increases, the pulse wave will also change, so some characteristic points on the pulse wave waveform It is difficult to locate accurately, so that the features cannot be extracted; in addition, because the pulse wave may be disturbed during the acquisition process, the waveform of some cardiac cycles will have noise. If such a waveform is used for blood pressure prediction, it is difficult for the two methods to achieve a relatively good precision

Images