ppg monitoring system based on multi-layer time-delay neural network to remove motion artifacts

Abstract

The invention belongs to the technical field of medical equipment, in particular to a PPG monitoring system for removing motion artifacts based on a multi-layer time-delay neural network. It includes PPG probe, transmission control main board and PC upper computer; PPG probe, transmission control main board form a wearable structure; PPG probe includes PPG sensor and IMU sensor, PPG sensor and LDO are fixed on the front of PCB, IMU sensor and FPC line are set on the back Connector; the interrupt prompt lines of the PPG sensor and the IMU sensor are respectively connected to GPIO; the transmission control main board is connected to the PPG probe through the FPC line, and the transmission control main board is connected to the PC host computer; the transmission control main board includes the main control board and the wireless data transmission module; The main control board includes a multi-layer delay network motion artifact removal module. The system can realize real-time, online and accurate PPG motion artifact removal under severe exercise.

Description

technical field [0001] The invention belongs to the technical field of medical equipment, and in particular relates to a PPG monitoring system for removing motion artifacts based on a multi-layer time-delay neural network. Background technique [0002] In recent years, transmissive and reflective pulse oximetry devices based on photoplethysmography (PPG) have been greatly promoted and developed in the health and medical industries because of their portability, cheapness and accuracy, and are widely used in Vital signs monitoring, wearable devices and other fields. [0003] The detection of PPG signal is usually achieved by injecting light of one or more different wavelengths from the body surface into the subcutaneous vascular network, and recording the intensity of transmitted or reflected light. Oxygenated hemoglobin and unoxygenated hemoglobin in blood have differences in light absorption of different wavelengths, the steady-state and fluctuating parts of the PPG signal ...

AI Technical Summary

Problems solved by technology

However, no matter which method is used, it faces problems and challenges such as the strength of the reference signal correlation, the time delay between the reference signal and the noisy PPG signal, the trade-off between the number of filter taps and the sampling rate, etc., and the effect is limited in practical applications.

[0007] In addition, methods based on classical wavelet transform [12,13] and dual-tree complex wavelet transform [14,15] have also been proposed and proved to be suitable for certain motion situations, but the methods based on this are different for different Devices and various PPG waveforms collected for different experimental objects may need to change different threshold selection strategies, which is not conducive to the promotion among smart devices

[0008] In recent years, some researchers have also proposed based on time domain or frequency domain ICA [10,16-18], SSA [19,20], cycle-by-cycle Fourier series [21], empirical mode decomposition [22], Kalman filter [23], Wigner-Ville distribution [24] and other methods to remove artifacts, but these methods are either difficult to apply in practice [21], or are only proven to be effective under specific experimental conditions

In fact, there is still no universal and robust solution to this problem

Images