Multi-position dynamic spectrum difference extraction method

A dynamic spectrum, multi-position technology, applied in the use of spectrum diagnosis, medical science, sensors, etc., to achieve the effect of suppressing random errors

Inactive Publication Date: 2019-03-26
TIANJIN UNIV
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

So far, there is no extraction method that can effectively suppress the nonlinear effects of blood scattering

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Multi-position dynamic spectrum difference extraction method
  • Multi-position dynamic spectrum difference extraction method
  • Multi-position dynamic spectrum difference extraction method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0036] A multi-position dynamic spectrum extraction method, see Figure 1-Figure 6 , the extraction method includes the following steps:

[0037] 101: Divide a period of the PPG signal into four areas, which are two areas in the upper half and two areas in the lower half;

[0038] Among them, see image 3 , a periodic PPG signal, the PPG signals at all wavelengths in a sample are superimposed as the PPG template signal, and the time difference between two adjacent peaks of the template PPG signal is taken as a cycle.

[0039] see Figure 4 , with the position of the mean value and the minimum value of a pulse wave cycle as the boundary, a cycle of PPG signal is divided into four areas, namely area I, area II, area III, and area IV.

[0040] 102: Perform dynamic spectral extraction independently on the two regions of the upper half and the two regions of the lower half, and perform dynamic spectral extraction on the overall spectrum, and finally superimpose and average these...

Embodiment 2

[0047] Combine below Figure 2-Figure 6 , and the specific calculation formula further introduces the scheme in embodiment 1, see the following description for details:

[0048] 201: Establish a PPG template;

[0049] That is, the logarithmic PPG signals at all wavelengths in a sample are superimposed as the PPG template signal of the sample. By superimposing the PPG signals, part of the high-frequency noise is suppressed, which is conducive to the establishment of the subsequent period and the search for peak and valley values. , PPG template such as image 3 shown.

[0050] 202: Find the peak-valley value of the PPG template, and use the interval between two adjacent peaks of the PPG template as a pulse period;

[0051] 203: Divide the PPG signal of one pulse cycle into 4 regions;

[0052] That is, take the average value in one period as the midline, divide the PPG signal of this period into the upper half and the lower half, and then use the respective average values ​​...

Embodiment 3

[0071] Below in conjunction with concrete experiment, the scheme in embodiment 1 and 2 is carried out feasibility verification, see the following description for details:

[0072] pass Figure 7 The experimental results show that the multi-position spectrum provided by the embodiment of the present invention can extract more scattering information, which meets various needs in practical applications.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a multi-position dynamic spectrum extraction method. The method comprises the following steps: dividing a PPG signal of one cycle into four regions, comprising two regions in the upper half part and two regions in the lower half part, by using the positions of the mean value and the minimum value of a pulse wave cycle as a boundary; carrying out data point number division on the two regions in the upper half part and the two regions in the lower half part according to a principle that the data point number of the dynamic spectrum of each region is divisible by 2 or not;independently performing dynamic spectrum extraction on all the regions in a sliding difference manner after the division; carrying out dynamic spectrum extraction on the whole spectrum, and overlaying and averaging above spectra to obtain an integral body; and carrying out sliding difference on collected logarithmic PPG signals at every wavelength, and forming a dynamic sub-spectrum by using obtained differences. The method can be used to extract situations representing different pieces of nonlinear information, and spectrum information carried by the nonlinear information is used to suppress nonlinear influences caused by blood scattering in order to improve the precision of dynamic spectrum noninvasive analysis of the blood component.

Description

technical field [0001] The invention relates to the field of dynamic spectrum extraction, in particular to a multi-position dynamic spectrum difference extraction method. Background technique [0002] Dynamic spectroscopy is a non-invasive measurement method for detecting blood components. By extracting the absorbance of pulsating blood at multiple wavelengths, it can greatly reduce the influence of individual differences and measurement environments on non-invasive detection in the human body. In the previous dynamic spectrum extraction, dynamic spectrum is constantly developing towards higher signal-to-noise ratio and higher data utilization. [0003] In the peak-to-peak method of dynamic spectroscopy [ 1 ], only use 2 points of the peak-valley value of each PPG (photoplethysmography) pulsation cycle to extract the dynamic spectrum, because the data utilization rate is low, this method cannot obtain higher signal-to-noise ratio; In the Chinese invention patent application...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): A61B5/0295
CPCA61B5/0075A61B5/0295
Inventor 李刚王玉宇汤伟林凌
Owner TIANJIN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products