Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Gas marker detection method based on radial basis function neural network and application

A neural network and detection method technology, applied in the field of model systems, can solve problems such as reduced reliability and versatility, complex types of human exhaled gas, and severe cross-sensitivity of sensor responses

Pending Publication Date: 2020-03-13
CHINA UNIV OF PETROLEUM (EAST CHINA)
View PDF5 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] However, the mathematical relationship between the direct response value collected by existing sensors and the concentration of exhaled gas to be collected is often not intuitive enough, the data volume is large and the relationship is complicated
At the same time, the types of exhaled gas from the human body are complex, difficult to preserve and easily polluted, and the cross-sensitivity between sensor responses is serious, resulting in poor fitting effect and reduced reliability and versatility

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
  • Gas marker detection method based on radial basis function neural network and application
  • Gas marker detection method based on radial basis function neural network and application
  • Gas marker detection method based on radial basis function neural network and application

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0123] The detection method of exhaled gas markers based on radial basis function neural network, first, use the gas sensor to detect and calibrate the human exhaled gas markers, build a multi-dimensional sensor array to test the exhaled gas when simulating the sick state, and obtain several samples data; then, use the principal component analysis-particle swarm optimization-radial basis function (referred to as PCA-PSO-RBF) neural network algorithm model to preprocess the sample data, reduce the variable dimension, and reduce the cross-sensitivity of the gas sensor; finally, Prediction of human exhaled gas concentration.

[0124] In this embodiment, commercially available gas sensors for ammonia, acetone, and hydrogen sulfide (Digi-Key Electronics Co., Ltd.) are used as multi-dimensional sensor array construction elements for raw sample data collection.

[0125]In the PCA-PSO-RBF algorithm model, the principal component analysis (PCA) model is used to conduct principal compon...

Embodiment 2

[0185] In this example, if Figure 14 , Figure 15As shown, the human disease diagnosis model system uses the cluster analysis module to analyze the preprocessed markers; specifically, the algorithm of the K-means cluster processing framework in the cluster analysis module is as follows:

[0186] (1) Suppose the sample data set X is X={x 1 ,x 2 ,...,x N}, the number of clusters is k; if I=1, the initial clustering center is {Z j :j=1,2,3,...,k};

[0187] (2) Calculate the distance from each data point in the sample data to the cluster center, D(X i ,Z j (I)), where i=1,2,...,N; j=1,2,...,k; when D(X i ,Z j (I))=min{D(X i ,Z j (I)):j=1,2,...,k}, then X i is classified into the t category, denoted as

[0188] (3) Calculate the new cluster center in the sample data:

[0189]

[0190] (4) If Z j (I+1)≠Z j (I), j=1, 2,...,k, then I=I+1, return to step (2) and restart the calculation, otherwise the algorithm ends.

[0191] Effect of disease diagnosis based on cl...

Embodiment 3

[0197] In this embodiment, the human disease diagnosis model system uses a Deep Belief Neural Network (DBN) module to analyze the preprocessed sample data; specifically, the restricted Boltzmann Machine (restricted Boltzmann Machine) in the module , RBM for short) energy function is:

[0198]

[0199] The formula for converting to energy is:

[0200]

[0201] The probability distribution is expressed as:

[0202]

[0203] Among them, n and m respectively represent the number of neurons in the visible layer and hidden layer of the DBN neural network;

[0204] v and h respectively represent the state vectors of the visible layer and hidden layer of the DBN neural network;

[0205] a and b represent the state vectors of the visible layer and the hidden layer of the DBN neural network, respectively;

[0206] w represents the weight matrix connecting the visible layer and the hidden layer of the DBN neural network, θ={w, a, b};

[0207] is the partition function; P(v...

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 provides an exhaled gas marker detection method based on a radial basis function neural network and application, and relates to the field of gas-sensitive sensing detection. The method comprises the following steps: to begin with, detecting and calibrating a human body exhaled gas marker by utilizing a gas-sensitive sensor; constructing a multi-dimensional sensor array to test exhaled gas in a simulated illness state and obtaining a large amount of sample data; then, carrying out pretreatment on the sample data by using principal component analysis-particle swarm optimization-radial basis function neural network algorithm model to reduce the variable dimension and reduce the cross sensitivity of a gas-sensitive sensor; finally, carrying out exact prediction on the concentration of the human exhaled gas, so that the reliability is good, and the universality is high; and finally, establishing a relationship between the obtained sample data and corresponding disease states,constructing a human body disease diagnosis database model system, and providing guidance for prediction of related diseases of the exhaled gas.

Description

technical field [0001] The invention relates to the technical field of gas sensor detection, in particular to a method for detecting markers in exhaled gas based on a radial basis function neural network and a model system for its application in disease diagnosis. Background technique [0002] With the development of society and the improvement of living standards, people's livelihood and health have become the focus of attention. Various gases are produced during the metabolic process of the human body, such as volatile organic gases, nitric oxide, hydrogen sulfide and other gases. These gases will enter the respiratory system through the blood circulation in the body, and then be excreted from the body. If the exhaled gas of one or several markers is sampled, and the concentration of the sample is found to exceed a certain range, it means that the metabolic mechanism of the human body has changed, and there is a possibility of suffering from a disease. From the relations...

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
IPC IPC(8): G16H50/20G01N33/48G06N3/00G06K9/62G06N3/04G06N3/08
CPCG16H50/20G01N33/48G06N3/006G06N3/08G06N3/044G06F18/23G06F18/2135G06F18/214
Inventor 张冬至薛庆忠吴振岭王兴伟张勇
Owner CHINA UNIV OF PETROLEUM (EAST CHINA)
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com