Oral cavity gas detection method and device for children

Abstract

The invention discloses an oral cavity gas detection method and device for children. The method comprises the following steps: acquiring oral cavity gas detection information sets of children with different types of health state information; the oral cavity gas detection information set comprises a plurality of pieces of oral cavity gas detection information corresponding to each type of health state information; the oral cavity gas detection information comprises a hydrogen sulfide content sequence value, a methyl mercaptan content sequence value and a dimethyl sulfur content sequence value; performing correlation analysis processing on the health state information and the oral cavity gas detection information set to obtain a gas correlation model; component analysis is conducted on the collected oral cavity gas of the children, and gas component detection result information is obtained; and performing identification processing on the gas component detection result information by using the gas correlation model to obtain child health state information. According to the invention, through full detection and feature extraction of the oral cavity gas information of the child, accurate detection of the physical state of the child is realized.

Description

Technical Field

[0001] The present invention relates to the field of oral cavity detection and medical big data processing, and in particular to an oral cavity gas detection method and device for children. Background Art

[0002] Bad breath is a common clinical phenomenon, with an incidence rate of about 30% in children. The main possible causes are problems with the oral cavity itself, such as caries, dental plaque, food residues, etc. In addition, tonsil stones in the ENT department, foreign bodies in the nasal cavity (erasers, pen caps, small toys, etc. accidentally enter the children's nasal cavity and are not removed in time, causing infection over time) sinusitis, etc., may also cause bad breath. Gastrointestinal reasons also account for a small part, indigestion, or food with odor itself, which causes bad breath in the mouth through burping and other forms. From the perspective of the composition of the odor, it is mainly some sulfides. Specifically, the sulfide components contained in the odor in different locations such as the mouth, nasal cavity, throat, and stomach are different. In general, children's oral gas contains a wealth of information about their health status and physical status. How to effectively detect and extract features of children's oral gas information to detect children's physical status is an important problem that needs to be solved. Summary of the invention

[0003] The present invention mainly solves the problem of how to effectively detect and extract features of children's oral gas information to detect the physical state of children. The present invention discloses a method and device for detecting oral gas for children.

[0004] In a first aspect of the embodiment of the present application, a method for detecting oral gas for children is disclosed, comprising:

[0005] S1, obtaining a set of oral gas detection information of children who have obtained different types of health status information; the set of oral gas detection information includes a number of oral gas detection information corresponding to each type of health status information; the oral gas detection information includes a sequence value of hydrogen sulfide content, a sequence value of methyl mercaptan content, and a sequence value of dimethyl sulfide content;

[0006] S2, performing correlation analysis on the health status information and the oral gas detection information set to obtain a gas correlation model;

[0007] S3, analyzing the components of the collected oral gas of the child to obtain gas component detection result information;

[0008] S4, using the gas association model to identify and process the gas composition detection result information to obtain the child's health status information.

[0009] The correlating analysis and processing of the health status information and the oral gas detection information set to obtain a gas correlation model includes:

[0010] S21, preprocessing the health status information and the oral gas detection information set respectively to obtain preprocessed health status information and preprocessed oral gas detection information set;

[0011] S22, representing all types of health status information as a state vector; representing a set of pre-processed oral gas detection information corresponding to each type of health status information as a corresponding gas detection matrix; the elements of the state vector are the encoding values ​​of each type of health status information;

[0012] S23, performing data reduction processing on the gas detection matrices corresponding to all types of health status information to obtain a data reduction matrix;

[0013] S24, using a data reduction matrix, transforming a gas detection matrix corresponding to each type of health status information to obtain a corresponding reduced-dimensionality gas detection matrix; the reduced-dimensionality gas detection matrix includes a reduced-dimensionality gas detection vector;

[0014] S25, performing state estimation processing on the state vector and the reduced-dimensional gas detection matrix corresponding to all types of health status information to obtain a state estimation vector;

[0015] S26, constructing a gas correlation model using the reduced-dimensional gas detection matrix and the state estimation vector.

[0016] The preprocessing of the health status information and the oral gas detection information set to obtain preprocessed health status information and preprocessed oral gas detection information set includes:

[0017] S211, normalizing the health status information and the oral gas detection information set respectively to obtain pre-processed health status information and a normalized oral gas detection information set;

[0018] S212, performing boundary check processing on the pre-processed health status information and the normalized oral gas detection information set respectively to obtain a pre-processed oral gas detection information set.

[0019] The pre-processed health status information and the normalized oral gas detection information set are subjected to boundary check processing respectively to obtain the pre-processed oral gas detection information set, including:

[0020] For the normalized oral gas detection information set corresponding to all types of health status information, the data value of each type of oral gas detection information is taken as the dependent variable, and the corresponding health status information of the data is taken as the independent variable. The dependent variable and the independent variable are subjected to autoregression-sliding average modeling to obtain the regression model of the data of the type;

[0021] Using the regression model, the independent variable is calculated to obtain regression data values;

[0022] Determine whether the absolute value of the difference between the regression data value and the corresponding dependent variable value is greater than a set first regression discrimination threshold; if it is greater than the first regression discrimination threshold, delete the data from the normalized oral gas detection information set; if it is less than or equal to the first regression discrimination threshold, do not process the data.

[0023] The state estimation process is performed on the state vector and the reduced-dimensional gas detection matrix corresponding to all types of health status information to obtain the state estimation vector, including:

[0024] An estimation error model is constructed using the reduced-dimensional gas detection matrix corresponding to the state vector and all types of health status information;

[0025] The estimation error model is solved to obtain a state estimation vector.

[0026] The estimation error model comprises:

[0027]

[0028] v T v = 1,

[0029] Among them, v is the state estimation vector, v is a column vector with dimension 2, and a k is the kth element of the state vector, and is also the encoding value of the kth type of health status information, r i k is the i-th row vector of the reduced-dimensionality gas detection matrix corresponding to the k-th type of health status information, P is the number of types of health status information, and N is the row dimension of the reduced-dimensionality gas detection matrix.

[0030] The gas correlation model comprises:

[0031] The data reduction matrix is ​​used to multiply the preprocessed oral gas detection information to obtain a dimension-reduced gas detection vector;

[0032] The state estimation vector is used to multiply the dimension-reduced gas detection vector to obtain the estimated value of the child's health state;

[0033] Rounding the estimated value of the child's health status to obtain a child's health status value;

[0034] According to the child health status value, the corresponding child health status information is determined.

[0035] In a second aspect of the embodiment of the present application, a device for detecting oral gas for children is disclosed, the device comprising:

[0036] A memory storing executable program code;

[0037] a processor coupled to the memory;

[0038] The processor calls the executable program code stored in the memory to execute the oral gas detection method for children.

[0039] In a third aspect of the embodiments of the present application, a computer storable medium is disclosed, wherein the computer storable medium stores computer instructions, and when the computer instructions are called by a computer, they are used to execute the oral gas detection method for children.

[0040] In a fourth aspect of the embodiments of the present application, an information data processing terminal is disclosed, wherein the information data processing terminal is used to implement the oral gas detection method for children.

[0041] The beneficial effects of the present invention are:

[0042] Through the device of the present invention, children blow into the gas collection device in the morning on an empty stomach, and the reagents in the device (or other measurement methods) can measure the content of certain target gases. The present invention establishes a gas association model by extracting and associating the characteristics of children's physical characteristics with the components of the blown air, and achieves accurate detection of children's physical conditions by effectively detecting and extracting the characteristics of children's oral gas information.

[0043] The present invention establishes a state vector, performs state estimation processing on the state vector and the reduced-dimensional gas detection matrix corresponding to all types of health status information, and obtains a state estimation vector, thereby reducing detection complexity, achieving effective suppression of noise and interference in collected information, and improving detection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a flow chart for implementing the method of the present invention. DETAILED DESCRIPTION

[0045] In order to better understand the content of the present invention, an embodiment is given here.

[0046] Figure 1 It is a flow chart for implementing the method of the present invention.

[0047] In order to solve the problem of how to effectively detect and extract features of children's oral gas information to detect the physical condition of children, the present invention discloses an oral gas detection method and device for children.

[0048] In a first aspect of the embodiment of the present application, a method for detecting oral gas for children is disclosed, comprising:

[0049] S1, obtaining a set of oral gas detection information of children who have obtained different types of health status information; the set of oral gas detection information includes a number of oral gas detection information corresponding to each type of health status information; the oral gas detection information includes a sequence value of hydrogen sulfide content, a sequence value of methyl mercaptan content, and a sequence value of dimethyl sulfide content;

[0050] S2, performing correlation analysis on the health status information and the oral gas detection information set to obtain a gas correlation model;

[0051] S3, analyzing the components of the collected oral gas of the child to obtain gas component detection result information;

[0052] S4, using the gas association model to identify and process the gas component detection result information to obtain the child's health status information;

[0053] The correlating analysis and processing of the health status information and the oral gas detection information set to obtain a gas correlation model includes:

[0054] S21, preprocessing the health status information and the oral gas detection information set respectively to obtain preprocessed health status information and preprocessed oral gas detection information set;

[0055] S22, representing all types of health status information as a state vector; representing a set of pre-processed oral gas detection information corresponding to each type of health status information as a corresponding gas detection matrix; the elements of the state vector are the encoding values ​​of each type of health status information;

[0056] S23, performing data reduction processing on the gas detection matrices corresponding to all types of health status information to obtain a data reduction matrix;

[0057] S24, using a data reduction matrix, transforming a gas detection matrix corresponding to each type of health status information to obtain a corresponding reduced-dimensionality gas detection matrix; the reduced-dimensionality gas detection matrix includes a reduced-dimensionality gas detection vector;

[0058] S25, performing state estimation processing on the state vector and the reduced-dimensional gas detection matrix corresponding to all types of health status information to obtain a state estimation vector;

[0059] S26, constructing a gas correlation model using the reduced-dimensional gas detection matrix and the state estimation vector.

[0060] The preprocessing of the health status information and the oral gas detection information set to obtain preprocessed health status information and preprocessed oral gas detection information set includes:

[0061] S211, normalizing the health status information and the oral gas detection information set respectively to obtain pre-processed health status information and a normalized oral gas detection information set;

[0062] S212, performing boundary check processing on the pre-processed health status information and the normalized oral gas detection information set respectively to obtain a pre-processed oral gas detection information set;

[0063] The pre-processed health status information and the normalized oral gas detection information set are subjected to boundary check processing respectively to obtain the pre-processed oral gas detection information set, including:

[0064] For the normalized oral gas detection information set corresponding to all types of health status information, the data value of each type of oral gas detection information is taken as the dependent variable, and the corresponding health status information of the data is taken as the independent variable. The dependent variable and the independent variable are subjected to autoregression-sliding average modeling to obtain the regression model of the data of the type;

[0065] Using the regression model, the independent variable is calculated to obtain regression data values;

[0066] Determine whether the absolute value of the difference between the regression data value and the corresponding dependent variable value is greater than a set first regression discrimination threshold; if it is greater than the first regression discrimination threshold, delete the data from the normalized oral gas detection information set; if it is less than or equal to the first regression discrimination threshold, do not process the data;

[0067] The first regression discrimination threshold may be set to 5.

[0068] The state estimation process is performed on the state vector and the reduced-dimensional gas detection matrix corresponding to all types of health status information to obtain the state estimation vector, including:

[0069] An estimation error model is constructed using the reduced-dimensional gas detection matrix corresponding to the state vector and all types of health status information;

[0070] The estimation error model is solved to obtain a state estimation vector.

[0071] The estimation error model comprises:

[0072]

[0073] v T v = 1,

[0074] Among them, v is the state estimation vector, v is a column vector with dimension 2, and a k is the kth element of the state vector, and is also the encoding value of the kth type of health status information, r i k is the i-th row vector of the reduced-dimensionality gas detection matrix corresponding to the k-th type of health status information, P is the number of types of health status information, and N is the row dimension of the reduced-dimensionality gas detection matrix.

[0075] The gas correlation model comprises:

[0076] The data reduction matrix is ​​used to multiply the preprocessed oral gas detection information to obtain a dimension-reduced gas detection vector;

[0077] The state estimation vector is used to multiply the dimension-reduced gas detection vector to obtain the estimated value of the child's health state;

[0078] Rounding the estimated value of the child's health status to obtain a child's health status value;

[0079] According to the child health status value, the corresponding child health status information is determined.

[0080] The data reduction matrix is ​​obtained by performing data reduction processing on the gas detection matrix corresponding to all types of health status information, using a principal component analysis method to perform principal component extraction processing on the gas detection matrix, and using the obtained principal component extraction matrix as the data reduction matrix; the dimension of the data reduction matrix is ​​3 rows and 2 columns. The data reduction processing can adopt a feature reduction algorithm.

[0081] The data reduction matrix is ​​used to implement data reduction processing on the data.

[0082] The autoregressive-moving average modeling may adopt the ARMA method.

[0083] All oral gas detection information of the same type of health status information is represented by the corresponding oral map

[0084] The health status information includes a code value 1 of the first health status information, a code value 2 of the second health status information, a code value 3 of the third health status information, and a code value 4 of the fourth health status information; the first health status information may be completely healthy; the second health status information may be an unhealthy oral cavity or a problem with dietary factors; the third health status information may be an unhealthy digestive system or a problem with living habits; the third health status information may be a genetic factor;

[0085] Determining the corresponding child health status information according to the child health status value is a process of determining the corresponding health status information according to the coded value of the health status information, and performing reverse decoding according to the composition of the health status information.

[0086] The collected oral gas of children is measured by gas chromatograph; the gas composition detection result information includes hydrogen sulfide content, methyl mercaptan content, and dimethyl sulfide content;

[0087] The health status information is digitally coded information of different health status types of children;

[0088] In this method, after executing S23 and executing S24 to obtain the data reduction matrix, the method further includes optimizing the data reduction matrix, and using the optimized data reduction matrix in S24;

[0089] The optimization process includes:

[0090] Calculate the cross-correlation matrix of all row vectors of the data reduction matrix;

[0091] Decompose the cross-correlation matrix to obtain the feature matrix;

[0092] The decomposition process is calculated as follows:

[0093] Y=UAV,

[0094] Among them, U is the left decomposition matrix, A is the characteristic matrix, V is the right decomposition matrix, U and V are both orthogonal matrices, and A is a diagonal matrix; the decomposition process can adopt QR decomposition process.

[0095] The optimized data reduction matrix is ​​obtained by multiplying the left decomposition matrix with the data reduction matrix.

[0096] In a second aspect of the embodiment of the present application, a device for detecting oral gas for children is disclosed, the device comprising:

[0097] A memory storing executable program code;

[0098] a processor coupled to the memory;

[0099] The processor calls the executable program code stored in the memory to execute the oral gas detection method for children.

[0100] In a third aspect of the embodiments of the present application, a computer storable medium is disclosed, wherein the computer storable medium stores computer instructions, and when the computer instructions are called by a computer, they are used to execute the oral gas detection method for children.

[0101] In a fourth aspect of the embodiments of the present application, an information data processing terminal is disclosed, wherein the information data processing terminal is used to implement the oral gas detection method for children.

[0102] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A method for detecting oral gas for children, characterized in that: include: S1, obtaining a set of oral gas detection information of children who have obtained different types of health status information; the set of oral gas detection information includes a number of oral gas detection information corresponding to each type of health status information; the oral gas detection information includes a sequence value of hydrogen sulfide content, a sequence value of methyl mercaptan content, and a sequence value of dimethyl sulfide content; S2, performing correlation analysis on the health status information and the oral gas detection information set to obtain a gas correlation model; S3, analyzing the components of the collected oral gas of the child to obtain gas component detection result information; S4, using the gas association model to identify and process the gas composition detection result information to obtain the child's health status information.

2. The oral gas detection method for children according to claim 1, characterized in that: The correlating analysis and processing of the health status information and the oral gas detection information set to obtain a gas correlation model includes: S21, preprocessing the health status information and the oral gas detection information set respectively to obtain preprocessed health status information and preprocessed oral gas detection information set; S22, representing all types of health status information as a state vector; representing a set of pre-processed oral gas detection information corresponding to each type of health status information as a corresponding gas detection matrix; the elements of the state vector are the encoding values ​​of each type of health status information; S23, performing data reduction processing on the gas detection matrices corresponding to all types of health status information to obtain a data reduction matrix; S24, using a data reduction matrix, transforming a gas detection matrix corresponding to each type of health status information to obtain a corresponding reduced-dimensionality gas detection matrix; the reduced-dimensionality gas detection matrix includes a reduced-dimensionality gas detection vector; S25, performing state estimation processing on the state vector and the reduced-dimensional gas detection matrix corresponding to all types of health status information to obtain a state estimation vector; S26, constructing a gas correlation model using the reduced-dimensional gas detection matrix and the state estimation vector.

3. The method for detecting oral gas for children according to claim 2, characterized in that: The preprocessing of the health status information and the oral gas detection information set to obtain preprocessed health status information and preprocessed oral gas detection information set includes: S211, normalizing the health status information and the oral gas detection information set respectively to obtain pre-processed health status information and a normalized oral gas detection information set; S212, performing boundary check processing on the pre-processed health status information and the normalized oral gas detection information set respectively to obtain a pre-processed oral gas detection information set.

4. The method for detecting oral gas for children according to claim 3, characterized in that: The pre-processed health status information and the normalized oral gas detection information set are subjected to boundary check processing respectively to obtain the pre-processed oral gas detection information set, including: For the normalized oral gas detection information set corresponding to all types of health status information, the data value of each type of oral gas detection information is taken as the dependent variable, and the corresponding health status information of the data is taken as the independent variable. The dependent variable and the independent variable are subjected to autoregression-sliding average modeling to obtain the regression model of the data of the type; Using the regression model, the independent variable is calculated to obtain regression data values; Determine whether the absolute value of the difference between the regression data value and the corresponding dependent variable value is greater than a set first regression discrimination threshold; if it is greater than the first regression discrimination threshold, delete the data from the normalized oral gas detection information set; if it is less than or equal to the first regression discrimination threshold, do not process the data.

5. The method for detecting oral gas for children according to claim 4, characterized in that: The state estimation process is performed on the state vector and the reduced-dimensional gas detection matrix corresponding to all types of health status information to obtain the state estimation vector, including: An estimation error model is constructed using the reduced-dimensional gas detection matrix corresponding to the state vector and all types of health status information; The estimation error model is solved to obtain a state estimation vector.

6. The method for detecting oral gas for children according to claim 5, characterized in that: The estimation error model comprises: v T ·v=1, Among them, v is the state estimation vector, v is a column vector with dimension 2, and a k is the kth element of the state vector, and is also the encoding value of the kth type of health status information, r i k is the i-th row vector of the reduced-dimensionality gas detection matrix corresponding to the k-th type of health status information, P is the number of types of health status information, and N is the row dimension of the reduced-dimensionality gas detection matrix.

7. The method for detecting oral gas for children according to claim 4, characterized in that: The gas correlation model comprises: The data reduction matrix is ​​used to multiply the preprocessed oral gas detection information to obtain a dimension-reduced gas detection vector; The state estimation vector is used to multiply the dimension-reduced gas detection vector to obtain the estimated value of the child's health state; Rounding the estimated value of the child's health status to obtain a child's health status value; According to the child health status value, the corresponding child health status information is determined.

8. An oral gas detection device for children, characterized in that: The device comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the oral gas detection method for children according to any one of claims 1 to 7.

9. A computer storable medium, characterized in that: The computer storable medium stores computer instructions, and when the computer instructions are called by a computer, they are used to execute the oral gas detection method for children according to any one of claims 1 to 7.

10. An information data processing terminal, characterized in that: The information data processing terminal is used to implement the oral gas detection method for children according to any one of claims 1 to 7.

Images