CN113288133A - Method, apparatus, storage medium, and processor for predicting blood glucose level - Google Patents
Method, apparatus, storage medium, and processor for predicting blood glucose level Download PDFInfo
- Publication number
- CN113288133A CN113288133A CN202110489717.7A CN202110489717A CN113288133A CN 113288133 A CN113288133 A CN 113288133A CN 202110489717 A CN202110489717 A CN 202110489717A CN 113288133 A CN113288133 A CN 113288133A
- Authority
- CN
- China
- Prior art keywords
- blood glucose
- user
- value
- pulse signals
- predicting
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000008280 blood Substances 0.000 title claims abstract description 157
- 210000004369 blood Anatomy 0.000 title claims abstract description 157
- WQZGKKKJIJFFOK-GASJEMHNSA-N Glucose Natural products OC[C@H]1OC(O)[C@H](O)[C@@H](O)[C@@H]1O WQZGKKKJIJFFOK-GASJEMHNSA-N 0.000 title claims abstract description 115
- 239000008103 glucose Substances 0.000 title claims abstract description 115
- 238000000034 method Methods 0.000 title claims abstract description 69
- 238000007781 pre-processing Methods 0.000 claims abstract description 17
- 238000012545 processing Methods 0.000 claims abstract description 15
- 238000004458 analytical method Methods 0.000 claims abstract description 13
- 239000011159 matrix material Substances 0.000 claims description 74
- 238000000513 principal component analysis Methods 0.000 claims description 24
- 238000001514 detection method Methods 0.000 claims description 11
- 238000010606 normalization Methods 0.000 claims description 6
- 238000013186 photoplethysmography Methods 0.000 description 15
- 238000010586 diagram Methods 0.000 description 10
- 230000006870 function Effects 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 238000012549 training Methods 0.000 description 7
- 235000021152 breakfast Nutrition 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 230000036541 health Effects 0.000 description 5
- 238000012937 correction Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 230000009467 reduction Effects 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 235000005911 diet Nutrition 0.000 description 2
- 230000037213 diet Effects 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 235000012054 meals Nutrition 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 206010020772 Hypertension Diseases 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000001684 chronic effect Effects 0.000 description 1
- 206010012601 diabetes mellitus Diseases 0.000 description 1
- 230000001631 hypertensive effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
- A61B5/1455—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue using optical sensors, e.g. spectral photometrical oximeters
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/024—Detecting, measuring or recording pulse rate or heart rate
- A61B5/02416—Detecting, measuring or recording pulse rate or heart rate using photoplethysmograph signals, e.g. generated by infrared radiation
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
- A61B5/14532—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue for measuring glucose, e.g. by tissue impedance measurement
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7275—Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Surgery (AREA)
- Medical Informatics (AREA)
- Biophysics (AREA)
- Animal Behavior & Ethology (AREA)
- Pathology (AREA)
- Public Health (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Heart & Thoracic Surgery (AREA)
- Veterinary Medicine (AREA)
- Biomedical Technology (AREA)
- Physiology (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Optics & Photonics (AREA)
- Psychiatry (AREA)
- Cardiology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Spectroscopy & Molecular Physics (AREA)
- General Physics & Mathematics (AREA)
- Emergency Medicine (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The embodiment of the invention provides a method, a device, a storage medium and a processor for predicting blood sugar values, and belongs to the field of biological signal processing. The method for predicting a blood glucose level includes: acquiring pulse signals and historical blood glucose data of a user in a plurality of preset time periods; preprocessing the pulse signals; performing characteristic analysis on the preprocessed pulse signals to obtain characteristic information of the user; establishing a blood glucose prediction model according to the characteristic information and the historical blood glucose data based on a least square method; and predicting the blood sugar value of the user through the blood sugar prediction model. The method can improve the accuracy of the blood sugar value prediction.
Description
Technical Field
The present invention relates to the field of bio-signal processing, and in particular, to a method, an apparatus, a storage medium, and a processor for predicting a blood glucose level.
Background
At present, China has more than 3 hundred million chronic patients, wherein about 1.6 to 1.7 million of hypertensive population, about 1 million of hyperlipoidemia population and about 9240 million of diabetic patients, and China has become the first three high countries, but the traditional blood sugar detection method is time-consuming and labor-consuming, and is inconvenient for the measurement of patients by collecting finger blood for detection. Therefore, a technology for collecting human body signals through a bracelet to achieve noninvasive detection of human blood glucose values has appeared. The existing non-invasive detection technology estimates health indexes such as blood sugar values and the like based on big data modeling, and has the problem of large prediction error of the blood sugar values.
Disclosure of Invention
An object of embodiments of the present invention is to provide a method, an apparatus, a storage medium, and a processor for predicting a blood glucose level, so as to solve the problem of a large error in blood glucose level prediction.
In order to achieve the above object, a first aspect of the present invention provides a method for predicting a blood glucose level, the method comprising:
acquiring pulse signals and historical blood glucose data of a user in a plurality of preset time periods;
preprocessing the pulse signals;
performing characteristic analysis on the preprocessed pulse signals to obtain characteristic information of the user;
establishing a blood glucose prediction model according to the characteristic information and the historical blood glucose data based on a least square method;
and predicting the blood sugar value of the user through the blood sugar prediction model.
In an embodiment of the present invention, the data preprocessing of the pulse signal includes: determining the mean value of pulse signals in a preset time period; and in the case that the pulse signal is null within the preset time period, replacing the null value with the average value.
In an embodiment of the present invention, the method further comprises: determining the minimum value and the maximum value of the pulse signals in a preset time period; and carrying out normalization processing on the pulse signals according to the minimum value and the maximum value to obtain preprocessed pulse signals.
In the embodiment of the present invention, the performing feature analysis on the preprocessed pulse signal to obtain feature information of the user includes: and performing principal component analysis on the preprocessed pulse signals based on a principal component analysis algorithm to obtain a characteristic sequence of the user.
In the embodiment of the present invention, the principal component analysis of the preprocessed pulse signal based on the principal component analysis algorithm to obtain the feature sequence of the user includes: based on a principal component analysis algorithm, obtaining a first non-negative diagonal matrix according to the preprocessed pulse signals; the ten numerical values with the largest numerical value in the first non-negative diagonal matrix are reserved to obtain a second non-negative diagonal matrix; obtaining a data observation matrix according to the second non-negative diagonal matrix; and eliminating a zero vector sequence in the data observation matrix to obtain a characteristic sequence of the user.
In the embodiment of the invention, a blood glucose prediction model is established according to characteristic information and historical blood glucose data based on a least square method, and the method comprises the following steps: determining a regression coefficient according to the characteristic information and a preset error; and determining a blood glucose prediction model according to the regression coefficient, the characteristic information and the historical blood glucose data based on a least square method.
In an embodiment of the present invention, the method further comprises: acquiring personal information of a user; and correcting the blood sugar prediction model according to the personal information.
A second aspect of the invention provides a processor configured to perform the method for predicting a blood glucose value as described above.
A third aspect of the present invention provides an apparatus for predicting a blood glucose value, comprising: a blood glucose detecting device for detecting a blood glucose value of a user; a pulse detection device for detecting a pulse signal of a user; and the processor.
A fourth aspect of the invention provides a machine-readable storage medium having stored thereon instructions which, when executed by a processor, cause the processor to perform the above-described method for predicting a blood glucose value.
According to the technical scheme, the pulse signals and the historical blood sugar data of the user in a plurality of preset time periods are obtained, data preprocessing is carried out on the pulse signals, feature analysis is carried out on the preprocessed pulse signals to obtain feature information of the user, a blood sugar prediction model is built according to the feature information and the historical blood sugar data based on a least square method, and the blood sugar value of the user is predicted through the blood sugar prediction model. The method utilizes the early-stage collected data of each user to carry out modeling, is different from the traditional big data modeling method, trains the corresponding blood sugar prediction model according to the early-stage measured data of each user, can improve the accuracy rate of blood sugar value prediction, and can improve the training speed of the model and realize accurate and rapid prediction of blood sugar value by extracting the characteristic information from the preprocessed pulse signals to carry out modeling.
Additional features and advantages of embodiments of the invention will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the embodiments of the invention without limiting the embodiments of the invention. In the drawings:
FIG. 1 schematically illustrates a flow chart of a method for predicting blood glucose values in an embodiment of the present invention;
fig. 2 is a block diagram schematically showing the configuration of an apparatus for predicting a blood glucose value in an embodiment of the present invention.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating embodiments of the invention, are given by way of illustration and explanation only, not limitation.
FIG. 1 schematically shows a flow chart of a method for predicting blood glucose values in an embodiment of the invention. As shown in fig. 1, in an embodiment of the present invention, a method for predicting a blood glucose value is provided, which is exemplified by being applied to a processor, and the method may include the following steps:
step S102, obtaining pulse signals and historical blood glucose data of a user in a plurality of preset time periods.
It is understood that the preset time period is a preset time period for obtaining early measurement data of the user, such as a time period before or after breakfast, lunch or dinner of a day. The historical blood glucose data is the true blood glucose value measured by the user at the same time of measuring the pulse signal in the early stage.
Specifically, pulse signals (PPG) can be acquired at a bracelet of a human body by utilizing an LED lamp and a photoelectric sensor, signal waveforms are recorded through wireless transmission equipment, photoplethysmography (PPG) signals of 14 days in the period are acquired for each user, the acquisition is divided into half an hour before breakfast, lunch and dinner, and early-stage training sample small data X of each user is obtainediThe dimension is 42 × P, i is 1,2, …, M, i is the ith user, M is the number of users, P is the number of sampling points of the PPG signal, and the actual value of blood glucose can be recorded by the blood glucose meter. That is, the processor may acquire, through the pulse detection device (e.g., a bracelet), pulse signals (PPG) measured by the user over a plurality of preset time periods (e.g., half an hour before breakfast, lunch, and dinner within 14 days) and historical blood glucose data of the user detected by the blood glucose meter.
Step S104, data preprocessing is carried out on the pulse signals.
In particular, the processor performs data pre-processing on the acquired pulse signal (PPG) of the user.
In one embodiment, the data preprocessing of the pulse signals comprises: determining the mean value of pulse signals in a preset time period; and in the case that the pulse signal is null within the preset time period, replacing the null value with the average value.
Since jitter and hardware errors may occur during data acquisition and null values may occur in read data, the processor may calculate a mean value of each pulse signal (PPG) by a statistical method, and use the mean value to fill a position where the null value in the pulse signal is located, that is, when a pulse signal that is a null value occurs within a preset time period (for example, half an hour before a meal in a certain day), the null value may be replaced by the mean value of the preset time period.
In another embodiment, the data preprocessing of the pulse signals further comprises: determining the minimum value and the maximum value of the pulse signals in a preset time period; and carrying out normalization processing on the pulse signals according to the minimum value and the maximum value to obtain preprocessed pulse signals.
Specifically, in order to find an optimal solution more easily in the training process of the model, the processor may obtain the pulse signals within a preset time period, so as to determine the minimum value and the maximum value of the pulse signals within the preset time period by using a preset maximum value function and a preset minimum value function, so as to perform normalization processing on the PPG signals of each preset time period according to the minimum value and the maximum value, where a specific formula is as follows:
wherein, XiIs a pulse signal, XminIs a minimum value of XmaxIs the maximum value.
And step S106, performing characteristic analysis on the preprocessed pulse signals to obtain characteristic information of the user.
Specifically, the processor performs feature analysis on the preprocessed pulse signals to obtain feature information of the user.
In one embodiment, performing feature analysis on the preprocessed pulse signals to obtain feature information of the user includes: and performing principal component analysis on the preprocessed pulse signals based on a principal component analysis algorithm to obtain a characteristic sequence of the user.
Specifically, the processor can perform dimensionality reduction processing and feature selection on the data through a principal component analysis algorithm, and perform preprocessing on the X'iThe signal is subjected to principal component analysis to extract the characteristic information of the user, more specifically, a principal component analysis algorithm is used for replacing a plurality of random characteristics with less comprehensive characteristics, and the information of the lost random characteristics is kept as much as possible, so that the correlation among the characteristics is the lowest. The principal component here represents a feature including information of a plurality of feature variables, and can retain information of data to the maximum extent.
In one embodiment, the principal component analysis of the preprocessed pulse signals based on a principal component analysis algorithm to obtain a feature sequence of the user includes: based on a principal component analysis algorithm, obtaining a first non-negative diagonal matrix according to the preprocessed pulse signals; the ten numerical values with the largest numerical value in the first non-negative diagonal matrix are reserved to obtain a second non-negative diagonal matrix; obtaining a data observation matrix according to the second non-negative diagonal matrix; and eliminating a zero vector sequence in the data observation matrix to obtain a characteristic sequence of the user.
After the data preprocessing is performed on the pulse signals, the pulse signals preprocessed by the ith user can be the following matrix X'i:
Wherein P is the sampling point number of the pulse signal, X'iIs the preprocessed pulse signal.
To the above matrix X'iAnd performing basis transformation, transforming the data into a new coordinate system, so that the maximum variance of the data projection falls on the coordinate, wherein the specific solution is as follows: for covariance matrixAnd (3) carrying out singular value decomposition, wherein a specific formula is as follows:
where W is a P × P eigenvector matrix, Δ is a first non-negative diagonal matrix of P × Q, VTIs a QxQ eigenvector matrix, Q is a matrix VTOf dimension, X'iIs a matrix of the preprocessed pulse signals,is a transposed matrix of the preprocessed pulse signals.
Performing feature screening according to the values in the first non-negative diagonal matrix delta, reserving the first ten numerical values with the largest numerical value in the non-negative diagonal matrix, and setting other numerical values in the non-negative diagonal matrix to zero to obtain a second non-negative diagonal matrix delta1By left-multiplying the eigenvector matrix W and right-multiplying the eigenvector matrix VTObtaining a data observation matrix X ″)iThe concrete formula is as follows:
X″i TX″i=WΔ1VT
wherein, Delta1Is a second non-negative diagonal matrix, X ″)iFor a data observation matrix, Xi″TIs a transposed matrix of the data observation matrix.
Eliminating a zero vector sequence in the data observation matrix to obtain a principal component observation matrix X 'after dimension reduction processing'iI.e. the signature sequence of the user, the matrix size is 42 x 10.
And S108, establishing a blood glucose prediction model according to the characteristic information and the historical blood glucose data based on a least square method.
Specifically, the processor may build a blood glucose prediction model based on least squares based on the characteristic information and the historical blood glucose data.
In one embodiment, the blood glucose prediction model is established based on a least square method according to the characteristic information and the historical blood glucose data, and the method comprises the following steps: determining a regression coefficient according to the characteristic information and a preset error; and determining a blood glucose prediction model according to the regression coefficient, the characteristic information and the historical blood glucose data based on a least square method.
Specifically, a least square AR prediction (linear prediction) regression model is established by using the characteristic information of the user and the historical blood glucose data, and the blood glucose value of the user is continuously predicted. The autoregressive AR model can construct a prediction model according to the linear relation between data, and the specific formula is as follows:
X″′it-b1X″′it-1-b2X″′it-2-…-bjX″′it-p=εt
wherein, X'itAs characteristic information, bjIs the regression coefficient, j is 1,2, …, p, epsilontThe preset error can be randomly set.
Ten characteristic sequences with autocorrelation can be obtained through the step S106, parameters of the blood glucose prediction model are estimated by adopting a least square method, and an error equation is constructed:
V=X″′itB-Y
wherein B is a matrix of regression coefficients, X'itIs a feature sequence (feature information), and Y is historical blood glucose data. The least squares solution of B is:
the blood glucose values can be predicted by solving the regression coefficient matrix B.
In step S110, the blood glucose level of the user is predicted by the blood glucose prediction model.
Specifically, after the blood glucose prediction model is determined, the processor may implement prediction of the blood glucose values of the users according to the blood glucose prediction model of each user.
According to the method for predicting the blood sugar value, the pulse signals and the historical blood sugar data of the user in a plurality of preset time periods are obtained, data preprocessing is carried out on the pulse signals, feature analysis is carried out on the preprocessed pulse signals to obtain feature information of the user, a blood sugar prediction model is built according to the feature information and the historical blood sugar data based on a least square method, and the blood sugar value of the user is predicted through the blood sugar prediction model. The method utilizes the early-stage collected data of each user to carry out modeling, is different from the traditional big data modeling method, trains a corresponding blood sugar prediction model according to the early-stage measured data of each user, can improve the accuracy of blood sugar value prediction, and can improve the training speed of the model by extracting the characteristic information from the preprocessed pulse signals to carry out modeling, thereby realizing accurate and rapid prediction of blood sugar values, helping diabetics or non-diabetics to carry out non-invasive prediction of blood sugar values and continuously monitor body health parameters for 24 hours, reducing the occurrence of emergencies of the patients, improving the health consciousness of people and promoting the development of society.
In one embodiment, the method for predicting a blood glucose value further includes: acquiring personal information of a user; and correcting the blood sugar prediction model according to the personal information.
Specifically, when early detection data, namely poor data, of a user is obtained, personal information such as the age, sex, weight and disease history of the user can be obtained, and reference is provided for subsequent model correction. Furthermore, the self-adaptive adjustment is carried out on the blood glucose prediction model by analyzing the age, the gender and the current diet condition of the user, so that the error between the predicted value and the true value of the blood glucose prediction model is minimum. The embodiment uses the exclusive blood sugar prediction model of each user and refers to personal information, namely basic indexes of each body to carry out correction, so that the accuracy rate of blood sugar prediction can be improved to the maximum extent.
Fig. 2 is a block diagram schematically showing the configuration of an apparatus for predicting a blood glucose value in an embodiment of the present invention. As shown in fig. 2, in an embodiment of the present invention, there is provided an apparatus 200 for predicting a blood glucose value, including: a blood glucose detecting device 210, a pulse detecting device 220, and a processor 230, wherein:
the blood glucose detecting apparatus 210 detects a blood glucose level of the user.
A pulse detection device 220 for detecting a pulse signal of the user.
A processor 230 configured to: acquiring pulse signals and historical blood glucose data of a user in a plurality of preset time periods; preprocessing the pulse signals; performing characteristic analysis on the preprocessed pulse signals to obtain characteristic information of the user; establishing a blood glucose prediction model according to the characteristic information and the historical blood glucose data based on a least square method; and predicting the blood sugar value of the user through the blood sugar prediction model.
It is understood that the preset time period is a preset time period for obtaining early measurement data of the user, such as a time period before or after breakfast, lunch or dinner of a day. The historical blood glucose data is the true blood glucose value measured by the user at the same time of measuring the pulse signal in the early stage.
Specifically, photoplethysmography (PPG) signal acquisition may be performed on each user for 14 days in the past, and the measurement is divided into half an hour before breakfast, lunch and dinner, so as to obtain early-stage training sample small data X of each useriThe dimension is 42 × P, i is 1,2, …, M, i is the ith user, M is the number of users, P is the number of sampling points of the PPG signal, and the actual value of blood glucose can be recorded by the blood glucose meter. That is, the processor may acquire, through the pulse detection device (e.g., a bracelet), pulse signals (PPG) measured by the user over a plurality of preset time periods (e.g., half an hour before breakfast, lunch, and dinner within 14 days) and historical blood glucose data of the user detected by the blood glucose meter. The processor performs data pre-processing on the acquired pulse signal (PPG) of the user. The processor performs characteristic analysis on the preprocessed pulse signals to obtain characteristic information of the user. The processor may build a blood glucose prediction model based on a least squares method from the characteristic information and the historical blood glucose data. After the blood glucose prediction model is determined, the processor can realize the prediction of the blood glucose value of the user according to the blood glucose prediction model of each user.
The apparatus 200 for predicting blood glucose values obtains pulse signals and historical blood glucose data of a user in a plurality of preset time periods through the blood glucose detecting device 210 and the pulse detecting device 220, performs data preprocessing on the pulse signals, performs feature analysis on the preprocessed pulse signals to obtain feature information of the user, establishes a blood glucose prediction model according to the feature information and the historical blood glucose data based on a least square method, and predicts the blood glucose value of the user through the blood glucose prediction model. The device utilizes the early-stage collected data of each user to carry out modeling, is different from the traditional big data modeling method, trains a corresponding blood sugar prediction model according to the early-stage measured data of each user, can improve the accuracy rate of blood sugar value prediction, and carries out modeling by extracting characteristic information from the pulse signals after pretreatment, can improve the training speed of the model, realizes accurate and rapid prediction of blood sugar value, can help diabetics or non-diabetics to carry out non-invasive prediction of blood sugar value and monitor body health parameters for 24 hours continuously, reduces the occurrence of emergency of patients, improves the health consciousness of people and promotes the development of society.
In one embodiment, the processor 230 is further configured to: determining the mean value of pulse signals in a preset time period; and in the case that the pulse signal is null within the preset time period, replacing the null value with the average value.
Since jitter and hardware errors may occur during data acquisition and null values may occur in read data, the processor may calculate a mean value of each pulse signal (PPG) by a statistical method, and use the mean value to fill a position where the null value in the pulse signal is located, that is, when a pulse signal that is a null value occurs within a preset time period (for example, half an hour before a meal in a certain day), the null value may be replaced by the mean value of the preset time period.
In one embodiment, the processor 230 is further configured to: determining the minimum value and the maximum value of the pulse signals in a preset time period; and carrying out normalization processing on the pulse signals according to the minimum value and the maximum value to obtain preprocessed pulse signals.
Specifically, in order to find an optimal solution more easily in the training process of the model, the processor may obtain the pulse signals within a preset time period, so as to determine the minimum value and the maximum value of the pulse signals within the preset time period by using a preset maximum value function and a preset minimum value function, so as to perform normalization processing on the PPG signals of each preset time period according to the minimum value and the maximum value, where a specific formula is as follows:
wherein, XiIs a pulse signal, XminIs a minimum value of XmaxIs the maximum value.
In one embodiment, the processor 230 is further configured to: and performing principal component analysis on the preprocessed pulse signals based on a principal component analysis algorithm to obtain a characteristic sequence of the user.
Specifically, processor 230 may perform dimensionality reduction and feature selection on the data through a principal component analysis algorithm, on preprocessed X'iThe signal is subjected to principal component analysis to extract the characteristic information of the user, more specifically, a principal component analysis algorithm is used for replacing a plurality of random characteristics with less comprehensive characteristics, and the information of the lost random characteristics is kept as much as possible, so that the correlation among the characteristics is the lowest. The principal component here represents a feature including information of a plurality of feature variables, and can retain information of data to the maximum extent.
In one embodiment, the processor 230 is further configured to: based on a principal component analysis algorithm, obtaining a first non-negative diagonal matrix according to the preprocessed pulse signals; the ten numerical values with the largest numerical value in the first non-negative diagonal matrix are reserved to obtain a second non-negative diagonal matrix; obtaining a data observation matrix according to the second non-negative diagonal matrix; and eliminating a zero vector sequence in the data observation matrix to obtain a characteristic sequence of the user.
After the data preprocessing is performed on the pulse signals, the pulse signals preprocessed by the ith user can be the following matrix X'i:
Wherein P is the sampling point number of the pulse signal, X'iIs the preprocessed pulse signal.
To the above matrix X'iPerforming a basis transformation to transform the data into a new coordinate system, so that the maximum variance of the projection of the data falls on the coordinate,the specific solution method is as follows: for covariance matrixAnd (3) carrying out singular value decomposition, wherein a specific formula is as follows:
where W is a P × P eigenvector matrix, Δ is a first non-negative diagonal matrix of P × Q, VTIs a QxQ eigenvector matrix, Q is a matrix VTOf dimension, X'iIs a matrix of the preprocessed pulse signals,is a transposed matrix of the preprocessed pulse signals.
Performing feature screening according to the values in the first non-negative diagonal matrix delta, reserving the first ten numerical values with the largest numerical value in the non-negative diagonal matrix, and setting other numerical values in the non-negative diagonal matrix to zero to obtain a second non-negative diagonal matrix delta1By left-multiplying the eigenvector matrix W and right-multiplying the eigenvector matrix VTObtaining a data observation matrix X ″)iThe concrete formula is as follows:
wherein, Delta1Is a second non-negative diagonal matrix, X ″)iIn order to observe the matrix for the data,is a transposed matrix of the data observation matrix.
Eliminating a zero vector sequence in the data observation matrix to obtain a principal component observation matrix X 'after dimension reduction processing'iI.e. the signature sequence of the user, the matrix size is 42 x 10.
In one embodiment, the processor 230 is further configured to: determining a regression coefficient according to the characteristic information and a preset error; and determining a blood glucose prediction model according to the regression coefficient, the characteristic information and the historical blood glucose data based on a least square method.
Specifically, a least square AR prediction (linear prediction) regression model is established by using the characteristic information of the user and the historical blood glucose data, and the blood glucose value of the user is continuously predicted. The autoregressive AR model can construct a prediction model according to the linear relation between data, and the specific formula is as follows:
X″′it-b1X″′it-1-b2X″′it-2-…-bjX″′it-p=εt
wherein, X'itAs characteristic information, bjIs the regression coefficient, j is 1,2, …, p, epsilontThe preset error can be randomly set.
Ten characteristic sequences with autocorrelation can be obtained through the step S106, parameters of the blood glucose prediction model are estimated by adopting a least square method, and an error equation is constructed:
V=X″′itB-Y
wherein B is a matrix of regression coefficients, X'itIs a feature sequence (feature information), and Y is historical blood glucose data. The least squares solution of B is:
the blood glucose values can be predicted by solving the regression coefficient matrix B.
In one embodiment, the processor 230 is further configured to: acquiring personal information of a user; and correcting the blood sugar prediction model according to the personal information.
Specifically, when early detection data, namely poor data, of a user is obtained, personal information such as the age, sex, weight and disease history of the user can be obtained, and reference is provided for subsequent model correction. Furthermore, the self-adaptive adjustment is carried out on the blood glucose prediction model by analyzing the age, the gender and the current diet condition of the user, so that the error between the predicted value and the true value of the blood glucose prediction model is minimum. The embodiment uses the exclusive blood sugar prediction model of each user and refers to personal information, namely basic indexes of each body to carry out correction, so that the accuracy rate of blood sugar prediction can be improved to the maximum extent.
The device for predicting the blood glucose value comprises a processor and a memory, wherein the processor comprises a kernel, and the kernel calls the corresponding program unit from the memory. The kernel can be set to be one or more, and the accuracy of the blood glucose value prediction is improved by adjusting the kernel parameters.
The memory may include volatile memory in a computer readable medium, Random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip.
Embodiments of the present invention provide a processor configured to execute the method for predicting blood glucose values according to the above embodiments.
Embodiments of the present invention provide a machine-readable storage medium having stored thereon instructions which, when executed by a processor, cause the processor to perform the method for predicting blood glucose values according to the above embodiments.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. 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 (10)
1. A method for predicting a blood glucose value, the method comprising:
acquiring pulse signals and historical blood glucose data of a user in a plurality of preset time periods;
performing data preprocessing on the pulse signals;
performing characteristic analysis on the preprocessed pulse signals to obtain characteristic information of the user;
establishing a blood glucose prediction model according to the characteristic information and the historical blood glucose data based on a least square method;
predicting a blood glucose value of the user by the blood glucose prediction model.
2. The method for predicting a blood glucose value as set forth in claim 1, wherein the data preprocessing the pulse signal comprises:
determining the mean value of the pulse signals in the preset time period;
and under the condition that the pulse signal is null value within the preset time period, replacing the null value with the mean value.
3. The method for predicting a blood glucose value of claim 2, further comprising:
determining the minimum value and the maximum value of the pulse signals in the preset time period;
and carrying out normalization processing on the pulse signals according to the minimum value and the maximum value to obtain preprocessed pulse signals.
4. The method for predicting a blood glucose value as set forth in claim 1, wherein the performing a feature analysis on the preprocessed pulse signals to obtain feature information of the user comprises:
and performing principal component analysis on the preprocessed pulse signals based on a principal component analysis algorithm to obtain the characteristic sequence of the user.
5. The method for predicting a blood glucose value according to claim 4, wherein the principal component analysis of the preprocessed pulse signal based on the principal component analysis algorithm to obtain the feature sequence of the user comprises:
based on a principal component analysis algorithm, obtaining a first non-negative diagonal matrix according to the preprocessed pulse signals;
reserving the ten numerical values with the largest numerical value in the first non-negative diagonal matrix to obtain a second non-negative diagonal matrix;
obtaining a data observation matrix according to the second non-negative diagonal matrix;
and eliminating the zero vector sequence in the data observation matrix to obtain the characteristic sequence of the user.
6. The method for predicting a blood glucose value according to claim 1, wherein the building a blood glucose prediction model from the characteristic information and the historical blood glucose data based on a least squares method comprises:
determining a regression coefficient according to the characteristic information and a preset error;
and determining the blood glucose prediction model according to the regression coefficient, the characteristic information and the historical blood glucose data based on a least square method.
7. The method for predicting a blood glucose value of claim 1, further comprising:
acquiring personal information of the user;
and correcting the blood sugar prediction model according to the personal information.
8. A processor configured to perform the method for predicting a blood glucose value according to any one of claims 1 to 7.
9. An apparatus for predicting a blood glucose value, comprising:
a blood glucose detecting device for detecting a blood glucose value of a user;
a pulse detection device for detecting a pulse signal of the user; and
the processor of claim 8.
10. A machine-readable storage medium having instructions stored thereon, which when executed by a processor causes the processor to perform a method for predicting a blood glucose value according to any one of claims 1 to 6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110489717.7A CN113288133A (en) | 2021-05-06 | 2021-05-06 | Method, apparatus, storage medium, and processor for predicting blood glucose level |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110489717.7A CN113288133A (en) | 2021-05-06 | 2021-05-06 | Method, apparatus, storage medium, and processor for predicting blood glucose level |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113288133A true CN113288133A (en) | 2021-08-24 |
Family
ID=77320831
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110489717.7A Pending CN113288133A (en) | 2021-05-06 | 2021-05-06 | Method, apparatus, storage medium, and processor for predicting blood glucose level |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113288133A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113729698A (en) * | 2021-08-26 | 2021-12-03 | 朱斌 | Noninvasive blood glucose detection method and system |
CN117612737A (en) * | 2024-01-24 | 2024-02-27 | 胜利油田中心医院 | Intelligent optimization method for diabetes care data |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107157492A (en) * | 2017-05-19 | 2017-09-15 | 国家电网公司 | A kind of embedded human physiologic information non-invasive detection system and data processing method |
CN107463766A (en) * | 2017-06-23 | 2017-12-12 | 深圳市中识创新科技有限公司 | Generation method, device and the computer-readable recording medium of blood glucose prediction model |
US20200187835A1 (en) * | 2018-12-13 | 2020-06-18 | Samsung Electronics Co., Ltd. | Apparatus and method for estimating blood glucose |
WO2020228084A1 (en) * | 2019-05-13 | 2020-11-19 | 深圳六合六医疗器械有限公司 | Personalized blood sugar interval statistical method and device |
-
2021
- 2021-05-06 CN CN202110489717.7A patent/CN113288133A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107157492A (en) * | 2017-05-19 | 2017-09-15 | 国家电网公司 | A kind of embedded human physiologic information non-invasive detection system and data processing method |
CN107463766A (en) * | 2017-06-23 | 2017-12-12 | 深圳市中识创新科技有限公司 | Generation method, device and the computer-readable recording medium of blood glucose prediction model |
US20200187835A1 (en) * | 2018-12-13 | 2020-06-18 | Samsung Electronics Co., Ltd. | Apparatus and method for estimating blood glucose |
WO2020228084A1 (en) * | 2019-05-13 | 2020-11-19 | 深圳六合六医疗器械有限公司 | Personalized blood sugar interval statistical method and device |
Non-Patent Citations (1)
Title |
---|
吕亚帅: ""无创血糖检测系统的研究与设计"", 《中国优秀博硕士学位论文全文数据库(硕士)工程科技Ⅱ辑》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113729698A (en) * | 2021-08-26 | 2021-12-03 | 朱斌 | Noninvasive blood glucose detection method and system |
CN117612737A (en) * | 2024-01-24 | 2024-02-27 | 胜利油田中心医院 | Intelligent optimization method for diabetes care data |
CN117612737B (en) * | 2024-01-24 | 2024-05-03 | 胜利油田中心医院 | Intelligent optimization method for diabetes care data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11562222B2 (en) | Systems and methods of identity analysis of electrocardiograms | |
US20180233227A1 (en) | Systems and methods of analyte measurement analysis | |
US20210153761A1 (en) | Method, system and non-transitory computer-readable recording medium for estimating arrhythmia by using artificial neural network | |
EP3558113A1 (en) | Determining wellness using activity data | |
US20160128638A1 (en) | System and method for detecting and quantifying deviations from physiological signals normality | |
US9759558B2 (en) | Device and method for automatically normalizing the physiological signals of a living being | |
CN113288133A (en) | Method, apparatus, storage medium, and processor for predicting blood glucose level | |
EP3803709A1 (en) | Detecting abnormalities in ecg signals | |
Page et al. | Visualization of health monitoring data acquired from distributed sensors for multiple patients | |
Park et al. | Prediction of daily mental stress levels using a wearable photoplethysmography sensor | |
WO2019069417A1 (en) | Biological information processing device, biological information processing system, biological information processing method, and storage medium | |
CN115702782A (en) | Heart rate detection method based on deep learning and wearable device | |
Kumar et al. | STSR: spectro-temporal super-resolution analysis of a reference signal less photoplethysmogram for heart rate estimation during physical activity | |
WO2021045616A1 (en) | A computer-implemented method and an apparatus for determining body positions and/or sleep stages of a person during sleep. | |
Mortensen et al. | Multi-class stress detection through heart rate variability: A deep neural network based study | |
CN113645901A (en) | Assessment of muscle fatigue | |
Mesin | Heartbeat monitoring from adaptively down-sampled electrocardiogram | |
Molkkari et al. | Non-linear heart rate variability measures in sleep stage analysis with photoplethysmography | |
JP2008253727A (en) | Monitor device, monitor system and monitoring method | |
Pankaj et al. | Blood pressure estimation and classification using a reference signal-less photoplethysmography signal: a deep learning framework | |
CN115770028A (en) | Blood pressure detection method, system, device and storage medium | |
CN112741623B (en) | Electrocardio-based glycosylated hemoglobin detection system and method | |
Park et al. | A study on the development of a day-to-day mental stress monitoring system using personal physiological data | |
CN108814580A (en) | A kind of method and relevant device of non-contact type heart rate test | |
CN115666376A (en) | System and method for hypertension monitoring |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20210824 |
|
WD01 | Invention patent application deemed withdrawn after publication |