CN109872821A - A kind of method and device constructing blood glucose prediction model - Google Patents
A kind of method and device constructing blood glucose prediction model Download PDFInfo
- Publication number
- CN109872821A CN109872821A CN201910332868.4A CN201910332868A CN109872821A CN 109872821 A CN109872821 A CN 109872821A CN 201910332868 A CN201910332868 A CN 201910332868A CN 109872821 A CN109872821 A CN 109872821A
- Authority
- CN
- China
- Prior art keywords
- blood glucose
- model
- value
- training
- user
- 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
Abstract
The present invention provides a kind of method and devices for constructing blood glucose prediction model, wherein method includes: the blood glucose value and corresponding infrared nondestructive test characteristic parameter for obtaining multiple users, wherein blood glucose value includes training blood glucose value and test blood glucose value；For each preset blood glucose model, the blood glucose model is trained using corresponding trained blood glucose value as the output of the blood glucose model using infrared nondestructive test training characteristics parameter as the input of the blood glucose model；For the blood glucose model of each training, the input of blood glucose model using infrared nondestructive test test feature parameter as the training, obtain the blood glucose value to be compared of the output of the blood glucose model of the training, based on the corresponding test blood glucose value of infrared nondestructive test test feature parameter and blood glucose value to be compared, the index value for the blood glucose model accuracy for characterizing the training is obtained；The blood glucose model of the corresponding training of the highest index value of accuracy is determined as blood glucose prediction model.It can effectively improve blood glucose measurement efficiency.
Description
Technical field
The present invention relates to blood glucose measurement technical field, in particular to a kind of method for constructing blood glucose prediction model and
Device.
Background technique
Blood glucose measurement has important directive significance to people's lives rule, activity, movement, diet.Especially to sugar
Urinate patient.Diabetic finds the problem and sees a doctor in time, reduce diabetes complicated by grasping own blood glucose variation in real time
The risk of disease improves the quality of life of patient, improves physical condition.In addition, doctor can also pass through the blood glucose of diabetic
Understand treatment condition, adjusts therapeutic scheme.
In the prior art, the method for blood glucose measurement is largely measured using invasive blood glucose meter, which exists
It needs frequently to be drawn blood in measurement process or finger acupuncture treatment takes blood, the above method takes blood using blood collecting pen thorn finger pulp, will take
Hemophoric test paper is placed on reserved area and carries out analysis measurement.This invasive blood sugar measuring method not only give diabetic with
Body and pain physiologically are come, but also the time needed for measurement blood glucose is longer, the disposable apparatus needed is more, blood glucose
Measurement efficiency is lower, measurement cost is higher, moreover, invasive blood sugar measuring method, can also increase the risk of infectious disease.
Summary of the invention
In view of this, the purpose of the present invention is to provide a kind of method and device for constructing blood glucose prediction model, to improve
Blood glucose measurement efficiency.
In a first aspect, the embodiment of the invention provides a kind of methods for constructing blood glucose prediction model, comprising:
Obtain the blood glucose value and the corresponding measurement ginseng including infrared nondestructive test characteristic parameter of the user of multiple users
Number, wherein the blood glucose value includes training blood glucose value and test blood glucose value, and the infrared nondestructive test characteristic parameter includes infrared
Line thermography training characteristics parameter and infrared nondestructive test test feature parameter；
For each preset blood glucose model, using the infrared nondestructive test training characteristics parameter as the blood glucose model
Input, using the corresponding trained blood glucose value of the infrared nondestructive test training characteristics parameter as the output of the blood glucose model,
The blood glucose model is trained；
For the blood glucose model of each training, using the infrared nondestructive test test feature parameter as the blood glucose of the training
The input of model obtains the blood glucose value to be compared of the output of the blood glucose model of the training, is tested based on the infrared nondestructive test
The corresponding test blood glucose value of characteristic parameter and the blood glucose value to be compared, obtain the blood glucose model accuracy for characterizing the training
Index value；
The blood glucose model of the corresponding training of the highest index value of accuracy is determined as blood glucose prediction model.In conjunction with first party
Face, the embodiment of the invention provides the first possible embodiments of first aspect, wherein the measurement parameter further include:
Blood pressure, heart rate and body temperature,
It is described using the infrared nondestructive test training characteristics parameter as the input of the blood glucose model, comprising:
Using the infrared nondestructive test training characteristics parameter, blood pressure, heart rate and body temperature as the input of the blood glucose model；
The input of the blood glucose model using the infrared nondestructive test test feature parameter as the training, comprising:
Using the infrared nondestructive test test feature parameter, blood pressure, heart rate and body temperature as the blood glucose model of the training
Input.
With reference to first aspect, the embodiment of the invention provides second of possible embodiments of first aspect, wherein institute
Blood glucose model is stated to be trained using neural network algorithm or regression algorithm.
With reference to first aspect, the embodiment of the invention provides the third possible embodiments of first aspect, wherein institute
State the gray value that infrared nondestructive test characteristic parameter includes face thermography.
The third possible embodiment with reference to first aspect, the embodiment of the invention provides the 4th kind of first aspect
Possible embodiment, wherein obtain the gray value of the face thermography, comprising:
Obtain the face thermography of user；
Noise processed is removed to the face thermography using Wavelet Algorithm, the face thermal imagery after being denoised
Figure；
The gray value of face thermography after extracting the denoising calculates the average value of the gray value of extraction, obtains
The gray value of the face thermography.
With reference to first aspect, the embodiment of the invention provides the 5th kind of possible embodiments of first aspect, wherein institute
State method further include:
Shoot the face thermography to be measured of user to be predicted；
Obtain the target gray value of the face thermography to be measured；
Using the target gray value as the input of the blood glucose prediction model, the blood glucose of the user to be predicted is predicted
Value.
With reference to first aspect, the embodiment of the invention provides the 6th kind of possible embodiments of first aspect, wherein institute
State method further include:
Shoot the face thermography to be measured of user to be predicted；
Obtain the target gray value of the face thermography to be measured and blood pressure, heart rate and the body of the user to be predicted
Temperature；
Using the target gray value, blood pressure, heart rate and body temperature as the input of the blood glucose prediction model, described in prediction
The blood glucose value of user to be predicted.
The 5th kind of possible embodiment or the 6th kind of possible embodiment with reference to first aspect, the embodiment of the present invention
Provide the 7th kind of possible embodiment of first aspect, wherein the method also includes:
The blood glucose value of the user to be predicted of prediction is compared with pre-set blood glucose normality threshold range, it will
Comparison result is prompted to the user to be predicted.
The 7th kind of possible embodiment with reference to first aspect, the embodiment of the invention provides the 8th kind of first aspect
Possible embodiment, wherein it is described by comparison result to the user to be predicted carry out prompt include:
If the blood glucose value of the user to be predicted of prediction is greater than the blood glucose in pre-set blood glucose normality threshold range
Upper bound threshold value prompts the user blood glucose to be predicted excessively high, and prompts the excessively high corresponding points for attention of blood glucose and avoid blood glucose mistake
High treatment method；
If the blood glucose value of the user to be predicted of prediction is less than the blood glucose in pre-set blood glucose normality threshold range
Lower bound threshold value prompts the user blood glucose to be predicted too low, and prompts the corresponding points for attention of hypoglycemia and avoid blood glucose mistake
Low treatment method；
If the blood glucose value of the user to be predicted of prediction is within the scope of pre-set blood glucose normality threshold, described in prompt
User blood glucose to be predicted is normal.
Second aspect, the embodiment of the invention also provides a kind of devices for constructing blood glucose prediction model, comprising:
User data acquisition module, the blood glucose value and the user for obtaining multiple users are corresponding including infrared heat
As the measurement parameter of figure characteristic parameter, wherein the blood glucose value includes training blood glucose value and test blood glucose value, the infrared heat
As figure characteristic parameter includes infrared nondestructive test training characteristics parameter and infrared nondestructive test test feature parameter；
Blood glucose model training module, it is for being directed to each preset blood glucose model, infrared nondestructive test training is special
Input of the parameter as the blood glucose model is levied, by the corresponding trained blood glucose value of the infrared nondestructive test training characteristics parameter
As the output of the blood glucose model, which is trained；
Accuracy comparison module, for being directed to the blood glucose model of each training, with the infrared nondestructive test test feature
Input of the parameter as the blood glucose model of the training, obtains the blood glucose value to be compared of the output of the blood glucose model of the training, is based on
The corresponding test blood glucose value of infrared nondestructive test test feature parameter and the blood glucose value to be compared, obtaining characterization should
The index value of trained blood glucose model accuracy；
Blood glucose prediction model determining module, for determining the blood glucose model of the corresponding training of the highest index value of accuracy
For blood glucose prediction model.
The third aspect, the embodiment of the present application provide a kind of computer equipment, including memory, processor and are stored in institute
The computer program that can be run on memory and on the processor is stated, the processor executes real when the computer program
The step of existing above method.
Fourth aspect, the embodiment of the present application provide a kind of computer readable storage medium, the computer-readable storage
Computer program is stored on medium, the computer program executes above-mentioned method when being run by processor the step of.
The method and device of building blood glucose prediction model provided in an embodiment of the present invention, by the blood glucose for obtaining multiple users
Value and the corresponding measurement parameter including infrared nondestructive test characteristic parameter of the user, wherein the blood glucose value includes training
Blood glucose value and test blood glucose value, the infrared nondestructive test characteristic parameter includes infrared nondestructive test training characteristics parameter and infrared
Line thermography test feature parameter；For each preset blood glucose model, the infrared nondestructive test training characteristics parameter is made
For the input of the blood glucose model, using the corresponding trained blood glucose value of the infrared nondestructive test training characteristics parameter as the blood
The output of sugared model is trained the blood glucose model；For the blood glucose model of each training, with infrared nondestructive test survey
Input of the characteristic parameter as the blood glucose model of the training is tried, the blood glucose to be compared of the output of the blood glucose model of the training is obtained
Value is based on the corresponding test blood glucose value of the infrared nondestructive test test feature parameter and the blood glucose value to be compared, obtains
To the index value for the blood glucose model accuracy for characterizing the training；By the blood glucose model of the corresponding training of the highest index value of accuracy
It is determined as blood glucose prediction model.In this way, blood glucose prediction model is constructed using infrared nondestructive test characteristic parameter, so as to utilize
The infrared nondestructive test characteristic parameter that the infrared nondestructive test of user's shooting obtains is predicted, the blood glucose value of the user is obtained,
It realizes noninvasive blood glucose measurement, avoids bringing body and pain physiologically to diabetic, also make needed for measuring blood glucose
Time it is short, avoid consumption disposable apparatus, blood glucose measurement is high-efficient, measurement cost is low, moreover, noninvasive blood glucose measurement side
Method, the risk for the disease that can avoid infection.
To enable the above objects, features and advantages of the present invention to be clearer and more comprehensible, preferred embodiment is cited below particularly, and cooperate
Appended attached drawing, is described in detail below.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only certain embodiments of the present invention, therefore is not construed as pair
The restriction of range for those of ordinary skill in the art without creative efforts, can also be according to this
A little attached drawings obtain other relevant attached drawings.
Fig. 1 shows the method flow schematic diagram of building blood glucose prediction model provided by the embodiment of the present invention；
Fig. 2 shows the apparatus structure schematic diagrams that blood glucose prediction model is constructed provided by the embodiment of the present invention；
Fig. 3 shows a kind of structural schematic diagram of computer equipment 300 provided by the embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
Middle attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is only
It is a part of the embodiment of the present invention, instead of all the embodiments.The present invention being usually described and illustrated herein in the accompanying drawings is real
The component for applying example can be arranged and be designed with a variety of different configurations.Therefore, of the invention to what is provided in the accompanying drawings below
The detailed description of embodiment is not intended to limit the range of claimed invention, but is merely representative of selected reality of the invention
Apply example.Based on the embodiment of the present invention, those skilled in the art institute obtained without making creative work
There are other embodiments, shall fall within the protection scope of the present invention.
Showing blood as an organic molecule by statistics, vibration frequency is close with the vibration frequency of infrared ray, because
And when using infrared radiation blood organic molecule, chemical bond or functional group in blood organic molecule can be made to occur
The absorption frequency of absorption of vibrations, different chemical bonds or functional group is different.When Blood Glucose concentration difference, corresponding absorption
Frequency is also different, so that the infrared nondestructive test formed is also different.Thus, it is right using infrared ray in the embodiment of the present application
Face carries out thermal imaging, makes infrared penetration face and has an effect with the blood in face, and generating includes blood sugar concentration information
Infrared nondestructive test, thus by analysis obtain infrared nondestructive test in include characteristic parameter, finally obtain blood sugar for human body
Concentration information.
Fig. 1 shows the method flow schematic diagram of building blood glucose prediction model provided by the embodiment of the present invention.Such as Fig. 1 institute
Show, this method comprises:
Step 101, blood glucose value and the user for obtaining multiple users are corresponding including infrared nondestructive test characteristic parameter
Measurement parameter, wherein the blood glucose value include training blood glucose value and test blood glucose value, the infrared nondestructive test characteristic parameter
Including infrared nondestructive test training characteristics parameter and infrared nondestructive test test feature parameter；
In the embodiment of the present application, as an alternative embodiment, infrared nondestructive test is shot using thermal imaging system.Wherein, make
For an alternative embodiment, thermal imaging system can be set and be mounted on mobile terminal, for example, in mobile phone, wearable device, from
And infrared nondestructive test can be shot in order to user.When work, thermal imaging system (is used the object in scene using optical device
Family) issue focus infrared energy on infrared detector, then the infrared data from each detector element is converted into
The picture format of standard, obtains infrared nondestructive test.
In the embodiment of the present application, the blood glucose value for obtaining user is to be obtained by invasive mode using blood analysis method.
As an alternative embodiment, when carrying out blood collection to user, by the thermal imaging system installed on mobile terminal to the user
Face shot.
Step 102, for each preset blood glucose model, using the infrared nondestructive test training characteristics parameter as the blood
The input of sugared model, using the corresponding trained blood glucose value of the infrared nondestructive test training characteristics parameter as the blood glucose model
Output, which is trained；
In the embodiment of the present application, as an alternative embodiment, blood glucose model using neural network algorithm or regression algorithm into
Row training.Wherein, neural network algorithm and regression algorithm belong to deep learning algorithm scope, and deep learning algorithm is low by combining
Layer feature forms more abstract high-rise expression attribute classification or feature and utilizes depth to find that the distributed nature of data indicates
Learning algorithm is spent, the feature learning and layered characteristic that can use non-supervisory formula or Semi-supervised or supervised extract highly effective algorithm
Obtain feature.For example, convolutional neural networks (Convolutional neural networks, CNNs) are a kind of supervised
Machine learning model, and depth confidence net (Deep Belief Nets, DBNs) is a kind of machine learning model of unsupervised formula.
Wherein, for neural network algorithm, LM algorithm, Regularization algorithms and quantization conjugate gradient algorithms etc. can be used,
And by comparison, most suitable algorithm is chosen.
As another alternative embodiment, blood glucose model can also be trained using curve-fitting method, wherein curve is quasi-
Conjunction includes but is not limited to: Gaussian function fitting, fourier function fitting, trigonometric function fitting etc..
It is corresponding in the blood glucose value and the user for obtaining multiple users as an alternative embodiment in the embodiment of the present application
The measurement parameter including infrared nondestructive test characteristic parameter before, this method further include:
Listener clustering is carried out to multiple users, obtains each corresponding blood glucose value of user and the user in each listener clustering
The corresponding measurement parameter including infrared nondestructive test characteristic parameter.
In the embodiment of the present application, as an alternative embodiment, each listener clustering corresponds to preset multiple blood glucose models,
Multiple blood glucose models are trained respectively based on the listener clustering of the classification, and from trained blood glucose model, it is accurate to choose
The highest model of property is as the corresponding blood glucose prediction model of the listener clustering.
In the embodiment of the present application, as an alternative embodiment, infrared nondestructive test characteristic parameter includes face thermography
Gray value.
In the embodiment of the present application, as an alternative embodiment, the gray value of the face thermography is obtained, comprising:
Obtain the face thermography of user；
Noise processed is removed to the face thermography using Wavelet Algorithm, the face thermal imagery after being denoised
Figure；
The gray value of face thermography after extracting the denoising calculates the average value of the gray value of extraction, obtains
The gray value of the face thermography.
Step 103, for the blood glucose model of each training, using the infrared nondestructive test test feature parameter as the instruction
The input of experienced blood glucose model obtains the blood glucose value to be compared of the output of the blood glucose model of the training, is based on the infrared heat
As the corresponding test blood glucose value of figure test feature parameter and the blood glucose value to be compared, the blood glucose mould for characterizing the training is obtained
The index value of type accuracy；
In the embodiment of the present application, as an alternative embodiment, index value includes but is not limited to: error rate, the number of iterations etc..
Step 104, the blood glucose model by the corresponding training of the highest index value of accuracy is determined as blood glucose prediction model.
In the embodiment of the present application, it is contemplated that the influence of various factors, personal extent of metabolism, physical difference all can be different
Sample, thus, blood glucose prediction model customizing can also be carried out to personal or similar population.
In the embodiment of the present application, measured to improve the precision of blood glucose prediction model of building as an alternative embodiment
Parameter can also include: blood pressure, heart rate and body temperature,
It is described using the infrared nondestructive test training characteristics parameter as the input of the blood glucose model, comprising:
Using the infrared nondestructive test training characteristics parameter, blood pressure, heart rate and body temperature as the input of the blood glucose model；
The input of the blood glucose model using the infrared nondestructive test test feature parameter as the training, comprising:
Using the infrared nondestructive test test feature parameter, blood pressure, heart rate and body temperature as the blood glucose model of the training
Input.
In the embodiment of the present application, after obtaining blood glucose prediction model, it can use the blood glucose prediction model and user clap
The face thermography taken the photograph carries out blood sugar concentration prediction, as an alternative embodiment, this method further include:
A11 shoots the face thermography to be measured of user to be predicted；
A12 obtains the target gray value of the face thermography to be measured；
A13 predicts the blood of the user to be predicted using the target gray value as the input of the blood glucose prediction model
Sugar value.
In the embodiment of the present application, after obtaining blood glucose prediction model, it can use the blood glucose prediction model and user clap
The face thermography taken the photograph carries out blood sugar concentration prediction, thus, as another alternative embodiment, this method further include:
A21 shoots the face thermography to be measured of user to be predicted；
A22, obtain the target gray value of the face thermography to be measured and the blood pressure of the user to be predicted, heart rate with
And body temperature；
A23, using the target gray value, blood pressure, heart rate and body temperature as the input of the blood glucose prediction model, prediction
The blood glucose value of the user to be predicted.
In the embodiment of the present application, user can also be carried out according to the blood glucose value (blood sugar concentration information) that prediction obtains corresponding
Prompt, thus, this method further include:
The blood glucose value of the user to be predicted of prediction is compared with pre-set blood glucose normality threshold range, it will
Comparison result is prompted to the user to be predicted.
In the embodiment of the present application, as an alternative embodiment, comparison result is subjected to prompt packet to the user to be predicted
It includes:
If the blood glucose value of the user to be predicted of prediction is greater than the blood glucose in pre-set blood glucose normality threshold range
Upper bound threshold value prompts the user blood glucose to be predicted excessively high, and prompts the excessively high corresponding points for attention of blood glucose and avoid blood glucose mistake
High treatment method；
If the blood glucose value of the user to be predicted of prediction is less than the blood glucose in pre-set blood glucose normality threshold range
Lower bound threshold value prompts the user blood glucose to be predicted too low, and prompts the corresponding points for attention of hypoglycemia and avoid blood glucose mistake
Low treatment method；
If the blood glucose value of the user to be predicted of prediction is within the scope of pre-set blood glucose normality threshold, described in prompt
User blood glucose to be predicted is normal.
In the embodiment of the present application, infrared nondestructive test can be uploaded to by cloud host by mobile phone and carry out data storage
It deposits, cloud host carries out the prediction of blood glucose value and progress of classifying according to the infrared nondestructive test and other measurement parameters that receive
It saves, the blood glucose value of prediction is notified to user.
As an alternative embodiment, it is also based on cloud host design website, user uploads infrared heat picture by website
Whether figure, reselection input blood pressure, heart rate, body temperature parameter to obtain different types of blood glucose numerical value.If do not select input blood pressure,
Heart rate, body temperature parameter, it is only necessary to which blood glucose value can be obtained by uploading infrared nondestructive test；If selection input blood pressure, heart rate, body temperature ginseng
Number, in conjunction with the infrared nondestructive test of upload, available more accurate blood glucose value.
The method of building blood glucose prediction model provided in an embodiment of the present invention, by obtain multiple users blood glucose value and
The corresponding measurement parameter including infrared nondestructive test characteristic parameter of the user, wherein the blood glucose value includes training blood glucose value
With test blood glucose value, the infrared nondestructive test characteristic parameter includes infrared nondestructive test training characteristics parameter and infrared heat picture
Figure test feature parameter；For each preset blood glucose model, using the infrared nondestructive test training characteristics parameter as the blood
The input of sugared model, using the corresponding trained blood glucose value of the infrared nondestructive test training characteristics parameter as the blood glucose model
Output, which is trained；For the blood glucose model of each training, with the infrared nondestructive test test feature
Input of the parameter as the blood glucose model of the training, obtains the blood glucose value to be compared of the output of the blood glucose model of the training, is based on
The corresponding test blood glucose value of infrared nondestructive test test feature parameter and the blood glucose value to be compared, obtaining characterization should
The index value of trained blood glucose model accuracy；The blood glucose model of the corresponding training of the highest index value of accuracy is determined as blood
Sugared prediction model.In this way, blood glucose prediction model is constructed using infrared nondestructive test characteristic parameter, so as to shoot using user
The obtained Infrared Thermogram characteristic parameter of infrared nondestructive test predicted, obtain the blood glucose value of the user, realize noninvasive
Blood glucose measurement avoids bringing body and pain physiologically to diabetic, and, frequently user's bring psychology is given in blood sampling
Problem；Also making the time needed for measuring blood glucose short, avoids consumption disposable apparatus, blood glucose measurement is high-efficient, measurement cost is low,
Moreover, noninvasive blood sugar measuring method, reduces contact of the doctor with patient, the risk for the disease that can avoid infection；In addition, by
In Infrared Thermogram shooting can be carried out at any time, is continuously monitored for blood glucose value and provide technical support；It is possible to further move
Blood glucose prediction model is integrated in dynamic terminal, is shot using mobile terminal, there is preferably intelligence and interactivity, Neng Goushi
Blood glucose value is Anywhere measured now, largely improves user experience；Still further, can be according to different people
Group constructs be suitble to the even different personal blood glucose prediction models of different crowd respectively, meets the different accuracy requirement of all types of user.
Fig. 2 shows the apparatus structure schematic diagrams that blood glucose prediction model is constructed provided by the embodiment of the present invention.Such as Fig. 2 institute
Show, which includes:
User data acquisition module 201, the blood glucose value and the user for obtaining multiple users are corresponding including infrared
The measurement parameter of line thermography characteristic parameter, wherein the blood glucose value includes training blood glucose value and tests blood glucose value, described infrared
Line thermography characteristic parameter includes infrared nondestructive test training characteristics parameter and infrared nondestructive test test feature parameter；
In the embodiment of the present application, as an alternative embodiment, shot using the thermal imaging system of installation in the terminal red
Outside line thermography.
In the embodiment of the present application, as an alternative embodiment, infrared nondestructive test characteristic parameter includes face thermography
Gray value.
In the embodiment of the present application, the gray value of the face thermography is obtained, comprising:
Obtain the face thermography of user；
Noise processed is removed to the face thermography using Wavelet Algorithm, the face thermal imagery after being denoised
Figure；
The gray value of face thermography after extracting the denoising calculates the average value of the gray value of extraction, obtains
The gray value of the face thermography.
Blood glucose model training module 202, for being directed to each preset blood glucose model, by infrared nondestructive test training
Input of the characteristic parameter as the blood glucose model, by the corresponding trained blood glucose of the infrared nondestructive test training characteristics parameter
It is worth the output as the blood glucose model, which is trained；
In the embodiment of the present application, as an alternative embodiment, blood glucose model using neural network algorithm or regression algorithm into
Row training.As another alternative embodiment, blood glucose model can also be trained using curve-fitting method, wherein curve is quasi-
Conjunction includes but is not limited to: Gaussian function fitting, fourier function fitting, trigonometric function fitting etc..
Accuracy comparison module 203 is tested special for being directed to the blood glucose model of each training with the infrared nondestructive test
Input of the parameter as the blood glucose model of the training is levied, the blood glucose value to be compared of the output of the blood glucose model of the training, base are obtained
In the corresponding test blood glucose value of the infrared nondestructive test test feature parameter and the blood glucose value to be compared, characterized
The index value of the blood glucose model accuracy of the training；
Blood glucose prediction model determining module 204, for by the blood glucose model of the corresponding training of the highest index value of accuracy
It is determined as blood glucose prediction model.
In the embodiment of the present application, as an alternative embodiment, measurement parameter can also include: blood pressure, heart rate and body temperature,
It is described using the infrared nondestructive test training characteristics parameter as the input of the blood glucose model, comprising:
Using the infrared nondestructive test training characteristics parameter, blood pressure, heart rate and body temperature as the input of the blood glucose model；
The input of the blood glucose model using the infrared nondestructive test test feature parameter as the training, comprising:
Using the infrared nondestructive test test feature parameter, blood pressure, heart rate and body temperature as the blood glucose model of the training
Input.
In the embodiment of the present application, as an alternative embodiment, the device further include:
Blood glucose value the first prediction module (not shown), for shooting the face thermography to be measured of user to be predicted；
Obtain the target gray value of the face thermography to be measured；
Using the target gray value as the input of the blood glucose prediction model, the blood glucose of the user to be predicted is predicted
Value.
In the embodiment of the present application, as another alternative embodiment, the device further include:
Blood glucose value the second prediction module (not shown), for shooting the face thermography to be measured of user to be predicted；
Obtain the target gray value of the face thermography to be measured and blood pressure, heart rate and the body of the user to be predicted
Temperature；
Using the target gray value, blood pressure, heart rate and body temperature as the input of the blood glucose prediction model, described in prediction
The blood glucose value of user to be predicted.
In the embodiment of the present application, as yet another alternative embodiment, the device further include:
Nformation alert module (not shown), for by prediction the user to be predicted blood glucose value with preset
Blood glucose normality threshold range be compared, comparison result is prompted to the user to be predicted.
It is described to mention comparison result to the user to be predicted as an alternative embodiment in the embodiment of the present application
Show and includes:
If the blood glucose value of the user to be predicted of prediction is greater than the blood glucose in pre-set blood glucose normality threshold range
Upper bound threshold value prompts the user blood glucose to be predicted excessively high, and prompts the excessively high corresponding points for attention of blood glucose and avoid blood glucose mistake
High treatment method；
If the blood glucose value of the user to be predicted of prediction is less than the blood glucose in pre-set blood glucose normality threshold range
Lower bound threshold value prompts the user blood glucose to be predicted too low, and prompts the corresponding points for attention of hypoglycemia and avoid blood glucose mistake
Low treatment method；
If the blood glucose value of the user to be predicted of prediction is within the scope of pre-set blood glucose normality threshold, described in prompt
User blood glucose to be predicted is normal.
As shown in figure 3, one embodiment of the application provides a kind of computer equipment 300, for executing the building blood in Fig. 1
The method of sugared prediction model, the equipment include memory 301, processor 302 and are stored on the memory 301 and can be at this
The computer program run on reason device 302, wherein above-mentioned processor 302 realizes above-mentioned building when executing above-mentioned computer program
The step of method of blood glucose prediction model.
Specifically, above-mentioned memory 301 and processor 302 can be general memory and processor, do not do have here
Body limits, and when the computer program of 302 run memory 301 of processor storage, is able to carry out above-mentioned building blood glucose prediction mould
The method of type.
Corresponding to the method for the building blood glucose prediction model in Fig. 1, the embodiment of the present application also provides a kind of computers can
Storage medium is read, computer program is stored on the computer readable storage medium, when which is run by processor
The step of executing the method for above-mentioned building blood glucose prediction model.
Specifically, which can be general storage medium, such as mobile disk, hard disk, on the storage medium
Computer program when being run, the method for being able to carry out above-mentioned building blood glucose prediction model.
In embodiment provided herein, it should be understood that disclosed device and method, it can be by others side
Formula is realized.The apparatus embodiments described above are merely exemplary, for example, the division of the unit, only one kind are patrolled
Function division is collected, there may be another division manner in actual implementation, in another example, multiple units or components can combine or can
To be integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed is mutual
Coupling, direct-coupling or communication connection can be INDIRECT COUPLING or communication link by some communication interfaces, device or unit
It connects, can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
In addition, each functional unit in embodiment provided by the present application can integrate in one processing unit, it can also
To be that each unit physically exists alone, can also be integrated in one unit with two or more units.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer readable storage medium.Based on this understanding, the technical solution of the application is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or network equipment etc.) execute each embodiment the method for the application all or part of the steps.
And storage medium above-mentioned includes: that USB flash disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), arbitrary access are deposited
The various media that can store program code such as reservoir (Random Access Memory, RAM), magnetic or disk.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, does not then need that it is further defined and explained in subsequent attached drawing, in addition, term " the
One ", " second ", " third " etc. are only used for distinguishing description, are not understood to indicate or imply relative importance.
Finally, it should be noted that embodiment described above, the only specific embodiment of the application, to illustrate the application
Technical solution, rather than its limitations, the protection scope of the application is not limited thereto, although with reference to the foregoing embodiments to this Shen
It please be described in detail, those skilled in the art should understand that: anyone skilled in the art
Within the technical scope of the present application, it can still modify to technical solution documented by previous embodiment or can be light
It is readily conceivable that variation or equivalent replacement of some of the technical features；And these modifications, variation or replacement, do not make
The essence of corresponding technical solution is detached from the spirit and scope of the embodiment of the present application technical solution.The protection in the application should all be covered
Within the scope of.Therefore, the protection scope of the application shall be subject to the protection scope of the claim.
Claims (10)
1. a kind of method for constructing blood glucose prediction model characterized by comprising
The blood glucose value and the corresponding measurement parameter including infrared nondestructive test characteristic parameter of the user of multiple users are obtained,
In, the blood glucose value includes training blood glucose value and test blood glucose value, and the infrared nondestructive test characteristic parameter includes infrared heat
As figure training characteristics parameter and infrared nondestructive test test feature parameter；
For each preset blood glucose model, using the infrared nondestructive test training characteristics parameter as the defeated of the blood glucose model
Enter, it is right using the corresponding trained blood glucose value of the infrared nondestructive test training characteristics parameter as the output of the blood glucose model
The blood glucose model is trained；
For the blood glucose model of each training, using the infrared nondestructive test test feature parameter as the blood glucose model of the training
Input, obtain the blood glucose value to be compared of the output of the blood glucose model of the training, be based on the infrared nondestructive test test feature
The corresponding test blood glucose value of parameter and the blood glucose value to be compared, obtain the finger for the blood glucose model accuracy for characterizing the training
Scale value；
The blood glucose model of the corresponding training of the highest index value of accuracy is determined as blood glucose prediction model.
2. the method according to claim 1, wherein the measurement parameter further include: blood pressure, heart rate and body
Temperature,
It is described using the infrared nondestructive test training characteristics parameter as the input of the blood glucose model, comprising:
Using the infrared nondestructive test training characteristics parameter, blood pressure, heart rate and body temperature as the input of the blood glucose model；
The input of the blood glucose model using the infrared nondestructive test test feature parameter as the training, comprising:
Using the infrared nondestructive test test feature parameter, blood pressure, heart rate and body temperature as the defeated of the blood glucose model of the training
Enter.
3. the method according to claim 1, wherein the blood glucose model is using neural network algorithm or returns calculation
Method is trained.
4. the method according to claim 1, wherein the infrared nondestructive test characteristic parameter includes face thermal imagery
The gray value of figure.
5. according to the method described in claim 4, it is characterized in that, obtaining the gray value of the face thermography, comprising:
Obtain the face thermography of user；
Noise processed is removed to the face thermography using Wavelet Algorithm, the face thermography after being denoised；
The gray value of face thermography after extracting the denoising calculates the average value of the gray value of extraction, obtains described
The gray value of face thermography.
6. the method according to claim 1, wherein the method also includes:
Shoot the face thermography to be measured of user to be predicted；
Obtain the target gray value of the face thermography to be measured；
Using the target gray value as the input of the blood glucose prediction model, the blood glucose value of the user to be predicted is predicted.
7. the method according to claim 1, wherein the method also includes:
Shoot the face thermography to be measured of user to be predicted；
Obtain the target gray value of the face thermography to be measured and blood pressure, heart rate and the body temperature of the user to be predicted；
Using the target gray value, blood pressure, heart rate and body temperature as the input of the blood glucose prediction model, prediction is described to pre-
Survey the blood glucose value of user.
8. method according to claim 6 or 7, which is characterized in that the method also includes:
The blood glucose value of the user to be predicted of prediction is compared with pre-set blood glucose normality threshold range, will be compared
As a result it is prompted to the user to be predicted.
9. according to the method described in claim 8, it is characterized in that, described mention comparison result to the user to be predicted
Show and includes:
If the blood glucose value of the user to be predicted of prediction is greater than the blood glucose upper bound in pre-set blood glucose normality threshold range
Threshold value prompts the user blood glucose to be predicted excessively high, and prompts the excessively high corresponding points for attention of blood glucose and avoid blood glucose excessively high
Treatment method；
If the blood glucose value of the user to be predicted of prediction is less than the blood glucose lower bound in pre-set blood glucose normality threshold range
Threshold value prompts the user blood glucose to be predicted too low, and prompts the corresponding points for attention of hypoglycemia and avoid hypoglycemic
Treatment method；
If the blood glucose value of the user to be predicted of prediction within the scope of pre-set blood glucose normality threshold, prompts described to pre-
It is normal to survey user blood glucose.
10. a kind of device for constructing blood glucose prediction model characterized by comprising
User data acquisition module, the blood glucose value and the user for obtaining multiple users are corresponding including infrared nondestructive test
The measurement parameter of characteristic parameter, wherein the blood glucose value includes training blood glucose value and test blood glucose value, the infrared nondestructive test
Characteristic parameter includes infrared nondestructive test training characteristics parameter and infrared nondestructive test test feature parameter；
Blood glucose model training module joins the infrared nondestructive test training characteristics for being directed to each preset blood glucose model
Inputs of the number as the blood glucose model, using the corresponding trained blood glucose value of the infrared nondestructive test training characteristics parameter as
The output of the blood glucose model is trained the blood glucose model；
Accuracy comparison module, for being directed to the blood glucose model of each training, with the infrared nondestructive test test feature parameter
The input of blood glucose model as the training obtains the blood glucose value to be compared of the output of the blood glucose model of the training, based on described
The corresponding test blood glucose value of infrared nondestructive test test feature parameter and the blood glucose value to be compared, obtain characterizing the training
Blood glucose model accuracy index value；
Blood glucose prediction model determining module, for the blood glucose model of the corresponding training of the highest index value of accuracy to be determined as blood
Sugared prediction model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910332868.4A CN109872821A (en) | 2019-04-24 | 2019-04-24 | A kind of method and device constructing blood glucose prediction model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910332868.4A CN109872821A (en) | 2019-04-24 | 2019-04-24 | A kind of method and device constructing blood glucose prediction model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109872821A true CN109872821A (en) | 2019-06-11 |
Family
ID=66923004
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910332868.4A Pending CN109872821A (en) | 2019-04-24 | 2019-04-24 | A kind of method and device constructing blood glucose prediction model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109872821A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111383758A (en) * | 2020-03-06 | 2020-07-07 | 三七二二(北京)健康咨询有限公司 | Method and device for predicting postprandial blood glucose based on multidimensional data |
| CN112133442A (en) * | 2020-09-22 | 2020-12-25 | 博邦芳舟医疗科技（北京）有限公司 | Continuous non-invasive blood glucose detection device and method |
| CN112635054A (en) * | 2020-11-30 | 2021-04-09 | 新绎健康科技有限公司 | System and method for predicting target blood glucose value based on pulse map parameters |
| CN113130078A (en) * | 2021-05-11 | 2021-07-16 | 首都医科大学附属北京天坛医院 | Method, device and equipment for predicting intracranial aneurysm occlusion |
| CN113470818A (en) * | 2021-07-08 | 2021-10-01 | 建信金融科技有限责任公司 | Disease prediction method, device, system, electronic device and computer readable medium |
| CN113576475A (en) * | 2021-08-02 | 2021-11-02 | 浙江师范大学 | Non-contact blood glucose measurement method based on deep learning |
| CN113855007A (en) * | 2021-08-27 | 2021-12-31 | 联卫医疗科技(上海)有限公司 | Method and device for obtaining machine learning model samples for blood glucose prediction |
| CN114420301A (en) * | 2022-01-28 | 2022-04-29 | 广东工业大学 | Method, system and storage medium for predicting blood glucose based on segmented domain RF modeling |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160097716A1 (en) * | 2014-09-29 | 2016-04-07 | Zyomed Corp. | Systems and methods for blood glucose and other analyte detection and measurement using collision computing |
| CN106980746A (en) * | 2016-12-16 | 2017-07-25 | 清华大学 | A kind of general Woundless blood sugar Forecasting Methodology based on Time-Series analysis |
| CN107192690A (en) * | 2017-05-19 | 2017-09-22 | 重庆大学 | Near infrared spectrum Noninvasive Blood Glucose Detection Methods and its detection network model training method |
| CN107203700A (en) * | 2017-07-14 | 2017-09-26 | 清华-伯克利深圳学院筹备办公室 | A kind of method and device monitored based on continuous blood sugar |
-
2019
- 2019-04-24 CN CN201910332868.4A patent/CN109872821A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160097716A1 (en) * | 2014-09-29 | 2016-04-07 | Zyomed Corp. | Systems and methods for blood glucose and other analyte detection and measurement using collision computing |
| CN106980746A (en) * | 2016-12-16 | 2017-07-25 | 清华大学 | A kind of general Woundless blood sugar Forecasting Methodology based on Time-Series analysis |
| CN107192690A (en) * | 2017-05-19 | 2017-09-22 | 重庆大学 | Near infrared spectrum Noninvasive Blood Glucose Detection Methods and its detection network model training method |
| CN107203700A (en) * | 2017-07-14 | 2017-09-26 | 清华-伯克利深圳学院筹备办公室 | A kind of method and device monitored based on continuous blood sugar |
Non-Patent Citations (2)
| Title |
|---|
| A.SELVARANI1 ET AL: "Infrared Thermal Imaging for Diabetes Detection and Measurement", 《SPRINGER》 * |
| 代娟: "近红外光谱无创血糖检测模型研究", 《中国优秀硕士学位论文全文数据库医药卫生科技辑》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111383758A (en) * | 2020-03-06 | 2020-07-07 | 三七二二(北京)健康咨询有限公司 | Method and device for predicting postprandial blood glucose based on multidimensional data |
| CN112133442A (en) * | 2020-09-22 | 2020-12-25 | 博邦芳舟医疗科技（北京）有限公司 | Continuous non-invasive blood glucose detection device and method |
| WO2022063048A1 (en) * | 2020-09-22 | 2022-03-31 | 博邦芳舟医疗科技（北京）有限公司 | Continuous non-invasive blood glucose measurement device and method |
| CN112133442B (en) * | 2020-09-22 | 2024-02-13 | 博邦芳舟医疗科技（北京）有限公司 | Continuous noninvasive blood glucose detection device and method |
| CN112635054A (en) * | 2020-11-30 | 2021-04-09 | 新绎健康科技有限公司 | System and method for predicting target blood glucose value based on pulse map parameters |
| CN113130078A (en) * | 2021-05-11 | 2021-07-16 | 首都医科大学附属北京天坛医院 | Method, device and equipment for predicting intracranial aneurysm occlusion |
| CN113470818A (en) * | 2021-07-08 | 2021-10-01 | 建信金融科技有限责任公司 | Disease prediction method, device, system, electronic device and computer readable medium |
| CN113576475A (en) * | 2021-08-02 | 2021-11-02 | 浙江师范大学 | Non-contact blood glucose measurement method based on deep learning |
| CN113576475B (en) * | 2021-08-02 | 2023-04-21 | 浙江师范大学 | Deep learning-based contactless blood glucose measurement method |
| CN113855007A (en) * | 2021-08-27 | 2021-12-31 | 联卫医疗科技(上海)有限公司 | Method and device for obtaining machine learning model samples for blood glucose prediction |
| CN114420301A (en) * | 2022-01-28 | 2022-04-29 | 广东工业大学 | Method, system and storage medium for predicting blood glucose based on segmented domain RF modeling |
| CN114420301B (en) * | 2022-01-28 | 2022-08-05 | 广东工业大学 | Method, system and storage medium for predicting blood glucose based on segmented domain RF modeling |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109872821A (en) | A kind of method and device constructing blood glucose prediction model | |
| CN107981858B (en) | Automatic electrocardiogram heart beat identification and classification method based on artificial intelligence | |
| TWI709852B (en) | System and method for anomaly detection via a multi-prediction-model architecture | |
| CN104665768B (en) | The monitoring of physiological parameter | |
| CN109119130A (en) | A kind of big data based on cloud computing is health management system arranged and method | |
| CN111870237B (en) | Blood pressure detection device, blood pressure detection apparatus, and blood pressure detection medium | |
| CN106659392A (en) | Unobtrusive skin tissue hydration determining device and related method | |
| CN109452935A (en) | The non-invasive methods and system from photoplethysmogram estimated blood pressure are post-processed using statistics | |
| Raja et al. | An automatic detection of blood vessel in retinal images using convolution neural network for diabetic retinopathy detection | |
| Jovanović et al. | A mobile crowd sensing application for hypertensive patients | |
| JP2019091498A (en) | System and method for correcting answers | |
| CN111126350B (en) | Method and device for generating heart beat classification result | |
| US11823385B2 (en) | Processing fundus images using machine learning models to generate blood-related predictions | |
| CN110141249A (en) | Woundless blood sugar monitoring method, system, equipment and medium based on PPG signal | |
| Haque et al. | A novel technique for non-invasive measurement of human blood component levels from fingertip video using DNN based models | |
| CN112382384A (en) | Training method and diagnosis system for Turner syndrome diagnosis model and related equipment | |
| CN114098724A (en) | Blood glucose prediction method and device based on optical signal characteristics and metabolic heat characteristics | |
| US11484199B2 (en) | System and method for predicting a blood glucose level of a user | |
| CN115512836A (en) | Wearable intelligent health management system based on embedded AI | |
| Brophy et al. | A machine vision approach to human activity recognition using photoplethysmograph sensor data | |
| Samyoun et al. | Stress detection via sensor translation | |
| Arul Kumar et al. | RETRACTED ARTICLE: Application of back propagation artificial neural network in detection and analysis of diabetes mellitus | |
| Prudêncio et al. | Physical activity recognition from smartphone embedded sensors | |
| CN116327133A (en) | Multi-physiological index detection method, device and related equipment | |
| Fazlur et al. | Integrated Deep Learning Model for Heart Disease Prediction Using Variant Medical Data Sets. |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190611 |
|
| RJ01 | Rejection of invention patent application after publication |