CN109935331A - A kind of blood glucose prediction method and system based on multi-model dynamic comprehensive - Google Patents
A kind of blood glucose prediction method and system based on multi-model dynamic comprehensive Download PDFInfo
- Publication number
- CN109935331A CN109935331A CN201910260148.1A CN201910260148A CN109935331A CN 109935331 A CN109935331 A CN 109935331A CN 201910260148 A CN201910260148 A CN 201910260148A CN 109935331 A CN109935331 A CN 109935331A
- Authority
- CN
- China
- Prior art keywords
- model
- physiological data
- blood glucose
- target
- blood sugar
- 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.)
- Granted
Links
- 239000008280 blood Substances 0.000 title claims abstract description 256
- 210000004369 blood Anatomy 0.000 title claims abstract description 254
- 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 130
- 239000008103 glucose Substances 0.000 title claims abstract description 130
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000009826 distribution Methods 0.000 claims description 43
- NOESYZHRGYRDHS-UHFFFAOYSA-N insulin Chemical compound N1C(=O)C(NC(=O)C(CCC(N)=O)NC(=O)C(CCC(O)=O)NC(=O)C(C(C)C)NC(=O)C(NC(=O)CN)C(C)CC)CSSCC(C(NC(CO)C(=O)NC(CC(C)C)C(=O)NC(CC=2C=CC(O)=CC=2)C(=O)NC(CCC(N)=O)C(=O)NC(CC(C)C)C(=O)NC(CCC(O)=O)C(=O)NC(CC(N)=O)C(=O)NC(CC=2C=CC(O)=CC=2)C(=O)NC(CSSCC(NC(=O)C(C(C)C)NC(=O)C(CC(C)C)NC(=O)C(CC=2C=CC(O)=CC=2)NC(=O)C(CC(C)C)NC(=O)C(C)NC(=O)C(CCC(O)=O)NC(=O)C(C(C)C)NC(=O)C(CC(C)C)NC(=O)C(CC=2NC=NC=2)NC(=O)C(CO)NC(=O)CNC2=O)C(=O)NCC(=O)NC(CCC(O)=O)C(=O)NC(CCCNC(N)=N)C(=O)NCC(=O)NC(CC=3C=CC=CC=3)C(=O)NC(CC=3C=CC=CC=3)C(=O)NC(CC=3C=CC(O)=CC=3)C(=O)NC(C(C)O)C(=O)N3C(CCC3)C(=O)NC(CCCCN)C(=O)NC(C)C(O)=O)C(=O)NC(CC(N)=O)C(O)=O)=O)NC(=O)C(C(C)CC)NC(=O)C(CO)NC(=O)C(C(C)O)NC(=O)C1CSSCC2NC(=O)C(CC(C)C)NC(=O)C(NC(=O)C(CCC(N)=O)NC(=O)C(CC(N)=O)NC(=O)C(NC(=O)C(N)CC=1C=CC=CC=1)C(C)C)CC1=CN=CN1 NOESYZHRGYRDHS-UHFFFAOYSA-N 0.000 claims description 43
- 102000004877 Insulin Human genes 0.000 claims description 21
- 108090001061 Insulin Proteins 0.000 claims description 21
- 230000015572 biosynthetic process Effects 0.000 claims description 21
- 229940125396 insulin Drugs 0.000 claims description 21
- 238000003786 synthesis reaction Methods 0.000 claims description 21
- 230000002218 hypoglycaemic effect Effects 0.000 claims description 15
- 238000001802 infusion Methods 0.000 claims description 15
- 208000013016 Hypoglycemia Diseases 0.000 claims description 14
- 238000004891 communication Methods 0.000 claims description 12
- 230000035790 physiological processes and functions Effects 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 9
- 238000003860 storage Methods 0.000 claims description 9
- 230000037396 body weight Effects 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 claims description 2
- 238000012360 testing method Methods 0.000 claims description 2
- 230000008859 change Effects 0.000 description 14
- 206010012601 diabetes mellitus Diseases 0.000 description 9
- 238000010586 diagram Methods 0.000 description 9
- 206010022489 Insulin Resistance Diseases 0.000 description 7
- 230000009230 endogenous glucose production Effects 0.000 description 5
- 230000006377 glucose transport Effects 0.000 description 4
- 150000001720 carbohydrates Chemical class 0.000 description 3
- 235000014633 carbohydrates Nutrition 0.000 description 3
- 230000037406 food intake Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- WQZGKKKJIJFFOK-VFUOTHLCSA-N beta-D-glucose Chemical compound OC[C@H]1O[C@@H](O)[C@H](O)[C@@H](O)[C@@H]1O WQZGKKKJIJFFOK-VFUOTHLCSA-N 0.000 description 2
- 201000001421 hyperglycemia Diseases 0.000 description 2
- 230000010355 oscillation Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 206010067584 Type 1 diabetes mellitus Diseases 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 230000001154 acute effect Effects 0.000 description 1
- 230000010455 autoregulation Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000001684 chronic effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000029087 digestion Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000004907 flux Effects 0.000 description 1
- 229940088597 hormone Drugs 0.000 description 1
- 239000005556 hormone Substances 0.000 description 1
- 230000003914 insulin secretion Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 208000030159 metabolic disease Diseases 0.000 description 1
- 230000002503 metabolic effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000008288 physiological mechanism Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 208000011580 syndromic disease Diseases 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
- 208000001072 type 2 diabetes mellitus Diseases 0.000 description 1
Landscapes
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Investigating Or Analysing Biological Materials (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The embodiment of the present invention provides a kind of blood glucose prediction method and system based on multi-model dynamic comprehensive, this method comprises: the several groups desired physiological data of acquisition target object；Construct the Comprehensive Model being made of several blood glucose dynamical foundation models；According to several groups desired physiological data and bayesian algorithm, the Comprehensive Model is updated；The blood glucose of the target object is predicted using updated Comprehensive Model.The embodiment of the present invention comprehensively considers the uncertainty of blood glucose basic model and the uncertainty of model parameter, the multiple blood glucose basic models of dynamic comprehensive, and constantly the model of real-time update and its physiological parameter information are established in Comprehensive Model according to bayesian algorithm, so that Comprehensive Model dynamic updates tracking blood glucose, to guarantee that Comprehensive Model has more steady, more accurate predictive ability.
Description
Technical Field
The embodiment of the invention relates to the technical field of medical treatment, in particular to a blood sugar prediction method and system based on multi-model dynamic synthesis.
Background
Diabetes is a metabolic disorder syndrome mainly manifested by fasting hyperglycemia or postprandial hyperglycemia due to absolute or relative insufficiency of insulin secretion, and is very likely to cause various acute and chronic complications of the whole body. Insulin is the only hypoglycemic hormone in the body, and insulin infusion is the main treatment method for maintaining normal blood sugar level of the patients with type I diabetes and severe type II diabetes at present.
In clinic, diabetics have poor ability to regulate blood sugar, and are often at risk of hypoglycemia due to excessive insulin, especially fragile diabetics. Hypoglycemia can result in serious consequences and even death. Particularly, at night, the reaction of the patient to the blood sugar concentration reduction is weakened, the early warning symptoms are reduced, and the safety of the patient is seriously threatened.
Active human intervention (e.g., ingestion of exogenous carbohydrates) is an effective method to rapidly increase the blood glucose level of a patient. However, ingestion of exogenous carbohydrates requires a period of digestion and absorption to effectively raise the blood glucose level in humans. Therefore, accurate prediction of the occurrence of hypoglycemia and ingestion of appropriate amounts of exogenous carbohydrates in an effective time before occurrence are important means to prevent the occurrence of hypoglycemia.
The key to realize accurate prediction of blood sugar is to establish an accurate, effective and reliable blood sugar prediction model. In recent years, there are various insulin-blood glucose models of varying degrees of complexity in the prior art that describe changes in blood glucose. However, the metabolic regulation of glucose has complex uncertainty, and no effective and accurate prediction model is applied clinically, which is specifically represented by model uncertainty, inter-individual difference and parameter uncertainty caused by different physiological states due to incomplete research on physiological mechanisms.
The models show different fitting and predicting capabilities for the blood glucose data of different patients at different moments, one model of a certain group of blood glucose data of the same patient has the highest predicting precision, the other model of the blood glucose data of the same patient has better fitting and predicting effects, and no model is obviously superior to other models.
On the other hand, and for the same set of blood glucose data, the prediction capabilities of the multiple insulin-blood glucose models are similar. The choice of model has uncertainty. The traditional method for predicting blood sugar based on fitting accuracy or information criteria selects a fixed model and has lower robustness and higher risk.
For one, observing small changes in blood glucose data may result in the selection of different models, resulting in large changes in blood glucose estimates.
Second, estimates based on the selected model may lose information contained in other models, e.g., fast time scale models may describe oscillations in minutes, while long-term models tend to capture slower speed dynamics, often ignoring faster oscillation dynamics.
Therefore, the dynamic change rule of blood sugar of different patients and different physiological states is difficult to accurately describe by a traditional fixed parameter model due to strong individual difference and variable physiological states.
Disclosure of Invention
In order to solve the above problems, embodiments of the present invention provide a blood glucose prediction method based on multi-model dynamic synthesis.
In a first aspect, an embodiment of the present invention provides a blood glucose prediction method based on multi-model dynamic synthesis, including:
acquiring a plurality of groups of target physiological data of a target object, wherein each group of target physiological data comprises one or more of blood sugar, insulin infusion amount and body weight;
acquiring a comprehensive prediction model, wherein the comprehensive prediction model is composed of a plurality of blood glucose basic models representing blood glucose changes and model probabilities corresponding to each blood glucose basic model, the model probabilities represent the probabilities that target physiological data are output by the blood glucose basic models, each blood glucose basic model comprises a plurality of model parameters variable along with physiological states, and model structures and model parameters contained in any two blood glucose basic models are not completely the same;
updating the comprehensive prediction model according to a plurality of groups of target physiological data and a Bayesian algorithm;
and predicting the blood sugar of the target object by using the updated comprehensive prediction model.
In a second aspect, an embodiment of the present invention provides a blood glucose prediction system based on multi-model dynamic synthesis, including:
the system comprises an acquisition module, a data processing module and a data processing module, wherein the acquisition module is used for acquiring a plurality of groups of target physiological data of a target object, and each group of target physiological data comprises one or more of blood sugar, insulin infusion amount and weight;
the model comprehensive module is used for obtaining a comprehensive prediction model, the comprehensive prediction model is composed of a plurality of blood sugar basic models representing blood sugar changes and model probabilities corresponding to each blood sugar basic model, the model probabilities represent the probabilities of target physiological data output by the blood sugar basic models, each blood sugar basic model comprises a plurality of model parameters variable along with physiological states, and model structures and model parameters contained in any two blood sugar basic models are not completely the same;
the updating module is used for updating the comprehensive prediction model according to a plurality of groups of target physiological data and a Bayesian algorithm;
and the prediction module is used for predicting the blood sugar of the target object by using the updated comprehensive prediction model.
In a third aspect, an embodiment of the present invention provides an electronic device, including:
at least one processor, at least one memory, a communication interface, and a bus; wherein,
the processor, the memory and the communication interface complete mutual communication through the bus;
the communication interface is used for information transmission between the test equipment and the communication equipment of the display device;
the memory stores program instructions executable by the processor, and the processor calls the program instructions to execute a blood glucose prediction method based on multi-model dynamic synthesis provided by the first aspect.
In a fourth aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium, which stores computer instructions, where the computer instructions cause the computer to execute the method for predicting blood glucose based on multi-model dynamic synthesis provided in the first aspect.
According to the blood sugar prediction method and system based on multi-model dynamic synthesis provided by the embodiment of the invention, the uncertainty of a blood sugar basic model and the uncertainty of model parameters are comprehensively considered, and the physiological parameter information updated in real time is continuously established in the comprehensive prediction model, so that the comprehensive prediction model can dynamically update and track blood sugar, and the change rule of the blood sugar is more accurately described and predicted. In addition, physiological parameter information contained in different blood sugar basic models is different, a plurality of blood sugar basic models describing the dynamic change rule of blood sugar are comprehensively considered, the risk of information loss caused by a single blood sugar basic model is reduced, and the comprehensive prediction model is ensured to have more stable and accurate prediction capability.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
FIG. 1 is a flowchart of a blood glucose prediction method based on multi-model dynamic synthesis according to an embodiment of the present invention;
fig. 2 is a schematic diagram of an analysis result of a blood glucose prediction method based on multi-model dynamic synthesis according to an embodiment of the present invention;
FIG. 3 is a schematic structural diagram of a blood glucose prediction system based on multi-model dynamic synthesis according to an embodiment of the present invention;
fig. 4 illustrates a physical structure diagram of an electronic device.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a flowchart of a blood glucose prediction method based on multi-model dynamic synthesis according to an embodiment of the present invention, as shown in fig. 1, the method includes:
s1, collecting a plurality of groups of target physiological data of the target object, wherein each group of target physiological data comprises one or more of blood sugar, insulin infusion amount and body weight;
s2, obtaining a comprehensive prediction model, wherein the comprehensive prediction model is composed of a plurality of blood glucose basic models representing blood glucose changes and model probabilities corresponding to each blood glucose basic model, the model probabilities represent the probabilities of target physiological data output by the blood glucose basic models, each blood glucose basic model comprises a plurality of model parameters variable along with physiological states, and the model structures and the model parameters contained in any two blood glucose basic models are not completely the same;
s3, updating the comprehensive prediction model according to a plurality of groups of target physiological data and a Bayesian algorithm;
and S4, predicting the blood sugar of the target object by using the updated comprehensive prediction model.
In the embodiment of the invention, the diabetes patient is taken as a target object, and the target physiological data of the diabetes patient is collected to predict the blood sugar of the diabetes patient.
In an embodiment of the invention, the target physiological data comprises blood glucose, insulin infusion amount and body weight of the diabetic patient. The blood sugar of the diabetic can be obtained by a blood sugar monitoring system, and real-time data of the diabetic is obtained at preset intervals.
The insulin infusion amount can be obtained by the rate of infusion of insulin and the time of infusion of insulin.
And then establishing a comprehensive prediction model, wherein the comprehensive prediction model is composed of a plurality of blood sugar basic models representing blood sugar changes and model probabilities corresponding to each blood sugar basic model, for any blood sugar basic model, the blood sugar basic model comprises a plurality of model parameters variable along with physiological states, and the model parameters can reflect the dynamic change rule of blood sugar. In addition, model structures and model parameters contained in any two blood sugar basic models are not completely the same, so that the comprehensive prediction model comprises blood sugar basic models with different complexity degrees, the model structures and model parameters contained in each blood sugar basic model are also different, and the description energy and prediction capability of each blood sugar basic model on blood sugar are also different.
Each blood sugar basic prediction model corresponds to a model probability, and the sum of the model probabilities corresponding to all the blood sugar basic prediction models is 1. Taking any group of target physiological data as an example, the model probability represents the possibility that the corresponding blood glucose basic model outputs the group of target physiological data, and the sum of the probabilities of all blood glucose basic models outputting the group of target physiological data is 1.
The comprehensive prediction model fully considers the uncertainty of the single blood sugar basic model and the uncertainty of the model parameters corresponding to the single blood sugar basic model by integrating the plurality of blood sugar basic models, can comprehensively reflect the change rule of blood sugar by integrating the plurality of blood sugar basic models, and can predict the change rule of blood sugar more accurately. The risk of information loss caused by a single blood sugar basic model is reduced, and the comprehensive prediction model is guaranteed to have more stable and accurate prediction capability.
And then, according to a plurality of groups of target physiological data collected in real time, reflecting the target physiological data updated in real time in the comprehensive prediction model by using a Bayesian algorithm, updating the value of the model probability in the comprehensive prediction model and the value of the model parameter contained in each blood sugar basic model, so that the updated comprehensive prediction model has more dynamic property and real-time property, and has more accurate prediction result when predicting the blood sugar of the target object by using the updated comprehensive prediction model.
According to the blood sugar prediction method based on multi-model dynamic synthesis provided by the embodiment of the invention, a plurality of blood sugar basic models are synthesized, the uncertainty of a single blood sugar basic model and the uncertainty of model parameters corresponding to the single blood sugar basic model are fully considered, and the change rule of blood sugar can be comprehensively reflected and more accurately predicted by synthesizing the plurality of blood sugar basic models. The risk of information loss caused by a single blood sugar basic model is reduced, and the comprehensive prediction model is guaranteed to have more stable and accurate prediction capability.
In addition, physiological parameter information contained in different blood sugar basic models is different, a plurality of blood sugar basic models describing the dynamic change rule of blood sugar are comprehensively considered, the risk of information loss caused by a single blood sugar basic model is reduced, and the comprehensive prediction model is ensured to have more stable and accurate prediction capability.
On the basis of the foregoing embodiment, preferably, the updating the comprehensive prediction model according to a plurality of sets of target physiological data and a bayesian algorithm specifically includes:
for current target physiological data, acquiring a likelihood function of each blood sugar basic model corresponding to the current target physiological data according to the current target physiological data and each blood sugar basic model;
acquiring a comprehensive prediction model corresponding to the current target physiological data according to the posterior information of the model probability corresponding to the last group of target physiological data of the current target physiological data and the posterior information of the model parameters corresponding to the last group of data of the current target physiological data;
according to the comprehensive prediction model corresponding to the current target physiological data and the likelihood function corresponding to the current target physiological data, obtaining posterior information of model probability corresponding to the current target physiological data and posterior information of model parameters corresponding to the current target physiological data, and taking the next group of target physiological data of the current target physiological data as the current target physiological data again;
and repeating the steps until the last group of target physiological data is taken as the current target physiological data, and updating the comprehensive prediction model by utilizing the posterior distribution information of the model probability corresponding to the last group of target physiological data and the posterior distribution information of the model parameters corresponding to the last group of target physiological data.
Specifically, updating the comprehensive prediction model by using the bayesian algorithm can be realized by the following steps:
before using the bayesian algorithm, it is necessary to determine an initial value of the comprehensive prediction model, that is, prior distribution information of model probability corresponding to each blood glucose basic model and prior distribution information of model parameters corresponding to each blood glucose basic model.
In the embodiment of the invention, if N blood sugar basic models exist, the prior distribution information of the model probability corresponding to each blood sugar basic model is 1/N.
The model parameters in each blood glucose base model obey a log-normal distribution, and the expectation and variance in the log-normal distribution are obtained by moment estimation.
In the calculation process, for a first group of target physiological data, the prior distribution information of the model probability corresponding to each blood glucose basic model is used as the posterior information of a last group of target physiological data of the first group of target physiological data, and the prior distribution information of the model parameter corresponding to each blood glucose basic model is used as the posterior information of the last group of target physiological data of the first group of target physiological data.
For the current target physiological data, the likelihood function corresponding to each blood sugar basic model under the condition of the current target physiological data can be determined by using the current target physiological data and each blood sugar basic model.
And obtaining posterior information of model probability corresponding to the previous group of target physiological data and posterior information of model parameters corresponding to the previous group of target physiological data, and obtaining a comprehensive prediction model corresponding to the current target physiological data.
And then, according to the comprehensive prediction model corresponding to the current target physiological data and the likelihood function corresponding to the current target physiological data, obtaining posterior information of model parameters corresponding to the current target physiological data and posterior information of model probability corresponding to the current target physiological data, and taking the next group of target physiological data of the current target physiological data as the current target physiological data again.
And repeating the steps until the last group of target physiological data is taken as the current target physiological data, and updating the comprehensive prediction model by using the obtained posterior distribution information of the model probability corresponding to the last group of target physiological data and the posterior distribution information of the model parameter corresponding to the last group of target physiological data.
According to the embodiment of the invention, the comprehensive prediction model is updated by utilizing the target physiological data acquired in real time, so that the relevant parameters in the updated comprehensive prediction model are more consistent with the blood sugar condition of the target object under the current condition, therefore, the updated comprehensive prediction model has more accurate blood sugar prediction capability, and the predicted blood sugar value is more accurate.
On the basis of the foregoing embodiment, preferably, the updating the comprehensive prediction model according to several sets of target physiological data and a bayesian algorithm further includes:
the prior distribution information of the model parameters of each blood glucose basic model is lognormal distribution, and the expectation and the variance of the lognormal distribution are obtained through moment estimation.
When the blood sugar of the diabetic patient at a target time in the future is to be predicted, on the basis of the foregoing embodiment, preferably, the predicting the blood sugar of the target subject by using the updated comprehensive prediction model specifically includes:
obtaining expectation of model parameters in each blood sugar basic model according to posterior distribution information of the model parameters in each blood sugar basic model;
obtaining the expected posterior probability of each blood sugar basic model according to the posterior distribution information of the probability of each blood sugar basic model;
acquiring the comprehensive prediction model corresponding to the target time point according to the expectation of the model parameters in each blood sugar basic model and the expectation of the posterior probability of each blood sugar basic model;
and predicting the blood sugar of the target object at the target time point by using the comprehensive prediction model corresponding to the target time point.
Firstly, posterior distribution information of model parameters in each blood glucose basic model is obtained, the model parameters are a distribution condition rather than a fixed value, in order to predict the blood glucose of a target object at a target time point, expectation of the model parameters in each blood glucose basic model is used as an estimated value of the model parameters of the comprehensive prediction model at the target time point, and a comprehensive prediction model corresponding to the target time point is obtained by utilizing the estimated value.
And then predicting the blood sugar of the target object by using the comprehensive prediction model corresponding to the target time point.
When the blood glucose condition of the target patient within a target time interval in the future is to be predicted, on the basis of the above embodiment, preferably, the predicting the blood glucose of the target subject by using the updated comprehensive prediction model further includes:
acquiring the distribution information of the physiological parameters in each blood glucose basic model in a preset time interval according to the posterior distribution information of the physiological parameters in each blood glucose basic model;
acquiring a comprehensive prediction model corresponding to a target time interval according to the distribution information of the physiological parameters in each blood glucose basic model in the preset time interval and the posterior probability of each blood glucose basic model;
and predicting the blood sugar distribution of the target object in the target time interval according to the comprehensive prediction model corresponding to the target time interval.
And taking a small section of the posterior distribution of the physiological parameters in each blood glucose basic model as a preset time interval, taking the posterior distribution of the physiological parameters in the preset time interval as the distribution condition of the physiological parameters in each blood glucose basic model in the target time interval, thereby obtaining a corresponding comprehensive prediction model in the target time interval, and predicting the blood glucose distribution condition of the target object in the target time interval by using the corresponding comprehensive prediction model in the target time interval.
On the basis of the above embodiment, it is preferable to further include:
and judging the probability of hypoglycemia of the target object in the target time interval according to the blood sugar distribution of the target object in the target time interval and a preset threshold value.
In order to obtain the probability of hypoglycemia of the target object in the target time interval, a hypoglycemia threshold value needs to be set, and the probability of hypoglycemia can be obtained by comparing the blood sugar distribution of the target object in the target time interval with the hypoglycemia threshold value.
In order to more clearly describe the blood glucose prediction method based on multi-model dynamic synthesis provided by the embodiment of the invention, the embodiment of the invention provides the following embodiments:
firstly, target physiological data of a diabetes patient in 5 hours at night is collected once every 15 minutes, and therefore multiple groups of target physiological data can be obtained.
Obtaining a comprehensive prediction model, wherein the number of the blood glucose basic models in the embodiment is 2, and the blood glucose basic models are respectively a Ruan model and a Hovorka two-chamber model, wherein a blood glucose expression of the Ruan model is as follows:
wherein G is1(t) denotes the blood glucose concentration at time t (mmol/l) in the Ruan model, U is the insulin infusion rate (U/h), Si is the insulin sensitivity (mmol/l/min per mU/l), xbIs the basal insulin concentration (mU/l), GbFor basal glucose levels (mmol/l), K is the rate of glucose autoregulation (/ min), C is a constant and is related to the initial value of blood glucose, and W represents body weight.
Thus, model parameters in the Ruan model include glucose self-regulation rate, insulin sensitivity, basal insulin concentration, and basal glucose level.
The model parameter vector in the Ruan model is recorded as theta1＝[Gb,Si,xb,K]。
The Hovorka two-chamber model can be expressed as:
wherein G is2(t) is the blood glucose concentration at time t (mmol/l) in the Hovorka model, U is the insulin infusion rate (U/h), k12The blood glucose transport Rate constant (/ min), EGP0Is the endogenous glucose production constant, F01Is independent of insulin blood glucose consumption rate (mmol/l), k1、k2And k3Insulin sensitivity affecting glucose transport, insulin sensitivity affecting glucose consumption and insulin sensitivity affecting endogenous glucose production are indicated, respectively.
The Hovorka two-chamber model parameters are: non-insulin dependent glucose flux, insulin sensitivity affecting glucose transport, insulin sensitivity affecting glucose consumption, insulin sensitivity affecting endogenous glucose production, a blood glucose transport rate constant, and an endogenous glucose production constant.
The model parameter vector of the Hovorka model is recorded as theta2＝[k12,k1,k2,k3,EGP,F01]。
The Ruan model corresponds to a model probability of P1The model probability corresponding to the Hovorka two-chamber model is P2。
The comprehensive predictive model may be expressed as: g (t) ═ P1·G1(t)+P2·G2(t)。
Updating the comprehensive prediction model by using the Bayesian algorithm can be realized by the following steps:
(1) prior distribution information of model probability and model parameters in the Bayesian algorithm is as follows:
prior distribution information P of model probabilities1＝0.5，P20.5. Table 1 is prior information of model parameters in the embodiment of the present invention, as shown in table 1:
TABLE 1
(2) And determining a likelihood function of each basic blood glucose model.
The likelihood function is determined by the target physiological data and the blood sugar basic model together, and the Ruan model in the embodiment of the invention is specificallyThe likelihood function for the Hovorka model is expressed as
Then updating the comprehensive prediction model by using a Bayesian algorithm and a plurality of groups of collected target physiological data, wherein the finally updated comprehensive prediction model is as follows:
in the embodiment of the invention, target physiological data within 60 minutes are collected once every 15 minutes, 5 groups of target physiological data are collected in total, and the preset threshold value of hypoglycemia is set to be 3.9 mmol/L. Comparing and analyzing the preset threshold value and the predicted value, and fig. 2 is a schematic diagram of an analysis result of the blood glucose prediction method based on multi-model dynamic synthesis provided by the embodiment of the invention, as shown in fig. 2, the abscissa in the diagram represents time in minutes, the ordinate on the left represents the blood glucose value in mmol/L, and the ordinate on the right represents the hypoglycemia probability.
First, the blood glucose distribution after 30 minutes (90 th minute) is predicted based on the target physiological data of the first 60 minutes, and is represented by a box diagram, the incidence of hypoglycemia is represented by a bar diagram, and the line is a blood glucose measurement value line. And continuously updating the effective data to obtain the subsequent prediction condition.
By comparing the actually measured blood sugar values, the change range of the blood sugar can be effectively predicted by the prediction model, and the incidence probability of hypoglycemia after 30 minutes is reasonably given. In the sample, the MSE prediction precision of the comprehensive prediction model is respectively improved by 10.3% and 83.2% compared with the Ruan model and the Hovorka model.
The invention comprehensively considers the uncertainty of the blood sugar model and the uncertainty of the model parameters thereof, and continuously establishes the real-time updated physiological information in the model, so that the model has the capability of individually and dynamically updating and tracking the blood sugar, and more accurately describes and predicts the change rule of the blood sugar. Different blood sugar models contain different blood sugar information, and the invention comprehensively considers a plurality of blood sugar models describing the dynamic change rule of blood sugar, reduces the risk of information loss caused by selecting a single model, and ensures that the comprehensive blood sugar prediction model has more stable and accurate prediction capability.
Fig. 3 is a schematic structural diagram of a blood glucose prediction system based on multi-model dynamic synthesis according to an embodiment of the present invention, as shown in fig. 3, the system includes: an acquisition module 301, a model synthesis module 302, an update module 303, and a prediction module 304, wherein:
the acquisition module 301 is configured to acquire a plurality of sets of target physiological data of a target subject, where each set of target physiological data includes one or more of blood glucose, insulin infusion amount, and body weight;
the model synthesis module 302 is configured to obtain a comprehensive prediction model, where the comprehensive prediction model is composed of a plurality of blood glucose basic models representing blood glucose changes and model probabilities corresponding to each blood glucose basic model, the model probabilities represent probabilities that target physiological data are output by the blood glucose basic models, each blood glucose basic model includes a plurality of model parameters that are variable according to physiological states, and model structures and model parameters included in any two blood glucose basic models are not completely the same;
the updating module 303 is configured to update the comprehensive prediction model according to a plurality of sets of target physiological data and a bayesian algorithm;
the prediction module 304 is configured to predict the blood glucose of the target subject using the updated integrated prediction model.
Firstly, the acquisition module 301 acquires a plurality of groups of target physiological data, the model integration module 302 acquires an integrated prediction model, the integrated prediction model is composed of a plurality of blood glucose basic models, for any blood glucose basic model, the blood glucose basic model comprises a plurality of model parameters, and the model parameters can reflect the change condition of blood glucose. In addition, model parameters contained in any two blood sugar basic models are not completely the same, so that the comprehensive prediction model comprises blood sugar basic models with different complexity degrees, model parameters contained in each blood sugar basic model are also different, and the description energy and prediction capability of each blood sugar basic model on blood sugar are also different.
Each blood sugar basic prediction model corresponds to a model probability, and the sum of the model probabilities corresponding to all the blood sugar basic prediction models is 1. Taking any group of target physiological data as an example, the model probability represents the possibility that the corresponding blood glucose basic model outputs the group of target physiological data, and the sum of the probabilities of all blood glucose basic models outputting the group of target physiological data is 1.
The comprehensive prediction model fully considers the uncertainty of the single blood sugar basic model and the uncertainty of the model parameters corresponding to the single blood sugar basic model by integrating the plurality of blood sugar basic models, can comprehensively reflect the change rule of blood sugar by integrating the plurality of blood sugar basic models, and can predict the change rule of blood sugar more accurately. The risk of information loss caused by a single blood sugar basic model is reduced, and the comprehensive prediction model is guaranteed to have more stable and accurate prediction capability.
The updating module 303 updates the comprehensive prediction model obtained by the model synthesis module 302 by using the sets of target physiological data and the bayesian algorithm acquired by the acquiring module 301, the updating module 303 finally obtains an updated comprehensive prediction model, and the predicting module 304 predicts the blood sugar of the target object by using the updated comprehensive prediction model.
The specific execution process of the embodiment of the system is the same as that of the embodiment of the method described above, and please refer to the embodiment of the method for details, which is not described herein again.
Fig. 4 illustrates a physical structure diagram of an electronic device, and as shown in fig. 4, the server may include: a processor (processor)410, a communication Interface 420, a memory (memory)430 and a bus 440, wherein the processor 410, the communication Interface 420 and the memory 430 are communicated with each other via the bus 440. The processor 410 may call logic instructions in the memory 430 to perform the following method:
acquiring a plurality of groups of target physiological data of a target object, wherein each group of target physiological data comprises one or more of blood sugar, insulin infusion amount and body weight;
acquiring a comprehensive prediction model, wherein the comprehensive prediction model is composed of a plurality of blood glucose basic models representing blood glucose changes and model probabilities corresponding to each blood glucose basic model, the model probabilities represent the probabilities that target physiological data are output by the blood glucose basic models, each blood glucose basic model comprises a plurality of model parameters variable along with physiological states, and model structures and model parameters contained in any two blood glucose basic models are not completely the same;
updating the comprehensive prediction model according to a plurality of groups of target physiological data and a Bayesian algorithm;
and predicting the blood sugar of the target object by using the updated comprehensive prediction model.
In addition, the logic instructions in the memory 430 may be implemented in the form of software functional units and stored in a computer readable storage medium when the software functional units are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The present embodiments provide a non-transitory computer-readable storage medium storing computer instructions that cause the computer to perform the methods provided by the above method embodiments, for example, including:
acquiring a plurality of groups of target physiological data of a target object, wherein each group of target physiological data comprises one or more of blood sugar, insulin infusion amount and body weight;
acquiring a comprehensive prediction model, wherein the comprehensive prediction model is composed of a plurality of blood glucose basic models representing blood glucose changes and model probabilities corresponding to each blood glucose basic model, the model probabilities represent the probabilities that target physiological data are output by the blood glucose basic models, each blood glucose basic model comprises a plurality of model parameters variable along with physiological states, and model structures and model parameters contained in any two blood glucose basic models are not completely the same;
updating the comprehensive prediction model according to a plurality of groups of target physiological data and a Bayesian algorithm;
and predicting the blood sugar of the target object by using the updated comprehensive prediction model.
Those of ordinary skill in the art will understand that: all or part of the steps for implementing the method embodiments may be implemented by hardware related to program instructions, and the program may be stored in a computer readable storage medium, and when executed, the program performs the steps including the method embodiments; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A blood sugar prediction method based on multi-model dynamic synthesis is characterized by comprising the following steps:
acquiring a plurality of groups of target physiological data of a target object, wherein each group of target physiological data comprises one or more of blood sugar, insulin infusion amount and body weight;
acquiring a comprehensive prediction model, wherein the comprehensive prediction model is composed of a plurality of blood glucose basic models representing blood glucose changes and model probabilities corresponding to each blood glucose basic model, the model probabilities represent the probabilities that target physiological data are output by the blood glucose basic models, each blood glucose basic model comprises a plurality of model parameters variable along with physiological states, and model structures and model parameters contained in any two blood glucose basic models are not completely the same;
updating the comprehensive prediction model according to a plurality of groups of target physiological data and a Bayesian algorithm;
and predicting the blood sugar of the target object by using the updated comprehensive prediction model.
2. The method of claim 1, wherein updating the comprehensive predictive model based on the sets of target physiological data and a bayesian algorithm comprises:
for current target physiological data, acquiring a likelihood function of each blood sugar basic model corresponding to the current target physiological data according to the current target physiological data and each blood sugar basic model;
acquiring a comprehensive prediction model corresponding to the current target physiological data according to the posterior information of the model probability corresponding to the last group of target physiological data of the current target physiological data and the posterior information of the model parameters corresponding to the last group of data of the current target physiological data;
according to the comprehensive prediction model corresponding to the current target physiological data and the likelihood function corresponding to the current target physiological data, obtaining posterior information of model probability corresponding to the current target physiological data and posterior information of model parameters corresponding to the current target physiological data, and taking the next group of target physiological data of the current target physiological data as the current target physiological data again;
and repeating the steps until the last group of target physiological data is taken as the current target physiological data, and updating the comprehensive prediction model by utilizing the posterior distribution information of the model probability corresponding to the last group of target physiological data and the posterior distribution information of the model parameters corresponding to the last group of target physiological data.
3. The method of claim 2, wherein updating the comprehensive predictive model based on the sets of target physiological data and a bayesian algorithm further comprises:
the prior distribution information of the model parameters of each blood glucose basic model is lognormal distribution, and the expectation and the variance of the lognormal distribution are obtained through moment estimation.
4. The method according to claim 1, wherein the predicting the blood glucose of the target subject using the updated global prediction model comprises:
obtaining expectation of model parameters in each blood sugar basic model according to posterior distribution information of the model parameters in each blood sugar basic model;
obtaining the expected posterior probability of each blood sugar basic model according to the posterior distribution information of the probability of each blood sugar basic model;
acquiring the comprehensive prediction model corresponding to the target time point according to the expectation of the model parameters in each blood sugar basic model and the expectation of the posterior probability of each blood sugar basic model;
and predicting the blood sugar of the target object at the target time point by using the comprehensive prediction model corresponding to the target time point.
5. The method of claim 1, wherein predicting the blood glucose of the target subject using the updated global prediction model further comprises:
acquiring the distribution information of the model parameters in each blood glucose basic model in a preset time interval according to the posterior distribution information of the model parameters in each blood glucose basic model;
acquiring a comprehensive prediction model corresponding to a target time interval according to the distribution information of the model parameters in each blood glucose basic model in the preset time interval and the posterior probability of each blood glucose basic model;
and predicting the blood sugar distribution of the target object in the target time interval according to the comprehensive prediction model corresponding to the target time interval.
6. The method of claim 5, further comprising:
and judging the probability of hypoglycemia of the target object in the target time interval according to the blood sugar distribution of the target object in the target time interval and a preset threshold value.
7. The method of claim 1, wherein the sum of model probabilities for all blood glucose basis models is 1.
8. A blood glucose prediction system based on multi-model dynamic synthesis is characterized by comprising:
the system comprises an acquisition module, a data processing module and a data processing module, wherein the acquisition module is used for acquiring a plurality of groups of target physiological data of a target object, and each group of target physiological data comprises one or more of blood sugar, insulin infusion amount and weight;
the multi-model comprehensive module is used for obtaining a comprehensive prediction model, the comprehensive prediction model is composed of a plurality of blood glucose basic models representing blood glucose changes and model probabilities corresponding to each blood glucose basic model, the model probabilities represent the probabilities that target physiological data are output by the blood glucose basic models, each blood glucose basic model comprises a plurality of model parameters variable along with physiological states, and the model structures and the model parameters contained in any two blood glucose basic models are not completely the same;
the updating module is used for updating the comprehensive prediction model according to a plurality of groups of target physiological data and a Bayesian algorithm;
and the prediction module is used for predicting the blood sugar of the target object by using the updated comprehensive prediction model.
9. An electronic device, comprising:
at least one processor, at least one memory, a communication interface, and a bus; wherein,
the processor, the memory and the communication interface complete mutual communication through the bus;
the communication interface is used for information transmission between the test equipment and the communication equipment of the display device;
the memory stores program instructions executable by the processor, the processor invoking the program instructions to perform the method of any one of claims 1-7.
10. A non-transitory computer-readable storage medium storing computer instructions that cause a computer to perform the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910260148.1A CN109935331B (en) | 2019-04-02 | 2019-04-02 | Blood glucose prediction method and system based on multi-model dynamic synthesis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910260148.1A CN109935331B (en) | 2019-04-02 | 2019-04-02 | Blood glucose prediction method and system based on multi-model dynamic synthesis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109935331A true CN109935331A (en) | 2019-06-25 |
| CN109935331B CN109935331B (en) | 2021-04-16 |
Family
ID=66989050
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910260148.1A Active CN109935331B (en) | 2019-04-02 | 2019-04-02 | Blood glucose prediction method and system based on multi-model dynamic synthesis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109935331B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110379488A (en) * | 2019-07-12 | 2019-10-25 | 深圳市预防宝科技有限公司 | A kind of pair of postprandial hyperglycemia carries out the device and method of early warning |
| CN112133439A (en) * | 2020-09-07 | 2020-12-25 | 北京理工大学 | Pre-meal insulin dose individualized decision making system based on Gaussian process |
| CN114999658A (en) * | 2022-08-04 | 2022-09-02 | 四川大学华西医院 | System and method for digital intervention on ICU (posterior ICU) syndrome |
| WO2022237162A1 (en) * | 2021-05-12 | 2022-11-17 | 中国科学院深圳先进技术研究院 | Blood glucose prediction method and application thereof, and blood glucose prediction system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107064800A (en) * | 2016-11-29 | 2017-08-18 | 北京交通大学 | The real-time predicting method of lithium ion battery remaining life |
| CN107133400A (en) * | 2017-05-03 | 2017-09-05 | 厦门大学 | A kind of aircraft structure fatigue reliability Bayes's combination forecasting method |
| CN107687872A (en) * | 2017-08-14 | 2018-02-13 | 深圳市智能机器人研究院 | Bridge structure health state monitoring method and system based on dynamic model renewal |
| EP2973095B1 (en) * | 2013-03-15 | 2018-05-09 | Animas Corporation | Insulin time-action model |
| CN109270842A (en) * | 2018-10-25 | 2019-01-25 | 浙江大学 | A kind of district heating model predictive control system and method based on Bayesian network |
-
2019
- 2019-04-02 CN CN201910260148.1A patent/CN109935331B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2973095B1 (en) * | 2013-03-15 | 2018-05-09 | Animas Corporation | Insulin time-action model |
| CN107064800A (en) * | 2016-11-29 | 2017-08-18 | 北京交通大学 | The real-time predicting method of lithium ion battery remaining life |
| CN107133400A (en) * | 2017-05-03 | 2017-09-05 | 厦门大学 | A kind of aircraft structure fatigue reliability Bayes's combination forecasting method |
| CN107687872A (en) * | 2017-08-14 | 2018-02-13 | 深圳市智能机器人研究院 | Bridge structure health state monitoring method and system based on dynamic model renewal |
| CN109270842A (en) * | 2018-10-25 | 2019-01-25 | 浙江大学 | A kind of district heating model predictive control system and method based on Bayesian network |
Non-Patent Citations (2)
| Title |
|---|
| 王斌等: "基于贝叶斯模型加权平均法的径流序列高频分量预测研究", 《水力发电学报》 * |
| 雍永强: "基于ARIMA和BPNN的组合预测模型在血糖预测中的应用", 《中国优秀硕士学位论文全文数据库 医药卫生科技辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110379488A (en) * | 2019-07-12 | 2019-10-25 | 深圳市预防宝科技有限公司 | A kind of pair of postprandial hyperglycemia carries out the device and method of early warning |
| CN112133439A (en) * | 2020-09-07 | 2020-12-25 | 北京理工大学 | Pre-meal insulin dose individualized decision making system based on Gaussian process |
| CN112133439B (en) * | 2020-09-07 | 2022-06-03 | 北京理工大学 | Pre-meal insulin dose individualized decision making system based on Gaussian process |
| WO2022237162A1 (en) * | 2021-05-12 | 2022-11-17 | 中国科学院深圳先进技术研究院 | Blood glucose prediction method and application thereof, and blood glucose prediction system |
| CN114999658A (en) * | 2022-08-04 | 2022-09-02 | 四川大学华西医院 | System and method for digital intervention on ICU (posterior ICU) syndrome |
| CN114999658B (en) * | 2022-08-04 | 2022-11-11 | 四川大学华西医院 | System and method for digital intervention on ICU (posterior ICU) syndrome |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109935331B (en) | 2021-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109935331B (en) | Blood glucose prediction method and system based on multi-model dynamic synthesis | |
| AU2021200374B2 (en) | Improved accuracy continuous glucose monitoring method, system, and device | |
| US20190318801A1 (en) | Tracking changes in average glycemia in diabetics | |
| US11951278B2 (en) | Insulin monitoring and delivery system and method for CGM based fault detection and mitigation via metabolic state tracking | |
| US12057233B2 (en) | Method, system, and computer readable medium for virtualization of a continuous glucose monitoring trace | |
| CN111540436B (en) | Self-adaptive glucose and insulin concentration prediction system and method | |
| US11839745B2 (en) | System and method for body mass index relation to patient differing psychological stress effect on blood glucose dynamics in patients with insulin dependent diabetes | |
| US11567062B2 (en) | System and method for tracking changes in average glycemia in diabetics | |
| Yang et al. | Multi-Scale Long Short-Term Memory Network with Multi-Lag Structure for Blood Glucose Prediction. | |
| WO2021007485A1 (en) | System and method for online domain adaptation of models for hypoglycemia prediction in type 1 diabetes | |
| US20210151190A1 (en) | Joint state estimation prediction that evaluates differences in predicted vs. corresponding received data | |
| Eren-Oruklu et al. | Self-tuning controller for regulation of glucose levels in patients with type 1 diabetes | |
| Shroff et al. | GlucoseAssist: Personalized Blood Glucose Level Predictions and Early Dysglycemia Detection | |
| Cinar et al. | Self-tuning controller for regulation of glucose levels in patients with type 1 diabetes |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |