CN111540436B - An adaptive glucose and insulin concentration prediction system and method - Google Patents

Abstract

The invention relates to a self-adaptive glucose and insulin concentration prediction system and a method, wherein the system acquires monitoring data of a person to be monitored through a data acquisition module; the physiological model set construction module constructs a blood sugar-insulin model according to the collected monitoring data of the person to be monitored; the state space conversion module converts the blood sugar-insulin model into a state space model; the discrete state space model building module converts the state space model into a discrete state space model; the particle wave observer module predicts and updates state variables in the discrete state space model; the automatic eating detection module determines the intake condition of the carbohydrate according to the updated carbohydrate intake factor; the blood glucose prediction module determines a concentration of insulin and a concentration of glucose in the plasma at a future fixed time interval based on the updated state variables. The invention can predict the concentration of glucose and insulin in blood plasma and improve the accuracy of predicting the concentration of glucose in blood plasma.

Description

An adaptive glucose and insulin concentration prediction system and method

technical field

The present invention relates to the field of blood sugar prediction for diabetics, in particular to an adaptive glucose and insulin concentration prediction system and method.

Background technique

Diabetes mellitus is a metabolic abnormal syndrome mainly manifested by high glucose concentration in the body on an empty stomach or after meals due to absolute or relative insufficient insulin secretion, which can easily lead to various acute and chronic complications throughout the body. Insulin is the only hypoglycemic hormone in the body. For patients with type 1 diabetes and severe type 2 diabetes who rely on exogenous insulin infusion as the main treatment method, the control of insulin injection volume directly determines the control level of glucose concentration in the body. Timely and appropriate insulin infusion to control blood sugar within a reasonable range without severe hyperglycemia or hypoglycemia events is the goal pursued by doctors and patients.

Blood sugar includes the concentration of glucose in the subcutaneous interstitium and the concentration of glucose in the plasma. Initially, blood glucose monitoring equipment can realize real-time monitoring of human subcutaneous glucose concentration, and insulin pump can realize subcutaneous continuous micro-injection of exogenous insulin. Based on the above data, it is helpful for doctors and patients to easily assess the dynamic changes of blood glucose. However, the daily blood glucose monitoring equipment collects the glucose concentration in the subcutaneous interstitial space, and the influence of insulin and carbohydrate intake on blood glucose directly affects the glucose concentration in the plasma, and the change of the glucose concentration in the plasma affects the transmission to the subcutaneous interstitium. Therefore, there is a large error between the blood glucose value measured by the blood glucose meter and the glucose concentration in the plasma. Similarly, there will be time lag and loss in the process of insulin infusion from subcutaneously delivered plasma to participate in hypoglycemia, and there is also a large error between the amount of exogenous insulin infusion and the insulin concentration in plasma. Therefore, establishing an accurate glycemic kinetic model to estimate the plasma glucose concentration and insulin concentration is the premise of accurate blood glucose prediction and control.

The blood sugar regulation system of the human body has great individual differences, and the individual physiological state is changeable. It is difficult for the mathematical model of the blood sugar regulation system with fixed parameters to accurately track and predict the dynamic change law of blood sugar. Therefore, adaptive adjustment of all blood glucose system parameters of patients based on real-time collected data is an important method to improve the prediction accuracy. In addition, external disturbances such as carbohydrate intake and exercise can cause violent fluctuations in the glucose concentration in the body. Manually inputting the carbohydrate intake time and intake will improve the prediction accuracy of the model, but it will bring inconvenience to the patient's life, and it is easy to be caused by Personal negligence increases the risk of abnormal blood sugar fluctuations. Automatically identify the patient's carbohydrate intake status and adaptively adjust the prediction model, which can provide patients with more intelligent and accurate predictions of blood glucose and insulin concentrations.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide an adaptive glucose and insulin concentration prediction system and method, so as to realize intelligent and accurate prediction of the concentration of glucose and insulin in plasma.

For achieving the above object, the present invention provides following scheme:

An adaptive glucose and insulin concentration prediction system comprising:

a data collection module for collecting monitoring data of the subject to be monitored; the monitoring data includes the initial subcutaneous intercellular glucose concentration, the exogenous insulin infusion rate and the body weight of the subject to be monitored;

a physiological model set building module for constructing a blood glucose-insulin model according to the monitoring data of the person to be monitored; the blood glucose-insulin model includes: an insulin transmission sub-model, a blood glucose-insulin kinetics sub-model, and a blood glucose transmission sub-model;

A state space conversion module, used to convert the blood glucose-insulin model into a state space model; the state variables in the state space model include the state variables converted from the blood glucose insulin model system variables and the system parameters in the blood glucose insulin model The state variable that the expansion converts to;

a discrete state space model building module for converting the state space model into a discrete state space model;

The particle wave observer module is used to predict and update the state variables in the discrete state space model; the updated state variables include the effective insulin amount in one insulin chamber, the effective insulin amount in the second insulin chamber, Plasma glucose concentration, subcutaneous interstitial glucose concentration, insulin sensitivity coefficient, basal effective insulin concentration, basal blood glucose level, blood glucose self-regulation rate and carbohydrate intake factor;

an automatic eating detection module, configured to determine the carbohydrate intake of the person to be monitored according to the updated carbohydrate intake factor; the intake includes carbohydrate intake and no carbohydrate intake;

The blood glucose prediction module is used to determine the insulin concentration in the plasma, the glucose concentration in the plasma and the glucose concentration in the subcutaneous interstitium within a fixed time interval in the future according to the updated state variables.

Optionally, the insulin transmission sub-model specifically adopts the following formula:

其中，

其中，x₁(t)为一胰岛素腔室中的有效胰岛素量，x₂(t)为二胰岛素腔室中的有效胰岛素量；u(t)为外源性胰岛素输注速率,t_I为有效胰岛素的浓度达到最大值的时间,x(t)为血浆中胰岛素的浓度，W为所述待监测者的体重,M为人体胰岛素清除速率，M＝0.017(l/kg/min)；

in,

where x ₁ (t) is the effective insulin amount in one insulin chamber, x ₂ (t) is the effective insulin amount in two insulin chambers; u(t) is the exogenous insulin infusion rate, and t _I is The time when the concentration of effective insulin reaches the maximum value, x(t) is the concentration of insulin in the plasma, W is the body weight of the person to be monitored, M is the insulin clearance rate of the human body, and M=0.017 (l/kg/min);

The blood glucose and insulin dynamics sub-model specifically adopts the following formula:

其中，G(t)为血浆中的葡萄糖的浓度，S_I为胰岛素作用的敏感系数，G_b为基础血浆中葡萄糖的浓度水平，K为血浆中葡萄糖的浓度自调节率，U(t)为引起的血浆中的葡萄糖的浓度变化的碳水化合物摄入因子；

Among them, G(t) is the concentration of glucose in plasma, S _I is the sensitivity coefficient of insulin action, G _b is the concentration level of glucose in basal plasma, K is the self-regulation rate of glucose concentration in plasma, and U(t) is Carbohydrate intake factor causing changes in plasma glucose concentration;

The blood glucose transmission sub-model specifically adopts the following formula:

其中，IG(t)为皮下细胞间质的葡萄糖的浓度，τ为时滞因子。

Here, IG(t) is the glucose concentration in the subcutaneous interstitium, and τ is the time lag factor.

Optionally, the state space conversion module specifically adopts the following formula:

其中，X为状态变量，X＝[x₁,x₂,G,IG,S_i,t_I,K,G_b,τ,U]，H＝[0,0,0,1,0,0,0,0,0,0]，ω(t)为过程噪声，υ(t)为测量噪声，z(t)为测量状态。

Among them, X is the state variable, X=[x ₁ ,x ₂ ,G,IG,S _i ,t _I ,K,G _b ,τ,U], H=[0,0,0,1,0,0 ,0,0,0,0], ω(t) is the process noise, υ(t) is the measurement noise, and z(t) is the measurement state.

Optionally, the discrete state space model building module specifically adopts the following formula:

其中，X_k-1为k-1时刻的状态变量，u_k-1为k-1时刻的外源性胰岛素输注速率，ω_k为过程噪声且服从期望为0，方差为σ_wk的高斯噪声，υ_k为测量噪声且服从期望为0，方差σ_vk的高斯噪声。

where X _k-1 is the state variable at time k-1, u _k-1 is the exogenous insulin infusion rate at time k-1, ω _k is the process noise and obeys the expectation of 0, and the variance is Gaussian with σ _wk Noise, υ _k is the measurement noise and obeys Gaussian noise with an expectation of 0 and a variance σ _vk .

Optionally, the particle wave observer module specifically uses the following formula to predict and update:

p(X _k |z _1:k-1 )=∫p(X _k |X _k-1 )p(X _k-1 |z _1:k-1 )dX _k-1

p(X _k |z _1:k )∝p(z _k |X _k-1 )p(X _k |z _1:k-1 ), where X _k is the state variable at time k, and X _k-1 is The state variables at time k-1, z _1:k-1 are all blood sugar data sequences collected before time k-1, and z _1:k are all blood sugar data sequences collected before time k.

Optionally, the automatic eating detection module specifically includes:

其中，

U_k-1为k-1时刻的引起的血浆中的葡萄糖的浓度变化的碳水化合物摄入因子，U_k为k时刻的引起的血浆中的葡萄糖的浓度变化的碳水化合物摄入因子；The forward difference acquisition unit is used to obtain the forward difference of the external interference factor at time k

in,

U _k-1 is the carbohydrate intake factor for the change in plasma glucose concentration at time k-1, and U _k is the carbohydrate intake factor for the change in plasma glucose concentration at time k;

判断是否患者是否摄入碳水化合物。a judgment unit for based on the forward difference

Determine whether the patient is consuming carbohydrates.

Optionally, the judging unit specifically includes:

以及

是否同时满足

若同时满足

则有碳水化合物摄入，否则，无碳水化合物摄入；其中，

为k时刻的外界干扰因子的前向差分，

为k-1时刻的外界干扰因子的前向差分，Threshold为设定阈值，Flag^k-i＝0为无碳水化合物摄入的标识。judge

as well as

Whether it is satisfied at the same time

If both meet

There is carbohydrate intake, otherwise, there is no carbohydrate intake; of which,

is the forward difference of the external disturbance factor at time k,

is the forward difference of external interference factors at time k-1, Threshold is the set threshold, and Flag ^ki =0 is the sign of no carbohydrate intake.

Optionally, the blood glucose prediction module specifically adopts the following formula:

表示未来一个周期后的血浆中胰岛素浓度预测值，

表示未来两个采样周期后的血浆中的胰岛素浓度预测值，

表示未来一个周期后的血浆中葡萄糖浓度预测值，

表示血浆中的葡萄糖浓度未来两个周期后的血浆中葡萄糖浓度预测值，

表示未来一个周期后的皮下细胞间质的葡萄糖浓度预测值，

表示未来两个周期后的皮下细胞间质的葡萄糖浓度。in,

represents the predicted value of plasma insulin concentration after one cycle in the future,

represents the predicted value of insulin concentration in plasma after two sampling periods in the future,

represents the predicted value of plasma glucose concentration after one cycle in the future,

is the predicted value of plasma glucose concentration after two cycles of plasma glucose concentration,

represents the predicted value of glucose concentration in the subcutaneous interstitium after one cycle in the future,

Indicates the glucose concentration in the subcutaneous interstitium after the next two cycles.

An adaptive glucose and insulin concentration prediction method, the prediction method comprising:

Collect monitoring data of the subject to be monitored; the monitoring data includes the initial subcutaneous interstitial glucose concentration, the exogenous insulin infusion rate and the body weight of the subject to be monitored;

A blood glucose-insulin model is constructed according to the monitoring data of the person to be monitored; the blood glucose-insulin model includes: an insulin transmission sub-model, a blood glucose-insulin kinetics sub-model, and a blood glucose transmission sub-model;

converting the blood glucose-insulin model into a state space model;

converting the state space model to a discrete state space model;

Predicting and updating the state variables in the discrete state space model; the updated state variables include the effective insulin amount in one insulin chamber, the effective insulin amount in two insulin chambers, the concentration of glucose in plasma, the subcutaneous cells Interstitial glucose concentration, insulin sensitivity coefficient, basal effective insulin concentration, basal blood glucose level, blood glucose self-regulation rate and carbohydrate intake factor;

Determine the carbohydrate intake of the subject to be monitored according to the updated carbohydrate intake factor; the intake includes carbohydrate intake and no carbohydrate intake;

The insulin concentration in plasma, the glucose concentration in plasma, and the glucose concentration in the subcutaneous interstitium for a fixed time interval in the future are determined from the updated state variables.

Optionally, the insulin transmission sub-model specifically adopts the following formula:

其中，x₁(t)为一胰岛素腔室中的有效胰岛素量，x₂(t)为二胰岛素腔室中的有效胰岛素量；u(t)为外源性胰岛素输注速率,t_I为有效胰岛素的浓度达到最大值的时间,x(t)为血浆中胰岛素的浓度，W为所述待监测者的体重,M为人体胰岛素清除速率，M＝0.017(l/kg/min)；

where x ₁ (t) is the effective insulin amount in one insulin chamber, x ₂ (t) is the effective insulin amount in two insulin chambers; u(t) is the exogenous insulin infusion rate, and t _I is The time when the concentration of effective insulin reaches the maximum value, x(t) is the concentration of insulin in the plasma, W is the body weight of the person to be monitored, M is the insulin clearance rate of the human body, and M=0.017 (l/kg/min);

The blood glucose and insulin dynamics sub-model specifically adopts the following formula:

其中，G(t)为血浆中的葡萄糖的浓度，S_I为胰岛素作用的敏感系数，G_b为基础血浆中葡萄糖的浓度水平，K为血浆中葡萄糖的浓度自调节率，U(t)为引起的血浆中的葡萄糖的浓度变化的碳水化合物摄入因子；

Among them, G(t) is the concentration of glucose in plasma, S _I is the sensitivity coefficient of insulin action, G _b is the concentration level of glucose in basal plasma, K is the self-regulation rate of glucose concentration in plasma, and U(t) is Carbohydrate intake factor causing changes in plasma glucose concentration;

The blood glucose transmission sub-model specifically adopts the following formula:

其中，IG(t)为皮下细胞间质的葡萄糖的浓度，τ为时滞因子。

Here, IG(t) is the glucose concentration in the subcutaneous interstitium, and τ is the time lag factor.

According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

An adaptive glucose and insulin concentration prediction system and method provided by the present invention dynamically update various parameters and system state variables through a particle wave observer module, so that the prediction of the concentrations of insulin and glucose in plasma is a real-time dynamic update rather than open-loop estimates of plasma glucose and insulin concentrations directly from fixed-parameter models and acquired data. The present invention can predict the concentration of insulin in the body by determining the concentrations of insulin and glucose in the plasma, and can improve the accuracy of the prediction of the concentration of glucose in the plasma, thereby realizing the precise control of the concentration of the glucose in the plasma.

Description of drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings required in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some of the present invention. In the embodiments, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without creative labor.

Fig. 1 is a kind of self-adaptive glucose and insulin concentration prediction system structure diagram provided by the present invention;

2 is a schematic diagram of the predicted results of glucose concentration in the plasma of a certain type of diabetes patient of the present invention in 7 days;

3 is a schematic diagram of the predicted results of insulin concentration in the plasma of a certain type of diabetic patient of the present invention for 7 days;

Fig. 4 is the prediction result of subcutaneous interstitial glucose concentration in 7 days of a certain type of diabetic patient of the present invention;

Fig. 5 is the automatic detection result of eating for 5 days of a certain type of diabetes patient of the present invention;

FIG. 6 is a schematic flowchart of an adaptive glucose and insulin concentration prediction method provided by the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

The purpose of the present invention is to provide an adaptive glucose and insulin concentration prediction system and method to achieve intelligent and accurate prediction of blood glucose and insulin concentrations.

In order to make the above objects, features and advantages of the present invention more clearly understood, the present invention will be described in further detail below with reference to the accompanying drawings and specific embodiments.

FIG. 1 is a structural diagram of an adaptive glucose and insulin concentration prediction system provided by the present invention. As shown in FIG. 1 , an adaptive glucose and insulin concentration prediction system provided by the present invention includes: a data acquisition module 101 , a physiological model set building module 102 , a state space conversion module 103 , a discrete state space model building module 104 , a particle wave observer module 105 , an automatic eating detection module 106 and a blood glucose prediction module 107 . The sampling period of the data acquisition system is T.

The data collection module 101 is used to collect monitoring data of the subject to be monitored; the monitoring data includes the initial subcutaneous intercellular glucose concentration, the exogenous insulin infusion rate, and the weight of the subject to be monitored.

The physiological model set building module 102 is configured to construct a blood glucose-insulin model according to the monitoring data of the person to be monitored; the blood glucose-insulin model includes an insulin transmission sub-model, a blood glucose-insulin kinetics sub-model, and a blood glucose transmission sub-model.

The state space conversion module 103 is used to convert the blood glucose-insulin model into a state space model; the state variables in the state space model include the state variables converted from the blood glucose insulin model system variables and the system parameters in the blood glucose insulin model The state variable that the expansion converts to.

The discrete state space model building module 104 is used to convert the state space model into a discrete state space model. The particle wave observer module 105 is used to predict and update the state variables in the discrete state space model; the updated state variables include the effective insulin amount in one insulin chamber, the effective insulin amount in the second insulin chamber, Plasma glucose concentration, subcutaneous interstitial glucose concentration, insulin sensitivity coefficient, basal effective insulin concentration, basal blood glucose level, blood glucose autoregulation rate and carbohydrate intake factor.

The automatic eating detection module 106 is configured to determine the carbohydrate intake of the subject to be monitored according to the updated carbohydrate intake factor; the intake includes carbohydrate intake and no carbohydrate intake.

The blood glucose prediction module 107 is used to determine the insulin concentration in the plasma, the glucose concentration in the plasma and the glucose concentration in the subcutaneous interstitium within a fixed time interval in the future according to the updated state variables.

The insulin delivery sub-model specifically adopts the following formula:

其中，x₁(t)为一胰岛素腔室中的有效胰岛素量，x₂(t)为二胰岛素腔室中的有效胰岛素量，u(t)为外源性胰岛素输注速率,t_I为有效胰岛素的浓度达到最大值的时间,x(t)为血浆中胰岛素的浓度，W为所述待监测者的体重,M为人体胰岛素清除速率，M＝0.017(l/kg/min)。

where x ₁ (t) is the effective insulin amount in one insulin chamber, x ₂ (t) is the effective insulin amount in two insulin chambers, u(t) is the exogenous insulin infusion rate, and t _I is The time when the effective insulin concentration reaches the maximum value, x(t) is the insulin concentration in plasma, W is the body weight of the subject to be monitored, M is the human insulin clearance rate, M=0.017 (l/kg/min).

The blood glucose and insulin dynamics sub-model specifically adopts the following formula:

其中，G(t)为血浆中的葡萄糖的浓度，S_I为胰岛素作用的敏感系数，G_b为基础血浆中葡萄糖的浓度水平，K为血浆中葡萄糖的浓度自调节率，U(t)为引起的血浆中的葡萄糖的浓度变化的碳水化合物摄入因子。

Among them, G(t) is the concentration of glucose in plasma, S _I is the sensitivity coefficient of insulin action, G _b is the concentration level of glucose in basal plasma, K is the self-regulation rate of glucose concentration in plasma, and U(t) is Carbohydrate intake factor caused by changes in plasma glucose concentration.

The blood glucose transmission sub-model specifically adopts the following formula:

其中，IG(t)为皮下细胞间质的葡萄糖的浓度，τ为时滞因子。

Here, IG(t) is the glucose concentration in the subcutaneous interstitium, and τ is the time lag factor.

The state space conversion module specifically adopts the following formula:

其中，X为状态变量，X＝[x₁,x₂,G,IG,S_i,t_I,K,G_b,τ,U]，H＝[0,0,0,1,0,0,0,0,0,0]，ω(t)为过程噪声，υ(t)为测量噪声，z(t)为测量状态。

Among them, X is the state variable, X=[x ₁ ,x ₂ ,G,IG,S _i ,t _I ,K,G _b ,τ,U], H=[0,0,0,1,0,0 ,0,0,0,0], ω(t) is the process noise, υ(t) is the measurement noise, and z(t) is the measurement state.

The discrete state space model building module specifically adopts the following formula:

其中，X_k-1为k-1时刻的状态变量，u_k-1为k-1时刻的外源性胰岛素输注速率，ω_k为过程噪声且服从期望为0，方差为σ_wk的高斯噪声，υ_k为测量噪声且服从期望为0，方差σ_vk的高斯噪声。

where X _k-1 is the state variable at time k-1, u _k-1 is the exogenous insulin infusion rate at time k-1, ω _k is the process noise and obeys the expectation of 0, and the variance is Gaussian with σ _wk Noise, υ _k is the measurement noise and obeys Gaussian noise with an expectation of 0 and a variance σ _vk .

In a specific embodiment, the variance vector composed of the variances of each state is expressed as σ _w =[10 ^-4 , 10 ^-4 , 0.01, 1, 10 ^-6 , 10 ^-6 , 10 ^-6 , 10 ^-6 , 10 ^{- 6,1} ], the variance of the observed noise is σ _vk =1.

According to the above formula, when the data acquisition module samples 4 times per hour, for discrete data, the continuous state variables of the state space conversion module are converted into continuous discrete state variables.

The particle wave observer module specifically adopts the following formula to predict and update:

p(X _k |z _1:k-1 )=∫p(X _k |X _k-1 )p(X _k-1 |z _1:k-1 )dX _k-1

p(X _k |z _1:k )∝p(z _k |X _k-1 )p(X _k |z _1:k-1 )

Among them, X _k is the state variable at time k, X _k-1 is the state variable at time k-1, z _1:k-1 is all the blood sugar data sequences collected before the k-1th time, z _1:k is All blood glucose data sequences collected before the kth time.

及其权重的组合近似，在获得一些加权随机样本后，可以根据测量结果调整权重和粒子的位置。The optimization model in the particle filter observer module is solved based on the Monte Carlo method of sequential importance sampling (SIS), that is, the probability density function p(X _k |z _1:k ) uses some random particles

and a combined approximation of its weights, after obtaining some weighted random samples, the weights and the positions of the particles can be adjusted based on the measurements.

The pseudocode of the particle filter observer module is:

The automatic eating detection module specifically includes: a forward differential acquisition unit and a judgment unit.

其中，

U_k-1为k-1时刻的引起的血浆中的葡萄糖的浓度变化的碳水化合物摄入因子，U_k为k时刻的引起的血浆中的葡萄糖的浓度变化的碳水化合物摄入因子。The forward difference acquisition unit is used to obtain the forward difference of the external interference factor at time k

in,

U _k-1 is a carbohydrate intake factor for the change in plasma glucose concentration at time k-1, and U _k is a carbohydrate intake factor for the change in plasma glucose concentration at time k.

判断是否患者是否摄入碳水化合物。The judgment unit is used for the forward difference based on the

Determine whether the patient is consuming carbohydrates.

The judging unit specifically includes:

以及

是否同时满足

若同时满足

则有碳水化合物摄入，否则，无碳水化合物摄入；其中，

为k时刻的外界干扰因子的前向差分，

为k-1时刻的外界干扰因子的前向差分，Threshold为设定阈值，Flag^k-i＝0为无碳水化合物摄入的标识。judge

as well as

Whether it is satisfied at the same time

If both meet

There is carbohydrate intake, otherwise, there is no carbohydrate intake; of which,

is the forward difference of the external disturbance factor at time k,

is the forward difference of external interference factors at time k-1, Threshold is the set threshold, and Flag ^ki =0 is the sign of no carbohydrate intake.

血浆中的葡萄糖浓度

以及皮下细胞间质的葡萄糖浓度

即：The blood glucose prediction module dynamically predicts the plasma insulin concentration of the target object in the next 30 minutes based on the real-time updated system state of the particle wave observer module

plasma glucose concentration

and glucose concentration in the subcutaneous interstitium

which is:

血浆中的葡萄糖浓度

以及皮下细胞间质的葡萄糖浓度

即：Based on the real-time updated system state of the particle wave observer module, dynamically predict the plasma insulin concentration of the target object in the next 30 minutes

plasma glucose concentration

and glucose concentration in the subcutaneous interstitium

which is:

In the embodiment of the present invention, the blood glucose simulation software certified by the U.S. Food and Drug Administration (FDA) is used to verify the method, and the detailed physiological parameters of 30 patients are recorded in the system. When inputting external conditions (insulin infusion volume, Information such as eating), the software simulates the blood sugar changes of these patients based on these physiological data. In this embodiment, the insulin infusion amount of all patients is the infusion amount during the collection of patient parameters, and the food intake is given according to the normal three meals a day, and the 7-day blood sugar system changes of all patients are generated according to the above conditions Happening. In the embodiment of the present invention, only the time series of the blood glucose measurement value and the exogenous insulin infusion amount are used to continuously input the above data into the system designed in this scheme to track and predict the insulin concentration and glucose concentration in the plasma. At the same time predict the blood sugar level in the next 30 minutes. Taking one of the simulated patients as an example, the results shown in Figure 2-Figure 5 are obtained by comparing the insulin concentration, glucose concentration, and blood glucose measurement values in the plasma generated by the blood glucose simulation software. Statistics of all patients' test results showed that the predicted root mean square error of plasma insulin concentration was 0.1283, the predicted root mean square error of plasma glucose concentration was 1.5602, and the predicted root mean square error of subcutaneous interstitial glucose concentration was 0.0145. The accuracy rate of eating detection was 86.67%, the false alarm rate was 18.75%, and the false alarm rate was 13.33%.

FIG. 6 of the present invention is a schematic flowchart of a method for predicting the concentration of an adaptive glucose and insulin provided by the present invention. As shown in FIG. 6, an adaptive method for predicting the concentration of glucose and insulin provided by the present invention includes:

S601 , collecting monitoring data of the subject to be monitored; the monitoring data includes the initial subcutaneous intercellular glucose concentration, the exogenous insulin infusion rate, and the weight of the subject to be monitored.

S602, constructing a blood glucose-insulin model according to the monitoring data of the person to be monitored; the blood glucose-insulin model includes an insulin transmission sub-model, a blood glucose-insulin kinetics sub-model, and a blood glucose transmission sub-model.

The insulin delivery sub-model specifically adopts the following formula:

其中，x₁(t)为一胰岛素腔室中的有效胰岛素量，x₂(t)为二胰岛素腔室中的有效胰岛素量，u(t)为外源性胰岛素输注速率,t_I为有效胰岛素的浓度达到最大值的时间,x(t)为血浆中胰岛素的浓度，W为所述待监测者的体重,M为人体胰岛素清除速率，M＝0.017(l/kg/min)；

where x ₁ (t) is the effective insulin amount in one insulin chamber, x ₂ (t) is the effective insulin amount in two insulin chambers, u(t) is the exogenous insulin infusion rate, and t _I is The time when the concentration of effective insulin reaches the maximum value, x(t) is the concentration of insulin in the plasma, W is the body weight of the person to be monitored, M is the insulin clearance rate of the human body, and M=0.017 (l/kg/min);

The blood glucose and insulin dynamics sub-model specifically adopts the following formula:

其中，G(t)为血浆中的葡萄糖的浓度，S_I为胰岛素作用的敏感系数，G_b为基础血浆中葡萄糖的浓度水平，K为血浆中葡萄糖的浓度自调节率，U(t)为引起的血浆中的葡萄糖的浓度变化的碳水化合物摄入因子。

Among them, G(t) is the concentration of glucose in plasma, S _I is the sensitivity coefficient of insulin action, G _b is the concentration level of glucose in basal plasma, K is the self-regulation rate of glucose concentration in plasma, and U(t) is Carbohydrate intake factor caused by changes in plasma glucose concentration.

The blood glucose transmission sub-model specifically adopts the following formula:

其中，IG(t)为皮下细胞间质的葡萄糖的浓度，τ为时滞因子。

Here, IG(t) is the glucose concentration in the subcutaneous interstitium, and τ is the time lag factor.

S603, convert the blood glucose-insulin model into a state space model.

其中，X为可以根据测量结果调整权重和粒子的位置。变量，X＝[x₁,x₂,G,IG,S_i,t_I,K,G_b,τ,U]，H＝[0,0,0,1,0,0,0,0,0,0]，ω(t)为过程噪声，υ(t)为测量噪声，z(t)为测量状态。

Among them, X is the position where the weight and particle can be adjusted according to the measurement result. Variable, X=[x ₁ ,x ₂ ,G,IG,S _i ,t _I ,K,G _b ,τ,U], H=[0,0,0,1,0,0,0,0, 0,0], ω(t) is the process noise, υ(t) is the measurement noise, and z(t) is the measurement state.

S604. Convert the state space model into a discrete state space model.

S605, predict and update the state variables in the discrete state space model; the updated state variables include the effective insulin amount in the first insulin chamber, the effective insulin amount in the second insulin chamber, the concentration of glucose in the plasma, Subcutaneous interstitial glucose concentration, insulin sensitivity coefficient, basal effective insulin concentration, basal blood glucose level, blood glucose autoregulation rate and carbohydrate intake factor.

S606: Determine the carbohydrate intake of the subject to be monitored according to the updated carbohydrate intake factor; the intake includes the intake of carbohydrates and the intake of no carbohydrates.

S607 , determining the insulin concentration in the plasma, the glucose concentration in the plasma, and the glucose concentration in the subcutaneous interstitium within a fixed time interval in the future according to the updated state variables.

The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same and similar parts between the various embodiments can be referred to each other.

In this paper, specific examples are used to illustrate the principles and implementations of the present invention. The descriptions of the above embodiments are only used to help understand the methods and core ideas of the present invention; meanwhile, for those skilled in the art, according to the present invention There will be changes in the specific implementation and application scope. In conclusion, the contents of this specification should not be construed as limiting the present invention.

Claims (10)

1. A concentration prediction system of self-adaptive glucose and insulin, wherein the prediction system comprises: a data collection module for collecting monitoring data of the subject to be monitored; the monitoring data includes the initial subcutaneous intercellular glucose concentration, the exogenous insulin infusion rate and the body weight of the subject to be monitored; a physiological model set building module for constructing a blood glucose-insulin model according to the monitoring data of the person to be monitored; the blood glucose-insulin model includes: an insulin transmission sub-model, a blood glucose-insulin kinetics sub-model, and a blood glucose transmission sub-model; A state space conversion module for converting the blood glucose-insulin model into a state space model, where the state variables in the state space model include the state variables converted from the blood glucose insulin model system variables and the system parameters in the blood glucose insulin model The state variable that the expansion converts to; a discrete state space model building module for converting the state space model into a discrete state space model; The particle wave observer module is used to predict and update the state variables in the discrete state space model; the updated state variables include the effective insulin amount in one insulin chamber, the effective insulin amount in two insulin chambers, Plasma glucose concentration, subcutaneous interstitial glucose concentration, insulin sensitivity coefficient, basal effective insulin concentration, basal blood glucose level, blood glucose self-regulation rate and carbohydrate intake factor; an automatic eating detection module, configured to determine the carbohydrate intake of the person to be monitored according to the updated carbohydrate intake factor; the intake includes carbohydrate intake and no carbohydrate intake; The blood glucose prediction module is used to determine the insulin concentration in the plasma, the glucose concentration in the plasma, and the glucose concentration in the subcutaneous interstitium within a fixed time interval in the future according to the updated state variables. 2. a kind of self-adaptive glucose and insulin concentration prediction system according to claim 1, is characterized in that, described insulin transmission sub-model specifically adopts following formula:

in,

where x ₁ (t) is the effective insulin amount in one insulin chamber, x ₂ (t) is the effective insulin amount in two insulin chambers; u(t) is the exogenous insulin infusion rate, and t _I is The time when the concentration of effective insulin reaches the maximum value, x(t) is the concentration of insulin in the plasma, W is the body weight of the person to be monitored, M is the insulin clearance rate of the human body, and M=0.017 (l/kg/min); The blood glucose insulin kinetic sub-model specifically adopts the following formula:

Among them, G(t) is the concentration of glucose in plasma, S _I is the sensitivity coefficient of insulin action, G _b is the concentration level of glucose in basal plasma, K is the self-regulation rate of glucose concentration in plasma, and U(t) is Carbohydrate intake factor causing changes in plasma glucose concentration; The blood glucose transmission sub-model specifically adopts the following formula:

Here, IG(t) is the glucose concentration in the subcutaneous interstitium, and τ is the time lag factor. 3. the concentration prediction system of a kind of self-adaptive glucose and insulin according to claim 2, is characterized in that, described state space conversion module specifically adopts following formula:

Among them, X is the state variable, X=[x ₁ ,x ₂ ,G,IG,S _i ,t _I ,K,G _b ,τ,U], H=[0,0,0,1,0,0 ,0,0,0,0], ω(t) is the process noise, υ(t) is the measurement noise, and z(t) is the measurement state. 4. a kind of self-adaptive glucose and insulin concentration prediction system according to claim 1, is characterized in that, described discrete state space model building module specifically adopts following formula:

where X _k-1 is the state variable at time k-1, u _k-1 is the exogenous insulin infusion rate at time k-1, ω _k is the process noise and obeys the expectation of 0, and the variance is Gaussian with σ _wk Noise, υ _k is the measurement noise and obeys Gaussian noise with an expectation of 0 and a variance σ _vk . 5. a kind of self-adaptive glucose and insulin concentration prediction system according to claim 4, is characterized in that, described particle wave observer module specifically adopts following formula to predict and update:

Among them, X _k is the state variable at time k, X _k-1 is the state variable at time k-1, z _1:k-1 is all the blood sugar data sequences collected before the k-1th time, z _1:k is All blood glucose data sequences collected before the kth time. 6. A kind of self-adaptive glucose and insulin concentration prediction system according to claim 1, is characterized in that, described eating automatic detection module specifically comprises: The forward difference acquisition unit is used to obtain the forward difference of the external interference factor at time k

in,

U _k-1 is the carbohydrate intake factor for the change in plasma glucose concentration at time k-1, and U _k is the carbohydrate intake factor for the change in plasma glucose concentration at time k; a judgment unit for based on the forward difference

Determine whether the patient is consuming carbohydrates. 7. A kind of self-adaptive glucose and insulin concentration prediction system according to claim 6, is characterized in that, described judging unit specifically comprises: judge

as well as

Whether it is satisfied at the same time

If both meet

There is carbohydrate intake, otherwise, there is no carbohydrate intake; of which,

is the forward difference of the external disturbance factor at time k,

is the forward difference of external interference factors at time k-1, Threshold is the set threshold, and Flag ^ki =0 is the sign of no carbohydrate intake. 8. a kind of self-adaptive glucose and insulin concentration prediction system according to claim 6, is characterized in that, described blood sugar prediction module specifically adopts following formula:

in,

represents the predicted value of plasma insulin concentration after one cycle in the future,

represents the predicted value of insulin concentration in plasma after two sampling periods in the future,

represents the predicted value of plasma glucose concentration after one cycle in the future,

is the predicted value of plasma glucose concentration after two cycles of plasma glucose concentration,

represents the predicted value of glucose concentration in the subcutaneous interstitium after one cycle in the future,

Indicates the glucose concentration in the subcutaneous interstitium after the next two cycles. 9. A method for predicting the concentration of adaptive glucose and insulin, wherein the predicting method comprises: Collect monitoring data of the subject to be monitored; the monitoring data include the initial subcutaneous interstitial glucose concentration, the exogenous insulin infusion rate and the body weight of the subject to be monitored; A blood glucose-insulin model is constructed according to the monitoring data of the person to be monitored; the blood glucose-insulin model includes: an insulin transmission sub-model, a blood glucose-insulin kinetics sub-model, and a blood glucose transmission sub-model; converting the blood glucose-insulin model into a state space model; converting the state space model to a discrete state space model; Predicting and updating the state variables in the discrete state space model; the updated state variables include the effective insulin amount in one insulin chamber, the effective insulin amount in two insulin chambers, the concentration of glucose in plasma, the subcutaneous cells Interstitial glucose concentration, insulin sensitivity coefficient, basal effective insulin concentration, basal blood glucose level, blood glucose self-regulation rate and carbohydrate intake factor; Determine the carbohydrate intake of the subject to be monitored according to the updated carbohydrate intake factor; the intake includes carbohydrate intake and no carbohydrate intake; The insulin concentration in plasma, the glucose concentration in plasma, and the glucose concentration in the subcutaneous interstitium for a fixed time interval in the future are determined from the updated state variables. 10. The method for predicting the concentration of adaptive glucose and insulin according to claim 9, wherein the insulin transfer sub-model specifically adopts the following formula:

in,

where x ₁ (t) is the effective insulin amount in one insulin chamber, x ₂ (t) is the effective insulin amount in two insulin chambers; u(t) is the exogenous insulin infusion rate, and t _I is The time when the concentration of effective insulin reaches the maximum value, x(t) is the concentration of insulin in the plasma, W is the body weight of the person to be monitored, M is the insulin clearance rate of the human body, and M=0.017 (l/kg/min); The blood glucose insulin kinetic sub-model specifically adopts the following formula:

Among them, G(t) is the concentration of glucose in plasma, S _I is the sensitivity coefficient of insulin action, G _b is the concentration level of glucose in basal plasma, K is the self-regulation rate of glucose concentration in plasma, and U(t) is Carbohydrate intake factor causing changes in plasma glucose concentration; The blood glucose transmission sub-model specifically adopts the following formula:

Here, IG(t) is the glucose concentration in the subcutaneous interstitium, and τ is the time lag factor.

Images