A method for improving the accuracy of non-invasive blood glucose metabolism thermal integration method detection

Abstract

The method of improving the accuracy of non-invasive blood glucose metabolic heat integration method detection of the present invention belongs to the field of non-invasive blood glucose detection. The present invention is a method for reducing the interference of hypoglycemic drugs on the metabolic heat integration method for detecting blood glucose by correcting the measured body temperature. The method establishes a metabolic kinetic equation of hypoglycemic drugs, establishes a general formula for temperature correction based on the metabolic kinetic equation, and then uses a pre-calibration method to establish a temperature correction amount calculation formula suitable for individuals and their drugs. Finally, the corrected body temperature after deducting the personal temperature correction amount is used in the metabolic heat integration method to calculate the personal blood glucose value, so as to eliminate the influence of hypoglycemic drugs on the accuracy of blood glucose detection. Clinical examples show that after the body temperature is corrected by the method of the present invention, the accuracy of blood glucose calculation using the non-invasive metabolic heat integration method can be significantly improved.

Description

Technical Field

[0001] The invention belongs to the field of non-invasive blood sugar detection, and specifically is a method for improving the accuracy of non-invasive blood sugar metabolism heat integration method detection. Background Art

[0002] The metabolic heat integration method is the mainstream method of non-invasive blood glucose detection technology. This method is based on the theory of energy metabolism conservation, and uses temperature sensors, humidity sensors, radiation sensors and blood oxygen sensors to collect dynamic signals of skin surface temperature or deep oral body temperature. According to the quantitative relationship between these signals and current blood glucose, DSP (digital signal processor) is used for processing to give the user's blood glucose value in real time. Based on the use of MEMS technology to integrate multiple sensors, reduce the volume and weight of detection circuits, improve reliability and signal-to-noise ratio, and optimize detection algorithms, some non-invasive detection products have passed China or EU certification and are about to begin commercialization.

[0003] However, like other non-invasive blood glucose measurement technologies, the metabolic heat integration method is also affected by many internal and external factors, making its accuracy fail to meet the standard. For this reason, some conditions restricting its use have to be set. For example, a product that has been certified by CFDA has limited the blood glucose measurement range, and specifically states that it cannot be used for insulin users. However, insulin is a very effective blood glucose control drug, and almost all type 1 and nearly half of type 2 diabetics use insulin. The above restrictions greatly reduce the practical value of this product.

[0004] According to medical common sense, using insulin to control blood sugar is not "addictive". In order to maintain long-term stable blood sugar, diabetic patients need to adhere to the doctor's prescription and inject the prescribed type of insulin at a fixed time and quantity every day. Insulin promotes the oxidation of glucose, causing blood sugar to drop rapidly. However, many patients will feel cold or even shiver for a period of time after the injection.

[0005] Therefore, studying the relationship between blood glucose and body temperature, and the relationship between insulin and body temperature is of great significance for improving the accuracy of non-invasive blood glucose metabolism thermal integration method detection. Summary of the invention

[0006] The object of the present invention is to provide a method for improving the accuracy of non-invasive blood glucose metabolism heat integration method detection, comprising the following steps:

[0007] Step 1: Establish the metabolic kinetic equation of hypoglycemic drugs;

[0008] Step 2: Establish a general formula for body temperature correction based on the metabolic kinetic equation;

[0009] Step 3: Use a pre-calibration method to establish a temperature correction calculation formula suitable for the individual and the medication he / she uses;

[0010] Step 4: Use the corrected body temperature after deducting the personal body temperature correction amount to calculate the personal blood glucose value in the metabolic heat integration method to eliminate the impact of hypoglycemic drugs on the accuracy of blood glucose testing.

[0011] Preferably, the hypoglycemic drug is premixed and mixed insulin, and the metabolic kinetic equation of the premixed and mixed insulin is established by superimposing the metabolic kinetic equations of rapid-acting insulin, short-acting insulin or intermediate-acting insulin, and long-acting insulin, and then weighted averaging their respective proportions.

[0012] Preferably, the hypoglycemic drug is rapid-acting insulin, and the metabolic kinetic equation of rapid-acting insulin from the onset of action to the peak stage is C(t)=ɑ+βt, where ɑ and β are proportional constants; the metabolic kinetic equation of rapid-acting insulin from the peak to the end of the drug effect is C(t)=λ+γe -kt , where k is the metabolic rate constant of rapid-acting insulin, and λ and γ are proportionality constants.

[0013] Preferably, the hypoglycemic drug is short-acting insulin or intermediate-acting insulin, and the metabolic kinetic equation of short-acting insulin or intermediate-acting insulin from the onset of action to the end of its effect is C(t)=ω(C f -C o )Exp[-(lnt–lnt f ) 2 / (2σ 2 )] / [(2π) 0.5 σt]+C o , where σ and ω are empirical parameters, C f is the peak insulin concentration, C o is the basal insulin concentration in the body, t f Peak occurrence time; for short-acting insulin, σ=0.9~1.1, ω=1.52~1.55; for intermediate-acting insulin, σ=0.4~0.6, ω=0.91~1.16, the onset time of intermediate-acting insulin is 0.

[0014] Preferably, the hypoglycemic drug is long-acting insulin, and the metabolic kinetic equation of long-acting insulin from the onset of action to the appearance of the plateau is C(t)=ρ+φt, where ρ and φ are proportional constants; the metabolic kinetic equation of long-acting insulin from the appearance of the plateau to the end of its efficacy is C(t)=C f , where C f is the insulin plateau concentration.

[0015] Preferably, the general formula for body temperature correction is ΔT(t)=a C(t) 2 +b C(t)+c, where ΔT(t) is the correction value of body temperature at time t after insulin injection, C(t) is the insulin concentration in the body at time t, and a(℃ / (IU)2 ), b (℃ / IU) and c (℃) are empirical parameters related to individuals and insulin products.

[0016] Preferably, the steps of establishing a temperature correction calculation formula suitable for an individual and the medicines used by the individual in a pre-calibrated manner include:

[0017] Step 1: Do not inject insulin on the first day. Select three time points within a certain period of time, with an interval of no less than 30 minutes between the time points, measure your body temperature and blood sugar, and record the measurement data and measurement time;

[0018] Step 2: Inject insulin at the same time on the second day, and measure body temperature and blood sugar three times according to the measurement time on the first day, record the measurement data, and convert the three measurement times into the time value after insulin injection;

[0019] Step 3: Based on the temperature data measured on the first day, the blood glucose values ​​at three time points are calculated using the metabolic heat integration method and compared with the invasive blood glucose values. If the relative errors do not exceed 15%, the blood glucose calculation method of the metabolic heat integration method is recognized, otherwise it needs to be adjusted;

[0020] Step 4: Based on the blood sugar data measured on the second day, the body temperature data at three time points are inferred using the metabolic heat integration method;

[0021] Step 5: Subtract the body temperature at the three time points inferred from the body temperature at the three time points measured on the second day. The difference is the correction amount of the individual's body temperature at the three time points.

[0022] Step 6: Find the corresponding insulin metabolism kinetic equation according to the type, dosage and injection time of injected insulin;

[0023] Step 7, substituting the obtained body temperature correction values ​​at three time points, the time value after insulin injection and the corresponding kinetic equation into the general body temperature correction formula to obtain the specific values ​​of the three empirical parameters a, b and c;

[0024] Step 8: Substitute the three empirical parameter values ​​of a, b and c into the general temperature correction formula to obtain a temperature correction calculation formula suitable for the insulin brand and individual constitution.

[0025] Preferably, the relationship between the insulin peak time and the insulin dose is t f =t fo +pL, where t f0 is the shortest peak time, L is the insulin dose, and p is the empirical coefficient, which represents the degree of influence of insulin dose on the delayed peak time of different types of insulin. For rapid-acting insulin, p=0.

[0026] Preferably, the relationship between the end time of drug effect and insulin dosage is t e =m+nL, where t e is the end time of drug effect, L is the insulin dose, and m and n are empirical coefficients.

[0027] Compared with the prior art, the characteristics and beneficial effects of the present invention are as follows: The present invention provides a method for reducing the interference of hypoglycemic drugs on the detection of blood glucose by the metabolic heat integration method by correcting the measured body temperature. The method establishes a metabolic kinetic equation for hypoglycemic drugs, establishes a general formula for temperature correction based on the metabolic kinetic equation, and then uses a pre-calibration method to establish a temperature correction calculation formula suitable for individuals and the drugs they use. Finally, the corrected body temperature after deducting the personal temperature correction amount is used to calculate the personal blood glucose value in the metabolic heat integration method to eliminate the influence of hypoglycemic drugs on the accuracy of blood glucose detection. Clinical examples have shown that after the body temperature is corrected by the method of the present invention, the accuracy of blood glucose calculation using the non-invasive metabolic heat integration method is significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 Different types of insulin metabolic kinetic curves (concentration-time curves) cited from literature

[0029] Figure 2 The log-normal distribution curve for different σ under the condition that μ remains unchanged. DETAILED DESCRIPTION

[0030] In order to make the technical means, innovative features, objectives and effects achieved by the present invention easy to understand, the present invention is further described below.

[0031] The embodiments described herein are specific embodiments of the present invention, which are used to illustrate the concept of the present invention, are illustrative and exemplary, and should not be construed as limiting the embodiments of the present invention and the scope of the present invention. In addition to the embodiments described herein, those skilled in the art can also adopt other obvious technical solutions based on the claims and the contents disclosed in the specification of this application, including technical solutions that adopt any obvious replacement and modification of the embodiments described herein.

[0032] 1. Mechanism of insulin-induced temperature distortion

[0033] In 2009, scientists at the Scripps Research Institute in the United States discovered a direct link between insulin and core body temperature. Since its discovery in the 1920s, insulin has been studied extensively, but this is the first time it has been found to be related to the basic process of temperature regulation. Scientists found that when insulin was injected directly into a specific area of ​​the rodent's brain, core body temperature rose, metabolism increased, and brown adipose (fat) tissue was activated to release heat. The research team also found that these effects were dose-related, and the more insulin, the more these metabolic indicators rose.

[0034] Normally, the human body's core temperature remains within a narrow range that is associated with key enzymatic reactions. When the core temperature exceeds this range for a long time (fever or hypothermia), it is harmful to the body. Moderate changes in core temperature are consistent with the 24-hour sleep-wake cycle each day.

[0035] The present invention proposes that insulin can direct brown adipose tissue to consume energy to generate heat through a specific pathway involving neurons from the preoptic area of ​​the brain to the spinal cord. The higher the insulin level, the higher the core body temperature. It can be deduced that patients with type 1 diabetes who are purely due to insufficient insulin supply have lower core body temperature. In fact, a body temperature below 97F is one of the earliest signs of the disease. Similarly, the elevated insulin levels associated with type 2 diabetes will make the body hotter rather than cooler. However, long-term high insulin status may deplete the signaling mechanism that triggers the fat burning process, resulting in a disorder of the heat production process and a decrease in core body temperature instead of an increase.

[0036] Why do patients feel cold all over or even shiver after injecting insulin (especially rapid-acting or short-acting insulin)? The present invention believes that the body temperature of homeothermic animals is maintained at a stable level under the control of the hypothalamic temperature regulation center through physiological regulation reactions such as increasing and decreasing blood flow to the skin, sweating, and shivering. This physiological temperature regulation is completed by the organism's own control system. The center of this system is the set point (Set Point) located in the hypothalamus. If insulin sends a signal to the hypothalamus through a certain mechanism, so that the temperature level originally preset at the set point is increased, such as from 36.8°C to 37.1°C, the human body will automatically generate more heat to maintain the increased body temperature by shivering, increasing the efficiency of glucose oxidation in skeletal muscles and liver, etc. Just as a high fever makes patients feel cold, the shivering after insulin injection is not due to a drop in body temperature but a reaction of the body to produce heat. The reason why insulin signals can raise the body temperature set point is that the inventor believes that insulin can promote the efficiency of glucose oxidation throughout the body and increase heat production. After the hypothalamus set point senses the insulin signal, it will automatically raise the body temperature immediately so that it can maintain a constant body temperature level while increasing heat production. However, if the set point overreacts and the body temperature adjustment point is too high, it is necessary to compensate for the heat production required at the new body temperature through shivering and other methods. As the insulin metabolism process proceeds, the signal of insulin entering the hypothalamus gradually weakens or disappears, the temperature setting of the set point automatically returns to the original level, and the patient's feeling of cold gradually disappears.

[0037] In summary, after insulin injection, due to the adjustment of the set point, the human body will experience a rise in core body temperature in a short period of time. This causes serious interference to the metabolic heat integration method for measuring blood sugar. The reasons are: (1) The metabolic heat integration method is based on the dynamic balance of heat production and heat dissipation in the body. During the current period of time, if the glucose concentration in the blood remains high, the metabolic heat production during this period of time will be high, and the heat dissipation conditions during this period of time will change relatively slowly. If the heat production is greater than the heat dissipation, the body temperature will rise. After a meal, since the glucose in the food continues to enter the blood, the postprandial blood sugar and oral temperature measured over a period of time will both increase, showing a significant positive correlation between the increase in blood sugar and the increase in body temperature. Therefore, the former can be calculated by measuring the latter. (2) After insulin injection, the heat production during the current period of time mainly comes from the oxidation of blood sugar promoted by insulin. Under the condition of unchanged heat dissipation conditions, the core body temperature will rise. However, since glucose is rapidly oxidized and decomposed but cannot be replenished in time, the blood sugar concentration will decrease. Therefore, during this period of time, the increase in blood sugar and the increase in temperature are inversely proportional. According to the original metabolic heat integration method, the former cannot be correctly calculated by measuring the latter. (3) As insulin is metabolized in the blood, the glucose oxidation rate gradually decreases, and the blood sugar concentration slowly decreases and remains at a low level. The metabolic heat production in the current period is less, and the heat dissipation conditions change relatively slowly during this period. If the heat production is less than the heat dissipation, the body temperature will decrease. There is a positive correlation between the decrease in blood sugar and the decrease in body temperature, and the metabolic heat integration method becomes effective again. Until the blood sugar drops to the basal blood sugar (fasting blood sugar) and the body temperature drops to the basal body temperature, the body's heat production and heat dissipation reach a new dynamic balance.

[0038] Therefore, within a period of time after insulin injection, the blood sugar concentration calculated according to the existing metabolic heat integration method formula will have a huge error. This is the fundamental reason why this method is not suitable for insulin users.

[0039] This change in body temperature and blood sugar level error caused by hypoglycemic drugs is referred to as temperature drug distortion in the present invention. Clinical trials have shown that temperature drug distortion has the following characteristics: (1) The distortion caused by insulin is much greater than that caused by other hypoglycemic drugs. (2) Among insulins, the distortion caused by rapid-acting or short-acting insulin is the most significant. (3) Temperature drug distortion is positively correlated with the amount of injected insulin. (4) Temperature drug distortion is related to the injection time, which is actually related to the concentration of the drug in the blood.

[0040] 2. Solution to body temperature and drug distortion

[0041] As mentioned above, hypoglycemic drugs (mainly insulin) will cause the body temperature set point to rise, resulting in a higher blood sugar value calculated based on the measured body temperature. The inventors believe that: if the measured body temperature remains unchanged, the set point will rise, which is equivalent to the set point remaining unchanged and the measured body temperature falling. Therefore, the method to solve the distortion of body temperature and drug is to appropriately lower the measured body temperature according to a certain rule before calculating blood sugar. This downward correction amount of the measured body temperature is referred to as the body temperature and drug correction amount. The body temperature and drug correction amount needs to consider the following factors: (1) the type of hypoglycemic drug; (2) the metabolic kinetics of the hypoglycemic drug in the body; (3) the amount of hypoglycemic drug used; (4) the use time of the hypoglycemic drug; (5) the relationship between the hypoglycemic drug and other factors affecting blood sugar.

[0042] (1) Types of Insulin

[0043] Insulin can be broadly divided into the following categories:

[0044] a) Rapid-acting insulin (ultra-short-acting insulin)

[0045] Drug name: Insulin lispro, insulin aspart or insulin glulisine, etc.; Trade name: NovoRapid, Humalog, etc. Usually injected before three meals to lower blood sugar after meals. The dosage is generally not more than 10 units at a time.

[0046] b) Regular insulin (short-acting insulin)

[0047] Trade names: Novolin R, Humulin R, etc.

[0048] c) Intermediate-acting insulin

[0049] Drug name: low-protamine insulin, etc. Trade name: Novolin N, Humulin N, etc.

[0050] d) Long-acting insulin

[0051] Drug name: Insulin detemir, insulin glargine, etc.

[0052] e) Premixed insulin:

[0053] Mixing short-acting insulin (R) and intermediate-acting insulin (N) in a certain ratio in advance is called premixed insulin. Trade names include Novolin 30R, Novolin 50R, and Humulin 70 / 30. 30R means mixing 30% of short-acting R and 70% of intermediate-acting N insulin; 50R means short-acting R and intermediate-acting N each account for 50%. The common feature of this type of insulin is the biphasic effect, that is, the two types of insulin work separately after mixing, which is equivalent to injecting short-acting and intermediate-acting insulin at one time, and only 1-2 injections are needed per day.

[0054] f) Mixed insulin

[0055] Before injection, diabetics mix insulins with different action times and use them together, which is called mixed insulin. Common formulas are: ① Short-acting + long-acting: The most commonly used is a mixture of regular insulin (RI) and protamine zinc insulin (PZI). The ratio of RI to PZI can be 2:1, 3:1 or 4:1 according to needs. Generally speaking, when RI:PZI=2:1, the mixed insulin is equivalent to medium-acting insulin; when the ratio is >2:1, it is equivalent to medium- and short-acting insulin; when the ratio is <2:1, it is equivalent to medium- and long-acting insulin. ② Short-acting + medium-acting: In addition to fixed mixed types such as Novolin 30R and Ubiquinol 70 / 30, regular insulin can be mixed with neutral protamine zinc insulin (NPH). Different ratios of RI and NPH have different peak times. Generally speaking, the larger the RI ratio, the earlier the peak, and vice versa.

[0056] (2) Insulin metabolic kinetics

[0057] The metabolic kinetics of drugs studies the change of drug concentration in the blood over time. Based on the metabolic kinetic curves of different insulins (i.e. the relationship between concentration and time), we can clearly understand the insulin concentration at different time points after insulin injection. Figure 1 These are different types of insulin metabolic kinetic curves (concentration-time curves) cited from literature.

[0058] from Figure 1 It can be seen that rapid-acting insulin takes effect the fastest (10-20 minutes), peaks 1-2 hours after injection, and the drug effect lasts for 4-6 hours. Therefore, for patients who inject this type of insulin, it is necessary to correct the body temperature downward (i.e., body temperature drug correction) from 15 minutes after injection until within 4 hours in order to obtain an accurate blood sugar value. And during this period, the correction range should match the plasma insulin level. In other words, the downward correction value of body temperature is the largest around 30-45 minutes after injection, and the correction value gradually approaches 0 after 3 hours.

[0059] In contrast, long-acting insulin takes effect a little longer, has no obvious peak, and has a low concentration in the body but lasts up to 24 hours. Therefore, for patients who inject this type of insulin, temperature medication correction is required from 3 minutes after injection to the whole day, but the correction range is very small and the correction value is basically unchanged after 2 hours of injection.

[0060] The kinetic curves of regular (short-acting) and intermediate-acting insulins are between rapid-acting and long-acting insulins. The former takes effect in 20 to 30 minutes, peaks in 2 to 4 hours, and lasts for 6 to 10 hours. Therefore, temperature drug correction is required from 20 minutes to about 8 hours after injection. The latter takes effect in 1.5 to 4 hours, peaks in 4 to 8 hours, and lasts for about 14 to 20 hours. Therefore, temperature drug correction is required from 1 hour to 18 hours after injection.

[0061] There are three main dosage forms of premixed insulin: premixed 30R, premixed 50R and premixed insulin analog 30 / 70.

[0062] 1) Premix 30R is a mixture of 30% short-acting insulin and 70% intermediate-acting insulin: onset time 0.5 hours, maximum duration of action 2-8 hours, duration 15-20 hours. Therefore, temperature medication correction should be performed from 30 minutes to 18 hours after injection.

[0063] 2) Premixed 50R is a mixture of 50% short-acting insulin and 50% intermediate-acting insulin: the onset of action is within 0.5 hours, the maximum concentration time is 2-8 hours, and the duration is up to 24 hours. Therefore, temperature medication correction should be performed from 30 minutes to 20 hours after injection.

[0064] 3) Herui 30-30% soluble insulin aspart and 70% protamine-bound crystalline insulin aspart: After subcutaneous injection, It takes effect within minutes, and the strongest effect is after injection The duration of use can be up to 24 hours. Therefore, temperature correction should be performed from 10 minutes to 20 hours after injection.

[0065] (3) Insulin dosage

[0066] In the present invention, the temperature drug correction amount is set to be proportional to the insulin dosage. And necessary corrections are made according to the influence of insulin dosage on the metabolic kinetic characteristics of different types of insulin. These influences include:

[0067] 1) For rapid-acting insulin, the onset and peak time remain the same regardless of the dose, but the duration of action is dose-dependent. A dose of less than 10 units (IU) lasts for 4 hours (or less), while an injection of 25-30 units (IU) lasts for 5-6 hours.

[0068] 2) For regular insulin and intermediate-acting insulin, if the dosage is small, the onset time is short, the peak value appears earlier, and the duration of drug effect is also short. If the dosage is large, the onset time is fast, the peak value appears later, and the duration of drug effect is long.

[0069] 3) For long-acting insulin, the amount of insulin used has little effect on the kinetic properties and can be ignored.

[0070] On the other hand, when calculating the temperature medication correction amount, personal factors such as insulin sensitivity also need to be considered. About 40-50% of the daily insulin dosage is used to supplement the needs at night, fasting and between meals. This dosage is called the background (or basal) insulin supplement. This basal insulin requirement basically remains unchanged every day. Another 50-60% of the daily insulin dosage is used to convert carbohydrates in the diet and correct high blood sugar. This dosage is called the additional insulin dosage. In clinical practice, long-acting insulin is generally used to supplement the basal insulin dosage. The insulin injection time and injection amount are fixed. When making temperature medication corrections, just substitute the insulin dosage into the temperature medication correction calculation formula. However, for the additional insulin dosage, its injection time and injection amount may be adjusted at any time. Special attention should be paid when calculating the temperature medication correction amount.

[0071] 1) Calculate additional insulin dose based on carbohydrate conversion needs

[0072] The additional amount of insulin to convert carbohydrates is calculated based on the ratio of insulin to carbohydrates, which represents how many grams of carbohydrates can be converted by one unit of insulin. Generally speaking, one unit of rapid-acting insulin can convert 12-15 grams of carbohydrates. However, depending on the insulin sensitivity of each person, it can range from 6 to 30 grams, or even more. Insulin sensitivity can vary from person to person and at different times, and can be affected by activity and emotional stress. Therefore, personal factors need to be considered when calculating the temperature medication correction amount.

[0073] 2) Calculate the additional amount of insulin required to correct high blood sugar (also known as the insulin sensitivity coefficient)

[0074] The amount of insulin added to correct high blood sugar is defined by how much one unit of rapid-acting insulin can lower blood sugar. Generally speaking, one unit of rapid-acting insulin can lower blood sugar by 50 mg / dl (2.78 mmol / L). However, depending on individual insulin sensitivity and other conditions, it can range from 30 to 100 mg / dl (1.67-5.56 mmol / L). This insulin dosage also needs to take personal factors into consideration.

[0075] 3) Calculate the additional insulin dose based on the carbohydrate ratio in the meal

[0076] Insulin boost amount (IU) = total carbohydrates in food (g) ÷ carbohydrates converted by unit insulin (g / IU) Assuming lunch contains 60g carbohydrates, the carbohydrates converted by unit insulin is (10g / IU). Then the pre-meal insulin boost amount = 60 ÷ 10 = 6 (IU).

[0077] However, meal sizes and food types are usually not fixed, making it difficult to determine the carbohydrate content of each meal. In addition, individuals have different levels of insulin sensitivity, and the pattern of sensitivity changes over time is also different. Some people are insulin resistant in the morning, but are more sensitive to insulin at noon, so the amount of carbohydrates converted per unit of insulin should be adjusted at different meal times. For example, the basal insulin may still be 20 units, but the amount of carbohydrates converted per unit of insulin for breakfast is 8 grams, lunch may be 15 grams, and dinner may be 12 grams. Therefore, this additional insulin calculation also needs to take personal factors into consideration. When calculating the temperature drug correction amount, this can be solved through the preliminary calibration process.

[0078] (4) Insulin injection time

[0079] As mentioned above, the time of insulin injection must be determined in order to calculate the temperature correction amount of the drug at each time point after the injection. Basal insulin usually has a fixed injection time, such as the fasting period before breakfast. However, the injection time of additional insulin will vary with meal time or blood sugar peak time, and may be different every day, so the injection time needs to be carefully confirmed.

[0080] (5) Interactions between different types of insulin

[0081] According to the needs, especially for diabetic patients with uncontrolled high blood sugar, doctors will use two or even three insulin injections at different time periods to control blood sugar. The present invention assumes that there is no obvious mutual interference between various insulins. It is assumed that when different insulins are used in combination, their respective hypoglycemic effects (the amount of carbohydrates converted per unit insulin or the blood sugar lowering value per unit insulin) remain unchanged, and their metabolic kinetic characteristics remain unchanged. Therefore, when calculating the temperature drug correction amount, the effects of multiple insulins can be linearly superimposed according to their respective dosages. This provision is also suitable for the use of premixed insulin.

[0082] 3. Methods for establishing insulin metabolism kinetics equation

[0083] (1) Pharmacokinetic equation

[0084] The drug concentration in the body changes over time due to continuous elimination. The law of blood drug concentration changing over time can be expressed by a differential equation:

[0085] dC / dt=-kC n (1)

[0086] In the formula, dC / dt is the drug elimination rate; C is the drug concentration (usually represented by plasma drug concentration); t is time; k is the elimination rate constant; the negative sign indicates that the drug concentration decreases over time; n determines the kinetic type of drug elimination, n = 1, called first-order kinetics, its elimination rate is proportional to the drug concentration, and a constant proportion of the drug is eliminated per unit time (constant ratio elimination). Its curve is a straight line on a semi-logarithmic coordinate, called linear kinetics. Most drugs, including insulin, are eliminated using first-order kinetics. Characteristics of first-order kinetic elimination: ① When the drug concentration in the body is low and completely within the capacity of metabolism and excretion, the drug is eliminated at a constant ratio; ② When ordinary coordinates are used, the kinetic curve is an exponential curve. ③ Constant half-life: t 1 / 2 =0.693 / k; ④After one dose, the withdrawal time reaches 5t 1 / 2 , the dosage can eliminate 96%.

[0087] When n = 0 in equation (1), it is called the zero-order kinetic elimination equation. A few drugs are eliminated according to zero-order kinetics. Its characteristics are: ① When the drug concentration in the body is high, it exceeds the metabolic excretion capacity, and the body can only eliminate it at the maximum amount, that is, constant amount; ② When using ordinary coordinates, the kinetic curve is a straight line; ③ The half-life is not constant: t 1 / 2 =0.5C0 / k0, which is related to the initial concentration C0; ④ After a single dose, the time it takes to basically eliminate the drug from the body increases with the increase in the dosage.

[0088] Some drugs, such as aspirin and phenytoin sodium, are eliminated by zero-order kinetics when administered in large doses, and by first-order kinetics when the blood concentration drops to a common dose. This mixed elimination is called nonlinear kinetic elimination. The present invention assumes that the increase in the concentration of rapid-acting insulin in the body presents a zero-order kinetic form, and after reaching the peak, it presents a first-order kinetic form.

[0089] (2) Conditions for establishing the insulin metabolism kinetics equation

[0090] The present invention will establish the kinetic equations of insulin concentration-time for different types of insulin. The establishment of these equations follows the following principles:

[0091] 1) Time period division principle:

[0092] The present invention divides the entire kinetic process into several time periods according to the insulin injection time, onset time, peak time and duration of drug effect. To this end, it is necessary to first determine these time points:

[0093] 1. Onset time t s Determination

[0094] t s =t so +tsp (2)

[0095] where t so is the shortest onset time, t sp is the onset adjustment time. For rapid-acting insulin, regular insulin, and intermediate-acting insulin, t so The onset time can be determined according to the instructions of different brands of insulin. For example, the t sp The optional value is 10 (min). The T of a certain brand of neutral protamine zinc human insulin (NPH) sp The options are 30 (min) and so on.

[0096] 2. Peak time t f Determination

[0097] t f =t fo +pL(min) (3)

[0098] where t f0 is the shortest peak time, L is the insulin dose (IU), and p is the empirical coefficient, which represents the degree of influence of insulin dose on the delayed peak time of different types of insulin. For rapid-acting insulin, the dose has little effect, so p = 0. For other types of insulin, you can refer to the product manual for adjustment and confirmation. For example, according to the manual of a certain brand of neutral protamine zinc human insulin (NPH), the larger the dose, the longer t f The longer. When 5 units (U) are injected, t f is 240 (min). When 20 units (IU) are injected, t f is 330 (min). Substituting this into formula (3) we can deduce the t of this brand of insulin: fo =210(min), p=6(min / IU).

[0099] 3. Duration of drug effect t e Determination

[0100] t e =m+nL(min) (4)

[0101] where t e is the duration of drug effect (from injection to elimination of drug effect), L is the insulin dose, m and n are empirical coefficients, which can be determined by referring to the product instructions and clinical trials. For example, according to the instructions of a certain brand of insulin aspart, when the dose is 5 units, the drug effect can last for 4 hours, but if it is increased to 30 units, it will last for 6 hours. Substituting into formula (4), it can be calculated that m = 216 (min), n = 4.8 (min / IU).

[0102] 2) The principle of keeping the insulin level unchanged

[0103] The kinetic properties of the secretion of the human body's own pancreatic islet cells are related to meals. Under healthy conditions, the amount of insulin secretion will rapidly increase to 3-8 times the basal value 2-5 minutes after a meal, and then quickly fall back to near the basal level within a few minutes. It then slowly increases again, reaching a peak in about 1 hour (secretion rate of about 4-10IU / h), then slowly falls back, and returns to the basal level again in about 1 hour. The previous short secretion peak is the result of the rapid release of insulin stored in the pancreatic islet cells. The inventor believes that the pancreatic islet cells of patients who inject insulin cannot secrete insulin normally, and the two peaks of self-insulin secretion cannot be formed because of meals. Therefore, it is assumed that before the injection of insulin, the serum insulin level of the human body itself is a constant value C0. After the injection of insulin, the serum insulin in the body gradually increases, gradually falls back after reaching the peak, and returns to the C0 constant level when the drug effect ends. Through clinical testing, the C0 value of diabetic patients can be found.

[0104] 3) Principles for determining insulin peak

[0105] The present invention assumes that the diabetic patient adopts a normal insulin treatment plan, and after injection, the insulin concentration and change characteristics in the body can basically reach the level of a healthy person. The injection dose of different types of insulin will affect the peak time t f and duration of efficacy t e (See formulas (3) and (4)), but does not affect the onset time t s . The t between different insulins s The difference is large and can be found by consulting the product instructions. Insulin peak C f It is directly proportional to the injected dose.

[0106] 4) Principle of linear superposition of insulin characteristics

[0107] The present invention assumes that after two or more insulins are mixed and injected, their metabolic kinetic characteristics are equivalent to the weighted average results of the characteristics of various insulins. For example, Novolin 30R is a mixture of 30% short-acting insulin and 70% intermediate-acting insulin: according to the weighted average method, the duration of drug effect t e =0.3t es +0.7t em =0.3*8+0.7*20=16.4(h), where t es is the duration of action of short-acting insulin, t em It is the duration of action of intermediate-acting insulin.

[0108] (3) Kinetic equation of insulin metabolism

[0109] The present invention describes the entire kinetic process by the mathematical relationship between the plasma insulin concentration and time in vivo.

[0110] 1) Kinetic equation of rapid-acting insulin

[0111] 1. Kinetic equation before onset

[0112] C(t)=C0(0≤t <t s ) (5)

[0113] Among them, C(t) is the insulin concentration at time t, and C0 is the basal insulin concentration in the body.

[0114] 2. Kinetic equation from onset to peak

[0115] The present invention assumes that the zero-order kinetic growth law is followed from the onset of insulin to the peak:

[0116] C(t)=ɑ+βt(t s ≤t <t f ) (6)

[0117] Among them, ɑ and β are proportional constants, which are related to t s ,t f , C0 and insulin peak C f Related.

[0118] ɑ=C0–(C f -C o )t s / (t f –t s ) (7)

[0119] β=(C f -C o ) / (t f –t s ) (8)

[0120] 3. Kinetic equation from peak to end of efficacy

[0121] The present invention assumes that the elimination rate of rapid-acting insulin after the peak appears follows the first-order kinetic law, and it can be deduced from equation (1):

[0122] C(t)=λ+γe -kt (t s ≤t <t e ) (9)

[0123] Where k is the insulin metabolism rate (i.e. drug elimination rate) constant. It can be determined by referring to the product manual and clinical trials. Λ and γ are proportional constants, which are proportional to k,t e ,t f,C0 and C f Related.

[0124] Λ=C f –(C f -C o ) -ktf / (e -ktf -e -kte ) (10)

[0125] Γ=(C f -C o ) -ktf / (e -ktf -e -kte ) (11)

[0126] 2) Kinetic equations of short-acting (regular) insulin and intermediate-acting insulin

[0127] 1. Kinetic equation before onset

[0128] C(t)=C0(0≤t <t s ) (12)

[0129] Among them, C(t) is the insulin concentration at time t, and C0 is the basal insulin concentration in the body.

[0130] 2. Kinetic equation from onset to end of efficacy

[0131] The present invention uses the mathematical form of the log-normal distribution function to describe the kinetic characteristics of short-acting and intermediate-acting insulin.

[0132] Assume that F(x) obeys lognormal distribution, and its density function is: F(x) = Exp[-(lnx–μ) 2 / (2σ 2 )] / [(2π) 0.5 σx](x>0) (13)

[0133] Its mathematical expectation and variance are: E(x) = Exp(μ+σ 2 / 2) (14)

[0134] D(x)=(Expσ 2 –1)Exp(2μ+σ 2 ) (15)

[0135] Among them, σ and μ are two important parameters, which determine the shape of the lognormal distribution curve. Figure 2 This is the log-normal distribution curve for different σ under the condition that μ remains unchanged. Figure 1 It can be seen that when σ is around 1, it is close to the short-acting insulin kinetic curve. When σ is around 1 / 2, it is close to the medium-acting insulin kinetic curve.

[0136] However, equation (13) cannot be used directly to describe the kinetic characteristics of insulin metabolism. The modified kinetic equation is:

[0137] C(t)=ω(C f -C o )Exp[-(lnt–lnt f ) 2 / (2σ 2 )] / [(2π) 0.5 σt]+C o (t s ≤t <t e ) (16)

[0138] Among them, σ and ω are empirical parameters. Their value range is related to the type of insulin. For short-acting (regular) insulin, σ = 0.9-1.1, ω = 1.52-1.55. For intermediate-acting insulin, σ = 0.4-0.6, ω = 0.91-1.16. In addition, for intermediate-acting insulin, there is a horizontal period close to 0 before the kinetic equation curve rises significantly (see Figure 2 The ratio of the duration to the peak time is about 1:4. The ratio of the onset time to the peak time of intermediate-acting insulin is also about 1:4. Therefore, it can be considered that the kinetic curve of intermediate-acting insulin completely covers the entire process after injection, and the horizontal period at the beginning of the curve is equal to the drug onset period. In other words, the kinetic equation before the onset of intermediate-acting insulin can be ignored and directly described by equation (16).

[0139] 3) Kinetic equation of long-acting insulin

[0140] Long-acting insulin takes effect slowly, and the concentration in the body gradually rises to a certain level within about 2 hours, and then presents a stable plateau, which lasts for up to 24 hours. Therefore, its kinetic equation is relatively simple and is also divided into three stages:

[0141] 1. Kinetic equation before onset

[0142] C(t)=C0(0≤t <t s ) (17)

[0143] Among them, C(t) is the insulin concentration at time t, and C0 is the basal insulin concentration in the body.

[0144] 2. Kinetic equation from onset to insulin plateau

[0145] The present invention assumes that the zero-order kinetic growth law is followed during this period:

[0146] C(t)=ρ+φt(ts ≤t <t f ) (18)

[0147] Among them, t f represents the time to reach the insulin concentration plateau, ρ and φ are proportional constants, and t s ,t f , C0 and insulin plateau concentration C f Related.

[0148] ρ=C0–(C f -C o )t s / (t f –t s ) (19)

[0149] φ=(C f -C o ) / (t f –t s ) (20)

[0150] 3. Kinetic equations from the appearance of insulin plateau to the end of drug efficacy

[0151] C(t)=C f (t f ≤t <t e ) (twenty one)

[0152] 4) Kinetic equations for premixed and mixed insulin

[0153] For this type of insulin, the kinetic equation can be established by using the kinetic equations of different insulins according to the aforementioned linear superposition principle of insulin characteristics and by weighted averaging.

[0154] 4. Calculation method of temperature drug correction amount

[0155] The above kinetic equations (1)-(21) are universal equations based on ideal conditions. For a specific individual and the insulin product used, one must consider one's own insulin sensitivity and product specificity. Directly using universal equations for temperature-drug correction may result in deviations. For this reason, a pre-calibration method is required for the individual and the drug used before temperature-drug correction can be performed. The steps are as follows:

[0156] (1) Establishing a general correction formula

[0157] ΔT(t)=a C(t) 2 +b C(t)+c (22)

[0158] Among them, ΔT(t) is the correction amount of body temperature at time t after insulin injection, and C(t) is the insulin level in the body at time t. According to the type of insulin used, the injection dose and the injection time, the appropriate equation can be selected from the above kinetic equation for calculation. a(℃ / (IU) 2 ), b (℃ / IU) and c (℃) are empirical parameters related to individuals and insulin products.

[0159] (2) Establishing a personal correction formula through pre-calibration

[0160] Pre-calibrate at the same time period for two consecutive days (e.g., from the injection of insulin before breakfast to several hours after the meal). The eating time, food type and amount of food should remain basically the same for the two days, and the daily routine (such as work, exercise, shower) should also be as similar as possible.

[0161] 1) Do not inject insulin on the first day. Select three time points within a certain period of time (with an interval of no less than 30 minutes). Use a thermometer with an accuracy of no less than 0.02°C (preferably oral temperature) to measure the individual's body temperature (preferably oral temperature). Use an invasive blood glucose meter with high accuracy to measure blood glucose at these three time points. Record the measurement data and measurement time.

[0162] 2) The next day, insulin is injected according to the doctor's instructions during this time period, and the body temperature and blood sugar are measured three times according to the measurement time of the first day, and the measurement data are recorded. The three measurement times are converted into the time value after the injection of insulin. For example, if insulin is injected at 7:00 and the body temperature and blood sugar are measured at 9:30, the time value this time is 150 minutes.

[0163] 3) Based on the temperature data measured on the first day, the metabolic heat integration method is used to calculate the blood glucose values ​​at three time points and compared with the invasive blood glucose values. If the relative errors do not exceed 15%, the blood glucose calculation method of the metabolic heat integration method is recognized, otherwise it needs to be adjusted.

[0164] 4) Based on the blood sugar data measured on the second day, the metabolic heat integration method was used to infer the body temperature data at three time points.

[0165] 5) Subtract the body temperature at the three time points inferred from the body temperature at the three times measured on the second day. The difference obtained is the correction amount of the individual's body temperature at these three time points.

[0166] 6) According to the type, dosage and injection time of injected insulin, find the corresponding insulin kinetic equation, look up the insulin product manual, and determine the empirical parameters in the equation.

[0167] 7) Substituting the body temperature correction values ​​at the three time points, the time value after insulin injection, and the corresponding kinetic equation into the general correction formula (22), three three-variable linear equations can be obtained. There are only three unknowns in these equations, namely a, b and c. Through simple solution, the specific values ​​of these three empirical parameters are obtained.

[0168] 8) Substituting the three empirical parameter values ​​into formula (22), a personal temperature medication correction formula suitable for the insulin brand and personal constitution is obtained.

[0169] 9) After calibration, the original metabolic heat integration method can be used to calculate blood sugar at any time based on the body temperature and drug correction, and the calculation result should be accurate enough.

[0170] 10) If you change the type of insulin, you need to recalibrate according to the above steps.

[0171] (3) Calculation example

[0172] Li, female, 59 years old, type 2 diabetes. She followed the doctor's advice to use intermediate-acting low-protamine insulin to control blood sugar. No insulin was used on the first day, and 18 units of insulin were injected before breakfast on the second day. The body temperature and blood sugar data measured at the same time on the two days are shown in Table 1.

[0173] Table 1. Body temperature and blood sugar data measured at the same time within two days

[0174]

[0175] Taking the oral temperature and blood glucose values ​​at 80, 270, and 360 minutes after insulin injection, the temperature correction values ​​at the three time points were calculated as ΔT(80) = 0.11°C, ΔT(270) = 1.34°C, and ΔT(360) = 1.17°C according to the above method. Substituting them into equation (22), we get the three-variable linear equation system:

[0176] 0.11 = a C (80) 2 +b C(80)+c

[0177] 1.34 = a C (270) 2 +b C(270)+c

[0178] 1.17 = a C (360) 2 +b C(360)+c

[0179] Set σ = 0.5, ω = 1.0, t = 80, t f =240, C f=18, C0=0 is substituted into equation (16), and we get C(80)=1.28. Similarly, substituting σ=0.5, ω=1.0, t=270, t f =240, C f =18, C0=0 is substituted into equation (16), and we get C(270)=13.97. Then, σ=0.5, ω=1.0, t=360, t f =240, C f =18, C0=0 are substituted into equation (16), and we get C(360)=10.33, thus obtaining a three-variable equation system containing only three parameters a, b, and c:

[0180] 0.11 = 1.64a + 1.28b + c

[0181] 1.34 = 195.16a + 13.97b + c

[0182] 1.17 = 106.7a + 10.33b + c

[0183] Solving the equation yields: a = -0.0055, b = 0.181, c = -0.113. Therefore, the temperature medication correction equation suitable for this volunteer using this brand of intermediate-acting insulin is:

[0184] ΔT(t)=-0.0055 C(t) 2 +0.181 C(t)–0.113 (23)

[0185] According to equation (23), the body temperature correction value at any time after the volunteer injected insulin can be calculated. The corrected body temperature is used in the metabolic heat integration method to calculate a more accurate blood sugar value. To verify the accuracy of equation (23), the volunteer underwent several body temperature and blood sugar tests on the third day. The results are shown in Table 2.

[0186] Table 2. Test results of body temperature and blood sugar of volunteers on the third day

[0187]

[0188] It can be seen from Table 2 that after the body temperature is corrected by the method of the present invention, the accuracy of blood glucose calculation using the non-invasive metabolic heat integration method is significantly improved.

[0189] 5. Calculation method of temperature correction dose for oral hypoglycemic drugs

[0190] Generally speaking, the effect of oral hypoglycemic drugs on body temperature is much smaller than that of insulin. You will not feel obvious chills after taking the medicine. Heat production and body temperature set point are basically unchanged. Therefore, the amount of temperature drug correction is very small. Nevertheless, if chills and changes in heat production and body temperature set point occur in the case of large doses of medication or intravenous infusion, you can still refer to the body temperature correction method after insulin injection to improve the accuracy of non-invasive blood glucose testing.

[0191] The above embodiments are merely descriptions of preferred implementation modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for improving the accuracy of non-invasive blood glucose metabolism heat integration method detection, characterized in that The following steps are involved: Step 1: Establish the metabolic kinetic equation of hypoglycemic drugs; Step 2: Establish a general formula for body temperature correction based on the metabolic kinetic equation; Step 3: Use a pre-calibration method to establish a temperature correction calculation formula suitable for the individual and the medication he / she uses; Step 4: Use the corrected body temperature after deducting the personal body temperature correction amount to calculate the personal blood glucose value in the metabolic heat integration method to eliminate the influence of hypoglycemic drugs on the accuracy of blood glucose detection; The hypoglycemic drug is premixed and mixed insulin, and the metabolic kinetic equation of the premixed and mixed insulin is established by superimposing the metabolic kinetic equations of rapid-acting insulin, short-acting insulin or intermediate-acting insulin, and long-acting insulin, and then weighted averaging the respective proportions; The general formula for body temperature correction is ΔT(t) = a C(t) 2 +b C(t)+c, where ΔT(t) is the correction value of body temperature at time t after insulin injection, C(t) is the insulin concentration in the body at time t, a, b and c are empirical parameters related to individuals and insulin products, and the units of a, b and c are ℃ / (IU) 2 , ℃ / IU and ℃; The steps of establishing a temperature correction calculation formula suitable for an individual and the medication he / she uses by pre-calibration include: Step 1: Do not inject insulin on the first day. Select three time points within a certain period of time, with an interval of no less than 30 minutes between the time points, measure your body temperature and blood sugar, and record the measurement data and measurement time; Step 2: Inject insulin at the same time on the second day, and measure body temperature and blood sugar three times according to the measurement time on the first day, record the measurement data, and convert the three measurement times into the time value after insulin injection; Step 3: Based on the temperature data measured on the first day, the blood glucose values ​​at three time points are calculated using the metabolic heat integration method and compared with the invasive blood glucose values. If the relative errors do not exceed 15%, the blood glucose calculation method of the metabolic heat integration method is recognized, otherwise it needs to be adjusted; Step 4: Based on the blood sugar data measured on the second day, the body temperature data at three time points are inferred using the metabolic heat integration method; Step 5: Subtract the body temperature at the three time points inferred from the body temperature at the three times measured on the second day, and the difference is the correction amount of the individual's body temperature at the three time points; Step 6: Find the corresponding insulin metabolism kinetic equation according to the type, dosage and injection time of injected insulin; Step 7, substituting the obtained body temperature correction values ​​at three time points, the time value after insulin injection and the corresponding kinetic equation into the general body temperature correction formula to obtain the specific values ​​of the three empirical parameters a, b and c; Step 8: Substitute the three empirical parameter values ​​of a, b and c into the general temperature correction formula to obtain a temperature correction calculation formula suitable for the insulin brand and individual constitution.

2. The method for improving the accuracy of non-invasive blood glucose metabolism heat integration method detection according to claim 1, characterized in that: The hypoglycemic drug is rapid-acting insulin, and the metabolic kinetic equation of rapid-acting insulin from the onset of action to the peak appearance stage is C(t)=ɑ+βt, where ɑ and β are proportional constants; the metabolic kinetic equation of rapid-acting insulin from the peak appearance to the end of the drug effect is C(t)=λ+γe -kt , where k is the metabolic rate constant of rapid-acting insulin, and λ and γ are proportionality constants.

3. The method for improving the accuracy of non-invasive blood glucose metabolism heat integration method detection according to claim 1, characterized in that: The hypoglycemic drug is short-acting insulin or intermediate-acting insulin. The metabolic kinetic equation of short-acting insulin or intermediate-acting insulin from the onset of action to the end of its effect is C(t)=ω(C f -C o )Exp[-(lnt–lnt f ) 2 / (2σ 2 )] / [(2π) 0.5 σt]+C o , where σ and ω are empirical parameters, C f is the peak insulin concentration, C o is the basal insulin concentration in the body, t f Peak occurrence time; for short-acting insulin, σ=0.9~1.1, ω=1.52~1.55; for intermediate-acting insulin, σ=0.4~0.6, ω=0.91~1.16, the onset time of intermediate-acting insulin is 0.

4. The method for improving the accuracy of non-invasive blood glucose metabolism heat integration method detection according to claim 1, characterized in that: The hypoglycemic drug is long-acting insulin, and the metabolic kinetic equation of long-acting insulin from the onset of action to the appearance of the platform area is: where ρ and is the proportional constant; the metabolic kinetic equation of long-acting insulin from the appearance of the plateau to the end of its efficacy is C(t) = C f , where C f is the insulin plateau concentration.

5. The method for improving the accuracy of non-invasive blood glucose metabolism heat integration method detection according to any one of claims 2 to 4, characterized in that: The relationship between the peak insulin time and the insulin dose is t f =t fo +pL, where t f0 is the shortest peak time, L is the insulin dose, and p is the empirical coefficient, which represents the degree of influence of insulin dose on the delayed peak time of different types of insulin. For rapid-acting insulin, p=0.

6. The method for improving the accuracy of non-invasive blood glucose metabolism heat integration method detection according to any one of claims 2 to 4, characterized in that: The relationship between the end time of drug effect and insulin dose is t e =m+nL, where t e is the end time of drug effect, L is the insulin dose, and m and n are empirical coefficients.

Images