Multi-parameter correction non-intrusive high-precision continuous blood glucose monitoring method

Abstract

The invention provides a multi-parameter correction non-intrusive high-precision continuous blood glucose monitoring method, which is used for implementing dual real-time correction on glucose signals by synchronously detecting pH and sodium ion concentration on the surface of skin. PH correction is divided into two parts: 1, the influence of pH on the activity of glucose oxidase is compensated by utilizing an in-vitro calibration curve, so that the sensitivity of the sensor is corrected; secondly, a quantitative model of pH-Zeta potential-electroosmosis rate is established, and fluctuation of glucose extraction flux caused by skin electrical property difference is corrected. Sodium ion correction takes the assumption that the glucose / sodium ion flux ratio is equal to the tissue fluid concentration ratio as the assumption, and sodium ion flux is introduced to correct extraction flux fluctuation caused by individual differences such as skin impedance in real time. Through pH-sodium ion dual joint correction, extraction flux fluctuation and enzyme activity drift are compensated synchronously under a non-invasive condition, and continuous blood glucose monitoring with clinical precision is expected to be realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of counter-ion electroosmosis and intercellular fluid biomarker detection, and particularly relates to a multi-parameter corrected non-invasive high-precision continuous blood glucose monitoring method. BACKGROUND

[0002] Early intervention and precise management of diabetes and its complications highly depend on continuous and accurate monitoring of blood glucose level. Traditional fingertip blood sampling detection not only causes continuous trauma to patients, but also hinders the implementation of high-frequency and personalized treatment strategies. Therefore, non-invasive, continuous and high-precision blood glucose monitoring technology has become the focus of global academia and industry. In recent years, with the development of flexible electronics, nanomaterials and other technologies, non-invasive schemes represented by counter-ion electroosmosis are considered to be one of the most clinically transformative technical routes. Counter-ion electroosmosis drives the migration of ions in subcutaneous intercellular fluid by applying an alternating / direct current composite electric field on the skin surface, and forms an electroosmotic flow, thereby extracting glucose containing intercellular fluid to the skin surface without puncture, and then realizing real-time detection by combining electrochemical methods. This non-invasive technology provides a long-term, wearable continuous monitoring means for patients with chronic diseases such as diabetes and metabolic syndrome.

[0003] However, counter-ion electroosmosis technology still faces a series of key bottlenecks from the laboratory to the clinic. First, the intercellular fluid extraction flux is strongly affected by the dynamic changes of the skin microenvironment: the thickness, hydration, lipid composition of the stratum corneum, sweat gland activity and the change of skin surface pH (related to skin electrification) will all cause fluctuations in ion mobility and electroosmotic coefficient, thereby affecting the extraction flux of the target molecule, i.e. glucose. Second, the skin properties vary significantly among individuals (mainly skin impedance), making it difficult to ensure the consistency of extraction for different subjects or different anatomical sites of the same subject with fixed current / voltage parameters, ultimately leading to uncontrollable amplification of blood glucose prediction error. Finally, traditional algorithms mostly rely on a single current signal for concentration inversion, and the relationship between the extracted glucose concentration and the subcutaneous intercellular fluid glucose concentration is simplified as a linear relationship, lacking a multi-dimensional compensation mechanism for skin physiological state and individual differences, thereby limiting the accuracy and reliability of counter-ion electroosmosis technology in real clinical scenarios. SUMMARY

[0004] The technical problem to be solved by the present application is to provide a multi-parameter corrected non-invasive high-precision continuous blood glucose monitoring method.

[0005] The technical solution adopted by the present application is:

[0006] The application discloses a multi-parameter correction non-invasive high-precision continuous blood glucose monitoring method, which comprises a skin surface pH correction and a sodium ion correction combined blood glucose prediction algorithm, wherein the skin surface pH correction comprises two parts of a glucose sensor sensitivity correction caused by enzyme activity change and a glucose extraction flux correction caused by skin electrification difference; the glucose sensor sensitivity correction caused by enzyme activity change is obtained by fitting a sensor sensitivity relation curve under different pH conditions through in-vitro experiment, and then the sensitivity correction is performed according to the relation curve; the glucose extraction flux correction caused by skin electrification difference is obtained by fitting a glucose extraction flux relation curve under different pH conditions through in-vitro experiment, and then the glucose extraction flux correction is performed; and the sodium ion correction is based on the assumption that the ratio of the glucose and sodium ion extraction flux and the ratio of the subcutaneous glucose and sodium ion concentration are constant, and a blood glucose prediction formula containing the sodium ion correction is obtained through theoretical derivation Then, the sodium ion correction is performed.

[0007] In the multi-parameter correction non-invasive high-precision continuous blood glucose monitoring method, the glucose sensor sensitivity correction is caused by the fact that pH can affect the activity of glucose oxidase and then affect the sensitivity of the glucose sensor, so that the sensitivity correction is needed; the glucose extraction flux correction caused by skin electrification difference is caused by the fact that the change of the skin surface pH can affect the Zeta potential of the skin horny layer and then affect the electroosmotic rate of the glucose in the intercellular fluid of the subcutaneous cells and then affect the glucose extraction flux, so that the correction is needed; and the sodium ion correction is caused by the fact that the glucose extraction depends on the electromigration of the counter ion (mainly sodium ion) opposite to the skin electrification property (skin electrification is negative), the sodium ion extraction flux indirectly reflects the glucose extraction flux, so that the correction is needed to solve the problem of the glucose extraction flux fluctuation caused by the sodium ion extraction flux fluctuation caused by the individual difference of the skin impedance, and the sodium ion correction solves the problem of the glucose extraction flux fluctuation caused by the sodium ion extraction flux fluctuation caused by the individual difference of the skin impedance.

[0008] Preferably, in the multi-parameter correction non-invasive high-precision continuous blood glucose monitoring method, the glucose sensor sensitivity correction compensates the influence of pH on the activity of glucose oxidase by using an in-vitro calibration curve to correct the sensor sensitivity, and the specific method is that the sensitivity of the glucose sensor is measured under different pH conditions (for example, pH 4, 5, 6, 7, 8, 9 and 10) through in-vitro experiment, then a relation curve of the sensitivity and pH is fitted by least square method, and finally the sensitivity under the pH condition at the moment is obtained by bringing the pH value into the relation curve to realize the glucose sensor sensitivity correction during human body test.

[0009] Preferably, the multi-parameter corrected non-invasive high-precision continuous blood glucose monitoring method corrects the glucose extraction flux to establish a quantitative model of pH-Zeta potential-osmotic velocity, and corrects the fluctuation of glucose extraction flux caused by the difference in skin electrical properties. The specific method is as follows: the Zetasizer Nano nanoparticle size potential instrument is used to measure the Zeta potential of human skin stratum corneum (keratin) under different pH conditions, and the relationship between pH and Zeta potential is obtained; the glucose extraction flux under different pH conditions is obtained through in vitro simulation extraction experiment, and the relationship between pH and glucose extraction flux (which can be further expressed as osmotic velocity) is fitted; and the quantitative model of pH-Zeta potential-osmotic velocity is established by comprehensively considering the two fitting relationship curves.

[0010] Preferably, the multi-parameter corrected non-invasive high-precision continuous blood glucose monitoring method corrects the sodium ion correction, which assumes that the ratio of the glucose / sodium ion flux ratio and the tissue fluid concentration ratio is constant, and introduces sodium ion flux to correct the extraction flux fluctuation caused by individual differences in skin impedance. The specific method is described in the reference: Anke Sieg, Richard H Guy, M Delgado-Charro, Noninvasive Glucose Monitoring by Reverse Iontophoresis in Vivo: Application of the Internal Standard Concept, Clinical Chemistry, Volume 50, Issue 8, 1 August 2004, Pages 1383-1390.

[0011] Preferably, the multi-parameter corrected non-invasive high-precision continuous blood glucose monitoring method has the following specific steps:

[0012] (1) The sodium ion sensor is calibrated through in vitro experiment to obtain the sensitivity formula of the sodium ion sensor; the pH sensor is calibrated through in vitro experiment to obtain the sensitivity formula of the pH sensor; the glucose sensor is calibrated under different pH conditions (for example, pH 4, 5, 6, 7, 8, 9, and 10) through in vitro experiment, and the sensitivity of the glucose sensor under different pH conditions is measured, and then the relationship curve between sensitivity and pH is obtained by least squares fitting to obtain the formula k1=f(pH);

[0013] (2) According to the sensitivity formulas of the pH sensor and the sodium ion sensor, the voltages measured by the pH sensor and the sodium ion sensor are converted into pH values and sodium ion concentrations, respectively;

[0014] (3) The glucose sensor sensitivity at this pH value is calculated according to the formula k1 = f (pH), and the measured current is converted into glucose concentration according to the sensitivity at this pH value, so as to realize the sensitivity correction of pH to the glucose sensor;

[0015] (4) The pH value, sodium ion concentration and glucose concentration in the two continuous extraction periods are measured.

[0016] (5) In the process of reverse iontophoretic extraction of subcutaneous interstitial fluid, the difference between the two adjacent extraction periods represents the net flux extracted in a single extraction cycle, the pH values measured in the two continuous periods are averaged, the sodium ion concentration is subtracted, the glucose concentration is subtracted, and the results are brought into the formula to obtain the high-precision blood glucose prediction value after multi-parameter correction.

[0017] The beneficial effects of the present application are:

[0018] The multi-parameter corrected non-invasive high-precision continuous blood glucose monitoring method corrects the glucose signal by the combination of skin surface pH and sodium ion, implements double real-time correction, synchronously compensates the extraction flux fluctuation and enzyme activity drift under non-invasive conditions, and is expected to realize non-invasive continuous blood glucose monitoring with clinical precision, overcome the fluctuation of glucose extraction flux caused by the change of skin surface pH and the difference of skin properties, and realize high-precision blood glucose prediction with multi-parameter correction. Specifically, in the process of reverse iontophoresis, the movement of hydrogen ions and the skin self-healing mechanism may cause the change of skin surface pH, thereby affecting the extraction rate of interstitial fluid and the accuracy of glucose concentration measurement. The pH correction not only solves the problem of the drift of enzyme glucose sensor sensitivity with pH, but also solves the problem of the fluctuation of glucose extraction flux caused by the change of skin electrical properties in the process of reverse iontophoretic extraction of interstitial fluid. On the other hand, due to the large fluctuation of skin impedance affected by hydration, skin thickness and environmental factors, even if the external extraction conditions (such as extraction current density, extraction time, etc.) are unchanged, the extraction flux of glucose will also fluctuate. As a relatively stable parameter in interstitial fluid, the sodium ion concentration can indirectly reflect the change of skin impedance in the extraction process, and then dynamically correct the glucose extraction flux, reduce the measurement deviation caused by individual differences such as impedance fluctuation, and further improve the accuracy of non-invasive blood glucose prediction based on reverse iontophoresis. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 is the principle diagram of the multi-parameter corrected non-invasive high-precision continuous blood glucose monitoring method of the present application;

[0020] Figure 2 is the flow chart of the multi-parameter corrected non-invasive high-precision continuous blood glucose monitoring method of the present application;

[0021] Figure 3 This is an experimental result diagram of the non-invasive high-precision continuous blood glucose monitoring method with multi-parameter correction described in this invention. Detailed Implementation

[0022] To further illustrate the present invention, the technical solution will be clearly and completely described in conjunction with the following embodiments.

[0023] Example 1

[0024] like Figure 1 As shown, the principle of the multi-parameter correction non-invasive high-precision continuous blood glucose monitoring method is to improve the accuracy of blood glucose prediction through combined pH correction and sodium ion correction. The principle of pH correction is as follows: during the reverse ion electroosmosis extraction of subcutaneous intercellular fluid, H... + / OH - The high mobility of glucose molecules causes them to accumulate rapidly on the skin surface, altering the skin's pH value. The skin's self-protective mechanisms hinder this change, resulting in fluctuations in skin surface pH during counterion electroosmosis extraction. These fluctuations affect the sensitivity of the enzyme glucose sensor, thus impacting subsequent measurement accuracy. Furthermore, they affect the zeta potential of the electrical double layer of the stratum corneum. This potential influences the concentration of oppositely charged ions (counterions, such as sodium ions) in the diffusion layer of the double layer, affecting the migration flux of counterions in the diffusion layer. Since the primary driving force for glucose electroosmosis extraction is the migration of counterions in the diffusion layer, this further affects the glucose extraction flux, thus impacting the accuracy of subsequent blood glucose prediction. The principle of sodium ion correction is as follows: During the counterion transdermal extraction of glucose molecules from subcutaneous intercellular fluid, skin impedance is a crucial factor affecting the transdermal extraction flux. Under the assumption that the ratio of glucose to sodium ion extraction flux and the ratio of subcutaneous glucose to sodium ion concentration are constant, monitoring sodium ion extraction flux can correct for fluctuations in glucose extraction flux caused by skin impedance, thereby improving the accuracy of subsequent blood glucose prediction.

[0025] Example 2

[0026] The derivation process of the non-invasive, high-precision continuous blood glucose monitoring method with multi-parameter correction described in this invention is as follows:

[0027] Steady-state response current I of glucose sensor glu (t) and glucose concentration c glu The relationship between (t) and the sensitivity curve of the glucose sensor can be expressed by equation (1).

[0028] I glu (t)=k1·c glu (t)+b glu (1)

[0029] Where k1 is the sensitivity of the glucose sensor, b glu This represents the baseline measured by the glucose sensor in blank PBS solution. Since the sensitivity of the enzyme-mediated glucose sensor is significantly affected by pH, the sensitivity k1 of the glucose sensor varies at different pH values, which can be expressed as:

[0030] k1=f(pH) (2) In the process of reverse ion electroosmosis extraction of subcutaneous intercellular fluid, the sodium ion extraction flux, that is, the molar amount of sodium ions extracted per unit time J Na+ , can be represented as:

[0031]

[0032] Where k2 represents the sodium ion concentration in the extracted intercellular fluid and c represents the sodium ion concentration in the subcutaneous intercellular fluid. iNa+ (Normal proportionality coefficient is generally 140mM), A is the area of ​​the extraction electrode, u Na+ U represents the sodium ion electromigration rate. Na+ It can be represented as:

[0033]

[0034] Among them, z Na+ The charge of the sodium ion is (+1), E is the electric field strength generated by the applied extraction current, and μ is the value of the sodium ion. Na+ is the sodium ion electromobility. According to the Nernst-Einstein equation, μ Na+ It can be represented as:

[0035]

[0036] Where e is the electron charge, D Na+ Let be the sodium ion diffusion coefficient, k be the Boltzmann constant, and T be the thermodynamic temperature. According to equations (4) and (5), when the temperature and the applied electric field strength remain constant, the electromigration velocity u of sodium ions is... Na+ The value is a constant. In the process of reverse ion electroosmosis extraction of subcutaneous intercellular fluid, the extraction mechanism of glucose is mainly electroosmosis, and its extraction flux J is [value missing]. glu It can be represented as

[0037] J glu =k3·A·u glu ·c iglu (6)

[0038] In equation (6), k3 represents the glucose concentration in the extracted intercellular fluid and c represents the glucose concentration in the subcutaneous intercellular fluid. iglu The proportionality coefficient, u glu The glucose electroosmotic rate can be expressed as:

[0039] u glu =μ glu ·E (7)

[0040] Where, μ glu Representing the electroosmotic mobility of glucose, i.e., the rate of electroosmosis of glucose under a given electric field, according to the Helmholtz-Smoluchowski equation, μ glu It can be represented as:

[0041]

[0042] In equation (8), ε₀ is the vacuum permittivity, ε₀ is the relative permittivity, ζ is the Zeta potential of the electrical double layer in the stratum corneum of the epidermis, and η is the viscosity of the intercellular fluid. The relationship between the Zeta potential and pH can be expressed as:

[0043] ζ=k4·pH (9)

[0044] Combined equations (7)(8)(9), glucose electroosmosis rate u glu It can be represented as:

[0045] u glu =k5·pH (10)

[0046] in, And it remains a constant in subsequent calculations. Sodium ion extraction flux (J) Na+ On the other hand, the amount of sodium ions extracted can be used to infer the result:

[0047]

[0048] Among them, c Na+ (t) represents the sodium ion concentration extracted to the sensor hydrogel collection area, V is the volume of the hydrogel, and T is the extraction time of a single extraction cycle. Similarly, the glucose extraction flux J in a single extraction cycle... glu It can also be expressed as:

[0049]

[0050] Among them, c glu (t) represents the glucose concentration extracted to the hydrogel collection area of ​​the sensor. Combining equations (3)(6)(10)(11)(12), the formula for predicting the glucose concentration in subcutaneous tissue fluid can be expressed as:

[0051]

[0052] in, The value is a constant during the calculation and requires initial calibration via finger-prick blood sampling. Na+ and cglu The difference between two consecutive measurements is represented by , and pH is the average of the two consecutive measurements. Since the glucose concentration in subcutaneous ISF closely correlates with blood glucose levels within a certain range, the high-precision blood glucose prediction algorithm, after pH and sodium ion joint correction, can be expressed as follows, without considering the molecular diffusion delay between blood and ISF:

[0053]

[0054] Example 3

[0055] like Figure 2 As shown, the specific steps of the multi-parameter correction non-invasive high-precision continuous blood glucose monitoring method are as follows:

[0056] Step S1: The sodium ion sensor is calibrated through in vitro experiments to obtain the sensitivity formula of the sodium ion sensor; the pH sensor is calibrated through in vitro experiments to obtain the sensitivity formula of the pH sensor; the glucose sensor is calibrated through in vitro experiments under different pH conditions (e.g., pH 4, 5, 6, 7, 8, 9, 10), and the sensitivity of the glucose sensor under different pH conditions is measured. Then, the relationship curve between sensitivity and pH is obtained by least squares fitting to obtain the formula k1 = f(pH).

[0057] Step S2: Based on the sensitivity formulas of the pH sensor and the sodium ion sensor, convert the voltages measured by the pH sensor and the sodium ion sensor into pH value and sodium ion concentration, respectively.

[0058] Step S3: Since the sensitivity of the enzyme glucose sensor is greatly affected by pH, the sensitivity of the glucose sensor at this pH value is first calculated according to formula (2), and the measured current is converted into glucose concentration according to the sensitivity at this pH value. This process is the pH sensitivity correction of the glucose sensor.

[0059] Step S4: Measure the pH, sodium ion concentration, and glucose concentration over two consecutive extraction cycles.

[0060] Step S5: During the reverse ion electroosmosis extraction of subcutaneous intercellular fluid, the difference between two adjacent extraction cycles represents the net throughput extracted in a single extraction cycle. Therefore, the average pH value measured in two consecutive cycles is taken, the difference in sodium ion concentration and glucose concentration is taken, and the results are substituted into equation (14) to obtain the high-precision blood glucose prediction value after multi-parameter correction.

[0061] Example 4

[0062] The accuracy and precision of the method described in Example 3 were verified using the non-invasive high-precision blood glucose sensor described in CN119846039A. The verification method is as follows:

[0063] The oral glucose tolerance test (OGTT) was used to test blood glucose levels in 6 healthy individuals and 15 patients with type 2 diabetes. Reference blood glucose values ​​for healthy individuals were obtained by collecting finger-prick blood using a commercial blood glucose meter (ACCU-CHEK, Roche, Switzerland), while reference values ​​for diabetic patients were determined by venous blood sampling. Before the test, the sensor attachment site was disinfected with an alcohol swab. Then, the non-invasive, high-precision blood glucose sensor, based on the patent CN119846039A, was attached to the skin surface, and continuous monitoring was initiated. Thirty minutes after the start of monitoring, each subject orally ingested 250 mL of an aqueous solution containing 75 g of glucose. The entire test lasted 2.5 hours. After the test, the predicted blood glucose value was obtained using the multi-parameter calibration method described in this invention. The performance of the sensor in actual blood glucose monitoring was evaluated by comparing the difference between the predicted value and the actual reference value. The mean absolute relative error (MARD) is a key indicator for evaluating the accuracy of blood glucose monitoring; the smaller the value, the higher the detection accuracy of the sensor. Figure 3 As shown in the test results, the MARD value of blood glucose obtained by the multi-parameter correction proposed in this invention is lower than that of no correction and sodium ion correction alone, which confirms that the method is more conducive to high-precision blood glucose prediction.

[0064] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A non-invasive, high-precision continuous blood glucose monitoring method with multi-parameter correction, characterized in that: This algorithm combines skin surface pH correction and sodium ion correction. Skin surface pH correction includes two parts: sensitivity correction of the glucose sensor due to changes in enzyme activity and correction of glucose extraction throughput due to differences in skin charge. The sensitivity correction due to changes in enzyme activity is achieved by fitting sensitivity curves of the sensor under different pH conditions in in vitro experiments, and then performing sensitivity correction based on these curves. The correction of glucose extraction throughput due to differences in skin charge is achieved by fitting glucose extraction throughput curves under different pH conditions in in vitro experiments, and then performing glucose extraction throughput correction. The algorithm is then used in a blood glucose prediction formula that includes sodium ion correction. Then sodium ion correction is performed.

2. The non-invasive, high-precision continuous blood glucose monitoring method with multi-parameter correction according to claim 1, characterized in that: The specific method for calibrating the sensitivity of the glucose sensor is to measure the sensitivity of the glucose sensor under different pH conditions through in vitro experiments, and then use least squares to fit the relationship curve between the sensitivity and pH; finally, in human testing, the pH value is measured first, and then the relationship curve is substituted to obtain the sensitivity under the current pH condition.

3. The non-invasive, high-precision continuous blood glucose monitoring method with multi-parameter correction according to claim 1, characterized in that: The specific method for correcting the glucose extraction flux is as follows: the zeta potential of the stratum corneum of human skin under different pH conditions is measured by a Zetasizer Nano nanoparticle potentiometer to obtain the relationship between pH and zeta potential; the glucose extraction flux under different pH conditions is obtained through in vitro simulated extraction experiments, and the relationship between pH and glucose extraction flux is fitted; a quantitative model of pH-Zeta potential-electroosmotic rate is established by combining the two fitted relationship curves.

4. The non-invasive, high-precision continuous blood glucose monitoring method with multi-parameter correction according to claim 1, characterized in that: The sodium ion correction assumes that "the ratio of glucose / sodium ion flux to tissue fluid concentration is constant" and introduces sodium ion flux to correct for extraction flux fluctuations caused by individual differences in real time.

5. The non-invasive, high-precision continuous blood glucose monitoring method with multi-parameter correction according to claim 1, characterized in that: The specific steps are as follows: (1) The sodium ion sensor was calibrated through in vitro experiments to obtain the sensitivity formula of the sodium ion sensor; the pH sensor was calibrated through in vitro experiments to obtain the sensitivity formula of the pH sensor; the glucose sensor was calibrated through in vitro experiments under different pH conditions, and the sensitivity of the glucose sensor under different pH conditions was measured. Then, the relationship curve between the sensitivity and pH was fitted by least squares to obtain the formula k1=f(pH). (2) Based on the sensitivity formulas of the pH sensor and the sodium ion sensor, the voltages measured by the pH sensor and the sodium ion sensor are converted into pH value and sodium ion concentration, respectively. (3) The sensitivity of the glucose sensor at this pH value is calculated according to the formula k1=f(pH), and the measured current is converted into glucose concentration according to the sensitivity at this pH value to achieve pH sensitivity correction of the glucose sensor. (4) Measure the pH value, sodium ion concentration and glucose concentration during two consecutive extraction cycles; (5) In the process of reverse ion electroosmosis extraction of subcutaneous intercellular fluid, the difference between two adjacent extraction cycles represents the net flux extracted in a single extraction cycle. The average pH value measured in two consecutive cycles is taken, and the difference in sodium ion concentration and glucose concentration is taken. The results are then substituted into the formula. A high-precision blood glucose prediction value is obtained after multi-parameter correction.

Images