Calibration method and system for real-time dynamic blood glucose monitoring

Abstract

The invention provides a calibration method for real-time dynamic blood glucose monitoring. The calibration method comprises the steps that 1, weak current signals generated by a glucose sensitive electrode are obtained regularly; 2, filtering the detected weak current signal by adopting a sliding window double-layer screening algorithm; 3, monitoring the body temperature of the human body, compensating the temperature based on the filtered weak current value, and estimating the glucose concentration after temperature compensation; 4, monitoring the PH value of the human blood, compensating the PH value according to the filtered weak current value, and estimating the glucose concentration after PH value compensation; and 5, combining the glucose concentrations after temperature and pH value compensation, and carrying out filtering processing on the combined glucose concentrations by adopting a parameter compensation filtering algorithm to obtain a final glucose concentration estimation result.

Description

Technical Field

[0001] The present invention relates to the technical field of blood glucose detection, and in particular to a calibration method and system for real-time dynamic blood glucose monitoring. Background Art

[0002] With changes in lifestyle and demographics, the number of people with diabetes continues to increase. Diabetes is a chronic disease that requires long-term blood sugar monitoring and management. Traditional blood sugar monitoring methods rely primarily on intermittent fingertip punctures and blood draws. This method is not only cumbersome and inconvenient, but also fails to provide real-time dynamic monitoring, making it difficult to meet the actual needs of diabetic patients for blood sugar monitoring.

[0003] In recent years, with the advancement of biosensor and wireless communication technologies, real-time continuous blood glucose monitoring (CGM) has matured and is now being applied in clinical practice. This monitoring system can monitor a patient's blood glucose level in real time through non-invasive or minimally invasive means and transmit the results to a smart terminal, facilitating real-time monitoring and management by patients and physicians. Real-time CGM technology is expected to improve blood glucose management outcomes for diabetic patients, reduce the risk of complications, and enhance their quality of life.

[0004] Real-time dynamic blood glucose monitoring systems can provide diabetic patients with personalized blood glucose management plans, helping them better understand their diabetes status and make timely adjustments to medication and lifestyle to achieve better blood glucose control. Furthermore, real-time dynamic blood glucose monitoring systems can provide doctors with detailed blood glucose data, assisting them in developing more scientific treatment plans and improving diabetes management effectiveness.

[0005] Real-time dynamic blood glucose monitoring systems have broad application prospects. They can not only be used in clinical practice for the personalized management of diabetic patients, but can also play an important role in scientific research, drug development, and other fields. Furthermore, the research and promotion of real-time dynamic blood glucose monitoring systems have positive social and economic significance for promoting the health management of diabetic patients in my country, improving diabetes prevention and control, and reducing medical and social burdens.

[0006] However, existing real-time dynamic blood glucose monitoring technology still has the following problems because the measurement results may be affected by various factors such as ambient temperature, humidity, and sensor position offset:

[0007] First, the detection accuracy is insufficient. The inherent errors in non-invasive blood glucose monitoring technology and the interference of environmental factors lead to deviations between the measured data and the actual blood glucose value, affecting the measurement accuracy.

[0008] Second, there is signal drift. Sensor signal drift is mainly caused by the influence of working conditions and environmental parameters, such as human body temperature, blood pH, glucose diffusion rate and permeability in tissue fluid, oxygen content, etc., which affects the user experience and data reliability. Summary of the Invention

[0009] In response to the shortcomings of the above-mentioned prior art, the present invention proposes a real-time dynamic blood glucose monitoring system and calibration algorithm. The system amplifies the tiny current of the glucose-sensitive electrode made by electrochemical method through an amplifier circuit, and uses temperature and pH value to calibrate the data generated by the sensor to obtain more accurate glucose concentration data.

[0010] In a first aspect, the present invention provides a calibration method for real-time dynamic blood glucose monitoring, comprising the following steps:

[0011] Step 1: Regularly obtain the weak current signal generated by the glucose-sensitive electrode;

[0012] Step 2: Use a sliding window double-layer screening algorithm to filter the detected weak current signal to obtain the weak current value;

[0013] Step 3: Monitor human body temperature, compensate for the temperature based on the weak current value, and estimate the temperature-compensated glucose concentration;

[0014] Step 4: Monitor the pH value of human blood, compensate the pH value based on the filtered weak current value, and estimate the glucose concentration after pH compensation;

[0015] Step 5: Combine the glucose concentrations obtained in steps 4 and 5 to obtain the final glucose concentration estimation result.

[0016] Optionally, the specific steps of step 2 include:

[0017] Step 2.1: Set a first window with a length of N and obtain N sampling current data in the first window;

[0018] Step 2.2: Delete the highest and lowest values ​​of the N sampling current data in sequence, calculate the average value of the remaining N-2 sampling current data, and use the average value as the median value;

[0019] Step 2.3: Keep the length of the first window unchanged, move it back 3 sampling current data points, and repeat step 2.2;

[0020] Step 2.4: Execute step 2.2 again for the obtained N-1 intermediate values, and calculate the average value of all the intermediate values, and use the average value as the final current storage value;

[0021] Step 2.5: Output the stored current value as a filtered weak current value.

[0022] Optionally, the specific steps of step 3 include:

[0023] Step 3.1: Obtain the temperature and weak current value of the human body;

[0024] Step 3.2: Perform temperature compensation by determining the functional relationship between temperature T and temperature sensor sensitivity S;

[0025] Step 3.3: Determine the temperature-compensated glucose concentration using the following formula:

[0026]

[0027] Where G represents glucose concentration, I represents current value, Ea represents reaction activation energy, R represents gas constant, T represents temperature, b represents linear correction value, k0 represents rate constant, and c represents linear offset;

[0028] Step 3.4: Obtain the unknown parameters Ea, k0, b, and c using the dropper method. Substitute the weak current value and temperature value obtained in step 3.1 into the glucose concentration calculation formula to obtain the current glucose concentration.

[0029] Optionally, the specific steps of step 4 include:

[0030] Step 4.1: Obtain the pH value of the solution based on the temperature-calibrated pH sensor and determine the OH content in the solution. - The first relationship between ion concentration and pH value:

[0031] PH=14+lg([OH - ]);

[0032] Step 4.2: Determine the nickel ions consumed and the OH consumed during the solution reaction. - The second variation relationship between ion concentrations:

[0033] Ni(III)=d·[OH - ];

[0034] Where d is the preset OH - coefficient of ion concentration;

[0035] Step 4.3: Determine the first conversion relationship between the consumed nickel ion concentration and the solution pH value based on the first change relationship and the second change relationship:

[0036] Ni(III)=f(PH)=d·10 PH-14

[0037] Step 4.4: Determine the second conversion relationship between the pH value of the solution and the glucose concentration:

[0038]

[0039] Wherein, kbos represents the reaction rate constant, Ni(III) is the nickel ion concentration, G is the glucose concentration, a and b are constants, and pH is the pH value;

[0040] Step 4.5: Determine the glucose concentration based on the pH value according to the first conversion equation and the second conversion equation:

[0041]

[0042] Wherein, b represents the linear correction value, I represents the current value, and c represents the linear offset;

[0043] Step 4.6: Obtain the unknown parameters a′, b′, c, d, k, and kobs by the dropper method;

[0044] Step 4.7: Based on the pH value obtained in step 4.1 and the parameters obtained in step 4.6, the glucose concentration at this time is calculated through step 4.5.

[0045] Optionally, the specific steps of step 5 include:

[0046] Obtaining n temperature-compensated glucose concentrations and m pH-compensated glucose concentrations, and establishing a mixed glucose concentration set; wherein the glucose concentrations of the mixed glucose concentration set are arranged in chronological order;

[0047] The average value of A glucose concentrations in the mixed glucose concentration set is calculated each time through an average filter of size A, and the average value is shifted backward by 2 glucose concentrations after each calculation;

[0048] The A+2 average values ​​output by the average filter are input into a root mean square filter of size B;

[0049] The RMS filter calculates the RMS of the previous B average values ​​each time and moves back 1 average value after each calculation;

[0050] The maximum and minimum values ​​are screened out from the obtained B+3 root mean square results, and the average value of the remaining root mean square results is calculated and used as the final glucose concentration estimation result.

[0051] In a second aspect, the present invention provides a system for real-time dynamic blood glucose monitoring, comprising:

[0052] A weak current signal acquisition module is used to periodically acquire the weak current signal generated by the glucose sensitive electrode;

[0053] The weak current signal filtering module uses a sliding window double-layer screening algorithm to filter the detected weak current signal;

[0054] A compensation module, for obtaining an estimated temperature-compensated glucose concentration and a pH-compensated glucose concentration;

[0055] The combined estimation module is used to combine the glucose concentrations after temperature and pH compensation, and filter the combined glucose concentrations using a parameter compensation filtering algorithm to obtain a final glucose concentration estimation result.

[0056] Optionally, the compensation module includes:

[0057] The temperature compensation submodule is used to monitor human body temperature, compensate for the temperature based on the filtered weak current value, and estimate the temperature-compensated glucose concentration;

[0058] The pH compensation submodule is used to monitor the pH value of human blood, compensate the pH value with the filtered weak current value, and estimate the glucose concentration after pH compensation.

[0059] Optionally, it also includes:

[0060] The display module is used to display the current pH value, glucose concentration and body temperature in line graphs.

[0061] Therefore, the present invention adopts the above-mentioned real-time dynamic blood glucose monitoring system and calibration algorithm, which has the following beneficial effects:

[0062] First, the present invention designs a weak current amplification circuit that can amplify the tiny current of the glucose-sensitive electrode made by electrochemical method;

[0063] Second, the present invention employs a combination of mean and root mean square (RMS) filtering techniques to design a self-adjusting filtering algorithm. This algorithm maximizes the effects of temperature and pH on the output current of the sensitive electrode, ultimately converting the results into equations that affect glucose concentration, thereby achieving self-adjustment of blood glucose concentration. Classical least squares curve fitting was then employed, and accuracy testing was conducted. The results demonstrated that the glucose predictions generated by the real-time self-adjusting algorithm designed by the present invention were highly accurate, capable of correcting deviations caused by temperature and pH variations to a certain extent, and maintaining a relatively close fluctuation within the true value, demonstrating high accuracy and reliability.

[0064] Third, the present invention implements a WeChat applet, which can accurately and real-timely display corresponding data through its visualization page, making it easier for patients to obtain their own blood sugar values.

[0065] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 This is a data interaction diagram of the present invention.

[0067] Figure 2 This is the technical route of the present invention.

[0068] Figure 3 This is the I / V conversion amplifier circuit diagram of the present invention.

[0069] Figure 4 This is a flow chart of the sliding window double-layer screening processing algorithm of the present invention.

[0070] Figure 5 This is a system flow chart of the temperature compensation module and pH compensation module of the present invention.

[0071] Figure 6 This is a diagram of the parameter compensation filtering algorithm of the present invention.

[0072] Figure 7 This is a diagram of the interaction between the OLED screen display module and ESP8266 of the present invention.

[0073] Figure 8 This is the overall logical architecture diagram of the host computer of the present invention.

[0074] Figure 9 This is the dynamic graph test diagram of the WeChat applet in the present invention.

[0075] Figure 10 This is a test chart showing the health record data in the present invention. DETAILED DESCRIPTION

[0076] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention. In addition, it should be understood that after reading the content taught by the present invention, those skilled in the art will make various changes or modifications to the present invention, and these equivalent forms also fall within the scope defined by the appended claims of the application.

[0077] Example 1

[0078] like Figure 1-Figure 2As shown, the present invention proposes a real-time dynamic blood glucose monitoring system and calibration algorithm. The main technical ideas are: first, based on the signal generated by the glucose sensitive electrode, the factors causing the current signal drift are analyzed and determined to be temperature and pH value; secondly, based on the influence of temperature and pH value on the current signal, a triple calibration mode of compensation, filtering and screening is proposed.

[0079] The real-time dynamic blood glucose monitoring method proposed by the present invention mainly includes the following steps:

[0080] Step 1: Regularly obtain the weak current signal generated by the glucose-sensitive electrode;

[0081] Step 2: Use a sliding window double-layer screening algorithm to filter the detected weak current signal;

[0082] Step 3: Monitor human body temperature, compensate for the temperature based on the filtered weak current value, and estimate the temperature-compensated glucose concentration;

[0083] Step 4: Monitor the pH value of human blood, compensate the pH value with the filtered weak current value, and estimate the glucose concentration after pH compensation;

[0084] Step 5: Combine the glucose concentrations after temperature and pH compensation, and use a parameter compensation filtering algorithm to filter the combined glucose concentrations to obtain the final glucose concentration estimation result.

[0085] in:

[0086] (1) I / V conversion amplifier circuit

[0087] like Figure 3 As shown, the present invention designs an I / V conversion amplifier circuit. In step 1, based on the I / V conversion amplifier circuit, the weak current signal generated by the glucose sensitive electrode is detected every 10 seconds.

[0088] from Figure 3 It can be seen that the input end of the I / V conversion amplifier circuit is connected to the glucose sensor, and the output end is used for signal output. The I / V conversion amplifier circuit includes operational amplifiers U1A, U1B, R1, R2, R3, R4, C1, C2, and C3.

[0089] Compared with the traditional amplifier circuit, R1, C1 and the integrated operational amplifier form a first-order active low-pass filter with an inverting input, which is beneficial for filtering out the high-frequency part. At the same time, the post-stage U1B operational amplifier adopts a voltage follower circuit, which can act as a voltage buffer.

[0090] (2) Sliding window double-layer screening algorithm

[0091] Since sampling a weak current every 10 seconds may result in a large sampling error, it is necessary to filter the sampling results. Therefore, the present invention proposes a sliding window double-layer screening algorithm in step 2, which runs every 3 minutes and can accurately reflect the changes in the detection current within 3 minutes;

[0092] like Figure 4 As shown, the sliding window double-layer screening algorithm includes the following steps:

[0093] Step 2.1: Set a first window with a length of N and obtain N sampling current data in the first window;

[0094] Step 2.2: Delete the highest and lowest values ​​of the N sampling current data in sequence, calculate the average value of the remaining N-2 sampling current data, and use the average value as the median value;

[0095] Step 2.3: Keep the length of the first window unchanged, move it back 3 sampling current data points, and repeat step 2.2;

[0096] Step 2.4: Execute step 2.2 again for the obtained N-1 intermediate values, and calculate the average value of all the intermediate values, and use the average value as the final current storage value;

[0097] Step 2.5: Output the stored current value as a filtered weak current value.

[0098] In an embodiment of the present application, a weak current may be collected every 10 seconds, and a total of 6 raw sampled currents may be collected within 1 minute. A first window of length 6 is set, and the first window contains 6 raw sampled current data, and the first raw sampled current data and the sixth raw sampled current data are the starting and ending positions of the first window, respectively.

[0099] In the first window, delete the highest and lowest values, and the remaining 4 original sampling currents are combined Figure 4 ,The points with dotted lines in the figure are the points removed during the sliding process;

[0100] Next, the average value of the remaining four original sampling currents is calculated and used as the first intermediate value F1. The first window is then moved backward by three sampling current data points (i.e., three time intervals, each 10 seconds being an interval). At this point, the data in the first window are the 7th to 12th original sampling currents from the start of sampling.

[0101] Similarly, the first window is repeatedly moved, and the average value of the sampled current after screening in the current first window is calculated to obtain the five intermediate values ​​F1, I1, L1, O1 and R1 within 3 minutes.

[0102] Specifically, combined Figure 4 , it can be seen that the first 10s sample is A, and the 6 sampling currents in the first minute are A, B, C, D, E, and F. Remove the lowest value A and the highest value D, and the remaining are B, C, E, and F. Calculate the average value of B, C, E, and F to get the intermediate value F1. Then move the window containing the four sampling points B, C, E, and F backward by 3 time intervals (10s is an interval), that is, the second window starts from D. The 6 data included in this window are D, E, F, G, H, and I. Remove the highest value H and the lowest value F, and the remaining are D, E, G, and I. Calculate the average value of these four sampling points to get the intermediate value I1, and then move the second window The third window starts from G. The third window includes G, H, I, J, K, and L. The highest value J and the lowest value G are removed, leaving H, I, K, and L. The average value of these four sampling points is calculated to obtain the intermediate value L1. Then, move the third window again. The fourth window contains J, K, L, M, N, and O. The highest value J and the lowest value O are removed, leaving K, L, N, and O. The average value of these four sampling points is calculated to obtain the intermediate value O1. Then, move the fourth window again. The fifth window contains M, N, O, P, Q, and R. The highest value M and the lowest value O are removed, leaving N, P, Q, and R. The average value of these four sampling points is calculated to obtain the intermediate value R1.

[0103] Then, the five intermediate values ​​F1, I1, L1, O1, and R1 are re-screened, the highest and lowest values ​​are removed from these five intermediate values, and the average of the remaining three intermediate values ​​is calculated. This average value is used as the final current storage value R2, which can accurately reflect the detection current within 3 minutes;

[0104] from Figure 3 It can be seen that after removing the highest value L1 and the lowest value R1 from F1, I1, L1, O1 and R1, F1, I1 and O1 remain. The average of these three intermediate values ​​is calculated and used as the final current storage value R2.

[0105] Finally, the current storage value R2 is output as the current value generated by the sensitive electrode after filtering.

[0106] (3) Temperature compensation

[0107] Since temperature affects many factors such as instantaneous glucose concentration, enzyme catalytic activity and physiological steady-state environment during the monitoring of human blood glucose concentration, it is necessary to monitor body temperature in real time and compensate for the temperature. Figure 5 , the specific temperature compensation method includes the following steps:

[0108] Step 3.1: Obtain the temperature value of the human body and the current value generated by the sensitive electrode after filtering;

[0109] Step 3.2: Temperature compensation of glucose concentration estimation is achieved by determining a functional relationship between temperature T and temperature sensor sensitivity S;

[0110] Since the slope of the curve (temperature sensor sensitivity) corresponding to different temperatures is different, there will be a certain error if only the slope of the curve is used to estimate the blood glucose concentration. In order to eliminate the influence of temperature on the fluctuation of current value, temperature compensation is required. Under temperature compensation, the glucose concentration has a linear relationship with the current. According to the slope of this linear relationship, the functional relationship between temperature T and temperature sensor sensitivity S can be determined, thereby achieving temperature compensation.

[0111] The functional relationship between the temperature T and the temperature sensor sensitivity S is:

[0112]

[0113] Among them, a and b represent linear correction values, k0 represents rate constant, c represents linear offset, and E a represents the reaction activation energy, S represents the sensitivity of the temperature sensor, R represents the gas constant, and T represents the temperature;

[0114] Step 3.3: Determine the temperature-compensated glucose concentration using the following formula:

[0115]

[0116] Where G represents the glucose concentration and I represents the current value;

[0117] Step 3.4: Determine the unknown parameters Ea, k0, b and c mentioned above, and substitute the current value and temperature value obtained in step 3.1 into the above formula to obtain the temperature-compensated human glucose concentration value at this time, thereby realizing temperature-compensated self-adjustment of glucose concentration.

[0118] In this embodiment, the unknown parameters Ea, k0, b and c are gradually obtained by the "dropper method" in the prior art, and then the current value and temperature value obtained in step 3.1 are substituted into the above formula to obtain the glucose concentration at this time, thereby realizing temperature compensation self-adjustment of the glucose concentration.

[0119] (4) pH value compensation

[0120] Since changes in pH value will affect the charge status of reactants and products, thus changing the equilibrium constant and direction of the reaction, and thus having a significant impact on the reaction rate, pH value will also affect the current measured by the glucose sensitive electrode. Therefore, the present invention also compensates for pH value. Figure 5 , the pH value compensation method includes the following steps:

[0121] Step 4.1: Obtain the pH value of the solution based on the temperature-calibrated pH sensor and determine the OH content in the solution. - The relationship between ion concentration and pH value;

[0122] The present invention uses a temperature-calibrated pH sensor to obtain the pH value of the solution. - The relationship between ion concentration and solution pH is as follows:

[0123] PH=14+lg([OH - ]);

[0124] Step 4.2: Determine the nickel ions consumed and the OH consumed during the solution reaction. - The changing relationship between ion concentrations;

[0125] As the solution reaction proceeds, the concentration of nickel ions Ni(III) consumed and the amount of OH consumed - There is a proportional relationship between the ion concentrations. Assume that the two have the following relationship:

[0126] Ni(III)=d·[OH - ]

[0127] Where d is the preset OH - coefficient of ion concentration;

[0128] Step 4.3: According to OH - The relationship between ion concentration and pH value and the relationship between consumed nickel ions and consumed OH - The changing relationship between ion concentrations, determine the conversion relationship between the consumed nickel ion concentration and the pH value of the solution;

[0129] Consumed OH - The concentration is proportional to the Ni(III) concentration consumed, and the pH value of the solution is proportional to the OH - The concentration is also related, so the concentration of Ni(III) can be expressed by a function related to the pH value: Ni(III) = f(PH). From steps 4.1 and 4.2, we can see that the conversion relationship between Ni(III) and pH is as follows:

[0130] Ni(III)=f(PH)=d·10 PH-14 ;

[0131] Step 4.4: Determine the conversion relationship between the pH value of the solution and the glucose concentration;

[0132] The conversion relationship between the pH value of the solution and the glucose concentration G is:

[0133]

[0134] Where kbos represents the reaction rate constant, Ni(III) is the nickel ion concentration, G is the glucose concentration, a and b are constants, and pH is the pH value. Here, f(I) = k·I + c, and the reaction orders x and y are quantities that describe the relative rates at which reactants participate in a chemical reaction.

[0135] The reaction order represents the amount of reactants in the reaction, and the exponents are the same as their amounts shown in the rate equation, usually with x and y set to 1. This gives the following simplified form of the above equation:

[0136]

[0137] Step 4.5: Combining steps 4.3 and 4.4, the formula for calculating glucose concentration by pH value is as follows:

[0138]

[0139] Where G is the glucose concentration, a and b are constants, b represents the linear correction value, I represents the current value, c represents the linear offset, and kobs represents the reaction rate constant;

[0140] Since a and b are both constants, the above formula can be simplified to:

[0141]

[0142] Step 4.6: Obtain the unknown parameters a′, b′, c, d, k, and kobs by the dropper method;

[0143] Step 4.7: Substitute the pH value obtained in step 4.1 and the parameters obtained in step 4.6 into the simplified calculation formula in step 4.5 to obtain the glucose concentration G at this time, thereby achieving pH value compensation self-adjustment of glucose concentration.

[0144] In summary, calibration is a real-time conversion based on the convertible relationship between the current signal I(t) generated by the glucose-sensing electrode at a specific time t and the estimated blood glucose concentration G(t). The magnitude of the current is not only related to temperature but also to pH. Chemical reactions are complex processes. Since temperature and pH are dominant among the many influencing factors, the conversion relationship between temperature and pH and the generated current is determined using the control variable method.

[0145] (5) Parameter compensation filtering algorithm

[0146] Since the glucose concentration estimated after temperature compensation is generally inconsistent with the glucose concentration estimated after pH compensation, it is necessary to perform corresponding filtering on these inconsistent data to obtain a glucose concentration with a relatively small error;

[0147] The glucose concentrations estimated in steps 3 and 4 are mixed together to form a set of data. The mixed data are then used to create a set. The data in the set are then subjected to parameter compensation filtering. The filtered result is affected by both temperature and pH value, which reduces the relative error of the final filtered result, thereby obtaining a glucose concentration with a reduced relative error. Figure 6 , the parameter compensation filtering algorithm includes the following steps:

[0148] Step 5.1: Obtain n temperature-compensated glucose concentrations and m pH-compensated glucose concentrations, and establish a mixed glucose concentration set; wherein the glucose concentrations in the mixed glucose concentration set are arranged in chronological order;

[0149] In the embodiment of the present application, during the period when the glucose sensitive electrode generates a weak current, the glucose concentration G can be estimated by continuously acquiring 10 temperature compensation values ​​according to step 3. T1 -G T10 , and obtain the estimated glucose concentration G after 10 pH value compensation according to step 4 PH1 -G PH10 , and G T1 , G T2 , G T3 ,…,G T10 and G PH1 , G PH2 , G PH3 ,…,G PH10 The data are arranged in chronological order to form a mixed glucose concentration set, and each glucose concentration in the mixed glucose concentration set is used as a sampling point.

[0150] Step 5.2: Set an average filter of size A and obtain A glucose concentrations from the mixed glucose concentration set. The average filter outputs the average of the A glucose concentrations.

[0151] In an embodiment of the present application, an average value filter of size 6 can be set, that is, A=6; then 6 glucose concentrations are read from the mixed glucose concentration set in chronological order, each glucose concentration is a sampling value, and the average value A1 of the first 6 sampling values ​​is calculated and output according to the average value filter.

[0152] Step 5.3: Shift the average filter back by 2 sampling points and repeat step 5.2.

[0153] After calculating A1, shift the average filter back two sampling points, that is, start the calculation from the third glucose concentration data in the mixed glucose concentration set, calculate the average value A2 of the six sampling values ​​again, and add A2 to the output of the average filter. Repeat steps 5.2-5.3 to obtain the average values ​​A3, A4, A5, A6, A7 and A8 in sequence. All eight average values ​​are added to the output of the average filter.

[0154] Step 5.4: Input the A+2 average values ​​output by the average filter into an RMS filter of size B;

[0155] You can set the size of the RMS filter to B=3 and get the average value A1-A8 of the average filter output;

[0156] Step 5.5: The RMS filter first calculates the RMS of the first B average values ​​and outputs the RMS result;

[0157] Calculate the sum of the squares of the first three average values, divide the sum by 3, take the square root of the result, and get the RMS result B1. Add B1 to the output of the RMS filter. The calculation formula for B1 is:

[0158]

[0159] Step 5.6: Shift the RMS filter back by 1 mean value, repeat step 5.5, and output B+3 RMS results.

[0160] Calculate the three average values ​​again to get the RMS result B2, and add B2 to the output of the RMS filter. The calculation formula for B2 is:

[0161]

[0162] Repeat the above steps to calculate the RMS results B3, B4, B5, and B6 in sequence, and add B3, B4, B5, and B6 to the output of the RMS filter;

[0163] Step 5.7: Filter out the maximum and minimum values ​​from all the obtained root mean square results, calculate the average of the remaining root mean square results, and use it as the final glucose concentration estimation result.

[0164] Finally, the maximum and minimum values ​​are removed from B1, B2, B3, B4, B5, and B6, and the average of the remaining four root mean square results is used as the final glucose concentration estimation result.

[0165] Example 2

[0166] Based on the above calibration method, the present invention also realizes a real-time dynamic blood glucose monitoring system, which can be realized based on WeChat applet. The data is visualized through WeChat applet. The overall logical architecture of the WeChat applet can be referred to Figure 7 The glucose concentration monitoring results are as follows: Figure 9-10 The display of WeChat applet is based on page (Page), and different pages are different display interfaces. The present invention mainly uses WeChat applet for visual page display.

[0167] The system may include:

[0168] A weak current signal acquisition module is used to periodically acquire the weak current signal generated by the glucose sensitive electrode;

[0169] The weak current signal filtering module uses a sliding window double-layer screening algorithm to filter the detected weak current signal;

[0170] A compensation module, for obtaining an estimated temperature-compensated glucose concentration and a pH-compensated glucose concentration;

[0171] The combined estimation module is used to combine the glucose concentrations after temperature and pH compensation, and filter the combined glucose concentrations using a parameter compensation filtering algorithm to obtain a final glucose concentration estimation result.

[0172] Wherein, the compensation module includes:

[0173] The temperature compensation submodule is used to monitor human body temperature, compensate for the temperature based on the filtered weak current value, and estimate the temperature-compensated glucose concentration;

[0174] The pH compensation submodule is used to monitor the pH value of human blood, compensate the pH value with the filtered weak current value, and estimate the glucose concentration after pH compensation.

[0175] In order to facilitate the display of results and facilitate user viewing, the system may further include:

[0176] The display module is used to display the current pH value, glucose concentration and body temperature in line graphs.

[0177] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the same. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solutions of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A calibration method for real-time dynamic blood glucose monitoring, characterized in that: The following steps are involved: Step 1: Regularly obtain the weak current signal generated by the glucose-sensitive electrode; Step 2: Use a sliding window double-layer screening algorithm to filter the detected weak current signal to obtain the weak current value; Step 3: Monitor human body temperature, compensate for the temperature based on the weak current value, and estimate the temperature-compensated glucose concentration; Step 4: Monitor the pH value of human blood, compensate the pH value based on the filtered weak current value, and estimate the glucose concentration after pH compensation; Step 5: Combine the glucose concentrations obtained in steps 4 and 5 to obtain the final glucose concentration estimation result.

2. A calibration method for real-time dynamic blood glucose monitoring according to claim 1, characterized in that: The specific steps of step 2 include: Step 2.1: Set a first window with a length of N and obtain N sampling current data in the first window; Step 2.2: Delete the highest and lowest values ​​of the N sampling current data in sequence, calculate the average value of the remaining N-2 sampling current data, and use the average value as the median value; Step 2.3: Keep the length of the first window unchanged, move it back 3 sampling current data points, and repeat step 2.2; Step 2.4: Execute step 2.2 again for the obtained N-1 intermediate values, and calculate the average value of all the intermediate values, and use the average value as the final current storage value; Step 2.5: Output the stored current value as a filtered weak current value.

3. A calibration method for real-time dynamic blood glucose monitoring according to claim 2, characterized in that: The specific steps of step 3 include: Step 3.1: Obtain the temperature and weak current value of the human body; Step 3.2: Perform temperature compensation by determining the functional relationship between temperature T and temperature sensor sensitivity S; Step 3.3: Determine the temperature-compensated glucose concentration using the following formula: Where G represents glucose concentration, I represents current value, Ea represents reaction activation energy, R represents gas constant, T represents temperature, b represents linear correction value, k0 represents rate constant, and c represents linear offset; Step 3.4: Obtain the unknown parameters Ea, k0, b, and c using the dropper method. Substitute the weak current value and temperature value obtained in step 3.1 into the glucose concentration calculation formula to obtain the current glucose concentration.

4. A calibration method for real-time dynamic blood glucose monitoring according to claim 3, characterized in that: The specific steps of step 4 include: Step 4.1: Obtain the pH value of the solution based on the temperature-calibrated pH sensor and determine the OH content in the solution. - The first relationship between ion concentration and pH value: PH=14+lg([OH - ]); Step 4.2: Determine the nickel ions consumed and the OH consumed during the solution reaction. - The second variation relationship between ion concentrations: Ni(III)=d·[OH - ]; Where d is the preset OH - coefficient of ion concentration; Step 4.3: Determine the first conversion relationship between the consumed nickel ion concentration and the solution pH value based on the first change relationship and the second change relationship: Ni(II)=f(PH)=d·10 PH-14 Step 4.4: Determine the second conversion relationship between the pH value of the solution and the glucose concentration: Wherein, kbos represents the reaction rate constant, Ni(III) is the nickel ion concentration, G is the glucose concentration, a and b are constants, and pH is the pH value; Step 4.5: Determine the glucose concentration based on the pH value according to the first conversion equation and the second conversion equation: Wherein, b represents the linear correction value, I represents the current value, and c represents the linear offset; Step 4.6: Obtain the unknown parameters a′, b′, c, d, k, and kobs by the dropper method; Step 4.7: Based on the pH value obtained in step 4.1 and the parameters obtained in step 4.6, the glucose concentration at this time is calculated through step 4.

5.

5. A calibration method for real-time dynamic blood glucose monitoring according to claim 4, characterized in that: The specific steps of step 5 include: Obtaining n temperature-compensated glucose concentrations and m pH-compensated glucose concentrations, and establishing a mixed glucose concentration set; wherein the glucose concentrations of the mixed glucose concentration set are arranged in chronological order; The average value of A glucose concentrations in the mixed glucose concentration set is calculated each time through an average filter of size A, and the average value is shifted backward by 2 glucose concentrations after each calculation; The A+2 average values ​​output by the average filter are input into a root mean square filter of size B; The RMS filter calculates the RMS of the previous B average values ​​each time and moves back 1 average value after each calculation; The maximum and minimum values ​​are screened out from the obtained B+3 root mean square results, and the average value of the remaining root mean square results is calculated and used as the final glucose concentration estimation result.

6. A real-time dynamic blood glucose monitoring system, characterized in that: A calibration method for real-time dynamic blood glucose monitoring according to any one of claims 1 to 5 is implemented, comprising: A weak current signal acquisition module is used to periodically acquire the weak current signal generated by the glucose sensitive electrode; The weak current signal filtering module uses a sliding window double-layer screening algorithm to filter the detected weak current signal; A compensation module, for obtaining an estimated temperature-compensated glucose concentration and a pH-compensated glucose concentration; The combined estimation module is used to combine the glucose concentrations after temperature and pH compensation, and filter the combined glucose concentrations using a parameter compensation filtering algorithm to obtain a final glucose concentration estimation result.

7. A real-time dynamic blood glucose monitoring system according to claim 6, characterized in that: The compensation module includes: The temperature compensation submodule is used to monitor human body temperature, compensate for the temperature based on the filtered weak current value, and estimate the temperature-compensated glucose concentration; The pH compensation submodule is used to monitor the pH value of human blood, compensate the pH value with the filtered weak current value, and estimate the glucose concentration after pH compensation.

8. A real-time dynamic blood glucose monitoring system according to claim 6, characterized in that: Also includes: The display module is used to display the current pH value, glucose concentration and body temperature in line graphs.

Images