Design of an Intelligent Blood Glucose Self-Monitoring Instrument

0 Introduction

Diabetes is a common metabolic endocrine disease caused by a lack of insulin or abnormal insulin receptors in the human body. It is characterized by high blood sugar and is a worldwide epidemic. In recent years, its incidence has shown a significant upward trend. Currently, about 10% of adults in the world suffer from this disease [1]. In China, there are about 40 million diabetic patients. The current treatment method is mainly to regulate the glucose metabolism in the patient's body. The important basis for clinical treatment and medication is the patient's blood glucose content. Therefore, it is important to track and evaluate the control of diabetes through self-monitoring blood glucose meters. In particular, self-monitoring blood glucose meters can detect results conveniently and quickly in hospitals or even at home. Currently, blood glucose meter products on the market can only give blood glucose values, and patients still have certain difficulties in adjusting treatment plans and diet control based on this. In addition, there are deficiencies in test accuracy and test range. Therefore, it is of great significance to study high-precision and intelligent blood glucose monitoring instruments.

1. Detection principle

Large-scale detection instruments used in clinical diagnosis are not suitable for small and medium-sized hospitals, emergency departments, and patients' long-term self-monitoring of their conditions due to their large size, high price, and cumbersome detection process. The practical application of enzyme electrodes in biosensors has made rapid progress. For example, enzyme electrodes for glucose, lactic acid, cholesterol, urea, and amino acids are widely used in medical testing. A common feature of these sensors is that they use the molecular recognition ability of biomaterials such as microorganisms, enzymes, and antibodies, and use biomaterials as sensors for molecular recognition elements; then use the electrons generated by the reaction between the detected substance contained in the sample liquid and the enzyme to reduce the electron acceptor, and the measuring device uses an electrochemical method to measure the reduction amount of the electron acceptor to perform quantitative analysis of the detected substance.

The detection principle of this instrument is based on the solidification of glucose oxidase (GOD) on the electrode surface. When blood drips into the blood glucose test electrode, a redox reaction occurs. During the reaction, the divalent iron ions in the transmission medium lose electrons and an oxidation reaction occurs. Glucose oxidase oxidizes glucose to generate H2O2 and gluconic acid. H2O2 oxidizes the divalent iron ions. During the redox process, electron gain and loss occur. Under the action of a certain voltage, an oxidation current is formed. The purpose of detecting blood glucose concentration is achieved by detecting that the current change is approximately linearly related to the glucose concentration. The specific reaction equation is as follows:

2 Instrument design and implementation

2.1 Hardware Structure

The entire instrument system consists of enzyme electrode sensing part, signal conditioning (current voltage conversion, amplification and filtering part), temperature compensation part, LCD display part, single chip, and intelligent chip, as shown in Figure 1. The microcurrent generated by dripping blood after the enzyme electrode is small, only reaching the microampere level, which is not convenient for measurement and analysis, so it is first converted into a voltage signal and then amplified. Since the system noise generated by the power supply and various interference signals affects the test accuracy, a filter circuit should be designed to remove the interference signal to make the test more accurate. The processed voltage value is transmitted to the single chip microcomputer MSP430 with built-in A/D conversion. The single chip microcomputer calculates the blood sugar concentration value and then displays the result using LCD.

2.2 Main testing process

The signal conditioning circuit in the detection system is shown in Figure 2. After the instrument is powered on, the single-chip microcomputer controls the multi-way switch S1 to close, and the input of the front-stage op amp is the voltage divider value of R1 and R2, which is set to 0.4V. According to the characteristics of the follower and the ideal op amp, the output of the rear stage remains unchanged at 0.4V at this time, and is input into the single-chip microcomputer after A/D conversion, and the control display screen displays "Please insert the test strip". When the system inserts the blood glucose test strip, S2 is closed and R4 is introduced into the circuit. The rear stage op amp constitutes a common-phase amplifier , and R4 is adjusted to make the circuit output value 0.6V. After A/D conversion, it is input into the single-chip microcomputer, and the display screen displays "Please drop blood". The current signal generated after dripping blood is input from "Input", and the output voltage signal suddenly changes and is converted and then input into the single-chip microcomputer. After the reaction is completed, the blood glucose value is calculated and displayed according to the voltage change.

2.3 Temperature compensation

Changes in ambient temperature cause zero drift and sensitivity changes in the detection system, resulting in measurement errors. To eliminate the influence of ambient temperature, the temperature compensation circuit in the system is implemented using a single-bus micro temperature sensor DS18B20 from DALLAS, which has a temperature measurement range of -55 to +125°C, a measurement resolution of 0.0625°C, and a temperature measurement accuracy of ±0.5°C. The temperature signal is input into the microcontroller through a multi-way switch, and the CPU automatically corrects the test result error based on the temperature characteristics of the blood glucose test electrode.

2.4 Smart Chip

In order to further improve the practicality of the instrument and make it as convenient as possible for patients, the instrument introduces the principle of artificial intelligence . Patients can get personal condition information and simple treatment methods and precautions while testing their own blood sugar concentration. This function is mainly achieved by calling the information in the solidified smart chip with diabetes knowledge base. The solidified knowledge base in the chip is made according to the treatment guidelines approved and published by the World Health Organization, the International Diabetes Federation and the Chinese Diabetes Association. The guidelines are authoritative in diabetes analysis. After measuring the blood sugar value, patients can select the "inference" function in the operation interface.

3. Instrument performance analysis and testing

3.1 Easy operation and low power consumption

[page]

After turning on the machine, the patient only needs to follow the prompts to complete the entire test process. The operation interface is simple and friendly. According to the output voltage conditioned by the test, as shown in Figure 2, it can be seen that the reaction tends to be stable about 20 seconds after the blood is dripped, and the patient's test time does not exceed 25 seconds under the correct operation steps. Since the blood is collected by siphon test paper, the single blood collection volume only needs about 3 to 5μL. The MSP430 series low-power microcontroller is selected. The instrument can automatically power off after the test paper is removed according to the voltage change during the test, effectively saving energy consumption.

3.2 Wide test range

According to the diagnostic criteria for non-insulin-dependent diabetes mellitus (NIDDW) recommended by the WTO in 1998, blood sugar levels less than 7 mmol/L are non-diabetic (fasting blood sugar levels of 3.5 to 7 mmol/L are normal), 7 to 11.1 mmol/L are impaired glucose tolerance, and greater than 11.1 mmol/L are diabetic. After testing different samples, the test range of this instrument can reach 2.2 to 27.8 mmol/L, completely covering the range of possible human blood sugar values.

3.3 Test data is reliable

In order to verify the authenticity and reliability of the data measured by the blood glucose detector, clinical tests were conducted in the laboratory of the First Hospital of Shanxi University. The test samples with different blood glucose concentrations were tested using an intelligent blood glucose meter and an Italian large-scale biochemical analyzer (BT-3000). The test results showed that the two had good consistency, and the error at high blood glucose values ​​was larger than that at low blood glucose and normal blood glucose, which required further calibration and calibration. The test values ​​of the large-scale biochemical analysis were used as standard values, and the test results of 30 groups are shown in Table 1. The test data were subjected to regression analysis, and the fitting results are shown in Figure 3. The regression equation is:

In the formula: regression coefficient b = 0.9851, intercept a = -0.0806, correlation coefficient R = 0.9962, standard deviation SD = 0.2378.

Table 1 Comparison results between smart blood glucose meter and standard blood glucose values

4 Conclusion

The blood glucose concentration results measured by the smart blood glucose self-monitoring instrument are significantly correlated with the results measured by the high-precision blood glucose meter. It has the advantages of simple operation, short measurement time, relatively accurate and reliable results, high intelligence, and low power consumption. After optimization and improvement, it can become an ideal instrument for diabetic patients to monitor and control their own blood glucose concentration at home.

Acknowledgements: We would like to thank the Taiyuan Science and Technology Bureau and North China University for providing funding for this project; the medical staff of the First Hospital of Shanxi University for their assistance and providing experimental equipment and venues. We would like to thank Professor Zhou Hanchang for his technical guidance and Beijing Soft Test Company for its technical support on artificial intelligence chip technology.