Gta-mn biosensor and preparation method and application thereof

Abstract

The application relates to the technical field of inspection, and discloses a GTA-MN biosensor and a preparation method and application thereof, which comprises the following steps: etching Ti3AlC2 powder in a mixed solution of HF, HCl and deionized water to synthesize multilayer MXene; stripping the multilayer MXene in deionized water with LiCl to obtain few-layer Ti3C2Tx_MXene; depositing a gold layer on the surface of the microneedle through 3D printing of PMMA microneedles to obtain Au-MNs; sequentially immersing and coating the Au-MNs in a Ti3C2Tx_MXene aqueous dispersion and drying, coating GOx solution to functionalize, and obtaining a GTA-MNs array; assembling the GTA-MNs array, an Ag / AgCl reference electrode, a platinum counter electrode and an integrated circuit board into a GTA-MNs biosensor, which can be used for continuous monitoring of biological blood glucose. The application promotes the development of a sensitive, robust and clinically reliable CGM system by fusing material science and data analysis, and helps to realize personalized diabetes management.

Description

Technical Field

[0001] This invention relates to the field of testing technology, specifically to the GTA-MN biosensor, its preparation method, and its application. Background Technology

[0002] Blood glucose testing is a core component of diabetes prevention, diagnosis, and management, playing an irreplaceable role in controlling abnormal blood glucose fluctuations and reducing the risk of acute and chronic complications. Diabetes, a major global health challenge, stems primarily from the various complications caused by blood glucose imbalances. Acute hypoglycemia can lead to coma, nerve damage, and increased cardiovascular risk, while chronic hyperglycemia damages microvessels, inducing retinopathy, nephropathy, and neuropathy. Hyperosmolar hyperglycemic coma carries an extremely high risk of death. Accurate, rapid, and stable blood glucose testing for early intervention and real-time monitoring is crucial for preventing complications, improving patient prognosis, and enhancing diabetes management. Therefore, developing efficient and reliable blood glucose testing technologies has significant clinical value and social implications.

[0003] In existing technologies, to achieve rapid and stable blood glucose detection, various technical approaches have been developed around the optimization of detection principles and equipment. These can be broadly categorized into three main types: First, traditional finger-prick blood sampling methods, which collect capillary blood through puncture and combine it with a blood glucose meter for rapid quantitative detection. This method is convenient, low-cost, and commonly used in clinical and home settings. Second, continuous glucose monitoring systems, which can track blood glucose trends in real time (up to two weeks) and integrate alarm functions to warn of hypoglycemia and hyperglycemia events. This eliminates the need for frequent finger-prick blood sampling and provides a more comprehensive reflection of blood glucose dynamics. Third, minimally invasive microneedle sensing technology, relying on an interstitial fluid (ISF)-based microneedle sensing platform. Its advantages of being minimally invasive and having good patient compliance have made it a research hotspot. It obtains interstitial fluid through microneedle penetration of the epidermis and combines it with electrochemical sensing principles to achieve rapid glucose detection. In addition, some technologies improve detection sensitivity and stability by functionalizing the sensing interface (e.g., loading glucose oxidase GOx) and optimizing sensing materials, attempting to overcome the limitations of traditional methods.

[0004] However, existing blood glucose detection technologies still have the following technical problems, making it difficult to meet the needs of accurate clinical monitoring and long-term management: (1) Traditional finger-prick blood sampling detection has inherent limitations, requiring repeated punctures to collect blood, which can easily cause pain and poor patient compliance. Moreover, it can only obtain single-point blood glucose data, which cannot reflect the dynamic changes in blood glucose and makes it difficult to capture the risk of sudden rises and falls in blood glucose in a timely manner; (2) Existing continuous glucose monitoring systems have obvious shortcomings. Probe implantation can easily cause discomfort to the human body, the detection accuracy is limited, and the equipment cost is high, making it difficult to popularize and apply; (3) Existing mainstream microneedle sensing platforms face severe challenges. The detection range is narrow and cannot cover the extremely high blood glucose concentration required for hyperosmolar hyperglycemic coma. At the same time, there is a physiological delay (5-10 min) in the glucose balance between blood and interstitial fluid. In addition, there are problems such as signal drift and the need for frequent calibration, which seriously affect the timeliness and accuracy of detection and cannot accurately reflect the real blood glucose dynamics; (4) The sensing materials and functional modifications of existing GTA-MN biosensors are not perfect, resulting in insufficient detection sensitivity and stability, making it difficult to achieve long-term stable monitoring. Furthermore, they are easily affected by complex internal environments, which further restricts the reliability of detection. In summary, existing technologies cannot simultaneously achieve the speed, stability, wide range, and minimally invasive nature of blood glucose testing. There is an urgent need for a new type of high-efficiency electrochemical microneedle sensor to overcome the current technological bottlenecks and achieve accurate, real-time, and long-term stable blood glucose testing. Summary of the Invention

[0005] The present invention aims to provide a GTA-MN biosensor, its preparation method and application, to solve the technical problems of limited detection range and physiological lag of interstitial fluid glucose level in existing technologies.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a method for preparing a GTA-MN biosensor, comprising the following steps: (a) Preparation of Ti3C2Tx_MXene: Ti3AlC2 powder was etched in a mixture of HF, HCl and deionized water to synthesize multilayer MXene; the multilayer MXene was dispersed with LiCl in deionized water to obtain a mixture, and the multilayer MXene in the mixture was peeled off to prepare few-layer Ti3C2Tx_MXene; (II) Preparation of GTA-MNs array: Au-MNs were obtained by 3D printing PMMA microneedles and depositing a gold layer on the surface of the microneedles; Au-MNs were dipped into Ti3C2Tx_MXene aqueous dispersion and dried to form Ti3C2Tx-Au-MNs; Ti3C2Tx-Au-MNs were functionalized by coating with GOx solution to obtain electrochemical microneedle sensor GTA-MNs array; (III) Preparation of GTA-MN biosensor: GTA-MN array, Ag / AgCl reference electrode, platinum counter electrode and integrated circuit board are assembled into GTA-MN biosensor.

[0007] Technical Effects: This method employs a standardized process of "Ti3C2Tx_MXene preparation - GTA-MNs array functionalization - sensor assembly," relying on precise design and synergistic cooperation in each step to prepare a GTA-MN biosensor with high sensitivity, strong stability, and good adaptability. This overcomes the shortcomings of traditional biosensors, such as poor conductivity, limited functionality, and low detection efficiency, providing reliable technical support for accurate detection. Specifically, in step (I), a two-step etching-stripping method is used to prepare few-layer Ti3C2Tx MXene with high purity, good dispersibility, and excellent conductivity, laying the core foundation for the sensor's electrochemical performance. Etching Ti3AlC2 powder with a mixture of HF, HCl, and deionized water precisely removes the Al layer, efficiently synthesizing multilayer MXene and ensuring product purity and structural integrity. LiCl-assisted stripping of multilayer MXene is simple to operate, yields good stripping results, and produces few-layer MXene with a large specific surface area and outstanding conductivity, significantly improving the signal transmission efficiency of the subsequent sensor and avoiding conductivity obstruction caused by multilayer structures. In step (ii), the microneedle array is structurally formed and functionalized through 3D printing and layer-by-layer modification, balancing structural stability, electrochemical performance, and detection specificity. 3D-printed PMMA microneedles not only offer high forming precision and regular structure but also accurately match the detection scenario. Furthermore, PMMA material exhibits good biocompatibility and high mechanical strength, preventing damage to the sample. A gold layer is deposited on the microneedle surface to further enhance conductivity and biocompatibility, providing a good substrate for MXene loading and signal transduction. Immersion coating with Ti3C2Tx_MXene aqueous dispersion and drying achieve a stable bond between MXene and the gold layer, enhancing electrochemical signal amplification. Finally, a GOx solution is coated for functionalization, endowing the sensor with specific detection capabilities, enabling accurate identification of target substances, while ensuring the stability of the functional layer and preventing detachment. In step (iii), through precise assembly of multiple components, a compact, stable, and directly applicable integrated biosensor is constructed, improving detection convenience and reliability. The GTA-MNs array (working electrode) is precisely assembled with the Ag / AgCl reference electrode and platinum counter electrode to form a complete electrochemical detection system, ensuring the accuracy and stability of the detection signal and avoiding signal interference caused by electrode misalignment. Combined with an integrated circuit board, it enables rapid acquisition, transmission and processing of detection signals, simplifying the detection operation process and improving detection efficiency. The overall assembly structure is compact and highly adaptable, which can meet the needs of rapid on-site detection and overcome the shortcomings of traditional sensor structures that are loose and cumbersome to detect.

[0008] Preferably, as an improvement, in (a) the preparation of Ti3C2Tx_MXene, the volume ratio of HF, HCl and deionized water in the mixture is 1:5~7:2~4; the concentration of HF is 28~30M, the concentration of HCl is 11~13M; and the mass-volume ratio of Ti3AlC2 powder to the mixture is 1~2g:5~9ml.

[0009] Preferably, as an improvement, the etching is carried out at 25±2℃ and stirred at 500~600rpm for 24h using a polytetrafluoroethylene rod; the etched product is repeatedly washed with deionized water, centrifuged until the pH of the supernatant is 5.5~6.5, and the resulting black precipitate is collected and vacuum dried at 50±2℃ for 24h to obtain multilayer MXene.

[0010] Preferably, as an improvement, the stripping involves vigorously shaking the mixture until bubbles are generated, then stirring at 300-400 rpm for 24 hours at room temperature, collecting the product by centrifugation, and repeating the washing cycle multiple times until a clear supernatant is obtained; the supernatant from multiple washings is collected and concentrated to obtain a few-layer Ti3C2Tx_MXene.

[0011] Preferably, as an improvement, in (ii) the preparation of the GTA-MNs array, the Ti3C2Tx_MXene aqueous dispersion comprises the following components with the following mass concentrations: Ti3C2Tx_MXene is 1~5 mg / mL, and the balance is deionized water.

[0012] Preferably, as an improvement, the GOx solution is prepared by mixing 8 mL of 1% bovine serum albumin, 1 mL of 2.5% glutaraldehyde, and 2 mL of GOx enzyme solution.

[0013] Preferably, as an improvement, the integrated circuit board is used to achieve real-time glucose monitoring by accurately reading the current signal of the three-electrode sensor system; it includes a three-port interface for stably connecting the working electrode, the counter electrode, and the reference electrode; the GTA-MNs array is connected to the working electrode.

[0014] Preferably, as an improvement, the core constant potential circuit of the integrated circuit board is based on a high-gain transimpedance amplifier, which maintains potential stability while converting the sensor current signal into voltage; a digital-to-analog converter is used to provide a precise bias voltage for the GTA-MN biosensor to ensure stable operating conditions; the generated analog signal is then digitized by a high-precision analog-to-digital converter and processed by a low-power microcontroller; the processed data is transmitted to a smartphone application for display via Bluetooth Low Energy technology; the entire GTA-MN biosensor is powered by a miniature lithium-ion battery and integrates a transient voltage suppression diode to provide electrostatic discharge protection.

[0015] Preferably, as an improvement, this solution also provides a GTA-MN biosensor prepared by the above method.

[0016] Preferably, as an improvement, this solution also provides an application of the GTA-MN biosensor in monitoring blood glucose levels, including (1) using the GTA-MNs biosensor to monitor blood glucose in a STZ-induced type 1 diabetic rat model, and (2) intervening by intraperitoneal injection of glucose or insulin during the monitoring period, and simultaneously evaluating the batch-to-batch stability and continuous monitoring performance of the GTA-MNs biosensor.

[0017] The principle and advantages of this scheme are: (1) This invention proposes a wearable continuous glucose monitoring platform using Ti3C2Tx MXene-gold microneedles functionalized with glucose oxidase, which can effectively monitor biological blood glucose continuously and improve the accuracy of physiological detection results of interstitial fluid glucose level.

[0018] (2) This invention utilizes the excellent electrochemical properties of MXene to enable the sensor to have an ultra-wide detection range of 10 nM to 50 mM, which can cover blood glucose concentrations from extremely low to extremely high (including HHS-related) levels, with a detection limit as low as nanomolar level (1.444 nM), effectively improving detection accuracy.

[0019] (3) This invention introduces a machine learning-assisted calibration strategy for the first time, which effectively compensates for the physiological time lag between ISF and blood glucose, achieves high-precision matching with reference values, and solves the key bottleneck of existing ISF monitoring.

[0020] (4) The present invention was evaluated under challenging conditions such as an environment rich in interfering substances, artificial ISF and in vivo rat model, and confirmed that the sensor has excellent working stability for more than 10 days and excellent anti-interference ability.

[0021] In summary, this invention, by integrating materials science and data analysis, promotes the development of sensitive, robust, and clinically reliable CGM systems, which helps to achieve personalized diabetes management. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the GTA-MNs biosensor in Embodiment 1 of the present invention.

[0023] Figure 2 This is a differential charge density and Bader charge analysis diagram of Au, Au-Ti3C2Tx in Embodiment 1 of the present invention (yellow: charge accumulation, cyan: charge depletion).

[0024] Figure 3This is a Gibbs free energy analysis diagram of the decomposition of hydrogen peroxide on the Au, Au-Ti3C2Tx surface in Example 1 of the present invention.

[0025] Figure 4 This is a cyclic voltammetry diagram of Au-GOx and Au-Ti3C2Tx-GOx in Embodiment 1 of the present invention.

[0026] Figure 5 This is a comparative graph showing the effect of Ti3C2Tx_MXene on the glucose redox reaction in Example 1 of the present invention.

[0027] Figure 6 This is a diagram illustrating the fabrication process of the GTA-MNs array in Embodiment 1 of the present invention.

[0028] Figure 7 The in vitro detection performance of the GTA-MNs array in Example 1 of this invention is shown in the graphs (a. Current-time (It) response curve of the GTA-MNs array in 10 nM-50 mM glucose solution; b. Calibration curve obtained by fitting data based on a).

[0029] Figure 8 The interstitial fluid glucose levels monitored by the GTA-MNs array were calibrated using machine learning in Example 1 of this invention (a. Correlation analysis before calibration; b. Correlation analysis after calibration).

[0030] Figure 9 This is a photo of the smartphone-based fully integrated continuous glucose monitoring (CGM) system in Embodiment 1 of the present invention.

[0031] Figure 10 This is a comparison of blood glucose concentration in hypoglycemic rats before and after calibration in Example 1 of the present invention (commercial blood glucose meter, blood glucose value measured by the sensor before machine learning calibration, and blood glucose value measured by the sensor after machine learning calibration).

[0032] Figure 11 This is a comparison of blood glucose concentration in hyperglycemic rats before and after calibration in Example 1 of the present invention (commercial blood glucose meter, blood glucose value measured by the sensor before machine learning calibration, and blood glucose value measured by the sensor after machine learning calibration).

[0033] Figure 12 This is a long-term stability evaluation of the biosensor in Embodiment 1 of the present invention, lasting up to 20 days. Detailed Implementation

[0034] The present invention will be further described in detail below with reference to embodiments, but the implementation of the present invention is not limited thereto. Unless otherwise specified, the technical means used in the following embodiments and experimental examples are conventional means well known to those skilled in the art, and the materials, reagents, etc. used can all be obtained commercially.

[0035] Example 1 This solution provides a method for preparing a GTA-MN biosensor, including the following steps: Step 1: Preparation of Ti3C2Tx_MXene: Multilayer MXenes were synthesized by etching 2 g of Ti3AlC2 powder in a mixed solution of 1 mL HF (29 M), 3 mL deionized water, and 6 mL HCl (12 M). The etching reaction was carried out at 25 °C for 24 h with continuous stirring at 500 rpm using a polytetrafluoroethylene (PTFE) rod. The resulting product was repeatedly washed with deionized water and centrifuged at 3500 rpm for 5 min each time, repeating the cycle until the pH of the supernatant reached approximately 6. The resulting black precipitate was collected and vacuum dried at 50 °C for 24 h.

[0036] Few-layer Ti3C2Tx_MXene was prepared by exfoliating 3g of the synthesized multilayer MXene with 1g of LiCl in 20 mL of deionized water. The mixture was first vigorously shaken for 10 min until bubbles were generated, then stirred at 300 rpm for 24 h at room temperature. The product was collected by centrifugation at 3500 rpm for 5 min, and the washing cycle was repeated six times until a clear supernatant was obtained. Finally, the supernatant was concentrated by centrifugation at 3500 rpm for 30 min to obtain the few-layer MXene product.

[0037] Step 2: The electrochemical sensing performance of Ti3C2Tx_MXene was tested using the following method: (1) The electrocatalytic performance of Ti3C2Tx was evaluated by density functional theory (DFT) calculation and experimental verification to examine its potential in improving the sensing performance of GOx-based sensors.

[0038] (2) Density functional theory calculations: The electrocatalytic performance of Ti3C2Tx was evaluated by using VASP to perform Gibbs free energy, differential charge density and Bader charge analysis.

[0039] (3) Experimental verification: Cyclic voltammetry curves and glucose detection kits were used to verify the electrocatalytic performance of Ti3C2Tx.

[0040] Step 3: Fabrication of GTA-MNs array: First, polymethyl methacrylate (PMMA) microneedles were prepared using 3D printing technology. Then, a gold layer was deposited on the surface of the PMMA microneedles using magnetron sputtering to obtain Au-MNs. Next, the Au-MNs were dipped into a Ti3C2Tx_MXene aqueous dispersion and dried at 30°C for 2 hours to form Ti3C2Tx-Au-MNs. Finally, the Ti3C2Tx-Au-MNs were functionalized by coating with a GOx solution; this GOx solution consisted of 8 mL of 1% bovine serum albumin, 1 mL of 2.5% glutaraldehyde, and 2 mL of GOx enzyme solution.

[0041] Step 4: In vitro performance testing of the GTA-MNs array: The characteristic response peak was determined using cyclic voltammetry. All subsequent chronoamperometry tests were performed at this peak potential of 0.75 V with a scan rate of 0.1 V / s. The linear detection range and detection limit of the sensor were determined by chronoamperometry tests in glucose solutions ranging from 1 nM to 50 mM, using the endpoint current as the response value. Continuous monitoring capability was validated in a biomimetic tissue model consisting of a 2 cm thick 10% agarose gel pre-equilibrated for at least 8 h in an artificial interstitial fluid containing a known glucose concentration. After inserting the GTA-MNs biosensor, glucose solutions of different concentrations were added sequentially for testing. Specificity was assessed by detecting the sensor's response to a 10 mM potential interfering solution. Finally, the intra-batch stability of the sensor was evaluated by chronoamperometry tests in 0.4 mM, 5 mM, and 10 mM glucose solutions.

[0042] Step 5: Machine learning to calibrate interstitial fluid glucose levels: A quantitative correlation between sensor current and blood glucose concentration was established using a support vector regression (SVR) model. From a total of 1000 pairs of measurements, the dataset was divided into training and test subsets in an 8:2 ratio. An SVR model was employed to capture this complex relationship. The model hyperparameters (regularization parameter C and kernel coefficient gamma) were optimized using grid search combined with cross-validation on the training set. The optimized model was then trained on the full training set, and the performance of the calibration model in converting current signals to reference blood glucose values ​​was evaluated on the test set using mean squared error and coefficient of determination.

[0043] Step Six: Fabrication of GTA-MNs Biosensors: A custom integrated circuit board was designed and developed for real-time glucose monitoring by accurately reading the current signal of a three-electrode sensor system. The core of the board is a three-port interface for stable connection of the working electrode, counter electrode, and reference electrode. Its core potentiostatic circuit is based on a high-gain transimpedance amplifier, maintaining potential stability while converting the sensor current signal into voltage. A digital-to-analog converter (DAC) provides a precise bias voltage to the GTA-MN biosensor, ensuring stable operating conditions. The generated analog signal is then digitized by a high-precision DAC and processed by a low-power microcontroller. Data is transmitted to a smartphone application for display via Bluetooth Low Energy. The entire GTA-MN biosensor is powered by a miniature lithium-ion battery and integrates a transient voltage suppression diode for electrostatic discharge protection. Finally, by integrating this integrated circuit board with a GTA-MN array, an Ag / AgCl reference electrode, and a platinum counter electrode, a complete GTA-MN biosensor is assembled (see schematic diagram of the resulting GTA-MN biosensor shown below). Figure 1 (As shown).

[0044] Step 7: In vivo performance testing of the GTA-MNs biosensor: The in vivo performance of the GTA-MNs biosensor was evaluated using an STZ-induced type 1 diabetic rat model. Rats with persistently high blood glucose levels (above 16.67 mmol / L) were selected for further research using commercially available blood glucose meters.

[0045] Before sensor application, the fur on the backs of rats was shaved. A GTA-MN array was then attached to the shaved area. Continuous glucose monitoring was performed using a chronoamperometry method at a fixed potential of 0.75 V, with a sampling frequency set every 20 minutes. Simultaneously, a reference blood glucose level was measured every 20 minutes using a FreeStyle_Libre blood glucose meter via tail vein sampling. To simulate different blood glucose states and evaluate sensor response, interventions were performed during monitoring via intraperitoneal injection of glucose or insulin. Furthermore, the inter-batch stability of the sensor and its continuous monitoring performance for up to 20 days were evaluated concurrently.

[0046] Results Display: 1. Density Functional Theory (DFT) Calculations to Evaluate the Electrocatalytic Performance of Ti3C2Tx: Density functional theory calculations were performed using the Vienna Ab initio simulation package (VASP). The exchange correlation potential was described using the Perdew-Burke-Ernzerhof functional (GGA-PBE) with the generalized gradient approximation. The interaction between the ionic core and valence electrons was handled using the projected appended plane wave (PAW) method. The plane wave cutoff energy was set to 450 eV. Relaxation calculations were performed on the given structural model until the interatomic Helerman-Feynman force was less than 0.02 eV / Å and the energy change was less than 10 eV / Å. -5 eV. Grimm's DFT-D3 method was used to describe the dispersive interactions between all atoms in the adsorption model. The grid at the center k-point of Gamma was set to 2×2×1. The vacuum layer thickness along the z-direction was set to 12 Å.

[0047] like Figure 2 As shown, density functional theory (DFT) calculations were used to determine the differential charge density of electrons and Bader charge analysis when isochloropyrazine (FAD core, which is also the core of glucose oxidase in redox reactions) adsorbed on Au and Au-Ti3C2Tx surfaces. The results revealed that the electron-rich region was mainly located near the isochloropyrazine, while the charge density of gold (Au) and Au-Ti3C2Tx active sites was significantly reduced, indicating a strong electron transfer from these active sites to the isochloropyrazine. Quantitative Bader charge analysis showed that the charge transfer at Au-Ti3C2Tx sites reached 0.737 e, significantly higher than the 0.334 e at Au sites. This difference indicates that isochloropyrazine exhibits a stronger adsorption tendency on the Au-Ti3C2Tx surface compared to single gold sites. This type of charge transfer characteristic provides crucial evidence for understanding the selective adsorption mechanism of catalytic active sites at the electronic structure level.

[0048] Figure 3Gibbs free energy diagrams for the decomposition of H2O2 on Au and Au-Ti3C2Tx surfaces were compared. Throughout the reaction pathway, each step on the Au-Ti3C2Tx surface exhibited a lower free energy than on the pure Au surface. The key difference lies in the initial H2O2 adsorption (Au-Ti3C2Tx: -0.64 eV, Au: 0.14 eV) and the final product formation stage (total ΔG: -3.11 eV, Au-Ti3C2Tx: -2.41 eV, Au: -0.74 eV). H2O2 is an intermediate product of the redox reaction of glucose, and its further decomposition is central to electron transfer; the more and faster the decomposition, the greater the electron transfer. These results indicate that the introduction of Ti3C2Tx significantly enhances the adsorption and dissociation capabilities of H2O2.

[0049] 2. The electrocatalytic performance of Ti3C2Tx was verified using electrochemical cyclic voltammetry. Figure 4 Cyclic voltammetry (CV) was used to analyze the cyclic voltammetric curves of Au-GOx and Au-Ti3C2Tx-GOx electrodes. A distinct redox peak pair was observed on the Au-Ti3C2Tx-GOx electrode at a potential of approximately 0.48 V (relative to Ag / AgCl), while no characteristic peak was observed on Au-GOx. This peak corresponds to the characteristic reaction at the FAD / FADH2 center of glucose oxidase (GOx) cofactors, thus confirming the existence of a direct electron transfer (DET) pathway. This result indicates that efficient, mediator-free direct electron transfer is achieved between GOx and Ti3C2Tx.

[0050] 3. The catalytic performance of Ti3C2Tx was verified using a glucose assay kit. Figure 5 The catalytic performance of Ti3C2Tx was tested using a glucose assay kit. The results showed that the absorbance value after the reaction was higher in the presence of Ti3C2Tx, and the difference was statistically significant, confirming that it has good catalytic performance.

[0051] 4. Construct GTA-MNs array First, polymethyl methacrylate (PMMA) microneedles were prepared using 3D printing technology. Then, a gold layer was deposited on the surface of the PMMA microneedles using magnetron sputtering to obtain Au-MNs. Next, the Au-MNs were dipped into a Ti3C2Tx MXene aqueous dispersion and dried at 30°C for 2 h to form Ti3C2Tx-Au-MNs. Finally, the Ti3C2Tx-Au-MNs were functionalized by coating with a GOx solution; this GOx solution was prepared by mixing 10 mL of 1% bovine serum albumin, 1 mL of 2.5% glutaraldehyde, and 2 mL of GOx enzyme solution. Figure 6 ).

[0052] 5. In vitro detection performance of GTA-MNs array Chronoamperometry tests were conducted in glucose solutions ranging from 1 nM to 50 mM, with the endpoint current as the response value, to determine the linear detection range and detection limit of the sensor. The results are as follows: Figure 7 As shown. The results indicate that the sensor exhibits excellent linear response (R0). 2 =0.9617), especially in the low concentration range (10 nM-1 mM; R 2 = 0.9834) outstanding performance (e.g. Figure 7 (As shown in a). This demonstrates its ultra-wide detection range of 10 nM to 50 mM, covering blood glucose concentrations from extremely low to extremely high (including HHS-related) levels, with a detection limit as low as the nanomolar level (1.444 nM) (e.g. Figure 7 (as shown in b).

[0053] 6. Machine learning to calibrate interstitial fluid glucose levels monitored by GTA-MNs arrays During continuous monitoring, we observed that although the sensor current signal generally reflected the trend of blood glucose level changes, there was still significant hysteresis and signal fluctuation. Therefore, we established a support vector regression (SVR) model to correct for the physiological hysteresis of glucose concentration between tissue fluid and blood.

[0054] For sensor calibration, we first divided the blood glucose values ​​and their corresponding current signals obtained from dynamic monitoring into training and validation sets in an 80:20 ratio. After filtering out outliers from the training set data using random forest and local outlier factor algorithms, preliminary analysis showed an approximately linear relationship between blood glucose values ​​and current signals. Based on this, we used an SVR model for calibration modeling. Before calibration, the correlation between blood glucose values ​​and the original sensor signal can be expressed as Y = 0.7949X 0.8123 (R) 2 = 0.9146); After calibration with the SVR model, the linear relationship was significantly improved, and the calibration curve was Y = 1.364X + 0.3622, with the coefficient of determination increasing to R. 2 = 0.9897 ( Figure 8 ).

[0055] 7. Integration of GTA-MNs biosensors A custom integrated circuit board was designed and developed for real-time glucose monitoring by accurately reading the current signal of a three-electrode sensor system. The core of the board is a three-port interface for stable connection of the working electrode, counter electrode, and reference electrode. Its core potentiostatic circuit is based on a high-gain transimpedance amplifier, maintaining potential stability while converting the sensor current signal into voltage. A digital-to-analog converter (DAC) provides a precise bias voltage to the GTA-MN biosensor, ensuring stable operating conditions. The generated analog signal is then digitized by a high-precision DAC and processed by a low-power microcontroller. Data is transmitted to a smartphone application for display via Bluetooth Low Energy. The entire GTA-MN biosensor is powered by a miniature lithium-ion battery and integrates a transient voltage suppression diode for electrostatic discharge protection. Finally, by integrating this integrated circuit board with a GTA-MN array, an Ag / AgCl reference electrode, and a platinum counter electrode, a complete GTA-MN biosensor is assembled. Figure 9 (Image showing the appearance of the GTA-MNs biosensor and its state when detecting blood glucose in rats).

[0056] 8. In vivo monitoring performance of GTA-MNs biosensors Before applying the sensor, the hair on the back of the rat was shaved. The GTA-MNs array was then attached to the shaved area. Figure 9 Continuous glucose monitoring was performed using the chronoamperometry method at a fixed potential of 0.75 V, with a sampling frequency set every 20 minutes. Simultaneously, a reference blood glucose level was measured every 20 minutes using a FreeStyle_Libre blood glucose meter via tail vein sampling. To simulate different blood glucose states and evaluate sensor response, interventions were performed during monitoring via intraperitoneal injection of glucose or insulin. Figure 10 , Figure 11 The results showed that, after machine learning calibration, the blood glucose levels measured by this sensor were essentially consistent with those of commercially available blood glucose meters. Furthermore, the inter-batch stability of the sensor and its continuous monitoring performance for up to 20 days were simultaneously evaluated. Figure 12 ).

[0057] The above descriptions are merely embodiments of the present invention, and common knowledge such as specific technical solutions and / or characteristics are not described in detail here. It should be noted that those skilled in the art can make various modifications and improvements without departing from the technical solutions of the present invention, and these should also be considered within the scope of protection of the present invention. These modifications and improvements will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A method for preparing a GTA-MN biosensor, characterized in that: Includes the following steps: (a) Preparation of Ti3C2Tx_MXene: Ti3AlC2 powder was etched in a mixture of HF, HCl and deionized water to synthesize multilayer MXene; the multilayer MXene was dispersed with LiCl in deionized water to obtain a mixture, and the multilayer MXene in the mixture was peeled off to prepare few-layer Ti3C2Tx_MXene; (II) Preparation of GTA-MNs array: Au-MNs were obtained by depositing a gold layer on the surface of PMMA microneedles through 3D printing; Au-MNs were dipped into Ti3C2Tx_MXene aqueous dispersion and dried to form Ti3C2Tx-Au-MNs; Ti3C2Tx-Au-MNs were functionalized by coating with GOx solution to obtain GTA-MNs array; The GOx solution was prepared by mixing 8 mL of 1% bovine serum albumin, 1 mL of 2.5% glutaraldehyde, and 2 mL of GOx enzyme solution. (III) Preparation of GTA-MN biosensor: GTA-MN array, Ag / AgCl reference electrode, platinum counter electrode and integrated circuit board are assembled into GTA-MN biosensor.

2. The method for preparing the GTA-MN biosensor according to claim 1, characterized in that: In the preparation of Ti3C2Tx_MXene (I), the volume ratio of HF, HCl and deionized water in the mixture is 1:5~7:2~4; the concentration of HF is 28~30M and the concentration of HCl is 11~13M; the mass-volume ratio of Ti3AlC2 powder to the mixture is 1~2g:5~9ml.

3. The method for preparing the GTA-MN biosensor according to claim 1, characterized in that: The etching was carried out at 25±2℃ and stirred at 500~600rpm for 24 h using a polytetrafluoroethylene rod. The etched product was repeatedly washed with deionized water and centrifuged until the pH of the supernatant was 5.5~6.

5. The black precipitate was collected and vacuum dried at 50±2℃ for 24 h to obtain multilayer MXene.

4. The method for preparing the GTA-MN biosensor according to claim 1, characterized in that: The stripping process involves vigorously shaking the mixture until bubbles are generated, followed by stirring at 300-400 rpm for 24 h at room temperature. The product is collected by centrifugation and the washing cycle is repeated multiple times until a clear supernatant is obtained. The supernatant from multiple washes is collected and concentrated to obtain a few-layer Ti3C2Tx_MXene.

5. The method for preparing the GTA-MN biosensor according to claim 1, characterized in that: In (ii) the preparation of GTA-MNs array, the Ti3C2Tx_MXene aqueous dispersion includes the following components with the following mass concentrations: Ti3C2Tx_MXene is 1~5 mg / mL, and the balance is deionized water.

6. The method for preparing the GTA-MN biosensor according to claim 1, characterized in that: The integrated circuit board is used to achieve real-time glucose monitoring by accurately reading the current signal of the three-electrode sensor system; it includes a three-port interface for stable connection of the working electrode, counter electrode and reference electrode; the GTA-MNs array is connected to the working electrode.

7. The method for preparing the GTA-MN biosensor according to claim 6, characterized in that: The core constant potential circuit of the integrated circuit board is based on a high-gain transimpedance amplifier, which maintains a stable potential while converting the sensor current signal into a voltage. A digital-to-analog converter is used to provide a precise bias voltage for the GTA-MN biosensor, ensuring stable operating conditions. The generated analog signal is then digitized by a high-precision analog-to-digital converter and processed by a low-power microcontroller. The processed data is transmitted to a smartphone application for display via Bluetooth Low Energy technology. The entire GTA-MN biosensor is powered by a miniature lithium-ion battery and integrates a transient voltage suppression diode to provide electrostatic discharge protection.

8. A GTA-MN biosensor, characterized in that: Prepared by the method described in any one of claims 1 to 7.

Images