Sensor integrating sweat induction and multi-parameter detection as well as preparation method and application of sensor

Abstract

The invention discloses a sweat induction and multi-parameter detection integrated sensor and a preparation method and application thereof, and belongs to the technical field of flexible biosensing, a PDMS flexible substrate is adopted to construct a multi-layer integrated structure, the sweat induction and multi-parameter detection integrated sensor comprises a sweat guide layer, a sweat collection layer, a detection electrode layer and a micro-fluidic chip layer, sweat secretion is promoted by accurately controlling the temperature to be less than or equal to 40 DEG C; the detection electrode layer adopts a silk-screen printing process to prepare a three-electrode system, comprises a specifically modified glucose working electrode and a beta-hydroxybutyric acid working electrode and can synchronously detect the concentration of glucose and beta-hydroxybutyric acid in sweat, and the micro-fluidic chip layer is integrated with a Bluetooth transmission module to realize real-time wireless transmission of detection data to a mobile terminal; the biosensor has the advantages of being good in biocompatibility, high in detection sensitivity and comfortable to wear and can tolerate bending with the curvature radius larger than 5 mm, and a non-invasive and continuous metabolism monitoring solution is provided for diabetes management.

Description

Technical Field

[0001] This invention belongs to the field of flexible biosensing technology, specifically relating to a sensor that integrates sweat induction and multi-parameter detection, its preparation method, and its application. Background Technology

[0002] Diabetes mellitus has become the fastest-growing chronic metabolic disease globally, and effective disease management heavily relies on continuous and reliable blood glucose monitoring technologies. Currently, routinely used single-point invasive testing methods (such as finger-prick blood sampling and glycated hemoglobin (HbA1c) testing) have significant limitations: these methods only provide blood glucose data at discrete time points and cannot fully reflect the patient's 24-hour blood glucose fluctuations. Of particular note is the insufficient sensitivity of HbA1c testing for acute metabolic events—recent clinical studies have shown that when the difference between the real-time blood glucose level and the estimated HbA1c value (Glucose Delta) exceeds 40 mg / dL in patients with myocardial infarction, their risk of cardiac injury increases significantly by 2.1 times. Furthermore, the pain and discomfort caused by repeated blood sampling, the potential risk of infection, and the resulting psychological resistance all severely impact patient compliance, a problem particularly pronounced in pediatric patients and patients with chronic kidney disease.

[0003] In terms of dynamic monitoring technology, while continuous glucose monitoring (CGM) systems can provide more comprehensive data on blood glucose changes, existing technologies still have many inherent drawbacks. Minimally invasive CGM requires subcutaneous implantation of rigid or soft microneedle sensors. Rigid microneedles are prone to displacement during patient movement, leading to measurement errors as high as 15%-20%; while soft microneedles typically have a lifespan of less than 14 days due to enzyme activity decay. Non-invasive optical detection technologies (such as near-infrared spectroscopy) avoid invasive procedures, but their results are easily affected by factors such as patient skin color, subcutaneous fat thickness, and ambient light, and their reliability in clinical trials still needs improvement. Electrochemical sweat sensors, as an emerging non-invasive detection technology, have significant application advantages but also face serious challenges: studies show that at the same blood glucose level, the difference in sweat glucose concentration between individuals can be as much as five times; in addition, environmental pollutants on the skin surface (such as sebum and sunscreen) can easily interfere with the detection results. More importantly, passive sweat collection methods have a failure rate exceeding 50% in the resting state (sweat secretion is typically below 0.1 μL / min), and current technologies generally lack the ability to simultaneously detect multiple biomarkers. Taking diabetic ketoacidosis, an acute complication of diabetes, as an example, the concentration threshold (>0.5 mM) of its core warning indicator, β-hydroxybutyrate (β-HB), needs to be monitored simultaneously with blood glucose levels. However, most current sensors can only detect glucose as a single indicator, which cannot meet the clinical needs for comprehensive risk assessment of complications.

[0004] In recent years, the field of flexible electrochemical sweat sensors has significantly improved wearability through material innovation (such as using PDMS substrates and PVA hydrogels) and structural optimization (flexible designs with a curvature radius > 5 mm). However, it still faces three key technical bottlenecks: First, there is a physiological lag of 10-15 minutes between sweat glucose concentration and blood glucose levels. Combined with individual differences in sweat gland density and skin thickness, this results in poor universality of calibration models. Second, passive sweat collection methods rely excessively on ambient temperature, leading to insufficient reliability of detection in resting states. Existing active induction technologies (such as iontophoresis) may alter the concentration of body fluid components. Third, multi-enzyme synergistic detection faces the challenge of biochemical environmental conflicts—the optimal pH range for glucose oxidase (GOx) is 6-8, while β-hydroxybutyrate dehydrogenase (HBD) requires a weakly alkaline environment (pH 7.5-9.0). This difference makes the mixed immobilization of the two enzymes highly susceptible to mutual inhibition of activity. Although the glucose sensor developed by the Shenzhen University research team has achieved a sensitivity of 12.69 μA·mM... -1 ·cm 2 The latest microfluidic SERS sensor reported in 2024 also successfully achieved dual detection of glucose and urea, but these technologies failed to cover the detection of β-hydroxybutyric acid (β-HB), and the stability of enzyme electrodes at 25 °C was generally less than two weeks (relative standard deviation RSD > 10%).

[0005] Therefore, developing a new type of automotive nonwoven material that retains the excellent mechanical properties and processing characteristics of traditional PET nonwoven fabrics while possessing higher heat resistance and lower heat shrinkage has become a key technical problem that the industry urgently needs to solve. This invention aims to introduce a filament component with specific low-temperature viscoelastic behavior to achieve effective bonding at relatively low temperatures while significantly improving the high-temperature dimensional stability and durability of the finished nonwoven fabric, thereby expanding its application potential in automotive components operating under high-temperature conditions. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this application provides a sensor integrating sweat induction and multi-parameter detection, its preparation method, and its application. This solves the problems of insufficient sensitivity of HbA1c detection to acute metabolic events, pain and discomfort from repeated blood sampling, potential infection risks, and resulting psychological resistance, all of which seriously affect patient compliance. Rigid microneedles have measurement errors as high as 15%-20%; while soft microneedles, due to enzyme activity decay, typically have a lifespan of less than 14 days. Furthermore, the detection results are easily affected by factors such as patient skin color, subcutaneous fat thickness, and ambient light. Environmental pollutants can easily interfere with detection results, and there is a lack of simultaneous detection capabilities for multiple biomarkers. Most sensors can only detect glucose as a single indicator, the universality of calibration models is poor, the reliability of detection under resting conditions is insufficient, and the mixed immobilization of two enzymes can easily lead to mutual inhibition of activity. This invention achieves real-time simultaneous detection of two parameters, glucose and β-HB, in sweat, and realizes early warning of metabolic abnormalities through intelligent algorithms. By optimizing the sweat collection and detection system, the detection sensitivity and efficiency are significantly improved. It has a low preparation cost, is suitable for large-scale production, and can meet the daily monitoring needs of diabetes monitoring devices.

[0007] The technical solution adopted in this invention is as follows: A method for fabricating a sensor integrating sweat induction and multi-parameter detection includes the following steps: Step 1, sweat-wicking layer preparation: PDMS is used as the base material. The PDMS base is engraved with CO2 laser to form an array of sweat collection holes with a diameter of 3.0 ± 0.1 mm and a hole spacing of 5 ± 0.1 mm. Then, a silver paste heating circuit is screen-printed. The silver paste heating circuit is connected to a temperature control module composed of a copper foil heating element and an NTC thermistor to achieve precise temperature control of 39 ± 0.5℃. Step 2, Preparation of the sweat collection layer: Using PDMS material, a microfluidic network is laser-engraved to form sweat collection channels, a set of sweat collection holes, and a reservoir. The sweat collection holes are connected to the circular sweat collection hole array of the sweat-guiding layer. Sweat passes through the sweat-guiding layer to the sweat collection holes and then flows through the sweat collection channels to the reservoir. The sweat collection channel is divided into a front end and a rear end. The front end of the sweat collection channel is connected to the reservoir, and the rear end is connected to the sweat collection holes. The front end of the sweat collection channel is 0.5±0.02mm wide, and the rear end is 1.0±0.02mm wide. The reservoir is cylindrical, and a sweat outlet is provided at the rear end of the reservoir. Step 3, Preparation of the detection electrode layer: The detection electrode layer is constructed using a screen printing process to create a three-electrode system, which includes a platinum auxiliary electrode, an Ag / AgCl reference electrode, and two functionalized working electrodes; the functionalized working electrodes are the glucose detection working electrode and the β-hydroxybutyric acid (β-HB) detection working electrode; Step 4, integrate microfluidic detection integrated chip layer: integrate the main controller, Bluetooth module, power module, TXS0102, ADC converter, GND, temperature sensor, signal conditioning, sensor glucose detection interface and sensor β-HB detection interface on flexible material PDMS. Step 5: The prepared microfluidic detection integrated chip layer, detection electrode layer, sweat collection layer, and sweat-conducting layer are bonded together layer by layer using medical adhesive to obtain an integrated sweat induction and multi-parameter detection sensor with an overall thickness of 1.5 ± 0.1 mm.

[0008] Preferably, the preparation method of the substrate material PDMS in step 1 is as follows: the main agent is Dow Corning Sylgard 184 silicone rubber prepolymer, and the curing agent is a platinum-catalyzed crosslinking agent that matches Dow Corning Sylgard 184 silicone rubber prepolymer. During preparation, the main agent and curing agent are accurately weighed in a mass ratio of 10:1, and after being thoroughly mixed by mechanical stirring, they are degassed under vacuum and then cast into shape. The molding and curing temperature is 80 ℃, the molding time is 2 h, and the final thickness is controlled at 200 ± 10 μm.

[0009] Preferably, in step 1, the CO2 laser engraving parameters are set as follows: wavelength 10.6 μm, 20±1W; line width of heating circuit 0.1 ± 0.01 mm, copper foil heating element thickness 35 μm, resistance 10 ± 1Ω; NTC thermistor 10 KΩ, B value 3950 K.

[0010] Preferably, in step 2, the diameter of the sweat collection hole is 3mm, and the contact angle of less than 30° is formed by the width difference between the front end and the rear end of the sweat collection channel, thereby forming capillary force drive. The depth of the sweat collection channel is 0.25±0.01mm; the diameter of the liquid storage pool is 8.0±0.2mm, the depth is 800±20μm, and the size of the sweat outlet is 2.0 × 3.0 mm.

[0011] Preferably, the specific preparation steps of the detection electrode layer in step 3 are as follows: The first step involves sequentially forming a carbon paste-based working electrode, an Ag / AgCl reference electrode, and a platinum auxiliary electrode on a PDMS substrate. The second step is to construct the rGO layer: on the working electrode of the carbon paste substrate, the reduced graphene oxide rGO after liquid metal gallium exfoliation is modified by drop casting, and the thickness of the rGO layer is 50±5 nm. The third step is to spin-coat the chitosan solution CS: spin-coat 2 wt% chitosan solution onto the rGO layer at a spin speed of 3000 rpm for 30 seconds. The fourth step involves immobilizing GOx and PB: A Prussian blue PB layer is formed by electrodeposition at a potential of -0.2V for 300 seconds. The electrolyte is a mixture of 0.1 M FeCl3 and 0.1 M K3[Fe(CN)6]. Subsequently, glucose oxidase with a concentration of 10 mg / mL and cross-linked with 1% glutaraldehyde is immobilized to form a glucose oxidase GOx composite layer, thus obtaining the glucose detection working electrode. Step 5, Preparation of the β-HB detection working electrode: On the working electrode of the carbon paste substrate, reduced graphene oxide (rGO) after liquid gallium exfoliation was modified by drop casting. The rGO layer thickness was 50±5 nm. Then, 2 wt% chitosan solution was spin-coated onto the rGO layer at 3000 rpm for 30 seconds. Next, 5±0.1 mM TBO solution was drop-coated to modify toluidine blue O (TBO). Crosslinking was performed at 25±1℃ for 4±0.5 h to fix 5 mM coenzyme NAD. + A β-HB detection working electrode was prepared by curing 2 U / cm² β-hydroxybutyrate dehydrogenase (HBDH) at 220±5 °C for 30±1 min.

[0012] Preferably, in step 4, the Altium Designer design integrates a 72MHz main controller AT89C52, a Bluetooth module HC-05 with a transmission distance of 10 m, a power module, TXS0102, an ADC converter ADS1115, GND, a temperature sensor, a signal conditioning AD623 instrumentation amplifier, a glucose detection interface for the sensor, and a β-HB detection interface for the sensor. The circuit connection between each module is realized using a pre-set ribbon cable interface.

[0013] A sensor for integrated sweat induction and multi-parameter detection prepared by any of the above preparation methods is composed of a microfluidic detection integrated chip layer, a detection electrode layer, a sweat collection layer and a sweat-conducting layer arranged sequentially from top to bottom; The microfluidic detection integrated chip layer integrates a Bluetooth module, a main controller, a power module, a TXS0102, an ADC converter, a GND, a temperature sensor, a signal conditioning AD623 instrumentation amplifier, a glucose detection interface, and a β-HB detection interface on a PDMS flexible substrate. The detection electrode layer is constructed using a screen printing process to create a three-electrode system, which includes a platinum auxiliary electrode, an Ag / AgCl reference electrode, and two functionalized working electrodes. The functionalized working electrodes are the glucose detection working electrode and the β-hydroxybutyric acid (β-HB) detection working electrode. The sweat collection layer is made of PDMS material and has laser-engraved sweat collection channels inside. The cross-section of the sweat collection channel has a gradient structure. The front end of the sweat collection channel is 0.5±0.02mm wide and the rear end is 1.0mm wide, forming a contact angle of less than 30° to generate capillary force. The channel depth is 0.25±0.01mm. The end of the channel is connected to a storage tank with a volume of 0.8±0.05μL. The tail end of the storage tank is equipped with an anti-evaporation sweat vent. The sweat-wicking layer is made of PDMS material, and its surface has an array of circular sweat collection holes with a diameter of 3.0 ± 0.1 mm and a spacing of 5 ± 0.1 mm corresponding to the sweat collection hole positions. It also integrates a temperature control module, which consists of a copper foil heating element and an NTC thermistor, for precise regulation of skin temperature. The copper foil heating element has a resistance of 10 ± 1 Ω, and the NTC thermistor has an accuracy of ± 0.1 ℃.

[0014] This application also discloses the application of the integrated sweat-inducing and multi-parameter detection sensor prepared by any of the above-mentioned methods in a flexible wearable monitoring device. In use, the integrated sweat-inducing and multi-parameter detection sensor is attached to the monitoring site. A temperature control module raises the local temperature to 39 ± 0.5 ℃. Sweat enters the sweat collection layer through the circular sweat collection hole array in the sweat-wicking layer, reaching a stable flow rate of 0.8 ± 0.1 μL / min within 5 minutes. In electrochemical detection, glucose is measured at a potential of 0.4 ± 0.02 V by detecting the current signal during the H2O2 oxidation process in the glucose reaction, with a sensitivity of 12.69 ± 0.5 μA·mM. -1 ·cm 2 β-HB is detected at a potential of -0.2 ± 0.02V, where the NADH reduction current during the β-HB reaction is measured, with a detection limit of 12.0 ± 2.0 μM. The sensor integrating sweat induction and multi-parameter detection measures the glucose detection signal by 100 times and the β-HB detection signal by 50 times through an AD623 instrumentation amplifier. The signal is then transmitted through various modules of the microfluidic chip and finally transmitted to the mobile APP via Bluetooth at a frequency of 10 ± 1Hz. The monitoring method of the sensor integrating sweat induction and multi-parameter detection covers the entire process of sweat induction and electrochemical analysis.

[0015] Preferably, the flexible wearable monitoring device is a non-invasive monitoring device.

[0016] Preferably, the flexible wearable monitoring device is a diabetes monitoring device.

[0017] Explanation of Principle: This invention utilizes flexible PDMS material in both the sweat collection layer and the sweat-wicking layer, allowing the sensor to better conform to the curved surfaces of the body. The sensor's circuitry is integrated into the microfluidic chip detection layer. This microfluidic chip layer integrates a power supply unit, an STM32 microcontroller, a Bluetooth 5.0 communication module, and a potentiostat circuit, enabling core control and data transmission. This flexible wearable sensor, integrating "active sweating promotion, simultaneous detection of dual-label substances, and AI dynamic early warning," employs a unique physically isolated dual-working-electrode design, modifying a glucose oxidase / Prussian blue composite electrode (detection range 0.01-2 mM) and a β-hydroxybutyrate dehydrogenase / nicotinamide adenine dinucleotide (NAD) electrode, respectively. + The cofactor composite electrode (detection limit 12.0 μM), combined with differential potential regulation technology (+0.4V / -0.2V), effectively avoids enzyme activity interference. By innovatively integrating a PID temperature-controlled heating module (temperature control accuracy 39 ± 0.5℃), it can induce sweat secretion of 0.5-2 μL / min within 5 min, successfully overcoming the detection bottleneck under resting conditions. A microfluidic layered structure design (including a 50 μm filter membrane and overflow microcavity) ensures sample purity and liquid surface stability. The accompanying mobile APP, based on the LSTM algorithm, constructs a glucose-ketone body kinetic model, immediately triggering a ketosis warning when β-HB > 0.5 mM and glucose > 1 mM is detected, improving response speed by 60%. This innovative technology is the first to simultaneously solve key technical challenges such as sample reliability, multi-standard detection, and intelligent decision-making in a single flexible device, providing a breakthrough solution for closed-loop diabetes management.

[0018] Beneficial effects: 1. This application discloses a sensor that integrates sweat induction and multi-parameter detection, as well as its preparation method and application, to achieve real-time synchronous detection of two parameters, glucose and β-HB, in sweat; 2. The sensor for integrated sweat induction and multi-parameter detection disclosed in this application uses a PDMS flexible substrate with a Young's modulus of 0.5-2 MPa. After 5000 bending tests, the conductivity decreases by less than 3%, and the breathability is as high as 400 g / m²·day, which is 20 times that of PET. This significantly improves wearing comfort while reducing material costs by 60%. 3. The sensor disclosed in this application that integrates sweat induction and multi-parameter detection features an innovative microcavity reservoir design with a volume of 0.5-1 μL. Combined with an optimized acquisition channel, it can complete the detection within 5 seconds with only 0.3 μL of sweat. The intelligent drainage structure ensures that the liquid level fluctuation is less than ±5%, effectively avoiding concentration deviation. 4. Compared with traditional methods, the sensor integrating sweat induction and multi-parameter detection, as well as its fabrication method and application, of this application show that the electrode fabricated using screen printing technology has lower impedance, with a sheet resistance <10 Ω / sq, compared to >100 Ω·cm for traditional sputtered gold electrodes. 2 Superior performance; the electrode specific surface area is increased to 520 m² using a reduced graphene oxide substrate exfoliated with liquid metal. 2 / g, which is three times that of traditional carbon electrodes, and the electron transfer rate constant for glucose detection reaches 8.7×10 -3 The enzyme activity retention rate in β-HB detection exceeded 90% at cm / s. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the microfluidic detection integrated chip layer of the sensor that integrates sweat induction and multi-parameter detection in this application; Figure 2 This is a schematic diagram of the sensor that integrates sweat induction and multi-parameter detection in this application; Figure 3 This is a schematic diagram of the three-electrode system of the sensor integrating sweat induction and multi-parameter detection in this application; Figure 4 This is a schematic diagram of the sweat collection layer of the sensor that integrates sweat induction and multi-parameter detection in this application. Figure 5 This is a schematic diagram of the glucose detection working electrode mechanism of the sensor integrating sweat induction and multi-parameter detection in this application; Figure 6 This is a schematic diagram of the β-HB detection working electrode mechanism of the sensor integrating sweat induction and multi-parameter detection in this application; Figure 7 This is a circuit diagram of the temperature control module of the sensor integrating sweat induction and multi-parameter detection in this application; Figure 8The graphs show the detection data of the sensor integrating sweat induction and multi-parameter detection in this application in a flexible wearable monitoring device; where a is the cyclic voltammetry (CV) curve of the glucose detection working electrode, characterizing the electrochemical performance of the electrode; b is the potentiostatic-current curve of the glucose detection working electrode, with a test range of 0.01-2 mmol / L glucose solution; c is the current-concentration relationship curve of the glucose detection working electrode, with the linear equation: y = -0.0765 - 1.005x; d is the cyclic voltammetry (CV) curve of the β-HB detection working electrode, reflecting the catalytic characteristics of the electrode; e is the potentiostatic-current curve of the β-HB detection working electrode, with a test range of 0.01-2 mmol / L β-HB solution; f is the current-concentration relationship curve of the β-HB detection working electrode, with a test range of 0.01-2 mmol / L β-HB solution, and the linear equation: y = 0.684 + 0.95x.

[0020] Explanation of reference numerals in the attached diagram: 1. Microfluidic detection integrated chip layer; 2. Detection electrode layer; 3. Sweat collection layer; 4. Sweat-conducting layer; 21. Platinum auxiliary electrode; 22. Glucose detection working electrode; 23. Ag / AgCl reference electrode; 24. β-HB detection working electrode; 31. Liquid reservoir; 33. Front end of sweat collection channel; 32. Rear end of sweat collection channel; 34. Sweat collection hole. Detailed Implementation

[0021] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings. The described embodiments are used to help understand the present invention, but do not constitute a limitation on the scope of protection. Other implementation methods obtained by those skilled in the art based on the embodiments of the present invention without creative effort are all within the scope of protection of the present invention.

[0022] Example 1: A sensor integrating sweat induction and multi-parameter detection, consisting of a microfluidic detection integrated chip layer 1, a detection electrode layer 2, a sweat collection layer 3, and a sweat-conducting layer 4 arranged sequentially from top to bottom; The microfluidic detection integrated chip layer 1 integrates a Bluetooth module, a main controller, a power module, a TXS0102, an ADC converter, a GND, a temperature sensor, a signal conditioning AD623 instrumentation amplifier, a glucose detection interface, and a β-HB detection interface on a PDMS flexible substrate. The detection electrode layer 2 is constructed using a screen printing process to create a three-electrode system, which includes a platinum auxiliary electrode 21, an Ag / AgCl reference electrode 23, and two functionalized working electrodes; the functionalized working electrodes are glucose detection working electrode 22 and β-hydroxybutyric acid (β-HB) detection working electrode 24. The sweat collection layer 3 is made of PDMS material and has laser-engraved sweat collection channels inside. The cross-section of the sweat collection channel has a gradient structure. The front end 33 of the sweat collection channel is 0.5±0.02mm wide and the rear end 32 of the sweat collection channel is 1.0mm wide, forming a contact angle of less than 30° to generate capillary force drive. The channel depth is 0.25±0.01mm. The end of the channel is connected to a liquid storage tank 31 with a volume of 0.8±0.05 μL. The end of the liquid storage tank 31 is equipped with an anti-evaporation sweat vent. The sweat-wicking layer 4 is made of PDMS material and has an array of circular sweat collection holes with a diameter of 3.0 ± 0.1 mm and a spacing of 5 ± 0.1 mm corresponding to the position of the sweat collection holes 34. It also integrates a temperature control module, which consists of a copper foil heating element and an NTC thermistor, for precise regulation of skin temperature. The copper foil heating element has a resistance of 10 ± 1 Ω and the NTC thermistor has an accuracy of ± 0.1 ℃.

[0023] The PDMS substrate has an elastic modulus of 0.8-1.2 MPa. After 5000 bending tests with a curvature radius of 5 mm, the electrode impedance change is <5% and the enzyme activity loss is <3%. The temperature control module integrated on the sweat-wicking layer 4 operates in three stages: during startup, the temperature is increased to 38 ± 0.5 ℃ at a rate of 0.5 ± 0.1 ℃ / s; during the constant temperature stage, the temperature fluctuation is controlled within ± 0.5 ℃ using a PID algorithm; and the safety mechanism automatically cuts off the power when the temperature is ≥40 ± 0.5 ℃.

[0024] A method for fabricating a sensor integrating sweat induction and multi-parameter detection includes the following steps: Step 1, Preparation of the sweat-wicking layer 4: PDMS is used as the base material. A CO2 laser is used to engrave the PDMS base to form an array of sweat collection holes with a diameter of 3.0 ± 0.1 mm and a hole spacing of 5 ± 0.1 mm. Then, a screen-printed silver paste heating circuit is used. This circuit is connected to a temperature control module composed of a copper foil heating element and an NTC thermistor to achieve precise temperature control of 39 ± 0.5℃, ensuring coverage of approximately 100 sweat gland units. The PDMS base material preparation method in Step 1 is as follows: the main agent is Dow Corning Sylgard 184 silicone rubber prepolymer, and the curing agent is a platinum-catalyzed crosslinking agent compatible with the Dow Corning Sylgard 184 silicone rubber prepolymer. During preparation, the main agent and curing agent are accurately weighed at a mass ratio of 10:1, thoroughly mixed mechanically, degassed under vacuum, and then cast into shape. The curing temperature is 80 ℃, the curing time is 2 h, and the final thickness is controlled at 200 ± 10 μm. The CO2 laser engraving parameters in Step 1 are set to a wavelength of 10.6 nm. μm, 20±1W; heating circuit line width 0.1 ±0.01 mm, copper foil heating element thickness 35 μm, resistance 10 ± 1Ω; NTC thermistor 10 KΩ, B value 3950 K; Step 2, Preparation of the sweat collection layer 3: Using PDMS material, a microchannel network is laser-engraved to form sweat collection channels, a set of sweat collection holes 34, and a reservoir 31. The sweat collection holes 34 are connected to the circular sweat collection hole array of the sweat-conducting layer 4. Sweat passes through the sweat-conducting layer 4 to the sweat collection holes 34 and then converges into the reservoir 31 through the sweat collection channels. The sweat collection channels are divided into a front end 33 and a rear end 32. The front end 33 is connected to the reservoir 31, and the rear end 32 is connected to the sweat collection holes 34. The sweat collection channel has a front end 33 with a width of 0.5±0.02mm and a rear end 32 with a width of 1.0±0.02mm. The storage tank 31 is cylindrical and has a sweat outlet at its rear end. The anti-evaporation design features a chamfered edge slit structure to ensure that excessive sweat storage does not overflow and damage the sensor. In step 2, the sweat collection hole 34 has a diameter of 3mm. The width difference between the front end 33 and the rear end 32 of the sweat collection channel creates a contact angle of less than 30°, thereby generating capillary force. The sweat collection channel has a depth of 0.25±0.01mm. The storage tank 31 has a diameter of 8.0±0.2mm, a depth of 800±20μm, and a sweat outlet size of 2.0 × 3.0 mm. Step 3, Preparation of detection electrode layer 2: Detection electrode layer 2 is constructed using a screen printing process to create a three-electrode system, comprising a platinum auxiliary electrode 21, an Ag / AgCl reference electrode 23, and two functionalized working electrodes; the functionalized working electrodes are glucose detection working electrode 22 and β-hydroxybutyric acid (β-HB) detection working electrode 24; the specific preparation steps of detection electrode layer 2 in step 3 are as follows: The first step involves sequentially forming a carbon paste-based working electrode, an Ag / AgCl reference electrode 23, and a platinum auxiliary electrode 21 on a PDMS substrate. The second step is to construct the rGO layer: on the working electrode of the carbon paste substrate, the reduced graphene oxide rGO after liquid metal gallium exfoliation is modified by drop casting, and the thickness of the rGO layer is 50±5 nm. The third step is to spin-coat the chitosan solution CS: spin-coat 2 wt% chitosan solution onto the rGO layer at a spin speed of 3000 rpm for 30 seconds. Step 4: Immobilize GOx and PB: Prussian blue PB layer is formed by electrodeposition at -0.2V for 300 seconds. The electrolyte is a mixture of 0.1 M FeCl3 and 0.1 M K3[Fe(CN)6]. Subsequently, glucose oxidase with a concentration of 10 mg / mL and 1% glutaraldehyde crosslinking is immobilized to form a glucose oxidase GOx composite layer, thus obtaining glucose detection working electrode 22. Step 5, Preparation of the β-HB detection working electrode 24: On the working electrode of the carbon paste substrate, reduced graphene oxide (rGO) after liquid gallium exfoliation was modified by drop casting. The rGO layer thickness was 50±5 nm. Then, a 2wt% chitosan solution was spin-coated onto the rGO layer at 3000 rpm for 30 seconds. Next, a 5±0.1 mM TBO solution was drop-coated to modify toluidine blue O (TBO). The mixture was cross-linked at 25±1℃ for 4±0.5 h to fix 5 mM coenzyme NAD. + The β-HB detection working electrode 24 was prepared by curing 2 U / cm² β-hydroxybutyrate dehydrogenase HBDH at 220 ± 5 ℃ for 30 ± 1 min. Step 4, Integrated Microfluidic Detection Chip Layer 1: Designed on flexible PDMS material using Altium Designer, this layer integrates a 72MHz main controller AT89C52 and a transmission distance of 10... The system comprises a Bluetooth module HC-05, a power module, a TXS0102, an ADS1115 ADC converter, GND, a temperature sensor, an AD623 signal conditioning instrumentation amplifier, a glucose detection interface, and a β-HB detection interface. Circuit connections between these modules are achieved using pre-configured ribbon cable interfaces. The weak current signals generated during glucose and β-HB detection are transmitted to the AD623 instrumentation amplifier in the signal conditioning module, connecting to the corresponding IN1+ / IN2+ interfaces. Gain resistors Rg1 and Rg2 are surface-mount soldered onto the AD623 instrumentation amplifier. The signal is amplified by the AD623 and transmitted to the ADS1115 ADC converter to convert the current signal into a voltage signal. The connection ports are A0 / A1. The ADS1115 connects to the main controller via the digital I2C protocol, with its SCL / SDA ports connected to the P3.0 / P3.1 pins of the main control chip AT89C52. The main control chip connects to the ADS1115 via the digital I2C protocol, and connects to a temperature sensor via ports P1.0 and / or P2.0. P1.0 connects to a DS18B20 as a temperature sensor driver, and P2.0 connects to the heating drive MOSFET (IRF540N) via a current-limiting resistor to control the PWM control signal. The TXD / RXD ports of the main control chip AT89C52 are connected to ports A3 / A4 of the TXS0102 to achieve voltage conversion (5V to 3.3V). Ports B3 / B4 of module 02 are connected to the TXD / RXD ports of the HC-05 Bluetooth module, providing 3.3V voltage. The power supply module consists of a 5V power supply and an AMS1117-3.3 regulator, which provides 3.3V power to ADS1115, TXS0102, and HC-05 through pin connections. The AT89C52, AD623, and heating element are powered by a filtered 5V power bus. Finally, each module is equipped with a ground terminal, which is connected to GND to ensure the stability of the entire circuit. Step 5: The prepared microfluidic detection integrated chip layer 1, detection electrode layer 2, sweat collection layer 3, and sweat-conducting layer 4 are bonded together layer by layer using 3M 1524 medical adhesive to obtain an integrated sweat induction and multi-parameter detection sensor with an overall thickness of 1.5 ± 0.1 mm.

[0025] The application of a sensor integrating sweat induction and multi-parameter detection in a flexible wearable monitoring device: During use, the sensor is attached to the monitoring site. A temperature control module raises the local temperature to 39 ± 0.5 ℃. Sweat enters the sweat collection layer 3 through the circular sweat collection hole array in the sweat-wicking layer 4, reaching a stable flow rate of 0.8 ± 0.1 μL / min within 5 minutes. In electrochemical detection, glucose is measured at a potential of 0.4 ± 0.02 V by detecting the current signal during the H2O2 oxidation process in the glucose reaction, with a sensitivity of 12.69 ± 0.5 μA·mM. -1 ·cm 2 β-HB is detected at a potential of -0.2 ± 0.02V, where the NADH reduction current during the β-HB reaction is measured, with a detection limit of 12.0 ± 2.0 μM. The sensor integrating sweat induction and multi-parameter detection measures the glucose detection signal by 100 times and the β-HB detection signal by 50 times through an AD623 instrumentation amplifier. The signal is then transmitted through various modules of the microfluidic chip and finally transmitted to the mobile APP via Bluetooth at a frequency of 10 ± 1Hz. The monitoring method of the sensor integrating sweat induction and multi-parameter detection covers the entire process of sweat induction and electrochemical analysis. The sweat-wicking layer 4 and the sweat collection layer 3 store and collect the sweat stimulated by the skin, allowing it to contact the detection electrode layer 2. The sweat-wicking layer 4 is designed with an array of equally spaced circular sweat collection holes. These 3.0 mm diameter holes connect to the storage tank 31 via sweat collection channels with a width of 0.5 ± 0.02 mm and a depth of 0.25 ± 0.01 mm. A drain outlet is located on the other side of the storage tank 31 to promptly drain the detected sweat and prevent sedimentation. This ensures that the lower layer of sweat can enter the sweat collection channels through the circular sweat collection hole array and ultimately collect in the storage tank 31, allowing the detection electrode layer 2 to immediately contact the sweat being tested.

[0026] In terms of material selection, PDMS is used as the base material. This material has excellent flexibility and can withstand bending deformation with a curvature radius of more than 5 mm, which allows it to adapt well to the curved surfaces of the body such as the wrist and forehead.

[0027] During detection by electrode layer 2, a detectable current signal is generated at a low potential by utilizing the mediating effect of NADH and TBO produced by the β-HB reaction in sweat.

[0028] Example 2: Application of the integrated sweat induction and multi-parameter detection sensor prepared in Example 1 in a flexible wearable monitoring device. The flexor side of the forearm is selected as the optimal detection site, where the sweat gland density is moderate (120-150 glands / cm²) and is less affected by movement. During use, the integrated sweat induction and multi-parameter detection sensor is attached to the monitoring site and secured with medical tape to ensure complete adhesion between the sweat-wicking layer and the skin. The temperature control module raises the local temperature to 39 ± 0.5℃, implementing a three-stage heating strategy: rapid heating at a rate of 1℃ / s for the initial 0-30 seconds; a gradual heating at 0.2℃ / s for 30-120 seconds; and finally, a stable constant temperature of 39 ± 0.5℃. Sweat enters the sweat collection layer 3 through the circular sweat collection hole array of the sweat-wicking layer 4, reaching 0.8 ± 0.1℃ within 5 minutes. A stable flow rate of μL / min is achieved; sweat transport after sweat production relies on the optimization of the microfluidic structure, in which the sweat collection channel is driven by capillary force with a contact angle of <30°, and the measured flow rate reaches 2.5 ± 0.3 mm / s, and a 0.3 μL sample volume can be filled within 5 ± 1 s.

[0029] Example 3: Application of the sensor integrating sweat induction and multi-parameter detection from Example 2 in a flexible wearable monitoring device. Sweat is in contact with the detection electrode layer 2. Sweat detection employs simultaneous dual-parameter detection of glucose and β-HB. The detection mechanism is as follows: In electrochemical detection, glucose at a potential of 0.4 ± 0.02 V reacts as follows: glucose + O2 → gluconolactone + H2O2; H2O2 → O2 + 2H+ + + 2e - The glucose concentration is determined by detecting the current signal during the H2O2 oxidation process in the glucose reaction, with a sensitivity of 12.69 ± 0.5 μA·mM. -1 ·cm 2 The current response model is: I (μA) = (0.28 ± 0.02) × [Glucose] + (0.05 ± 0.01) (R² = 0.993, n = 20).

[0030] β-HB at a potential of -0.2 ± 0.02V: β-HB + NAD + → Acetoacetic acid + NADH NADH + TBO ox → NAD + + TBO re d Current response model: I (μA) = (1.15 ± 0.05) × [β-HB] + (0.02 ± 0.005) (R² = 0.987, n = 20) The detection limit for NADH reduction current during the β-HB reaction was 12.0 ± 2.0 μM. Signal processing flow: The sensor test signal integrating sweat induction and multi-parameter detection is amplified by an AD623 instrumentation amplifier to increase the glucose detection signal by 100 times and the β-HB detection signal by 50 times, amplifying the current signal and converting it into voltage. Then, an ADS1115 ADC converter converts the analog voltage into a 16-bit digital signal. The main controller AT89C52 reads the ADC data and constructs a data frame (6-byte format). Part of its code is as follows: TTT typedef struct { uint8_t header; / / 0xAA int16_t glucose; / / Magnify by 100 times int16_t beta_hb; / / Magnify by 1000 times uint8_t checksum; / / Checksum SensorData; The data is then sent to the Bluetooth module via serial port. The HC-05 Bluetooth module transmits the data to the mobile app at a frequency of 10 ± 1Hz. The calibration coefficients are then applied by reading from the EEPROM during data processing in the main control chip.

[0031] Example 4: Application of the integrated sweat induction and multi-parameter detection sensor prepared in Example 1 in a flexible wearable monitoring device. The detection section automatically issues a risk warning when the detected human glucose / β-HB concentration is too high. The core of the artificial intelligence warning algorithm described in this invention lies in a metabolic state risk assessment model based on multi-parameter fusion. This model comprehensively calculates the ketosis risk index by analyzing the static concentration and dynamic trends of glucose and β-hydroxybutyrate (β-HB) in real time, and provides graded warnings accordingly.

[0032] The input parameters for the algorithm are acquired in real time and preprocessed using a constructed glucose / β-HB detection sensor. The code for the input section is as follows: import math class MetabolicParams: def __init__(self, glucose, beta_hb, glucose_rate, beta_hb_rate): self.glucose = glucose # Current glucose concentration (mM) self.beta_hb = beta_hb # Current β-HB concentration (mM) self.glucose_rate = glucose_rate # Glucose rate of change (mM / min) self.beta_hb_rate = beta_hb_rate # β-HB change rate (mM / min) def calculate_ketosis_risk(params): After the detection data is input, the concentration of the detected substance is mapped to a range of 0-1 through function simulation, and a risk threshold is determined. When the detection index is higher than this threshold, it indicates an increased risk. Calculations use a sigmoid function to map the β-HB concentration and glucose concentration to the set ranges respectively; a rate of change calculation is also established. The code for this part is as follows: Calculate and compare with the risk score R, where for # Basic Risk Score base_risk = 0 # Contribution of β-HB concentration (Sigmoid function) hb_contribution = 1.0 / (1.0 + math.exp(-10.0 * (params.beta_hb -0.5))) # Contribution to glucose concentration glu_contribution = 0.7 * (1.0 / (1.0 + math.exp(-5.0 *(params.glucose - 1.0)))) # Contribution of rate of change rate_contribution = 0.3 * math.tanh(params.beta_hb_rate * 10.0) +\ 0.2 * math.tanh(params.glucose_rate * 5.0) # Overall Risk Score (range 0-1) total_risk = 0.5 * hb_contribution + 0.3 * glu_contribution + 0.2* rate_contribution return max(0, min(1, total_risk)) # Ensures the risk is within the range of 0-1 # Define risk level enumeration class RiskLevel: LOW_RISK = 0 MEDIUM_RISK = 1 HIGH_RISK = 2 Finally, the calculated risk score (R) is mapped to discrete risk levels, with three risk levels in total, so that users can take appropriate measures based on their physical examination results. The code for the risk determination part is as follows: # Risk Level Assessment def evaluate_risk_level(risk_score): if risk_score < 0.3: return RiskLevel.LOW_RISK elif risk_score<0.6: return RiskLevel.MEDIUM_RISK else: return RiskLevel.HIGH_RISK The above algorithm formula for the root domain provides an example for detecting β-HB concentration: # Usage Example if __name__ == "__main__": # Create a metabolic parameter instance params = MetabolicParams( glucose = 1.2, # mM beta_hb=0.6, # mM glucose_rate=0.1, # mM / min beta_hb_rate=0.05 # mM / min # Calculate risk score risk_score = calculate_ketosis_risk(params) print(f"Risk score: {risk_score:.3f}") # Assess risk level risk_level = evaluate_risk_level(risk_score) if risk_level == RiskLevel.LOW_RISK: print("Risk level: Low risk") elif risk_level == RiskLevel.MEDIUM_RISK: print("Risk Level: Medium Risk") else: print("Risk level: High risk").

[0033] Example 5, Clinical Validation: The above Examples 1-4 constructed a complete flexible wearable sensor capable of simultaneously detecting glucose and β-HB concentrations in the human body, and the following clinical tests were conducted on it.

[0034] First, the participants were divided into two groups: a type 2 diabetes mellitus (T2DM) patient group (n = 15, age 45-65 years, HbA1c 7.5-10%) and a control group of healthy individuals (n = 10, age-matched). After the testing personnel wear the sensors on the flexor side of the forearm in batches, the mobile APP is activated to start the testing.

[0035] Simultaneous data collection and testing were conducted, with blood samples collected from each participant every 10 minutes (glucose meter: Rocon Superior; blood ketone meter: FreeStyle Optium).

[0036] Records glucose / β-HB concentrations displayed in the app in real time.

[0037] Comparative test data showed that the average relative error (MARD) for glucose detection was 8.7%, and the average deviation for β-HB was 0.08 mmol / L. The response times were 12 ± 3 min for glucose and 8 ± 2 min for β-HB. This sensor demonstrated a detection and warning sensitivity of 92.3% and a specificity of 89.5%, with an average early warning time of 55 ± 8 min. Stability tests showed that the signal attenuation was <5% after 14 days of storage at room temperature, it remained functional after 5000 bending cycles, and the error was <3% within a temperature range of 25-40℃.

[0038] This invention achieves medical-grade precision in non-invasive metabolic dynamic monitoring through systematic innovation. Its technological breakthroughs are concentrated in the synergistic optimization of material systems, structural design, algorithm models, and manufacturing processes. At the material level, a PDMS / rGO composite substrate is used. While maintaining flexibility of 0.8-1.2 MPa, the porous structure of rGO enhances conductivity stability. After 5000 bends, the electrode impedance change is <5%, overcoming the challenge of the mutual incompatibility between mechanical deformation and electrochemical performance in wearable devices. In terms of structural design, an innovative gradient wettability microchannel (30° contact angle at the front → 60° at the rear) is constructed, coupling capillary force drive and an anti-evaporation reservoir, achieving highly efficient sweat transmission at 2.5 ± 0.3 mm / s. The filling time for a 0.3 μL sample is shortened to 5 ± 1 s, significantly improving detection timeliness. At the algorithm level, a multi-parameter fusion ketosis early warning model was developed, integrating dynamic data from dual channels: glucose (lag 12 ± 3 min) and β-HB (lag 8 ± 2 min). After Kalman filtering and noise reduction, a sensitivity of 92.3% and a specificity of 89.5% were achieved, resulting in an average clinical early warning lead of 55 ± 8 minutes. The manufacturing process utilizes 300-mesh screen printing technology, combined with liquid gallium stripping and electrodeposition modification processes (Prussian blue deposition time accuracy ± 10 s), and medical adhesive encapsulation to achieve an ultra-thin device thickness of 1.5 ± 0.1 mm, with a yield exceeding 95%, laying the foundation for large-scale mass production. This four-dimensional innovation system collectively supports medical-grade performance with an 8.7% MARD accuracy for glucose detection and a β-HB deviation of 0.08 mmol / L, promoting the translation of non-invasive diabetes monitoring technology from the laboratory to clinical practice.

[0039] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for the preparation of a sensor integrating sweat induction and multi-parameter detection, characterized in that, Comprising the following steps: Step 1, sweat guide layer (4) preparation: PDMS is used as the base material, CO2 laser is used to carve the PDMS base to form an array of sweat collection holes with a diameter of 3.0 ± 0.1 mm, the hole spacing is 5 ± 0.1 mm, then silver paste heating circuit is printed by silk screen printing, the temperature control module composed of copper foil heating sheet and NTC thermistor is connected through the silver paste heating circuit, and accurate temperature control of 39 ± 0.5℃ is realized; Step 2, preparation of sweat collection layer (3): PDMS material is used, laser carving is used to form a micro-channel network to form a sweat collection channel, a group of sweat collection holes (34) and a liquid reservoir (31), the sweat collection holes (34) are connected in position corresponding to the array of circular sweat collection holes of the sweat guide layer (4), the sweat passes through the sweat guide layer (4) to the sweat collection holes (34), and then gathers to the liquid reservoir (31) through the sweat collection channel, the sweat collection channel is divided into a front end (33) and a rear end (32) of the sweat collection channel; the front end (33) of the sweat collection channel is connected with the liquid reservoir (31), and the rear end (32) of the sweat collection channel is connected with the sweat collection hole (34); the front end (33) of the sweat collection channel is 0.5 ± 0.02 mm wide, the rear end (32) of the sweat collection channel is 1.0 ± 0.02 mm wide, the liquid reservoir (31) is cylindrical, and the tail end of the liquid reservoir (31) is provided with a sweat outlet; Step 3, preparation of detection electrode layer (2): the detection electrode layer (2) is constructed into a three-electrode system by silk screen printing process, including a platinum auxiliary electrode (21), an Ag / AgCl reference electrode (23) and two functionalized working electrodes; the functionalized working electrodes are glucose detection working electrode (22) and β-hydroxybutyric acid, i.e. β-HB detection working electrode (24); Step 4, integrated microfluidic detection integrated chip layer (1): integrating a main controller, a Bluetooth module, a power module, TXS0102, an ADC conversion, a GND, a temperature sensor, a signal conditioning, a sensor glucose detection interface and a sensor β-HB detection interface on a flexible material PDMS; Step 5: the prepared microfluidic detection integrated chip layer (1), detection electrode layer (2), sweat collection layer (3) and sweat guide layer (4) are pasted layer by layer using a medical adhesive to obtain an integrated sweat induction and multi-parameter detection sensor with a total thickness of 1.5 ± 0.1 mm.

2. The method of claim 1, wherein the sensor is prepared by the steps of: The preparation method of the base material PDMS in step 1 is as follows: the main agent is Dow Corning Sylgard 184 silicone rubber prepolymer, and the curing agent is platinum catalyzed crosslinking agent matched with Dow Corning Sylgard 184 silicone rubber prepolymer; during preparation, the main agent and the curing agent are accurately weighed in a mass ratio of 10:1, uniformly mixed by mechanical stirring, vacuum degassed, and then poured into a mold for shaping; the shaping and curing temperature is 80℃, the shaping time is 2h, and the final thickness is controlled to be 200 ± 10 μm.

3. The method of claim 1, wherein the sensor is integrated with a wearable device. The CO2 laser engraving parameters in step 1 are set as follows: wavelength 10.6 μm, 20±1 W; line width of heating circuit 0.1±0.01 mm, thickness of copper foil heating sheet 35 μm, resistance 10±1 Ω; NTC thermistor 10 KΩ, B value 3950 K.

4. The method of claim 1, wherein the sensor is integrated with a wearable device. In step 2, the diameter of the sweat collection hole (34) is 3 mm, the width difference between the front end (33) of the sweat collection channel and the rear end (32) of the sweat collection channel forms a contact angle of less than 30° to form capillary force driving, and the depth of the sweat collection channel is 0.25±0.01 mm; the diameter of the liquid storage pool (31) is 8.0±0.2 mm, the depth is 800±20 μm, and the size of the sweat outlet is 2.0×3.0 mm.

5. The method of claim 1, wherein the sensor is integrated with a wearable device. In step 3, the specific preparation steps of the detection electrode layer (2) are as follows: Firstly, a working electrode of carbon paste substrate, an Ag / AgCl reference electrode (23) and a platinum auxiliary electrode (21) are sequentially formed on the PDMS substrate; Secondly, an rGO layer is constructed: rGO after liquid metal gallium exfoliation treatment is modified on the working electrode of the carbon paste substrate by drop casting method, and the thickness of the rGO layer is 50±5 nm; Thirdly, a chitosan solution CS is spin-coated: a 2 wt% chitosan solution is spin-coated on the rGO layer at a spin-coating speed of 3000 rpm for 30 seconds; Fourthly, GOx and PB are co-immobilized: a Prussian blue PB layer is formed by electrodeposition at a potential of -0.2 V for 300 seconds, the electrolyte is a mixture of 0.1 M FeCl3 and 0.1 M K3[Fe(CN)6], then a glucose oxidase GOx composite layer is formed by immobilizing glucose oxidase with 1% glutaraldehyde cross-linking and a concentration of 10 mg / mL, thereby obtaining a glucose detection working electrode (22); The fifth step, preparation of the β-HB detection working electrode (24): On the working electrode of the carbon paste substrate, the reduced graphene oxide rGO treated by liquid metal gallium stripping was modified by drop casting method, the thickness of the rGO layer was 50±5 nm, then 2 wt% chitosan solution was spin-coated on the rGO layer, the spin-coating speed was 3000 rpm, the time was 30 seconds, further 5±0.1 mM TBO solution was drop-coated, toluene blue O, i.e. TBO, was modified, and 5 mM coenzyme NAD was fixed by cross-linking at 25±1 ℃ for 4±0.5 h + The β-HB detection working electrode (24) was prepared at 220±5 ℃ for 30±1 min with 2 U / cm² β-hydroxybutyric acid dehydrogenase HBDH.

6. The method of claim 1, wherein the sensor is integrated with a wearable device. In step 4, by using Altium Designer, a 72MHz main controller AT89C52, a Bluetooth module HC-05 with a transmission distance of 10 m, a power module, TXS0102, ADC conversion ADS1115, GND, a temperature sensor, a signal conditioning AD623 instrument amplifier, a sensor glucose detection interface and a sensor β-HB detection interface are integrated, and the circuit connection between the modules is realized by using a pre-set flat cable interface.

7. The sensor for integrated sweat induction and multi-parameter detection prepared by the method of any one of claims 1-4. The microfluidic detection integrated chip layer (1), the detection electrode layer (2), the sweat collection layer (3) and the sweat guiding layer (4) are sequentially arranged from top to bottom; The microfluidic detection integrated chip layer (1) integrates a Bluetooth module, a main controller, a power module, TXS0102, ADC conversion, GND, a temperature sensor, a signal conditioning AD623 instrument amplifier, a sensor glucose detection interface and a sensor β-HB detection interface on a PDMS flexible substrate The detection electrode layer (2) adopts a screen printing process to construct a three-electrode system, including a platinum auxiliary electrode (21), an Ag / AgCl reference electrode (23), and two functionalized working electrodes; the functionalized working electrodes are a glucose detection working electrode (22) and a beta-hydroxybutyric acid (beta-HB) detection working electrode (24); The sweat collection layer (3) is made of PDMS material, and is internally provided with a laser-engraved sweat collection channel, which has a gradient structure in cross section, a front end (33) of the sweat collection channel is 0.5 ± 0.02 mm wide, a rear end (32) of the sweat collection channel is 1.0 mm wide, a contact angle less than 30° is formed to form capillary force driving, and the channel depth is 0.25 ± 0.01 mm; the channel end is connected to a liquid storage pool (31) with a volume of 0.8 ± 0.05 μL, and the tail end of the liquid storage pool (31) is provided with an anti-evaporation sweat outlet; The sweat guide layer (4) is made of PDMS material, and is provided on the surface with a circular sweat collection hole array corresponding to the position of the sweat collection hole (34) with a diameter of 3.0 ± 0.1 mm and a spacing of 5 ± 0.1 mm, and is integrated with a temperature control module composed of a copper foil heating sheet and an NTC thermistor, which is used for accurately regulating the skin temperature, the copper foil heating sheet has a resistance of 10 ± 1 Ω, and the NTC thermistor has an accuracy of ± 0.1 ℃.

8. Use of the sensor for integrated sweat induction and multi-parameter detection prepared by the method of any one of claims 1-4 in a flexible wearable monitoring device. In use, the sensor integrated with sweat induction and multi-parameter detection is attached to the monitoring site, the local temperature is raised to 39 ± 0.5 ℃ by the temperature control module, the sweat enters the sweat collection layer (3) through the circular sweat collection hole array of the sweat guide layer (4), and reaches a stable flow rate of 0.8 ± 0.1 μL / min within 5 minutes; in electrochemical detection, the glucose is detected by detecting the current signal in the oxidation process of H2O2 in the glucose reaction at a potential of 0.4 ± 0.02 V, and the sensitivity reaches 12.69 ± 0.5 μA·mM -1 ·cm 2 ; the β-HB is detected by detecting the NADH reduction current in the β-HB reaction process at a potential of -0.2 ± 0.02 V, and the detection limit is 12.0 ± 2.0 μM; the test signal of the sensor integrated with sweat induction and multi-parameter detection is amplified by 100 times for the glucose detection signal and by 50 times for the β-HB detection signal by an AD623 instrument amplifier, then the signals are transmitted by each block of the microfluidic chip, and finally transmitted to the mobile terminal APP at a frequency of 10 ± 1 Hz by a Bluetooth module; the monitoring method of the sensor integrated with sweat induction and multi-parameter detection covers the whole process of sweat induction and electrochemical analysis.

9. Use according to claim 6, characterized in that, The flexible wearable monitoring device is a non-invasive monitoring device.

10. Use according to claim 6, characterized in that, The flexible wearable monitoring device is a diabetes monitoring device. The flexible wearable monitoring device is a diabetes monitoring device.

Images