A zinc-air fuel cell self-powered hydrogel sensor for detecting pathogens, a sensing device, and a preparation method and application thereof

Abstract

The application relates to the field of self-powered electrochemical biosensing technology, and particularly discloses a zinc-air fuel cell self-powered hydrogel sensor for detecting pathogenic bacteria, a sensing device, a preparation method and application thereof. The sensor takes a zinc-air fuel cell as a self-powered module, combines DNA molecular recognition technology and nucleic acid amplification reaction, realizes targeted release of an oxygen reduction catalyst, obtains an electric current signal required for electrochemical detection, and constructs a regression model based on the above-mentioned sensor and a machine learning algorithm module, taking electrochemical signals as input and the concentration logarithm of a pathogenic target gene fragment as output, so that intelligent analysis and accurate quantification of signals are realized. The application combines DNA molecular recognition, nucleic acid amplification reaction and a machine learning intelligent model, realizes integrated detection of 'biological recognition-signal conversion-intelligent analysis', and simultaneously provides application of the application in detection of the brown rot, and has potential field application value.

Description

Technical Field

[0001] This invention relates to the field of self-powered electrochemical biosensing technology, and in particular to a zinc-air fuel cell self-powered hydrogel sensor, sensing device, preparation method and application for detecting pathogens. Background Technology

[0002] Sugarcane top rot is a significant disease of sugarcane caused by pathogenic fungi such as Fusarium verticillioides, severely impacting yield and sugar quality. Currently, detection of this disease mainly relies on traditional isolation and culture methods, polymerase chain reaction (PCR), and quantitative real-time PCR (qPCR). The core drawbacks of existing technologies are: traditional methods are cumbersome and time-consuming (usually requiring several hours to days), relying on specialized thermal cycling equipment and fluorescently labeled probes, making rapid on-site detection difficult; and they are easily affected by environmental temperature fluctuations and complex sample matrices, leading to decreased detection accuracy and failing to meet the needs of early warning and precise control in sugarcane cultivation.

[0003] In recent years, biosensor technology has shown potential in pathogen detection due to its advantages such as fast response and ease of operation. However, existing biosensors still have significant limitations: First, most sensors require external power supply, making them difficult to adapt to resource-scarce environments in the field; second, sensing elements have poor stability and are prone to failure after long-term storage; third, traditional signal analysis methods (such as direct current reading) are difficult to eliminate environmental noise interference, resulting in insufficient detection accuracy in complex scenarios. Although self-powered sensing devices can use biorecognition reactions to regulate output signals, existing self-powered systems mostly rely on traditional biocatalysts such as glucose oxidase, which suffer from low catalytic efficiency and weak signal amplification capabilities, thus limiting the sensitivity and specificity of detection. Although zinc-air fuel cells (ZAFCs) have become an ideal self-powered platform due to their high energy density and good biocompatibility, how to improve the sensitivity and specificity of detection under ZAFC power supply conditions remains a research challenge.

[0004] Nucleic acid amplification can amplify the detection signal of a single molecule by more than 1000 times. Due to its high sensitivity and large signal amplification capability, it can effectively improve detection specificity and sensitivity. In addition, machine learning is a process of mining deep patterns through data. It is currently also used in pathogen detection, which can improve detection efficiency and accuracy. However, the application of machine learning algorithms in sensor detection is mostly limited to data post-processing and lacks deep integration with self-powered hardware, thus failing to fully utilize its advantages in noise suppression and nonlinear relationship modeling.

[0005] Currently, there are no reports on the integrated application of zinc-air fuel cells with nucleic acid amplification and intelligent algorithms in pathogen detection to improve the sensitivity and specificity of detection. Summary of the Invention

[0006] This invention provides a zinc-air fuel cell self-powered hydrogel sensor, sensing device, preparation method, and application for detecting pathogens. By combining DNA molecular recognition technology with nucleic acid amplification reaction, it achieves targeted release of oxygen reduction catalyst and obtains more accurate electrochemical signals. Furthermore, it further uses machine learning algorithms to screen the optimal model, realizing intelligent signal analysis and accurate quantification, thereby improving the sensitivity and specificity of pathogen detection.

[0007] To achieve the above objectives, the technical solution of the present invention is as follows:

[0008] The first aspect of this invention provides a zinc-air fuel cell self-powered hydrogel sensor for detecting pathogens, comprising:

[0009] A zinc-air fuel cell includes a zinc sheet anode, a carbon cloth cathode, and a hydrogel electrolyte disposed between the anode and the cathode;

[0010] The nucleic acid amplification product α, which is loaded onto the carbon cloth cathode, is prepared by the following steps:

[0011] The test solution was added dropwise to the catalyst limiter at a volume ratio of (1-3):(5-8), and incubated at 35-39℃ for 20-80 min. Then, F chain was added at a volume ratio of (1-3):(5-10), and incubated at 35-39℃ for another 20-80 min. After centrifugation, the supernatant was collected to obtain the nucleic acid amplification product α.

[0012] The nucleotide sequence of the F chain is shown in SEQ ID No. 1.

[0013] Preferably, the method for preparing the catalyst limiter includes the following steps:

[0014] 50-100 μL of S-chains and 50-100 μL of platinum nanoparticles were mixed and incubated at 2-4 °C for 5-10 h to obtain an S-Pt solution. 50-100 μL of W-chains and 50-100 μL of D-chains were added sequentially to the S-Pt solution and incubated at 35-39 °C for 1-2 h. 5-25 μL of gold nanoparticles@magnetic beads composite was added and incubated at 2-4 °C for 5-10 h. The supernatant was discarded, and the precipitate was resuspended in sterile, enzyme-free water to a final volume of 50-100 μL to obtain the catalyst limiter.

[0015] The nucleotide sequences of the S, W, and D chains are shown in SEQ ID No. 2, SEQ ID No. 3, and SEQ ID No. 4, respectively.

[0016] Preferably, the method for preparing the platinum nanoparticles includes the following steps:

[0017] Add 1-5 mL of 0.5-4% chloroplatinic acid solution and 100-500 mg of polyvinylpyrrolidone to 15-45 mL of methanol solution and stir for 10-30 min to form a precursor solution; heat the precursor solution to 60-80℃, add 1-2 mL of 0.5-2% sodium citrate solution, and continue stirring for 10-30 min to obtain the platinum nanoparticles.

[0018] Preferably, the preparation method of the gold nanoparticles@magnetic beads composite includes the following steps:

[0019] Carboxylated magnetic beads with a concentration of 4-5 mg / mL were mixed with chloroauric acid with a concentration of 0.5-1%, and the volume ratio of carboxylated magnetic beads to chloroauric acid was (1-4): (1-4). The mixture was ultrasonically treated for 20-80 min to obtain pretreated magnetic beads. Sodium citrate solution with a concentration of 0.5-2% was added to the pretreated magnetic beads, and the volume ratio of pretreated magnetic beads to sodium citrate was (2-8): (1-4). The mixture was gently shaken for 4-6 h to separate the precipitate, which was the gold nanoparticle@magnetic bead composite.

[0020] Preferably, the hydrogel electrolyte is prepared by the following steps:

[0021] Heat 100-200 mL of 0.5-1 M KOH aqueous solution to 50-70℃ and stir. Add 1-5 g potato starch and 1-5 g κ-carrageenan and continue stirring for 30-60 min. Then add 0.1-0.5 g titanium carbonitride and 100-500 μL of 35-40 wt% glutaraldehyde and continue stirring for 0.5-2 h to obtain a mixed solution. Pour the obtained mixed solution into a mold, cool to room temperature, and let stand for 6-10 h to form the hydrogel electrolyte.

[0022] A second aspect of this invention provides a method for preparing the above-described self-powered hydrogel sensor for a zinc-air fuel cell, comprising the following steps:

[0023] (1) Prepare platinum nanoparticles and gold nanoparticles@magnetic beads composites, and then prepare a catalyst confinement device; use the prepared catalyst confinement device to prepare nucleic acid amplification product α; load the prepared nucleic acid amplification product α onto carbon cloth;

[0024] (2) Preparation of hydrogel electrolytes;

[0025] (3) Using a zinc sheet as the anode, a carbon cloth loaded with the nucleic acid amplification product α as the cathode, and the hydrogel electrolyte as the electrolyte, the zinc-air fuel cell self-powered hydrogel sensor is assembled.

[0026] A third aspect of the present invention provides a sensing device based on the above-described zinc-air fuel cell self-powered hydrogel sensor and machine learning, comprising:

[0027] The zinc-air fuel cell self-powered hydrogel sensor described above;

[0028] The machine learning algorithm module is used to receive the electrochemical signal generated by the self-powered hydrogel sensor of the zinc-air fuel cell, and calculate the predicted value of the pathogen target gene fragment concentration based on a pre-trained regression model. Specifically, the establishment of the regression model includes: selecting a specific algorithm, using a specific electrochemical signal as input features, and the pathogen target gene fragment concentration or the logarithm of the pathogen target gene fragment concentration as output labels, for training to obtain the regression model; the specific algorithm is selected from one of the following: linear regression, ridge regression, Lasso regression, elastic network algorithm, decision tree algorithm, random forest algorithm, gradient boosting algorithm, support vector regression algorithm, LightGBM algorithm, neural network algorithm, community learning algorithm, and ensemble algorithm.

[0029] The fourth aspect of this invention provides the application of the zinc-air fuel cell self-powered hydrogel sensor or the sensing device described above in the detection of sugarcane top rot.

[0030] The fifth aspect of this invention provides a method for detecting sugarcane top rot using the zinc-air fuel cell self-powered hydrogel sensor or the sensing device described above, comprising the following steps:

[0031] (1) Prepare test solutions containing different concentrations of target gene fragments of sugarcane top rot pathogen;

[0032] (2) The test liquid is added to the catalyst limiter for incubation to prepare nucleic acid amplification product α; the zinc-air fuel cell is assembled; the corresponding output current value generated by the self-powered hydrogel sensor of the zinc-air fuel cell is collected under the test liquid conditions of using target gene fragments of sugarcane top rot pathogen at various concentrations;

[0033] (3) Using the output current value as the input feature and the logarithm of the concentration of the target gene fragment of sugarcane top rot pathogen as the output label, construct a standard curve or regression model of current-concentration logarithm;

[0034] (4) Take an actual sample, prepare the actual sample test solution, obtain the corresponding output current value according to the method in step (2), input the output current value into the current signal-concentration logarithmic standard curve or regression model, and calculate the predicted value of the concentration of the target gene fragment of the sugarcane top rot pathogen.

[0035] Preferably, the construction of the regression model includes the following steps: using the output current value as the input feature and the logarithm of the concentration of the target gene fragment of sugarcane top rot pathogen as the output label, training is performed to obtain the regression model;

[0036] The establishment of the regression model specifically includes: selecting a specific algorithm, using a specific electrochemical signal as the input feature, and the concentration of pathogen target gene fragments or the logarithm of pathogen target gene fragment concentrations as the output label, and training to obtain the regression model;

[0037] The specific algorithm selection is based on one of the following algorithms: linear regression, ridge regression, Lasso regression, elastic network, decision tree, random forest, gradient boosting, support vector regression, and LightGBM.

[0038] The present invention has the following advantages:

[0039] (1) Improved detection sensitivity and specificity: The sensor of the present invention combines DNA molecular recognition technology with nucleic acid amplification reaction to achieve targeted release of oxygen reduction catalyst; by specifically regulating the output signal of zinc air fuel cell through nucleic acid amplification product α, combined with the signal amplification effect of catalyst limiter, it can achieve efficient target recognition and signal cascade amplification, effectively distinguishing target pathogens from non-target interferences, and significantly reducing the detection limit.

[0040] (2) Self-powered and portable: The zinc-air fuel cell is the core self-powered module, which does not require an external power source. At the same time, the hydrogel electrolyte improves the storage environment of the sensing element. Unlike traditional sensors that use biological substances as reaction catalysts, the sensor of this invention uses pure inorganic reaction, which does not have the problem of biological substances being deactivated at high temperatures. This expands the storage temperature range, allowing it to be stored even in environments where storage conditions cannot be strictly controlled. It can better adapt to changes in indoor and outdoor environments, and is especially suitable for the preparation of portable devices, providing the possibility for on-site detection in resource-scarce environments such as potential fields.

[0041] (3) Environmental adaptability and precise quantification: The sensing device of the present invention further uses a machine learning intelligent screening model to analyze the target concentration of pathogen genes with electrochemical signals as input, effectively reducing environmental interference such as temperature fluctuations, realizing intelligent signal analysis and precise quantification, and improving the model's generalization ability and prediction accuracy (predicted values ​​include confidence intervals).

[0042] (4) Simple and efficient operation: This invention does not require complex thermal cycling equipment and fluorescent labeled probes. The sample processing and detection steps are simplified, and the overall cycle is significantly shortened compared with traditional PCR / qPCR, making it suitable for large-scale screening.

[0043] (5) Low cost and easy to promote: The preparation conditions of key materials such as PtNPs, hydrogel electrolytes, and Au@MB are mild and the raw materials are readily available. The sensor assembly is standardized and the power consumption is low, which is conducive to industrial application.

[0044] (6) Excellent application technology foundation: Experiments have proven that the sensor and sensing device of the present invention can be used to detect low concentrations of target gene fragments of sugarcane top rot pathogens, providing a good technical foundation for early warning and precise control of sugarcane top rot, and also providing a broader idea for its application and promotion in the field. Attached Figure Description

[0045] Figure 1 This is a schematic diagram of the detection process of nucleic acid amplification in the self-powered hydrogel sensor device for zinc-air fuel cells of the present invention; A: Schematic diagram of the preparation process of nucleic acid amplification product α; B: Schematic diagram of the assembly process of zinc-air fuel cells.

[0046] Figure 2 This is a signal response diagram of the zinc-air fuel cell self-powered hydrogel sensor of the present invention for detecting the presence or absence of top rot target objects.

[0047] Figure 3 The intelligent model algorithm of the sensing device of the present invention is used to select response diagrams;

[0048] Figure 4 This diagram illustrates the mean squared error of 11 specific algorithms used in the regression model selection process.

[0049] Figure 5 Cross-validation of 11 specific algorithms for regression model screening using R 2 Fraction diagram.

[0050] Figure 6 This is a signal-concentration response diagram of the sensing device of the present invention. Detailed Implementation

[0051] The following is in conjunction with the attached diagram ( Figure 1-6 The present invention will be described in detail with respect to a preferred embodiment, but it should be understood that the scope of protection of the present invention is not limited to this embodiment. Unless otherwise expressly stated, the term "comprising" and its variations (such as "including" or "comprises") in this specification mean that the stated elements or components are included, but do not exclude other elements or components.

[0052] Example 1: Fabrication of a zinc-air fuel cell self-powered hydrogel sensor for pathogen detection

[0053] 1.1 Preparation of nucleic acid amplification product α

[0054] like Figure 1 As shown in A, 10 μL of the test solution (containing suspected nucleic acid of sugarcane top rot pathogen) was added to an 80 μL catalyst limiter and incubated at 37°C for 80 min; then 100 μL of F chain (sequence shown in Appendix 1 of the instruction manual) was added and incubated at 37°C for another 80 min; centrifuged and the supernatant was collected to obtain the nucleic acid amplification product α (containing the release unit of the targeted oxygen reduction catalyst).

[0055] The preparation method of the nucleic acid amplification product α includes the following steps:

[0056] 100 μL of S chain (sequence shown in Appendix 1 of the instruction manual) was mixed with 100 μL of platinum nanoparticles (PtNPs) and incubated at 4 °C for 10 h to prepare an S-Pt solution. 100 μL of W chain and 100 μL of D chain (sequence shown in Appendix 1 of the instruction manual) were added to the S-Pt solution in sequence and incubated at 37 °C for 2 h. 25 μL of gold nanoparticles@magnetic beads complex (Au@MB) was added and incubated at 4 °C for 10 h. After centrifugation, the supernatant was discarded, and the precipitate was resuspended in sterile, enzyme-free water to 100 μL to obtain the catalyst limiter.

[0057] The preparation method of the catalyst limiter, specifically the preparation method of platinum nanoparticles (PtNPs), includes the following steps:

[0058] 5 mL of 0.5-4% chloroplatinic acid (H2PtCl6·6H2O) solution and 500 mg of polyvinylpyrrolidone (PVP) were added to 45 mL of methanol and stirred for 30 min until PVP was completely dissolved to form a precursor solution. The precursor solution was transferred to a three-necked flask, heated to 80 °C in an oil bath, and 2 mL of 2% sodium citrate solution was added to initiate the reduction reaction. The mixture was stirred for another 30 min, cooled, centrifuged, and washed to obtain platinum nanoparticles (PtNPs).

[0059] The preparation method of the gold nanoparticle@magnetic bead composite (Au@MB) used in the preparation method of the catalyst limiter includes the following steps:

[0060] In a glass test tube, 4 mL of 5 mg / mL carboxylated magnetic beads were mixed with 4 mL of 1% chloroauric acid (HAuCl4) and sonicated for 80 min to obtain pretreated magnetic beads. 4 mL of 2% sodium citrate solution was quickly added to the pretreated magnetic beads, and the mixture was gently shaken for 6 h. After magnetic separation, the precipitate was collected to obtain Au@MB.

[0061] The nucleotide sequences of the F, S, W, and D chains are shown in SEQ ID No. 1, SEQ ID No. 2, SEQ ID No. 3, and SEQ ID No. 4, respectively.

[0062] 1.2 Preparation of hydrogel electrolytes

[0063] Heat 200 mL of pure water containing 1 M KOH to 70 °C and stir. Add 5 g of potato starch and 5 g of κ-carrageenan and stir for 60 min until completely dissolved. Then add 0.5 g of titanium carbonitride (TiCN) and 500 μL of 40 wt% glutaraldehyde and stir for 2 h to obtain a mixed solution. Pour the mixed solution into a mold while hot, cool to room temperature and let stand for 10 h. Cut into 1 cm × 1 cm × 0.1 cm pieces to obtain the hydrogel electrolyte.

[0064] 1.3 Assembly of Zinc-Air Fuel Cells and Preparation of Self-Powered Hydrogel Sensors for Pathogen Detection in Zinc-Air Fuel Cells

[0065] Using a zinc sheet (1 cm × 1 cm) as the anode, carbon cloth (CC, 1 cm × 1 cm) as the air cathode, and a hydrogel electrolyte as the ion-conducting layer; 60 μL of nucleic acid amplification product α was dropped onto the surface of the carbon cloth cathode and incubated at 25°C to dry, thereby immobilizing the nucleic acid amplification product α on the cathode surface (e.g., ...). Figure 1 As shown in B), a zinc-air fuel cell self-powered hydrogel sensor for detecting pathogens was obtained.

[0066] Example 2: Construction of Machine Learning Algorithm Module

[0067] Model input and output: The electrochemical signal (current value) output by the zinc-air fuel cell is used as the input feature, and the logarithm of the target gene concentration (lg Conc.) is used as the output label;

[0068] Algorithm selection: Eleven algorithms, including Linear Regression, Ridge Regression, Lasso Regression, ElasticNet, Decision Tree, Random Forest, Boosting, Support Vector Regression (SVR), Neural Networks, Bagging, and Ensemble algorithms, were trained using full computation. The coefficient of determination (R²) was compared through experiments in Example 3. 2 The system uses metrics such as mean squared error (MSE) to intelligently select the optimal model (in this embodiment, the random forest model is preferred, R0). 2 = 0.98).

[0069] Predictive application: Input the electrochemical signal of an unknown sample into the optimal model, and output the predicted target concentration and its 95% confidence interval.

[0070] Example 3: Performance Characterization Experiment of Zinc-Air Fuel Cell Self-Powered Hydrogel Sensor for Pathogen Detection

[0071] The test solution containing the target gene fragment of the sugarcane top rot pathogen used in the experimental group of this embodiment was obtained by treating sugarcane samples infected with top rot using a fungal DNA kit (purchased). The nucleotide sequence of the target gene fragment of the sugarcane top rot pathogen is shown in SEQ ID No. 5.

[0072] 3.1 Characterization of specific responses (corresponding to) Figure 2 )

[0073] A control group (without the target gene for sugarcane top rot) and an experimental group (containing the target gene) were set up. Zinc-air fuel cell self-powered hydrogel sensors for detecting the pathogen were assembled in both groups, and electrochemical signals (output current values) were collected. The results showed that the signal intensity of the experimental group (average 320 μA) was significantly higher than that of the control group (average 65 μA), indicating that the sensor has good recognition ability and selectivity for the target gene of sugarcane top rot, providing experimental evidence for its quantitative detection.

[0074] 3.2 Intelligent Regression Model Screening Characteristic (Corresponding) Figure 3 , Figure 4 , Figure 5 )

[0075] (1) Prepare test solutions containing different concentrations of target gene fragments of sugarcane top rot pathogen;

[0076] (2) Following the preparation method of Example 1, the test liquid was added to the catalyst limiter for incubation to prepare nucleic acid amplification product α; a zinc-air fuel cell was assembled; and the corresponding output current value generated by the self-powered hydrogel sensor of the zinc-air fuel cell was collected under the condition of using test liquid with target gene fragments of sugarcane top rot pathogen at various concentrations.

[0077] (3) Using the output current value as the input feature and the logarithm of the concentration of the target gene fragment of sugarcane top rot pathogen as the output label, the results are input into the machine learning model, and the output results are as follows: Figure 3 As shown, according to Figure 3 Calculate the mean square error and R of the model 2 The result is as follows Figure 4 and Figure 5 As shown, the random forest algorithm R 2 After the model automatically matches the optimal algorithm (random forest) to the minimum, it outputs a signal fingerprint (feature importance ranking) to obtain the regression model.

[0078] The results show that the regression model is more flexible in analyzing complex signals and significantly enhances the environmental adaptability of the detection.

[0079] 3.3 Linear performance characterization (corresponding to) Figure 6 )

[0080] At a target gene concentration of 10 -15 -10 -6 Within the nM range, the sensor output current (Current) and the logarithm of the concentration of the target gene fragment of sugarcane top rot pathogen (lg Conc.) increase exponentially, with the fitted equation being: Current = 15.594 lgConc. + 307.466 (R 2 = 0.99). This ultrasensitive response characteristic indicates that the zinc-air fuel cell self-powered hydrogel sensor for detecting pathogens can achieve accurate quantification over a wide concentration range, providing data support for early warning of sugarcane top rot disease.

[0081] Example 4: Application steps of a zinc-air fuel cell self-powered hydrogel sensor for pathogen detection.

[0082] 4.1 Sample pretreatment: Prepare a test solution containing pathogen target gene fragments, add the test solution to the catalyst limiter, and obtain nucleic acid amplification product α by referring to step 1.1 in Example 1;

[0083] 4.2 Device Assembly and Signal Acquisition: A zinc-air fuel cell self-powered hydrogel sensor for detecting pathogens was prepared according to steps 1.2 and 1.3 in Example 1, and connected to an electrochemical workstation to acquire current signals;

[0084] 4.3 Intelligent quantitative prediction: The current signal output by the self-powered hydrogel sensor of the zinc-air fuel cell is input into a pre-trained machine learning regression model. The regression model outputs the predicted concentration of pathogen target gene fragments and the confidence interval.

[0085] Table 1. Related sequences in the examples

[0086]

[0087] Note: The gene sequence names mentioned are for convenience of writing and have no actual meaning.

[0088] This embodiment provides the construction and performance verification of a self-powered hydrogel sensor detection device for zinc-air fuel cells regulated by a machine learning intelligent screening model, which is used for the accurate detection of target fragments of sugarcane top rot pathogen. The core innovation of this embodiment is: (1) Constructing a self-powered sensor based on ZAFC, combining DNA molecular recognition technology and nucleic acid amplification reaction to achieve targeted release of oxygen reduction catalyst; the system converts chemical energy into the current signal required for electrochemical detection, and completes the efficient amplification and signal cascade amplification of pathogen target gene fragments. (2) The introduction of hydrogel electrolyte (prepared from pure water containing KOH, potato starch, κ-carrageenan, titanium carbonitride and glutaraldehyde) significantly improves the storage environment of the sensing element; (3) The optimal model is intelligently screened by using machine learning algorithms (linear regression, ridge regression, Lasso regression, elastic network, decision tree, random forest, gradient boosting, support vector regression, and LightGBM full operation), with electrochemical signal as input feature and logarithm of pathogen target gene fragment concentration as output to construct a regression model, thereby realizing intelligent signal analysis and accurate quantification; that is, the sensing device in this embodiment combines DNA molecule recognition, nucleic acid amplification reaction and machine learning intelligent model to realize integrated detection of "biological recognition-signal conversion-intelligent analysis".

[0089] The foregoing description of specific exemplary embodiments of the invention is for illustrative and explanatory purposes. These descriptions are not intended to limit the invention to the precise forms disclosed, and it will be apparent that many changes and variations can be made in accordance with the foregoing teachings. The exemplary embodiments were chosen and described in order to explain the specific principles of the invention and its practical application, thereby enabling those skilled in the art to implement and utilize various different exemplary embodiments of the invention, as well as various different choices and variations. The scope of the invention is intended to be defined by the claims and their equivalents.

Claims

1. A zinc-air fuel cell self-powered hydrogel sensor for detecting pathogens, characterized in that: include: A zinc-air fuel cell includes a zinc sheet anode, a carbon cloth cathode, and a hydrogel electrolyte disposed between the anode and the cathode; The nucleic acid amplification product α, loaded on the carbon cloth cathode, is prepared by the following steps: adding the test solution dropwise into a catalyst confinement device, with a volume ratio of test solution to catalyst confinement device of (1-3):(5-8), and incubating at 35-39℃ for 20-80 min; then adding F chain, with a volume ratio of F chain to test solution of (1-3):(5-10), and continuing incubation at 35-39℃ for 20-80 min; centrifuging, and collecting the supernatant to obtain the nucleic acid amplification product α; The nucleotide sequence of the F chain is shown in SEQ ID No.

1.

2. The zinc-air fuel cell self-powered hydrogel sensor according to claim 1, characterized in that: The method for preparing the catalyst limiter includes the following steps: 50-100 μL of S-chains and 50-100 μL of platinum nanoparticles were mixed and incubated at 2-4 °C for 5-10 h to obtain an S-Pt solution. 50-100 μL of W-chains and 50-100 μL of D-chains were added sequentially to the S-Pt solution and incubated at 35-39 °C for 1-2 h. 5-25 μL of gold nanoparticles@magnetic beads composite was added and incubated at 2-4 °C for 5-10 h. The supernatant was discarded, and the precipitate was resuspended in sterile, enzyme-free water to a final volume of 50-100 μL to obtain the catalyst limiter. The nucleotide sequences of the S, W, and D chains are shown in SEQ ID No. 2, SEQ ID No. 3, and SEQ ID No. 4, respectively.

3. The zinc-air fuel cell self-powered hydrogel sensor according to claim 2, characterized in that: The preparation method of the platinum nanoparticles includes the following steps: Add 1-5 mL of 0.5-4% chloroplatinic acid solution and 100-500 mg of polyvinylpyrrolidone to 15-45 mL of methanol solution and stir for 10-30 min to form a precursor solution; heat the precursor solution to 60-80℃, add 1-2 mL of 0.5-2% sodium citrate solution, and continue stirring for 10-30 min to obtain the platinum nanoparticles.

4. The zinc-air fuel cell self-powered hydrogel sensor according to claim 2, characterized in that: The preparation method of the gold nanoparticles@magnetic beads composite includes the following steps: Carboxylated magnetic beads with a concentration of 4-5 mg / mL were mixed with chloroauric acid with a concentration of 0.5-1%, and the volume ratio of carboxylated magnetic beads to chloroauric acid was (1-4): (1-4). The mixture was ultrasonically treated for 20-80 min to obtain pretreated magnetic beads. Sodium citrate solution with a concentration of 0.5-2% was added to the pretreated magnetic beads, and the volume ratio of pretreated magnetic beads to sodium citrate was (2-8): (1-4). The mixture was gently shaken for 4-6 h to separate the precipitate, which was the gold nanoparticle@magnetic bead composite.

5. The zinc-air fuel cell self-powered hydrogel sensor according to claim 1, characterized in that: The hydrogel electrolyte is prepared by the following steps: Heat 100-200 mL of 0.5-1 M KOH aqueous solution to 50-70℃ and stir. Add 1-5 g potato starch and 1-5 g κ-carrageenan and continue stirring for 30-60 min. Then add 0.1-0.5 g titanium carbonitride and 100-500 μL of 35-40 wt% glutaraldehyde and continue stirring for 0.5-2 h to obtain a mixed solution. Pour the obtained mixed solution into a mold, cool to room temperature, and let stand for 6-10 h to form the hydrogel electrolyte.

6. The method for preparing the zinc-air fuel cell self-powered hydrogel sensor according to any one of claims 2-4, characterized in that... Includes the following steps: (1) Prepare platinum nanoparticles and gold nanoparticles@magnetic beads composites, and then prepare a catalyst confinement device; use the prepared catalyst confinement device to prepare nucleic acid amplification product α; load the prepared nucleic acid amplification product α onto carbon cloth; (2) Preparation of hydrogel electrolytes; (3) Using a zinc sheet as the anode, a carbon cloth loaded with the nucleic acid amplification product α as the cathode, and the hydrogel electrolyte as the electrolyte, the zinc-air fuel cell self-powered hydrogel sensor is assembled.

7. A sensing device based on a zinc-air fuel cell self-powered hydrogel sensor and machine learning, characterized in that... include: The zinc-air fuel cell self-powered hydrogel sensor according to any one of claims 1-5; The machine learning algorithm module is used to receive the electrochemical signal generated by the self-powered hydrogel sensor of the zinc-air fuel cell, and calculate the predicted value of the pathogen target gene fragment concentration based on a pre-trained regression model. Specifically, the establishment of the regression model includes: selecting a specific algorithm, using a specific electrochemical signal as input features, and the pathogen target gene fragment concentration or the logarithm of the pathogen target gene fragment concentration as output labels, for training to obtain the regression model; the specific algorithm is selected from one of the following: linear regression, ridge regression, Lasso regression, elastic network algorithm, decision tree algorithm, random forest algorithm, gradient boosting algorithm, support vector regression algorithm, and LightGBM algorithm.

8. The application of the zinc-air fuel cell self-powered hydrogel sensor according to any one of claims 1-5 or the sensing device according to claim 7 in the detection of sugarcane top rot.

9. A method for detecting sugarcane top rot using the zinc-air fuel cell self-powered hydrogel sensor according to any one of claims 1-5 or the sensing device according to claim 7, characterized in that... Includes the following steps: (1) Prepare test solutions containing different concentrations of target gene fragments of sugarcane top rot pathogen; (2) The test liquid is added to the catalyst limiter for incubation to prepare nucleic acid amplification product α; the zinc-air fuel cell is assembled; the corresponding output current value generated by the self-powered hydrogel sensor of the zinc-air fuel cell is collected under the test liquid conditions of using target gene fragments of sugarcane top rot pathogen at various concentrations; (3) Using the output current value as the input feature and the logarithm of the concentration of the target gene fragment of sugarcane top rot pathogen as the output label, construct a standard curve or regression model of current-concentration logarithm; (4) Take an actual sample, prepare the actual sample test solution, obtain the corresponding output current value according to the method in step (2), input the output current value into the current signal-concentration logarithmic standard curve or regression model, and calculate the predicted value of the concentration of the target gene fragment of the sugarcane top rot pathogen.

10. The detection method according to claim 9, characterized in that: The construction of the regression model includes the following steps: using the output current value as the input feature and the logarithm of the concentration of the target gene fragment of sugarcane top rot pathogen as the output label, the model is trained to obtain the regression model. The establishment of the regression model specifically includes: selecting a specific algorithm, using a specific electrochemical signal as the input feature, and the concentration of pathogen target gene fragments or the logarithm of pathogen target gene fragment concentrations as the output label, and training to obtain the regression model; The specific algorithm selection is based on one of the following algorithms: linear regression, ridge regression, Lasso regression, elastic network, decision tree, random forest, gradient boosting, support vector regression, LightGBM, neural network, community learning, and ensemble algorithms.

Images