Multi-modal detection method for pathogenic bacteria based on nano composite material and neural network

Abstract

The invention relates to the technical field of food safety detection, and discloses a multi-modal detection method for pathogenic bacteria based on a nano composite material and a neural network, and the method comprises the following steps: preparing the nano composite material based on the metal chelating property of cuttlefish ink source nano particles and an interface regulation and control technology; constructing a dual-mode immunosensing detection platform based on the nano composite material and a lateral flow immunoassay detection platform; the method comprises the following steps: primarily detecting pathogenic bacteria based on a dual-mode immunosensing detection platform to obtain a dual-mode detection signal; extracting a multi-mode signal feature of the dual-mode detection signal based on a fusion architecture of a multi-layer sensor and a radial basis function; and analyzing the multi-modal signal characteristics based on a neural network algorithm, and outputting a detection result and a prediction result of the pathogenic bacteria. Nano-composite engineering, multi-modal signal transduction and machine learning are organically combined, the bottleneck of a traditional detection technology is broken through, and bimodal detection of rapid and accurate detection and risk assessment of the food-borne pathogenic bacteria is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of food safety detection, and in particular to a multimodal detection method for pathogenic bacteria based on nanocomposites and neural networks. Background Art

[0002] Foodborne pathogens (such as Salmonella) are the main pathogens causing food poisoning incidents worldwide. Current detection methods (such as culture and PCR) have problems such as long cycle time (24-72 hours), high equipment dependence, and complex operation, which make it difficult to meet the needs of rapid on-site screening. At the same time, although lateral flow immunochromatography (LFIA) has become a mainstream point-of-care (POCT) tool due to its portability, its inherent defects still limit its practical application: (1) Insufficient sensitivity, relying on traditional markers such as colloidal gold, making it difficult to meet the needs of trace pathogen detection; (2) Lack of quantitative ability, only visual interpretation based on the color intensity of the test strip, lacking accurate quantitative analysis function; (3) Single result interpretation, unable to associate test data with risk level, and unable to provide decision support information.

[0003] Nanozymes (such as Prussian blue nanoparticles (PBNPs)) are widely used for LFIA signal amplification due to their peroxidase-like activity. However, there is an inherent conflict between nanozyme activity and stability: reducing the size of PBNPs improves catalytic activity but increases surface energy, exacerbating particle aggregation. Existing stabilization strategies (such as polymer coating) often sacrifice active site accessibility, limiting the potential for sensitivity improvement.

[0004] Existing technologies often focus on optimizing detection sensitivity, while neglecting the ability to deeply analyze data and predict risks. Supervised learning in machine learning can achieve "end-to-end" predictions from detection data to output results through pre-training of neural network algorithms, enabling the analysis of test results.

[0005] In view of this, it is necessary to design a multimodal detection method for pathogenic bacteria based on nanocomposites and neural networks to solve the above problems. Summary of the Invention

[0006] In view of this, the present invention proposes a multimodal detection method for pathogenic bacteria based on nanocomposite materials and neural networks. The method aims to break through the bottleneck of traditional detection technology by organically combining nanocomposite engineering, multimodal signal transduction and machine learning, and realize rapid and accurate detection and risk assessment of foodborne pathogens, providing a new solution for food and public health safety risk prevention and control.

[0007] In one aspect, the present invention proposes a multimodal detection method for pathogenic bacteria based on nanocomposites and neural networks, comprising:

[0008] Preparation of nanocomposites based on the metal chelation properties of cuttlefish ink-derived nanoparticles and interface regulation technology;

[0009] Determining whether the nanocomposite material is usable based on the performance test data of the nanocomposite material and a preset qualified performance threshold of the nanocomposite material, and constructing a dual-mode immunosensor detection platform based on the judgment result and the lateral flow immunoassay detection platform;

[0010] Based on the dual-mode immune sensing detection platform, preliminary detection of pathogenic bacteria was performed to obtain dual-mode detection signals;

[0011] Analyzing the dual-mode detection signal based on a fusion architecture of a multi-layer perceptron and a radial basis function, and extracting multi-modal signal features;

[0012] An end-to-end analysis model constructed based on a neural network algorithm analyzes the multimodal signal features and outputs detection results and prediction results of pathogens based on the analysis results.

[0013] Furthermore, the nanocomposite material is prepared based on the metal chelation and interface control technology of cuttlefish ink source nanoparticles, including:

[0014] Mechanically stirring and centrifuging the cuttlefish ink to obtain a cuttlefish ink source nanoparticle dispersion;

[0015] Cuttlefish ink nanoparticles were added to the Fe3+ mixed solution to inhibit the nucleation growth of Prussian blue nanoparticles through competitive metal chelation.

[0016] The droplet acceleration rate of the cuttlefish ink-derived nanoparticle dispersion was 160 μL / 30 s;

[0017] The interface self-assembly was completed by stirring at room temperature for 6 hours;

[0018] After centrifugal washing, the mixture is redispersed in an aqueous solution to obtain a nanocomposite material.

[0019] Furthermore, judging whether the nanocomposite material is usable based on the performance test data of the nanocomposite material and a preset qualified performance threshold of the nanocomposite material includes:

[0020] Performance data on colorimetric properties, antibody conjugation efficiency, and catalytic properties of the nanocomposites were measured;

[0021] If the performance data reaches a preset qualified performance threshold of the nanocomposite material, it is determined that the nanocomposite material can be used to detect pathogenic bacteria;

[0022] If the performance data does not reach a preset qualified performance threshold of the nanocomposite material, it is determined that the nanocomposite material cannot be used for detecting pathogenic bacteria.

[0023] Furthermore, the dual-mode immunosensor detection platform is constructed based on the judgment results and the lateral flow immunoassay detection platform, including:

[0024] When it is determined that the nanocomposite material can be used to detect pathogenic bacteria, the nanocomposite material is combined with a lateral flow immunoassay detection platform to construct a dual-mode immunosensor detection platform;

[0025] Among them, two parallel detection line assemblies coated with antibody probes are set on the test strip of the lateral flow immunoassay detection platform, which are set as T1 detection line and T2 quality control line;

[0026] disposing the pathogenic bacteria antibody labeled with the nanocomposite material on the conjugate pad of the lateral flow immunoassay detection platform;

[0027] Co-incubating the nanocomposite material with a sample to be tested to obtain a complex to be tested;

[0028] combining the complex to be tested with the pathogenic bacteria antibody labeled with the nanocomposite material on the binding pad to form a mixture to be tested;

[0029] The test mixture flows to two detection lines on the lateral flow immunochromatography platform, forming a dual-mode immunosensor detection platform.

[0030] Furthermore, the test strip on the lateral flow immunoassay detection platform is provided with two parallel detection line assemblies coated with antibody probes, which are set as T1 detection line and T2 quality control line, including: the T1 detection line is coated with specific antibodies against pathogenic bacteria, and the T2 quality control line is coated with quality control antibodies against pathogenic bacteria; the T1 detection line is also provided with a colorimetric signal capture area and a catalytic reaction signal amplification area.

[0031] Furthermore, the dual-mode immune sensing detection platform is used to perform preliminary detection of pathogens to obtain dual-mode detection signals, including:

[0032] The test mixture first flows to the T1 detection line, where the nanocomposite material aggregates to form a black-brown strip visible to the naked eye, forming a colorimetric detection platform. Based on the colorimetric detection platform, the pathogenic bacteria are visually and qualitatively detected, and a colorimetric detection signal is generated.

[0033] 3,3',5,5'-tetramethylbenzidine substrate is added to the catalytic reaction signal amplification area of ​​the T1 detection line for catalytic reaction, and the test mixture flows to the T2 quality control line to form a catalytic mode detection platform;

[0034] Pathogenic bacteria are quantitatively detected based on the catalytic signals recorded by the mobile device and the catalytic data of the catalytic pattern detection platform, and a catalytic detection signal is generated.

[0035] Furthermore, the dual-mode detection signal is analyzed based on the fusion architecture of the multi-layer perceptron and radial basis function, and multi-modal signal features are extracted, including:

[0036] The multilayer perceptron extracts the signal intensity gradient change characteristics of the catalytic detection signal to generate a global structured signal feature;

[0037] The radial basis function network extracts the mutation points and boundary features of the colorimetric detection signal to generate local discriminant signal features.

[0038] Furthermore, the end-to-end analysis model constructed based on the neural network algorithm analyzes the multimodal signal features and outputs the detection results and prediction results of pathogens according to the analysis results, including:

[0039] Construct data training set and data validation set based on neural network algorithm;

[0040] Constructing a classification prediction model and a quantitative prediction model based on the data training set and the data validation set respectively;

[0041] Analyzing local discriminative signal features based on the classification prediction model to obtain a detection result;

[0042] The global structured signal characteristics are analyzed based on the quantitative prediction model to obtain a prediction result.

[0043] Furthermore, the detection result is a concentration value of pathogenic bacteria, and the prediction result is a risk assessment level of pathogenic bacteria, and the risk assessment level is divided into potential risk, medium risk and high risk.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] The present invention's multimodal pathogen detection method based on nanocomposites and neural networks utilizes cuttlefish ink-derived nanoparticles as carriers, leveraging their metal chelation properties to generate a low-crystallinity Prussian blue nanoparticle composite system (i.e., a nanocomposite) through interfacial self-assembly. This overcomes the technical bottleneck of traditional Prussian blue nanoparticles, which lack both catalytic activity and stability. The resulting nanocomposite, combined with a lateral flow immunochromatography detection platform, significantly improves detection sensitivity. Combining the lateral flow immunochromatography detection platform with a neural network analysis and prediction system, the present invention achieves dual-modal biosensor detection and risk warning for pathogens.

[0046] The present invention organically combines nanocomposite engineering, multimodal signal transduction and machine learning, realizes end-to-end analysis through integrated neural network algorithms, directly links raw data with risk levels, provides a basis for measures to be taken after detection, enhances interpretability, and can improve the prediction accuracy of pathogens, thus realizing rapid and accurate detection and prediction of foodborne pathogens, and providing a new solution with advanced technology and reliable results for food and public health safety risk prevention and control.

[0047] In another aspect, the present invention further provides an application of the multimodal pathogen detection method based on nanocomposites and neural networks as described in any one of claims 1 to 9, including application of the detection method to rapid on-site detection and contamination tracing of foodborne Salmonella Typhimurium in food processing, cold chain transportation terminals, or clinical samples.

[0048] It is understandable that the above-mentioned multimodal detection method and application of pathogenic bacteria based on nanocomposite materials and neural networks of the present invention have the same beneficial effects and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0050] Figure 1 A flow chart of a multimodal detection method for pathogenic bacteria based on nanocomposites and neural networks provided in an embodiment of the present invention;

[0051] Figure 2 The synthesis mechanism and characterization diagram of the nanocomposite material provided by the embodiment of the present invention;

[0052] Figure 3 A comparison chart of the performance testing and evaluation of the nanocomposite material and Prussian blue nanoparticles provided in an embodiment of the present invention;

[0053] Figure 4 Schematic diagram of a test strip and conjugate pad of a lateral flow immunoassay detection platform provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0054] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, unless there is a conflict, the embodiments of the present disclosure and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0055] Foodborne pathogens (such as Salmonella) are the main pathogens causing food poisoning incidents worldwide. Current detection methods (such as culture and PCR) have problems such as long cycle time (24-72 hours), high equipment dependence, and complex operation, which make it difficult to meet the needs of rapid on-site screening. At the same time, although lateral flow immunochromatography (LFIA) has become a mainstream point-of-care (POCT) tool due to its portability, its inherent defects still limit its practical application: (1) Insufficient sensitivity, relying on traditional markers such as colloidal gold, making it difficult to meet the needs of trace pathogen detection; (2) Lack of quantitative ability, only visual interpretation based on the color intensity of the test strip, lacking accurate quantitative analysis function; (3) Single result interpretation, unable to associate test data with risk level, and unable to provide decision support information.

[0056] Nanozymes (such as Prussian blue nanoparticles (PBNPs)) are widely used for LFIA signal amplification due to their peroxidase-like activity. However, there is an inherent conflict between nanozyme activity and stability: reducing the size of PBNPs improves catalytic activity but increases surface energy, exacerbating particle aggregation. Existing stabilization strategies (such as polymer coating) often sacrifice active site accessibility, limiting the potential for sensitivity improvement.

[0057] Existing technologies often focus on optimizing detection sensitivity, while neglecting the ability to deeply analyze data and predict risks. Supervised learning in machine learning can pre-train models through neural network algorithms, achieving "end-to-end" prediction from detection data to output results, and providing the ability to analyze test results.

[0058] Therefore, the present invention proposes a multimodal detection method for pathogenic bacteria based on nanocomposites and neural networks to solve the above problems.

[0059] Reference Figure 1 As shown, in some embodiments of the present application, a multimodal detection method for pathogenic bacteria based on nanocomposites and neural networks includes:

[0060] S1. Preparation of nanocomposites based on the metal chelation properties and interface regulation of cuttlefish ink-derived nanoparticles.

[0061] S2. Determine whether the nanocomposite material is usable based on the performance test data of the nanocomposite material and a preset qualified performance threshold of the nanocomposite material, and construct a dual-mode immunosensor detection platform based on the judgment result and the lateral flow immunoassay detection platform.

[0062] S3. Conduct preliminary detection of pathogenic bacteria based on the dual-mode immune sensing detection platform to obtain dual-mode detection signals.

[0063] S4. Analyze the dual-mode detection signal based on a fusion architecture of a multi-layer perceptron and radial basis functions, and extract multi-modal signal features.

[0064] S5. An end-to-end analysis model constructed based on a neural network algorithm analyzes the multimodal signal features, and outputs detection results and prediction results of pathogens based on the analysis results.

[0065] It can be seen that this application organically combines nanomaterial engineering, multimodal biosensing and machine learning to form a complete technology chain of "material preparation-signal detection-intelligent analysis".

[0066] It can be seen that the dual-mode detection of the present application takes into account both qualitative convenience and accuracy, and solves the single signal defect of the traditional lateral flow immunoassay detection platform.

[0067] The neural network of this application realizes the intelligent conversion from raw signals to risk levels through end-to-end analysis, thereby improving the accuracy and interpretability of detection results.

[0068] Reference Figure 1-2 As shown, in some embodiments of the present application, the nanocomposite material is prepared based on the metal chelation and interface regulation technology of cuttlefish ink source nanoparticles, including:

[0069] S11, mechanically stirring and centrifuging the cuttlefish ink to obtain a cuttlefish ink source nanoparticle dispersion;

[0070] S12, adding cuttlefish ink source nanoparticles to the Fe3+ mixed solution to inhibit the nucleation growth of Prussian blue nanoparticles through competitive metal chelation;

[0071] S13, the droplet acceleration rate of the cuttlefish ink-derived nanoparticle dispersion was 160 μL / 30 s;

[0072] S14, stirring at room temperature for 6 hours to complete the interface self-assembly;

[0073] S15, after centrifugal washing, redisperse in an aqueous solution to obtain a nanocomposite material.

[0074] Specifically, the cuttlefish ink nanoparticles were purified using a reagent-free process involving sequential washing and centrifugation. The steps are as follows: First, food-grade cuttlefish ink was diluted with deionized water in a volume ratio greater than 15:1 and stirred thoroughly to ensure uniform mixing. The resulting suspension was then centrifuged five times at 10,000 rpm for 15 minutes each, with fresh deionized water replaced after each centrifugation. The resulting purified cuttlefish ink nanoparticles were freeze-dried under vacuum for 12 hours and stored for later use.

[0075] The nanocomposite was prepared by depositing Prussian blue nanoparticles on the surface of cuttlefish ink nanoparticles via interfacial self-assembly in solution. The synthesis process is as follows: First, 5 mg of cuttlefish ink nanoparticles were dispersed in 10 mL of deionized water to form solution A. Simultaneously, solution B was prepared, consisting of 25 mg of FeCl₃ and 28.8 mg of K₃Fe(CN)₆ dissolved in 5 mL of deionized water. Subsequently, 0.8 mL of solution B was diluted with deionized water to a total volume of 10 mL and slowly injected into solution A at a controlled rate (160 μL / 30 seconds) via a syringe pump at ambient temperature (25°C) while maintaining continuous stirring. After a 6-hour reaction, the system was centrifuged at 10,000 rpm for 10 minutes and rinsed five times to obtain the desired nanocomposite.

[0076] It is understood that the nanocomposite is a naturally occurring nanoparticle composed mainly of melanin, which provides multiple biosorption sites for metal ions. Therefore, the synthesis of the nanocomposite can be attributed to the preferential chelation of Fe3+ on the surface of cuttlefish ink nanoparticles, followed by interfacial self-assembly with [Fe(CN)6]4- to form the nanocomposite deposited on the surface of cuttlefish ink nanoparticles. The extent of interfacial growth of the nanocomposite can be adjusted by adjusting the concentration of various precursor solutions. Figure 2 As shown in the TEM (transmission electron microscopy) images of A and 2B, the surface of individual CINPs is relatively smooth, while the surface particle size of the assembled composite material is enhanced, and the clearly visible small-sized PBNPs present a curved square morphology. CINPs are mainly distributed between 120-140nm, with an average diameter of 131.75nm. In contrast, the size distribution of PBNPs formed by interfacial self-assembly is more uniform, with an average diameter of 26.25nm, which is called PBNPs. (30) This phenomenon may be related to the rapid nucleation during the self-assembly process. In addition, larger sized PBNPs (designated as PBNPs) were synthesized in the absence of CINPs. (200) ). Electron diffraction analysis based on TEM ( Figure 2 C) shows that PBNPs (200)The self-assembled PBNPs exhibited a single-crystal structure with well-defined lattice fringes, indicating high crystallinity. (30) It shows low crystallinity. Fourier transform infrared (FTIR) spectrum ( Figure 2 D) shows a broad peak around 3427 cm-1 and a distinct peak at 1600 cm-1, which correspond to the OH stretching vibration and CO bond vibration of the hydroxyl groups in CINPs, respectively. The peak at 2090 cm-1 is attributed to the C≡N stretching vibration of the cyanide bridge, which is a characteristic of PBNPs. CPBNPs and PBNPs (30) The comparative analysis confirmed that PBNPs (30) Successfully synthesized on the surface of CINPs, CPBNPs exhibited enhanced CO vibrational signals, which could be attributed to the presence of CINPs.

[0077] As can be seen, this application suppresses the crystallization rate of Prussian blue nanoparticles through interface regulation, forming an amorphous structure, and the resulting nanocomposite material achieves both activity and stability. Furthermore, this application utilizes waste cuttlefish ink sacs to prepare Prussian blue nanoparticles, aligning with the principles of a circular economy, reducing material costs, and minimizing environmental pollution.

[0078] Reference Figure 1 and 3 As shown, in some embodiments of the present application, judging whether the nanocomposite material is usable based on the performance test data of the nanocomposite material and a preset qualified performance threshold of the nanocomposite material includes:

[0079] Performance data on colorimetric properties, antibody conjugation efficiency, and catalytic properties of the nanocomposites were measured;

[0080] If the performance data reaches a preset qualified performance threshold of the nanocomposite material, it is determined that the nanocomposite material can be used to detect pathogenic bacteria;

[0081] If the performance data does not reach a preset qualified performance threshold of the nanocomposite material, it is determined that the nanocomposite material cannot be used for detecting pathogenic bacteria.

[0082] Specifically, the colorimetric properties of the nanocomposites and Prussian blue nanoparticles at equal concentrations were compared under serial dilution conditions. Figure 3 As shown in Figure A, the nanocomposite reached an undetectable visual signal threshold at a lower concentration than that of Prussian blue nanoparticles. Furthermore, the nanocomposite exhibited superior colorimetric signal intensity across every dilution gradient. These findings suggest that the combined structure of cuttlefish ink nanoparticles and Prussian blue nanoparticles in the nanocomposite enhances colorimetric responsiveness, despite a slower color development rate at equivalent dilution levels.

[0083] It is understandable that this phenomenon may be related to the broad-spectrum absorption properties of cuttlefish ink nanoparticles in the visible light region.

[0084] Specifically, in order to ensure the specific recognition ability of the target, the coupling relationship between the nanocomposite and the antibody recognition molecule must be considered when constructing the sensing platform. Gold nanoparticles (AuNPs) were selected as a control to evaluate the antibody coupling efficiency of the nanocomposite. Three dose groups (1, 5, and 10 μg) were coupled with AuNPs and nanocomposites, respectively. After centrifugation, the residual antibodies in the supernatant were quantified by enzyme-linked immunosorbent assay (ELISA) to calculate the coupling efficiency. Figure 3 As shown in Figures B and 3C, the conjugation efficiencies of AuNPs under the three gradients were 99.82%, 97.18%, and 83.14%, respectively, while the conjugation efficiencies of the nanocomposites under the corresponding conditions were 99.94%, 94.90%, and 80.41%, respectively. Complete binding of the two signal tags to the antibodies was observed at low antibody concentrations, while excessive antibody may induce binding saturation, resulting in an increase in the free antibody content.

[0085] Understandably, the nanocomposite exhibited comparable binding efficiency to AuNPs for antibody recognition molecules in immunoassays, which could be attributed to the presence of cuttlefish ink nanoparticles. Studies have shown that melanin in cuttlefish ink nanoparticles imparts multiple functional groups to the nanocomposite, enabling covalent or non-covalent interactions with various molecules.

[0086] Specifically, the catalytic properties of the nanocomposite were tested. First, the source of catalytic activity was investigated by comparing the color development ability of the supernatant after centrifugation with that of the nanocomposite precipitate. Figure 3 As shown in Figure D, only the substrate catalyzed by the nanocomposite precipitation showed obvious characteristic absorption peaks in the UV-visible spectrum, confirming that the catalytic function came from the nanocomposite composite rather than ion leaching. In addition, two color-developing substrates, TMB and OPD, were selected to evaluate the catalytic performance of the nanocomposite. Figure 3 As can be seen from E, the signal intensity obtained with TMB as the catalytic substrate significantly exceeds that of the reaction based on oxidase activity. Therefore, TMB was used as the optimal catalytic substrate for subsequent experiments. Figure 3 F) showed that when the nanocomposite interacted with TMB substrate alone, the color signal was weak, while the color intensity was significantly enhanced in the presence of H2O2. This observation suggests that the nanocomposite exhibited significantly better peroxidase (POD)-like activity compared to its oxidase-like function. Based on these findings, we systematically optimized the pH conditions for nanocomposite-mediated catalysis ( Figure 3G). The catalytic system is most active under weakly alkaline conditions, and its performance decreases significantly under strongly acidic or alkaline environments. Therefore, pH 5 was selected as the optimal reaction parameter for subsequent experimental verification. To quantitatively characterize the enzymatic performance, we determined the Michaelis-Menten kinetic parameters (Km and Vmax) using TMB and H2O2 as substrates, respectively. Comparative analysis of these indicators provides important insights into the catalytic affinity and turnover capacity of the nanocomposite under optimized reaction conditions. When using TMB as a substrate, the nanomaterial had a Km of 0.49 mM and a Vmax of 31.89 × 10 -8 Ms -1 ( Figure 3 H). When H2O2 is used as substrate, these parameters become Km = 21.75 mM and Vmax = 27.7 × 10 -8 Ms -1 ( Figure 3 1). The Km value of TMB is significantly lower, indicating a higher substrate binding affinity, while the Vmax value indicates efficient catalytic rate for both substrates.

[0087] It is understandable that nanocomposites possess superior catalytic activity and have significant potential as signal tags in catalytic applications.

[0088] It can be seen that multi-dimensional performance testing ensures that nanocomposites meet the testing requirements, avoids testing errors caused by material defects, and improves the reliability of the testing method. This application provides a quantitative standard for material screening by pre-setting qualified thresholds, facilitating quality control and batch-to-batch consistency verification in industrial production.

[0089] Reference Figure 1 and 4 As shown, in some embodiments of the present application, the dual-mode immunosensor detection platform is constructed based on the judgment result and the lateral flow immunoassay detection platform, including:

[0090] When it is determined that the nanocomposite material can be used to detect pathogenic bacteria, the nanocomposite material is combined with a lateral flow immunoassay detection platform to construct a dual-mode immunosensor detection platform;

[0091] Among them, two parallel detection line assemblies coated with antibody probes are set on the test strip of the lateral flow immunoassay detection platform, which are set as T1 detection line and T2 quality control line;

[0092] disposing the pathogenic bacteria antibody labeled with the nanocomposite material on the conjugate pad of the lateral flow immunoassay detection platform;

[0093] Co-incubating the nanocomposite material with a sample to be tested to obtain a complex to be tested;

[0094] combining the complex to be tested with the pathogenic bacteria antibody labeled with the nanocomposite material on the binding pad to form a mixture to be tested;

[0095] The test mixture flows to two detection lines on the lateral flow immunochromatography platform, forming a dual-mode immunosensor detection platform.

[0096] Reference Figure 1 and 4 As shown, in some embodiments of the present application, the test strip on the lateral flow immunoassay detection platform is provided with two parallel detection line assemblies coated with antibody probes, which are set as T1 detection line and T2 quality control line, including: the T1 detection line is coated with specific antibodies against pathogenic bacteria, and the T2 quality control line is coated with quality control antibodies against pathogenic bacteria; the T1 detection line is also provided with a colorimetric signal capture area and a catalytic reaction signal amplification area.

[0097] Reference Figure 1 and 4 As shown, in some embodiments of the present application, the dual-mode immunosensor detection platform is used to perform preliminary detection of pathogens to obtain a dual-mode detection signal, including:

[0098] The test mixture first flows to the T1 detection line, where the nanocomposite material aggregates to form a black-brown strip visible to the naked eye, forming a colorimetric detection platform. Based on the colorimetric detection platform, the pathogenic bacteria are visually and qualitatively detected, and a colorimetric detection signal is generated.

[0099] 3,3',5,5'-tetramethylbenzidine substrate is added to the catalytic reaction signal amplification area of ​​the T1 detection line for catalytic reaction, and the test mixture flows to the T2 quality control line to form a catalytic mode detection platform;

[0100] Pathogenic bacteria are quantitatively detected based on the catalytic signals recorded by the mobile device and the catalytic data of the catalytic pattern detection platform, and a catalytic detection signal is generated.

[0101] Specifically, the T1 detection line of the test strip is coated with Salmonella monoclonal antibody (1 mg / mL), and the streaking speed is 1 μL / cm; the T2 quality control line is coated with goat anti-mouse IgG (1 mg / mL), and the streaking speed is 1 μL / cm; the catalytic solution is 4 μL of 20 mM TMB and 8 μL of 3% H2O2.

[0102] Specifically, 100 μL of different concentrations (10-10 7A Salmonella typhimurium solution containing 100 CFU / mL (0.1 μg / mL) was mixed with 2 μL of the nanocomposite-labeled probe; the mixture was added dropwise to the sample well of the test strip, reacted at room temperature for 25 minutes, and then the T1 line signal intensity was measured. The detection limit criterion was recorded as the lowest bacterial concentration at which a color reaction could be visually identified; the colored test strip was injected with 180 μL of a 3,3',5,5'-tetramethylbenzidine substrate catalytic solution; after a 1-minute catalytic reaction, the catalytic signal was recorded and data was acquired using a mobile device.

[0103] It can be understood that the dual-zone division of the T1 detection line realizes "qualitative-quantitative" signal separation, which facilitates subsequent feature extraction and model analysis; the T2 quality control line excludes false positive or false negative results and improves the credibility of the test.

[0104] It can be understood that the nanocomposite material aggregates at the T1 detection line to form a black-brown strip, and a colorimetric signal is generated by visual observation or image acquisition, which can be used to qualitatively determine whether the pathogenic bacteria exist.

[0105] It can be understood that when 3,3',5,5'-tetramethylbenzidine substrate is added to the catalytic reaction zone of the T1 detection line, the nanocomposite material catalyzes the reaction to generate a blue product, and the catalytic signal (absorbance value) is recorded by a mobile device (such as a smartphone) and combined with the standard curve to generate a quantitative detection signal.

[0106] It can be seen that the colorimetric mode detection platform of this application is suitable for rapid on-site screening, and the catalytic mode detection platform is suitable for precise quantification. This application can meet the needs of different scenarios.

[0107] Reference Figure 1 As shown, in some embodiments of the present application, the fusion architecture based on the multilayer perceptron and radial basis function analyzes the dual-mode detection signal and extracts multimodal signal features, including:

[0108] The multilayer perceptron extracts the signal intensity gradient change characteristics of the catalytic detection signal to generate a global structured signal feature;

[0109] The radial basis function network extracts the mutation points and boundary features of the colorimetric detection signal to generate local discriminant signal features.

[0110] It can be understood that the global analysis of the multilayer perceptron is suitable for quantitative models, such as concentration prediction, and the local analysis of the radial basis function network is suitable for classification models, such as existence judgment.

[0111] It is understandable that in order to address possible visual interpretation errors of colorimetric signals, the radial basis function network reduces the impact of noise by extracting boundary features; the multilayer perceptron filters random fluctuations in the catalytic signal through gradient analysis to improve feature reliability.

[0112] Reference Figure 1 As shown, in some embodiments of the present application, the end-to-end analysis model constructed based on the neural network algorithm analyzes the multimodal signal features and outputs the detection results and prediction results of pathogens according to the analysis results, including:

[0113] Construct data training set and data validation set based on neural network algorithm;

[0114] Constructing a classification prediction model and a quantitative prediction model based on the data training set and the data validation set respectively;

[0115] Analyzing local discriminative signal features based on the classification prediction model to obtain a detection result;

[0116] The global structured signal characteristics are analyzed based on the quantitative prediction model to obtain a prediction result.

[0117] Specifically, the historical detection data is divided into training set and validation set at a ratio of 7:3, and then input into the model after standardization.

[0118] Specifically, the classification prediction model inputs local discriminative signal features and outputs pathogen concentration values; the quantitative prediction model inputs global structured signal features and outputs risk assessment levels.

[0119] Specifically, the parameters are adjusted by cross-validation to ensure that the training set R 2 ≥0.942, test set R 2 ≥0.962, mean absolute error (MAE) ≤0.21.

[0120] Reference Figure 1 As shown, in some embodiments of the present application, the detection result is the concentration value of the pathogenic bacteria, and the prediction result is the risk assessment level of the pathogenic bacteria, and the risk assessment level is divided into potential risk, medium risk and high risk.

[0121] Specifically, the classification prediction model outputs the specific concentration value of pathogens, and the quantitative prediction model maps the concentration value to three risk levels:

[0122] Specifically, the potential risk is the specific concentration value of pathogens ≤10 3 When the CFU / mL is high, the risk to the normal population is low, but susceptible populations need to be careful;

[0123] The specific concentration value of medium risk pathogens is 10 3 -10 5 CFU / mL, the risk of gastroenteritis is 60%, requiring initial intervention;

[0124] High risk means the specific concentration of pathogens is >10 5CFU / mL, the risk of severe diarrhea is 80%, requiring emergency treatment.

[0125] It can be seen that this application can provide precise concentration values ​​and risk levels at the same time. The former is used for scientific research and standard compliance judgment, and the latter is used for regulatory decision-making and public health warnings to meet the needs of different users.

[0126] It can be seen that converting complex concentration data into intuitive risk levels makes it easier for non-professionals (such as food processing practitioners) to quickly understand and take action, thereby improving the practical application value of the test results.

[0127] The multimodal pathogen detection method based on nanocomposites and neural networks in the above-mentioned embodiments uses cuttlefish ink-derived nanoparticles as carriers, exploits the metal chelation properties of cuttlefish ink-derived nanoparticles, and generates a low-crystallinity Prussian blue nanoparticle composite system, i.e., a nanocomposite, through interfacial self-assembly technology. This overcomes the technical bottleneck of traditional Prussian blue nanoparticles, which cannot achieve both catalytic activity and stability. The nanocomposite obtained in this application and the lateral flow immunochromatography detection platform significantly improve detection sensitivity; this application combines the lateral flow immunochromatography detection platform with a neural network analysis and prediction system to achieve dual-modal biosensor detection and risk warning of pathogens.

[0128] The present invention organically combines nanocomposite engineering, multimodal signal transduction and machine learning, realizes end-to-end analysis through integrated neural network algorithms, directly links raw data with risk levels, provides a basis for measures to be taken after detection, enhances interpretability, and can improve the prediction accuracy of pathogens, thus realizing rapid and accurate detection and prediction of foodborne pathogens, and providing a new solution with advanced technology and reliable results for food and public health safety risk prevention and control.

[0129] In another preferred embodiment based on the above embodiment, refer to Figure 1 As shown, this embodiment provides an application of a multimodal detection method for pathogenic bacteria based on nanocomposites and neural networks as described in any one of claims 1 to 9, including: the detection method is applied to on-site rapid detection and contamination tracing of foodborne Salmonella typhimurium in food processing links, cold chain transportation terminals or clinical samples.

[0130] It is understandable that in the above embodiments of the present invention, the multimodal detection method and application of pathogenic bacteria based on nanocomposites and neural networks have the same beneficial effects, which will not be described in detail.

[0131] It should be noted that:

[0132] In the description provided herein, a large number of specific details are described. However, it is understood that the embodiments of the present application can be practiced without these specific details. In some instances, well-known structures and technologies are not shown in detail so as not to obscure the understanding of this description.

[0133] Similarly, it should be understood that in order to streamline the present application and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present application, various features of the present application are sometimes grouped together into a single embodiment, figure, or description thereof. However, the disclosed system should not be interpreted as reflecting a schematic representation that the claimed application requires more features than are expressly recited in each claim.

[0134] Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment.The claims following the Detailed Description are thus expressly incorporated into this Detailed Description, with each claim standing on its own as a separate embodiment of this application.

[0135] Furthermore, those skilled in the art will appreciate that although some embodiments described herein include certain features and not other features included in other embodiments, the combination of features from different embodiments is meant to be within the scope of this application and to form different embodiments.

[0136] For example, in the following claims, any of the claimed embodiments can be used in any combination.

[0137] The above description is merely a preferred embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A multimodal detection method for pathogenic bacteria based on nanocomposites and neural networks, characterized in that: include: Preparation of nanocomposites based on the metal chelation properties of cuttlefish ink-derived nanoparticles and interface regulation technology; Determining whether the nanocomposite material is usable based on the performance test data of the nanocomposite material and a preset qualified performance threshold of the nanocomposite material, and constructing a dual-mode immunosensor detection platform based on the judgment result and the lateral flow immunoassay detection platform; Based on the dual-mode immune sensing detection platform, preliminary detection of pathogenic bacteria was performed to obtain dual-mode detection signals; Analyzing the dual-mode detection signal based on a fusion architecture of a multi-layer perceptron and a radial basis function, and extracting multi-modal signal features; An end-to-end analysis model constructed based on a neural network algorithm analyzes the multimodal signal features and outputs detection results and prediction results of pathogens based on the analysis results.

2. The multimodal detection method for pathogenic bacteria based on nanocomposites and neural networks according to claim 1, characterized in that: The nanocomposite material is prepared based on the metal chelation and interface control technology of cuttlefish ink source nanoparticles, including: Mechanically stirring and centrifuging the cuttlefish ink to obtain a cuttlefish ink source nanoparticle dispersion; Cuttlefish ink nanoparticles were added to the Fe3+ mixed solution to inhibit the growth of Prussian blue nanoparticle nuclei through competitive metal chelation. The droplet acceleration rate of the cuttlefish ink-derived nanoparticle dispersion was 160 μL / 30 s; The interface self-assembly was completed by stirring at room temperature for 6 hours; After centrifugal washing, the mixture is redispersed in an aqueous solution to obtain a nanocomposite material.

3. The multimodal detection method for pathogenic bacteria based on nanocomposites and neural networks according to claim 1, characterized in that: The method of judging whether the nanocomposite material is usable based on the performance test data of the nanocomposite material and a preset qualified performance threshold of the nanocomposite material includes: Performance data on colorimetric properties, antibody conjugation efficiency, and catalytic properties of the nanocomposites were measured; If the performance data reaches a preset qualified performance threshold of the nanocomposite material, it is determined that the nanocomposite material can be used to detect pathogenic bacteria; If the performance data does not reach a preset qualified performance threshold of the nanocomposite material, it is determined that the nanocomposite material cannot be used for detecting pathogenic bacteria.

4. The multimodal detection method for pathogenic bacteria based on nanocomposites and neural networks according to claim 3, characterized in that: The dual-mode immunosensor detection platform is constructed based on the judgment results and the lateral flow immunoassay detection platform, including: When it is determined that the nanocomposite material can be used to detect pathogenic bacteria, the nanocomposite material is combined with a lateral flow immunoassay detection platform to construct a dual-mode immunosensor detection platform; Among them, two parallel detection line assemblies coated with antibody probes are set on the test strip of the lateral flow immunoassay detection platform, which are set as T1 detection line and T2 quality control line; disposing the pathogenic bacteria antibody labeled with the nanocomposite material on the conjugate pad of the lateral flow immunoassay detection platform; Co-incubating the nanocomposite material with a sample to be tested to obtain a complex to be tested; combining the complex to be tested with the pathogenic bacteria antibody labeled with the nanocomposite material on the binding pad to form a mixture to be tested; The test mixture flows to two detection lines on the lateral flow immunochromatography platform, forming a dual-mode immunosensor detection platform.

5. The multimodal detection method for pathogenic bacteria based on nanocomposites and neural networks according to claim 4, characterized in that: The test strip on the lateral flow immunoassay detection platform is provided with two parallel detection line assemblies coated with antibody probes, namely the T1 detection line and the T2 quality control line, comprising: the T1 detection line is coated with specific antibodies against pathogenic bacteria, and the T2 quality control line is coated with quality control antibodies against pathogenic bacteria; the T1 detection line is also provided with a colorimetric signal capture area and a catalytic reaction signal amplification area.

6. The multimodal detection method for pathogenic bacteria based on nanocomposites and neural networks according to claim 5, characterized in that: The dual-mode immune sensing platform is used to perform preliminary detection of pathogens to obtain dual-mode detection signals, including: The test mixture first flows to the T1 detection line, where the nanocomposite material aggregates to form a black-brown strip visible to the naked eye, forming a colorimetric detection platform. Based on the colorimetric detection platform, the pathogenic bacteria are visually and qualitatively detected, and a colorimetric detection signal is generated. 3,3',5,5'-tetramethylbenzidine substrate is added to the catalytic reaction signal amplification area of ​​the T1 detection line for catalytic reaction, and the test mixture flows to the T2 quality control line to form a catalytic mode detection platform; Pathogenic bacteria are quantitatively detected based on the catalytic signals recorded by the mobile device and the catalytic data of the catalytic pattern detection platform, and a catalytic detection signal is generated.

7. The multimodal detection method for pathogenic bacteria based on nanocomposites and neural networks according to claim 6, characterized in that: The fusion architecture based on multi-layer perceptron and radial basis function analyzes the dual-mode detection signal and extracts multi-modal signal features, including: The multilayer perceptron extracts the signal intensity gradient change characteristics of the catalytic detection signal to generate a global structured signal feature; The radial basis function network extracts the mutation points and boundary features of the colorimetric detection signal to generate local discriminant signal features.

8. The multimodal detection method for pathogenic bacteria based on nanocomposites and neural networks according to claim 7, characterized in that: The end-to-end analysis model constructed based on the neural network algorithm analyzes the multimodal signal features and outputs the detection results and prediction results of pathogens based on the analysis results, including: Construct data training set and data validation set based on neural network algorithm; Constructing a classification prediction model and a quantitative prediction model based on the data training set and the data validation set respectively; Analyzing local discriminative signal features based on the classification prediction model to obtain a detection result; The global structured signal characteristics are analyzed based on the quantitative prediction model to obtain a prediction result.

9. The multimodal detection method for pathogenic bacteria based on nanocomposites and neural networks according to claim 8, characterized in that: The detection result is the concentration value of the pathogenic bacteria, and the prediction result is the risk assessment level of the pathogenic bacteria. The risk assessment level is divided into potential risk, medium risk and high risk.

10. An application of the multimodal detection method for pathogenic bacteria based on nanocomposites and neural networks according to any one of claims 1 to 9, characterized in that: The detection method is applied to on-site rapid detection and contamination tracing of foodborne Salmonella typhimurium in food processing links, cold chain transportation terminals or clinical samples.

Images