Multi-dimensional bacterial spectrum acquisition method, bacterial identification method and application device

Abstract

The application belongs to the technical field of bacteria identification, and discloses a multi-dimensional bacteria spectrum acquisition method, a bacteria identification method and an application device, wherein the multi-dimensional bacteria spectrum acquisition method comprises the following steps: cracking a bacteria suspension by using ultrasonic waves to obtain a bacteria lysate, mixing a plurality of surface-enhanced Raman scattering concentrated suspension substrates with the bacteria lysate respectively to obtain a plurality of mixed liquids, forming a liquid film by using the plurality of mixed liquids, performing surface-enhanced Raman scattering analysis on the liquid film, and obtaining multi-dimensional surface-enhanced Raman scattering bacteria spectrum data. The application can acquire multi-dimensional spectrum information of bacteria, and then realizes rapid and accurate detection and identification of bacteria samples.

Description

Technical Field

[0001] This invention belongs to the field of bacterial identification technology, and particularly relates to a multi-dimensional bacterial spectrum acquisition method, bacterial identification method, and application device. Background Technology

[0002] In the field of life and health, rapid and accurate detection of different types of bacteria is of great significance. Existing microbial detection technologies, such as polymerase chain reaction, enzyme-linked immunosorbent assay (ELISA), and mass spectrometry, can accurately identify microorganisms. However, the identification process is time-consuming and labor-intensive, and for the detection of complex samples containing multiple bacteria, current technologies can generally only detect a few bacterial species in the complex sample. Summary of the Invention

[0003] This application provides a method for acquiring multi-dimensional bacterial spectra, a method for identifying bacteria, and an application device, which can acquire multi-dimensional spectral information of bacteria and achieve rapid and accurate detection and identification of bacterial samples.

[0004] In a first aspect, embodiments of this application provide a method for obtaining multi-dimensional bacterial spectra, including:

[0005] The bacterial suspension was lysed using ultrasound to obtain bacterial lysate;

[0006] Various surface-enhanced Raman scattering concentrated suspension substrates and the bacterial lysate were mixed to obtain various mixtures;

[0007] A liquid film is formed using the aforementioned mixtures;

[0008] Surface-enhanced Raman scattering analysis was performed on the liquid film to obtain multidimensional surface-enhanced Raman scattering bacterial spectral data.

[0009] Secondly, embodiments of this application provide a bacterial identification method based on the multi-dimensional bacterial spectral acquisition method described in the first aspect, comprising:

[0010] Using the aforementioned multidimensional surface-enhanced Raman scattering (SMR) bacterial spectral data, a multidimensional SMR bacterial spectral database was constructed.

[0011] Build deep learning models;

[0012] The deep learning model is trained using the multidimensional bacterial surface-enhanced scattering spectral database to obtain the trained deep learning model.

[0013] Obtain the single-dimensional surface-enhanced Raman scattering bacterial spectrum of the bacteria to be detected;

[0014] The single-dimensional surface-enhanced Raman scattering data of the bacteria to be detected is input into the trained deep learning model, and the trained deep learning model outputs the species and subtype of the bacteria to be detected.

[0015] Thirdly, embodiments of this application provide a bacterial identification device, including:

[0016] The database construction module is used to construct a multi-dimensional surface-enhanced Raman scattering bacterial spectral database using the multi-dimensional surface-enhanced Raman scattering bacterial spectral data.

[0017] The model building module is used to build deep learning models;

[0018] The model training module is used to train the deep learning model using the multi-dimensional bacterial surface-enhanced scattering spectrum database to obtain the trained deep learning model.

[0019] The data acquisition module is used to acquire the single-dimensional surface-enhanced Raman scattering bacterial spectrum of the bacteria to be detected;

[0020] The bacterial identification module is used to input the single-dimensional surface-enhanced Raman scattering data of the bacteria to be detected into a trained deep learning model, and the trained deep learning model outputs the species and subtype of the bacteria to be detected.

[0021] Fourthly, embodiments of this application provide a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the bacterial identification method as described in the second aspect of the present invention.

[0022] Fifthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the bacterial identification method as described in the second aspect of the present invention.

[0023] In a sixth aspect, embodiments of this application provide a computer program product that, when run on a terminal device, causes the terminal device to execute the steps of the bacterial identification method described in the second aspect above.

[0024] The first aspect of this invention provides a method for acquiring multidimensional bacterial spectra. This method involves using ultrasound to lyse a bacterial suspension to obtain bacterial lysates. These lysates are then mixed with various surface-enhanced Raman scattering (SERS) concentrated suspension substrates to obtain multiple mixtures. A liquid film is formed using these mixtures, and SERS analysis is performed on the liquid film to obtain multidimensional SERS bacterial spectral data. This invention can acquire multidimensional spectral information of bacteria, enabling rapid and accurate detection and identification of bacterial samples.

[0025] It is understood that the beneficial effects of the second to sixth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 This is a schematic diagram of the first process of the multi-dimensional bacterial spectrum acquisition method provided in this embodiment of the invention;

[0028] Figure 2 This is a schematic diagram of a purchase list of 17 types of bacteria, including bacterial subtypes, provided in an embodiment of the present invention.

[0029] Figure 3 This is a schematic diagram of the SERS spectra of Shigella dysenteriae type 1 on different substrates provided in the embodiments of the present invention;

[0030] Figure 4 This is a schematic diagram of the SERS spectra of Shigella dysenteriae type 2 on different substrates provided in the embodiments of the present invention;

[0031] Figure 5 This is a schematic diagram of the SERS spectra of Escherichia coli on different substrates provided in the embodiments of the present invention;

[0032] Figure 6 This is a schematic diagram of SERS spectra of *Faecidobacterium faecalis* on different substrates provided in an embodiment of the present invention;

[0033] Figure 7 This is a schematic diagram of principal component analysis of Shigella dysenteriae type 1 on different substrates provided in the embodiments of the present invention;

[0034] Figure 8 This is a schematic diagram of principal component analysis of Shigella dysenteriae type 2 on different substrates provided in the embodiments of the present invention;

[0035] Figure 9 This is a schematic diagram of principal component analysis of Escherichia coli on different substrates provided in an embodiment of the present invention;

[0036] Figure 10 This is a schematic diagram of principal component analysis of *Faecidus flavus* on different substrates provided in an embodiment of the present invention;

[0037] Figure 11 This is a schematic diagram of the first process for preparing various surface-enhanced Raman scattering concentrated suspension substrates provided in this embodiment of the invention;

[0038] Figure 12 This is a schematic diagram of the first process of the bacterial identification method provided in the embodiments of the present invention;

[0039] Figure 13 This is a schematic diagram illustrating the model recognition accuracy of 2-dimensional data provided in an embodiment of the present invention;

[0040] Figure 14 This is a schematic diagram illustrating the model recognition accuracy of 5-dimensional data provided in this embodiment of the invention;

[0041] Figure 15 This is a schematic diagram illustrating the model recognition accuracy of 6-dimensional data provided in an embodiment of the present invention;

[0042] Figure 16 This is a schematic diagram of the first process for obtaining a single-dimensional surface-enhanced Raman scattering bacterial spectrum of bacteria to be detected, provided in an embodiment of the present invention.

[0043] Figure 17 This is a schematic diagram of the structure of the bacterial identification device provided in an embodiment of the present invention;

[0044] Figure 18 This is a schematic diagram of the structure of the terminal device provided in an embodiment of the present invention. Detailed Implementation

[0045] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0046] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0047] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0048] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0049] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0050] The bacterial identification method provided in this application, based on the multi-dimensional bacterial spectrum acquisition method described in the first aspect, is executed by the processor of a terminal device when running a computer program with corresponding functions. It constructs a multi-dimensional surface-enhanced Raman scattering (SERS) bacterial spectrum database using the SERS bacterial spectral data, builds a deep learning model, trains the deep learning model using the SERS spectral database, obtains the trained deep learning model, acquires the single-dimensional SERS bacterial spectrum of the bacteria to be detected, inputs the single-dimensional SERS data of the bacteria to be detected into the trained deep learning model, and the trained deep learning model outputs the species and subtype of the bacteria to be detected. This enables rapid and accurate detection and identification of bacterial samples.

[0051] In applications, terminal devices can be computing devices capable of data processing, such as tablet PCs, laptops, personal computers (PCs), and cloud servers. This application does not impose any restrictions on the specific type of terminal device.

[0052] like Figure 1As shown, in one embodiment, the multi-dimensional bacterial spectrum acquisition method provided in this application includes the following steps S101 to S104:

[0053] Step S101: Use ultrasound to lyse the bacterial suspension to obtain bacterial lysate, then proceed to step S102.

[0054] In applications, an ultrasonic cell disruptor can be used to lyse bacteria using ultrasound to obtain bacterial lysate.

[0055] In one embodiment, the bacterial suspension comprises a suspension of at least one of the following bacteria: Shigella flexneri, Shigella dysenteriae, Staphylococcus aureus, Bacillus cereus, Bacillus thuringiensis, Escherichia coli, and Bacillus faecalis.

[0056] In application, the bacteria in the bacterial suspension were purchased from the China Industrial Microbial Culture Collection Center, including a total of 17 bacterial species, such as... Figure 2 The image shows a schematic diagram of a purchase list for 17 types of bacteria, including bacterial subtypes.

[0057] Step S102: Mix various surface-enhanced Raman scattering concentrated suspension substrates and the bacterial lysate separately to obtain various mixtures, and proceed to step S103.

[0058] In application, a certain number of surface-enhanced Raman scattering concentrated suspensions can be selected (e.g., 2, 3, 6, etc., the specific number can be adjusted according to actual needs). The certain number of surface-enhanced Raman scattering concentrated suspensions and the bacterial lysate are poured into containers (e.g., centrifuge tubes, etc.) and mixed to obtain multiple mixtures.

[0059] Step S103: Using the various mixed liquids, form a liquid film, and proceed to step S104.

[0060] In applications, the various mixtures can be dropped onto the hydrophobic surface of a hydrophobic carrier (e.g., Teflon tape) to obtain droplets of the mixture. The droplets are then adsorbed onto the surface of a liquid film preparation plate (e.g., a steel plate surface with circular through holes) to form a liquid film.

[0061] Step S104: Perform surface-enhanced Raman scattering analysis on the liquid film to obtain multi-dimensional surface-enhanced Raman scattering bacterial spectral data.

[0062] In applications, a Raman spectrometer can be used to select a certain area of ​​the liquid film, set test points, and automatically collect all surface-enhanced Raman scattering (SERS) data of the sample within the selected area. For example, a 5*5 micrometer area can be selected on the formed liquid film, a certain step size can be set (e.g., 0.1 μm, 0.2 μm, etc.), and the Raman spectrometer can be started. The Raman spectrometer will automatically collect all SERS data of the sample points within the selected area, achieving rapid acquisition of a large amount of SERS data.

[0063] Figure 3 , Figure 4 , Figure 5 , Figure 6 SERS spectra of Shigella dysenteriae type 1, Shigella dysenteriae type 2, Escherichia coli, and Escherichia coli on different substrates are shown as examples. The horizontal axis of each figure represents the Raman shift, and the unit is cm. -1 The vertical axis represents the normalized intensity, which is dimensionless. The corresponding substrates for the four spectral curves, from top to bottom, are silver nano-substrates modified with 11-mercaptoundecanoic acid (MUA), silver nano-substrates modified with 11-mercapto-1-undecanol (MUO), silver nano-substrates modified with 4-mercaptophenylboronic acid (4B), and silver nano-substrates modified with 3-mercaptopropionic acid (3COOH).

[0064] Figure 7 , Figure 8 , Figure 9 , Figure 10 Principal component analysis diagrams of Shigella dysenteriae type 1, Shigella dysenteriae type 2, Escherichia coli, and Escherichia coli on different substrates are shown as examples. The analysis results on different substrates are marked with MUA, MUO, 4B, and 3COOH in each diagram.

[0065] In one embodiment, before step S102, the following is included:

[0066] Prepare various surface-enhanced Raman scattering concentrated suspension substrates.

[0067] In applications, various functionalized surface-enhanced Raman scattering concentrated suspension substrates were prepared using a variety of noble metal nanomaterials of different types or structures and various modifiers.

[0068] like Figure 11 As shown in one embodiment, the preparation of various surface-enhanced Raman scattering concentrated suspension substrates includes the following steps S201 to S202:

[0069] Step S201: Obtain noble metal nanomaterials of different types or structures, then proceed to step S102.

[0070] In applications, the types of noble metal nanomaterials can include silver nanoparticles, gold nanoparticles, etc., and the structures of noble metal nanomaterials can include gold-silver alloys, silver cores and gold shells, etc.

[0071] Step S202: Modify the different types or structures of noble metal nanomaterials using various self-assembled monolayers to obtain various surface-enhanced Raman scattering concentrated suspension substrates.

[0072] In applications, self-assembled monolayers can be 11-mercapto-1-undecanol (MUO), 11-mercaptoundecanoic acid (MUA), 1-dodecanethiol (DT), 4-mercaptophenylboronic acid (4B), 3-mercaptopropionic acid (3COOH), etc.

[0073] In applications, taking the modification of silver nanoparticles using multiple self-assembled monolayers as an example, the process can include the following:

[0074] A certain amount of 1 mM Ag NPs suspension was prepared;

[0075] Add 1 mL of 1 mM Ag NPs suspension to each centrifuge tube, and then add appropriate amounts of various modifier solutions to multiple centrifuge tubes.

[0076] Rotate the centrifuge tube for 20 seconds, then let it stand for 30 to 40 minutes.

[0077] After standing, centrifuge the centrifuge tubes at 10,000 rpm for 10 minutes.

[0078] After centrifugation, use a pipette to transfer 980 μL of supernatant from the centrifuge tube, and then add 1 mL of ethanol to the centrifuge tube to wash away the remaining modifying agent.

[0079] Centrifuge the centrifuge tubes at a speed of 10,000 rpm for 10 minutes.

[0080] After centrifugation, use a pipette to transfer 980 μL of supernatant from the centrifuge tube, and then add 1 mL of deionized water to resuspend the precipitate.

[0081] Centrifuge the centrifuge tubes at a speed of 10,000 rpm for 10 minutes.

[0082] After centrifugation, 960 μL of supernatant was transferred from the centrifuge tube using a pipette to obtain various surface-enhanced Raman scattering concentrated suspension substrates.

[0083] For example, modifying silver nanoparticles using 4-mercaptophenylboronic acid can include the following process:

[0084] A certain amount of 1 mM silver nanoparticle colloidal solution reduced by sodium citrate was prepared, and a 0.1% 4-mercaptophenylboronic acid ethanol solution was prepared.

[0085] Add 1 mL of 1 mM silver nanocolloid solution and 10 μL of 0.1% 4-mercaptophenylboronic acid ethanol solution to a 1.5 mL centrifuge tube. After adding the solution, immediately place the centrifuge tube into a vortex mixer and vortex for 10 to 15 seconds.

[0086] After vortexing, let the centrifuge tubes stand for 10 minutes.

[0087] After settling, the centrifuge tubes were placed in a high-speed refrigerated centrifuge. The centrifuge speed was set to 10,000 rpm and the centrifugation time was 10 min to complete the modification and concentration of silver nanoparticles with 4-mercaptophenylboronic acid, and the modified silver nanoparticles were obtained.

[0088] After centrifugation, the supernatant in the centrifuge tube was removed using a pipette, and then 1 mL of ultrapure water was added to the centrifuge tube. The centrifuge tube was then placed in an ultrasonic cleaner, and the modified silver nanoparticles were uniformly dispersed in the ultrapure water through ultrasonic treatment.

[0089] After ultrasonic treatment, the centrifuge tubes were placed back into a high-speed refrigerated centrifuge. The centrifuge speed was set to 10,000 rpm and the centrifugation time was 10 min to wash away the unreacted 4-mercaptophenylboronic acid modifier in the centrifuge tubes.

[0090] After centrifugation, the supernatant in the centrifuge tube is removed again using a pipette, and then 1 mL of ultrapure water is added to the centrifuge tube. The centrifuge tube is then placed in a high-speed refrigerated centrifuge, and the speed of the high-speed refrigerated centrifuge is set to 10,000 rpm and the centrifugation time is 10 min to complete the washing of the anhydrous ethanol in the centrifuge tube that cannot form a liquid film through step S103.

[0091] After centrifugation, the supernatant in the centrifuge tube was transferred again using a pipette to obtain silver nanoparticles modified with 4-mercaptophenylboronic acid.

[0092] In one embodiment, before step S102, the method further includes:

[0093] Bacteria are cultured to obtain a bacterial suspension.

[0094] In applications, culturing bacteria to obtain a bacterial suspension can include the following processes:

[0095] The bacteria were cultured in 100 mL of sterile nutrient broth for 24 hours at a temperature of 37°C to obtain the cultured bacteria.

[0096] After the culture was completed, the cultured bacteria were mixed with a certain amount of 0.9% NaCl solution using a vortex, and then the cultured bacteria were centrifuged at a speed of 4000 rpm for 5 min.

[0097] After centrifugation, the bacteria were rinsed twice with 10 mL of 0.9% NaCl aqueous solution to remove the growth medium.

[0098] After rinsing, the bacteria were stored in a 0.9% NaCl solution, with the bacterial cell density based on optical density (OD) 600, where the OD 600 value was 1.0, and the storage temperature was 4℃, resulting in a bacterial suspension.

[0099] In application, the order in which the preparation of multiple surface-enhanced Raman scattering (SERS) concentrated suspension substrates and the culture of bacteria are carried out to obtain bacterial suspensions is not limited. For example, the preparation of multiple SERS concentrated suspension substrates and the culture of bacteria can be carried out simultaneously to obtain bacterial suspensions; alternatively, the preparation of multiple SERS concentrated suspension substrates can be carried out first, followed by the culture of bacteria to obtain bacterial suspensions; or the culture of bacteria can be carried out first to obtain bacterial suspensions, followed by the preparation of multiple SERS concentrated suspension substrates.

[0100] like Figure 12 As shown, in one embodiment, the bacterial identification method based on the multi-dimensional bacterial spectrum acquisition method provided in this application includes the following steps S301 to S305:

[0101] Step S301: Using the multi-dimensional surface-enhanced Raman scattering bacterial spectral data, construct a multi-dimensional surface-enhanced Raman scattering bacterial spectral database, and proceed to step S302.

[0102] In applications, each substrate can acquire surface-enhanced Raman scattering (SERS) data in one dimension, while multiple substrates can acquire multi-dimensional SERS data. Therefore, by combining each bacterium with multiple substrates, multi-dimensional bacterial SERS data can be obtained. By summarizing the multi-dimensional bacterial SERS data of multiple bacteria, a multi-dimensional SERS bacterial spectral database can be constructed.

[0103] Step S302: Build a deep learning model, proceed to step S303.

[0104] In one embodiment, the deep learning model includes a one-dimensional convolutional structure and a multilayer perceptron structure;

[0105] The one-dimensional convolutional structure is used to extract bacterial spectral features;

[0106] The multilayer sensor structure is used for classifying bacteria.

[0107] In applications, deep learning models can include a main framework of one-dimensional convolutional structures and multilayer perceptron structures. The one-dimensional convolutional structure is used to extract high-dimensional spatial features from the input spectral data, and the multilayer perceptron structure is used for nonlinear transformation of features and outputting bacterial classification results.

[0108] In applications, deep learning models can include 29-layer network structures. The one-dimensional convolutional structure can include one-dimensional convolutional layers, activation layers, one-dimensional pooling layers, dropout layers, flattening layers, and other network structures. The multilayer perceptron can include two-layer perceptrons and two-layer activation layers, and other network structures.

[0109] In the application, the number of convolutional kernels in the two input layers of the one-dimensional convolutional structure is set to be the same as the dimension of the input spectral data, which is used to learn the features of multi-dimensional spectra. At the same time, a larger convolutional kernel size is set, and dilated convolution technique is used to facilitate the representation of the relationship between adjacent spectral signals over a larger range. The convolutional kernels of the remaining convolutional layers of the one-dimensional convolutional structure are set to a smaller size of 3*3, which makes it easier for the network model to incorporate more nonlinear representation processes, further improves the network model's ability to extract high-dimensional features, and greatly reduces the number of parameters and computational complexity of the network model, which is conducive to the further deepening of the network.

[0110] In the application, a Flatten layer is set in the last layer of the one-dimensional convolutional structure to expand the high-dimensional features along one dimension; an activation layer with a Softmax activation function is set in the last layer of the multilayer perceptron to complete the bacterial classification.

[0111] In the application, Dropout layers were added to the deeper layers of the one-dimensional convolutional structure and the multilayer perceptron structure to improve the effectiveness of gradient backpropagation during network model training, making the optimization process of network model parameters easier to converge.

[0112] Step S303: Using the multi-dimensional bacterial surface-enhanced scattering spectral database, train the deep learning model to obtain the trained deep learning model, and proceed to step S304.

[0113] In applications, spectral data composed of at least two or more different substrates can be obtained from the multi-dimensional bacterial surface-enhanced scattering spectral database as a dataset for training a deep learning model, resulting in a trained deep learning model. For example, taking 6-dimensional spectral data as an example, six different substrates can be randomly selected to form a 6-dimensional spectral dataset. 150 spectral data points can be obtained for each substrate material, resulting in a total of 15,300 spectral data points across the six dimensions. These 15,300 data points can be divided into a training set and a validation set, with the ratio set to 7:3, 6:4, etc., depending on actual needs.

[0114] In applications, such as Figure 13 , Figure 14 , Figure 15 As shown, exemplary diagrams illustrate model recognition accuracy for 2-dimensional, 5-dimensional, and 6-dimensional data, respectively. In the first row of the table, 2D, 5D, and 6D represent 2-dimensional, 5-dimensional, and 6-dimensional data, respectively. The values ​​in the second row represent the model accuracy using a single basis, and the chemical formulas in the third row represent different bases, such as Ag NP. S 11COOH represents silver nanoparticles reduced with sodium citrate, 11COOH represents silver nanoparticles modified with 11-mercaptoundecanoic acid (11COOH), 11OH represents silver nanoparticles modified with 11-mercapto-1-undecanol (11OH), 12CH3 represents silver nanoparticles modified with 1-dodecanethiol (12CH3), 3COOH represents silver nanoparticles modified with 3-mercaptopropionic acid, 3OH represents silver nanoparticles modified with 3-mercapto-1-propanol (3OH), and 4B represents silver nanoparticles modified with 4-mercaptophenylboronic acid. The values ​​in the last column of the table represent the model accuracy when multiple substrates are used. The multiple substrates used are the multiple substrates corresponding to the shading in each row.

[0115] Step S304: Obtain the single-dimensional surface-enhanced Raman scattering bacterial spectrum of the bacteria to be detected, and proceed to step S305.

[0116] like Figure 16 As shown, in one embodiment, obtaining the single-dimensional surface-enhanced Raman scattering bacterial spectrum of the bacteria to be detected includes the following steps S401 to S404:

[0117] Step S401: Obtain the bacterial suspension to be tested, and proceed to step S402.

[0118] In application, bacteria to be tested can be collected on-site where they need to be detected, and then a suspension of the bacteria to be tested can be prepared.

[0119] Step S402: Use ultrasound to lyse the bacterial suspension to be tested to obtain the bacterial lysate to be tested, and proceed to step S403.

[0120] In applications, an ultrasonic cell disruptor can be used to lyse bacteria using ultrasound to obtain bacterial lysate.

[0121] Step S403: Mix the bacterial lysate to be tested with any pre-prepared surface-enhanced Raman scattering concentrated suspension substrate to form a liquid film, and proceed to step S404.

[0122] Step S404: Perform surface-enhanced Raman scattering analysis on the liquid film to obtain a single-dimensional surface-enhanced Raman scattering bacterial spectrum.

[0123] In application, the specific implementation steps of steps S403 and S404 can be referred to the relevant descriptions of steps S102, S103 and S104, and will not be repeated here.

[0124] Step S305: Input the single-dimensional surface-enhanced Raman scattering data of the bacteria to be detected into the trained deep learning model, and the trained deep learning model outputs the species and subtype of the bacteria to be detected.

[0125] In the application, the single-dimensional surface-enhanced Raman scattering data of the bacteria to be detected is input into the trained deep learning model. The trained deep learning model can output the specific species of the bacteria to be detected with an accuracy of subtype level.

[0126] This application also provides a bacterial identification device for performing the steps described in the bacterial identification method embodiments above. This device can be a virtual appliance within a bacterial identification device, operated by the processor of the bacterial identification device, or it can be the bacterial identification device itself.

[0127] like Figure 17 As shown, the bacterial identification device 100 provided in this application embodiment includes:

[0128] The database construction module 101 is used to construct a multi-dimensional surface-enhanced Raman scattering bacterial spectral database using the multi-dimensional surface-enhanced Raman scattering bacterial spectral data, and then enters the model building module 102.

[0129] Model building module 102 is used to build deep learning models, then proceed to model training module 103;

[0130] The model training module 103 is used to train the deep learning model using the multi-dimensional bacterial surface-enhanced scattering spectrum database, obtain the trained deep learning model, and then enter the data acquisition module 104.

[0131] The data acquisition module 104 is used to acquire the single-dimensional surface-enhanced Raman scattering bacterial spectrum of the bacteria to be detected, and then enter the bacterial identification module 105.

[0132] The bacterial identification module 105 is used to input the single-dimensional surface-enhanced Raman scattering data of the bacteria to be detected into the trained deep learning model, and the trained deep learning model outputs the species and subtype of the bacteria to be detected.

[0133] In applications, each unit in the above-mentioned device can be a software program module, or it can be implemented by different logic circuits integrated in the processor or by independent physical components connected to the processor, or it can be implemented by multiple distributed processors.

[0134] like Figure 18 As shown, this application embodiment also provides a terminal device 200, including: at least one processor 201 (only one processor is shown in the figure), a memory 202, and a computer program 203 stored in the memory 202 and executable on the at least one processor 201. When the processor 201 executes the computer program 203, it implements the steps in any of the above-described bacterial identification method embodiments.

[0135] In applications, bacterial identification devices may include, but are not limited to, processors and memory. Those skilled in the art will understand that... Figure 18 This is merely an example of a bacterial identification device and does not constitute a limitation on bacterial identification devices. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, it may also include input / output devices, network access devices, etc.

[0136] In applications, the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0137] In applications, the memory may be an internal storage module of the bacterial identification device in some embodiments, such as a hard drive or memory of the bacterial identification device. In other embodiments, the memory may be an external storage device of the bacterial identification device, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the bacterial identification device. Furthermore, the memory may include both internal and external storage modules of the bacterial identification device. The memory is used to store the operating system, applications, boot loader, data, and other programs, such as the program code of the computer program. The memory can also be used to temporarily store data that has been output or will be output.

[0138] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0139] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0140] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0141] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the above-described bacterial identification method embodiments.

[0142] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various bacterial identification method embodiments above.

[0143] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographic device / testing device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0144] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0145] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0146] In the embodiments provided in this application, it should be understood that the disclosed apparatus / network devices and methods can be implemented in other ways. For example, the apparatus / network device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0147] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0148] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A bacterial identification method based on multidimensional bacterial spectra, characterized in that, include: The bacterial suspension was lysed using ultrasound to obtain bacterial lysate; The preparation of various surface-enhanced Raman scattering (SRF) concentrated suspension substrates specifically includes: obtaining noble metal nanomaterials of different types or structures; and modifying the noble metal nanomaterials of different types or structures using various self-assembled monolayers to obtain various surface-enhanced Raman scattering (SRF) concentrated suspension substrates. Various surface-enhanced Raman scattering concentrated suspension substrates and the bacterial lysate were mixed to obtain various mixtures; A liquid film is formed using the aforementioned mixtures; Surface-enhanced Raman scattering (SERS) analysis is performed on the liquid film to obtain multi-dimensional SERS bacterial spectral data. A multi-dimensional SERS bacterial spectral database is constructed by summarizing the multi-dimensional SERS bacterial spectral data of various bacteria. Spectral data composed of at least two different substrates are obtained from the multi-dimensional SERS bacterial spectral database and used as a dataset for training a deep learning model. The trained deep learning model includes a one-dimensional convolutional structure and a multilayer perceptron structure. The one-dimensional convolutional structure is used to extract bacterial spectral features; the multilayer perceptron structure is used to classify bacteria. The multilayer perceptron structure includes convolutional layers, activation layers, one-dimensional pooling layers, dropout layers, and flattening layers. It comprises two perceptron layers and two activation layers. The number of kernels in the two input convolutional layers of the one-dimensional convolutional structure is the same as the dimension of the input spectral data, used to learn multi-dimensional spectral features. The kernels of the remaining convolutional layers in the one-dimensional convolutional structure are 3x3. The flattening layer is placed at the last layer of the one-dimensional convolutional structure to expand high-dimensional features along one dimension. The activation layer using the Softmax activation function is placed at the last layer of the multilayer perceptron structure to perform bacterial classification. Dropout layers are added to both the one-dimensional convolutional structure and the multilayer perceptron structure. This allows for the acquisition of a single-dimensional surface-enhanced Raman scattering (SERS) spectrum of the bacteria to be detected. The SERS data of the bacteria to be detected is then input into a trained deep learning model, which outputs the species and subtype of the bacteria to be detected.

2. The bacterial identification method based on multidimensional bacterial spectra as described in claim 1, characterized in that, The bacterial suspension includes a suspension of at least one of the following bacteria: Shigella flexneri, Shigella dysenteriae, Staphylococcus aureus, Bacillus cereus, Bacillus thuringiensis, Escherichia coli, and Bacillus faecalis.

3. The bacterial identification method based on multidimensional bacterial spectra as described in claim 1, characterized in that, The acquisition of the single-dimensional surface-enhanced Raman scattering bacterial spectrum of the bacteria to be detected includes: Obtain the bacterial suspension to be tested; The bacterial suspension to be tested was lysed using ultrasound to obtain a lysate of the bacterial strain to be tested. The bacterial lysate to be tested is mixed with any pre-prepared surface-enhanced Raman scattering concentrated suspension substrate to form a liquid film; Surface-enhanced Raman scattering analysis was performed on the liquid film to obtain a single-dimensional surface-enhanced Raman scattering bacterial spectrum.

4. A bacterial identification device, characterized in that, include: The database construction module is used to construct a multi-dimensional surface-enhanced Raman scattering (SMR) bacterial spectral database using multi-dimensional SMR bacterial spectral data. The multi-dimensional SMR bacterial spectral data is obtained by the following method: using ultrasonic lysis of bacterial suspension to obtain bacterial lysate. The preparation of various surface-enhanced Raman scattering (SERS) concentrated suspension substrates specifically includes: obtaining noble metal nanomaterials of different types or structures; modifying the noble metal nanomaterials of different types or structures using various self-assembled monolayers to obtain various SERS concentrated suspension substrates; mixing the various SERS concentrated suspension substrates and the bacterial lysate to obtain various mixtures; forming a liquid film using the various mixtures; and performing SERS analysis on the liquid film to obtain multi-dimensional SERS bacterial spectral data. A model building module is used to build a deep learning model, which includes a one-dimensional convolutional structure and a multilayer perceptron structure. The one-dimensional convolutional structure is used to extract bacterial spectral features; the multilayer perceptron structure is used to classify bacteria. The one-dimensional convolutional structure includes convolutional layers, activation layers, one-dimensional pooling layers, dropout layers, and flattening layers. The multilayer perceptron structure includes two perceptron layers and two activation layers. The number of convolutional kernels in the two input layers of the one-dimensional convolutional structure is the same as the dimension of the input spectral data, used to learn multi-dimensional spectral features. The kernels of the remaining convolutional layers of the one-dimensional convolutional structure are 3*3. The flattening layer is placed at the last layer of the one-dimensional convolutional structure to expand high-dimensional features along one dimension. The activation layer using the Softmax activation function is placed at the last layer of the multilayer perceptron structure to complete bacterial classification. Dropout layers are added to both the one-dimensional convolutional structure and the multilayer perceptron structure. The model training module is used to train the deep learning model using the multi-dimensional surface-enhanced Raman scattering bacterial spectral database to obtain the trained deep learning model. The data acquisition module is used to acquire the single-dimensional surface-enhanced Raman scattering bacterial spectrum of the bacteria to be detected; The bacterial identification module is used to input the single-dimensional surface-enhanced Raman scattering data of the bacteria to be detected into the trained deep learning model, and the trained deep learning model outputs the species and subtype of the bacteria to be detected. The database construction module is specifically used for: By summarizing multidimensional surface-enhanced Raman scattering (SERS) bacterial spectral data of various bacteria, a multidimensional SERS bacterial spectral database was constructed. The model training module is specifically used for: Spectral data composed of at least two or more different substrates are obtained from the multidimensional surface-enhanced Raman scattering bacterial spectral database and used as the dataset for training the deep learning model to obtain the trained deep learning model.

5. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the bacterial identification method as described in claim 1 or 3.

6. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the bacterial identification method as described in claim 1 or 3.

Images