Intelligent detection method for dairy cow mastitis pathogenic bacteria AI

Abstract

The invention relates to the technical field of pathogenic bacteria detection, and discloses a dairy cow mastitis pathogenic bacteria AI intelligent detection device, which comprises a sample collection device, a sample pretreatment module, an AI detection module, an AI detection module, an AI detection module, an AI detection module, an AI detection module and an AI detection module, the centrifugal operation module is used for carrying out centrifugal operation on the collected sample to remove supernate, adding sterile normal saline, resuspending and oscillating to obtain a pathogenic bacterium sample, and the data collection and feature extraction module is used for injecting the preprocessed sample into the detection device. According to the method, learning is carried out by using a large amount of known pathogenic bacteria sample data, so that the model gradually and accurately captures the internal relation between the feature combination and the pathogenic bacteria type, and a reliable mapping relation is established, so that the model can accurately map the pathogenic bacteria type to which a new unknown sample belongs when facing the new unknown sample; powerful support is provided for accurate detection of the dairy cow mastitis pathogenic bacteria, and the detection accuracy and efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of pathogen detection, and in particular to an AI intelligent detection method for bovine mastitis pathogens. Background Art

[0002] Cow mastitis is a common and serious disease in the dairy farming industry. First of all, it has a serious impact on the health of dairy cows, causing inflammation that leads to congestion and edema of mammary tissue, destroying the structure and function of mammary cells, reducing the milk production of dairy cows, and consuming the energy and nutrition of dairy cows, reducing immunity, and causing a variety of complications. At the same time, it causes pain to dairy cows and changes their behavior patterns, providing a basis for targeted treatment, helping to improve treatment effects, reduce treatment costs and time. Through detection, mastitis can be detected early, and preventive measures can be taken to reduce the incidence rate. With the development of artificial intelligence technology, compared with the traditional microscopic observation method of pathogen detection, the use of AI intelligent detection methods, through advanced data processing, model construction and analysis technology, can achieve faster, more accurate and efficient detection of pathogens of cow mastitis, providing strong support for dairy cow health management and the development of the farming industry.

[0003] In current technology, the traditional method of pathogen detection through microscopic observation has a long detection cycle, low accuracy, high technical requirements for operators, and difficulty in coping with the diversity and complexity of pathogens, which affects the accuracy and efficiency of detection. Summary of the invention

[0004] In order to make up for the above shortcomings, the present invention provides an AI intelligent detection method for bovine mastitis pathogens, aiming to improve the problems that the detection has high technical requirements for operators and a long detection cycle, which affects the accuracy and efficiency of the detection.

[0005] In the first aspect, the present invention provides the following technical solution: an AI intelligent detection device for bovine mastitis pathogens, comprising:

[0006] The sample collection device is used to collect milk samples after cleaning and disinfecting the cow's nipples, and mark them to provide the original data source for subsequent testing;

[0007] The sample pretreatment module is used to centrifuge the collected samples to remove the supernatant, add sterile physiological saline to resuspend and oscillate, and obtain pathogen samples;

[0008] The data acquisition and feature extraction module is used to inject the pre-processed samples into the detection device, use multiple sensors to detect the samples at the same time, and pre-process the electrical signals obtained by the detection and extract features;

[0009] The model training and mapping module is used to establish a convolutional neural network model and a training model, and learn the corresponding relationship between feature combinations and pathogen types during the training process to establish a pathogen type mapping;

[0010] The pathogen detection and classification module is used to predict the results of the data set after sample feature extraction through a convolutional neural network model, and map the prediction results to pathogen categories based on the sample data characteristics;

[0011] The result reporting and feedback module is used to send the analysis results to the interactive Web application, calculate the ranking value of the detected pathogens through the built-in intelligent ranking algorithm, allow manual review through the interactive interface, and publish the bovine mastitis pathogen detection report. Finally, it provides report download and API interface to send report data to the third-party system.

[0012] Preferably, the sample collection device includes a sterile sampling device and a sample marking device, wherein the sterile sampling device is used to collect and store milk samples collected from the teats of cows, and the sample marking device is used to mark the collected milk samples, and the sample mark includes the cow number, sampling time and teat position.

[0013] Preferably, the sample pretreatment module includes a centrifugal unit, a supernatant removal unit and a resuspension unit, wherein the centrifugal unit is used to generate centrifugal force through high-speed rotation to stratify different components in the milk sample according to density differences, the supernatant removal unit is used to remove the supernatant of the sample after centrifugation, and the resuspension unit is used to disperse pathogens in the resuspension through mechanical vibration.

[0014] Preferably, the data acquisition and feature extraction module includes a multi-sensor detection unit and a signal conversion and extraction unit. The sensor detection unit is used to synchronously detect samples using a variety of different types of sensors to obtain physical and chemical property data of the samples in multiple dimensions, and the multiple different types include optical sensors and electrochemical sensors. The signal conversion and extraction unit is used to convert various physical and chemical signals collected by the sensors into a unified electrical signal form, and extract valuable feature information related to pathogens from the signal data.

[0015] Preferably, the model training and mapping module includes a model construction unit and a model training unit, the model construction unit is used to construct a convolutional neural network, and the model training unit is used to input the prepared training data into the constructed convolutional neural network model, perform forward propagation calculations to obtain output results, and after multiple iterations, allow the model to learn the features and mapping relationships in the data.

[0016] Preferably, the pathogen detection and classification module includes a model prediction unit and a classification decision unit. The model prediction unit is used to input the preprocessed sample feature data set into a trained convolutional neural network model, and obtain the predicted probability distribution of the samples belonging to different pathogen categories through the forward propagation calculation of the model. The classification decision unit is used to map the prediction result into a clear pathogen category based on the predicted probability distribution obtained by the model prediction unit.

[0017] Preferably, the result report and feedback module includes an interactive Web application module, a report generation unit and a report publishing and sharing unit. The interactive Web application module is used to sort and output the analysis results obtained by the pathogen detection and classification module. The report generation unit is used to generate a formal bovine mastitis pathogen detection report based on the audited pathogen detection results and related information. The report publishing and sharing unit is used to publish the generated bovine mastitis pathogen detection report to a designated platform or system, and provide multiple ways for users to obtain reports, and send report data to a third-party system through an API interface.

[0018] In a second aspect, the present invention provides the following technical solution, an AI intelligent detection method for bovine mastitis pathogens, comprising the following steps:

[0019] S1, sample collection: select the sampling cow's nipple through a special sterile sampling device, clean and disinfect it, collect the milk, quickly put it into a sterile container, and mark it;

[0020] S2, sample pretreatment: centrifuge the sample, discard the supernatant with a pipette, then add sterile saline resuspension solution and oscillate to obtain a pathogen sample;

[0021] S3, data acquisition and feature extraction: inject the pre-processed samples into the sample slot of the detection device, use multiple sensors to detect the samples at the same time, then pre-process the electrical signals obtained by the detection, and then use the data analysis algorithm to extract features from the pre-processed data to obtain a data set;

[0022] S4, model training and mapping: Establish a convolutional neural network model, learn a large amount of data from known pathogen samples, and use the back propagation algorithm to update the weights of the fully connected layer based on the difference between the predicted results and the real sample labels. During the training process, learn which pathogen type the feature combination corresponds to, so as to establish a stable mapping relationship;

[0023] S5, pathogen detection and classification: input the data set in S3 into the learned convolutional neural network model, calculate the prediction results through forward propagation, and then map the prediction results to pathogen categories based on the sample data characteristics;

[0024] S6, result report and feedback: the analysis results are sent to the interactive Web application module, and the ranking value of the detected pathogens is calculated by the built-in intelligent ranking algorithm. After manual review, the bovine mastitis pathogens are released and a test report is generated. The report is available for download or the report data is sent to a third-party system through the API interface.

[0025] In the third aspect, the invention provides the following technical solution: a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned AI intelligent detection method for bovine mastitis pathogens when executing the computer program.

[0026] In a fourth aspect, the present invention provides the following technical solution: a readable storage medium having a computer program stored thereon, and the computer program, when executed by a processor, implements the above-mentioned AI intelligent detection method for bovine mastitis pathogens.

[0027] The present invention has the following beneficial effects:

[0028] 1. In the present invention, by using a large amount of known pathogen sample data for learning, the model gradually and accurately captures the intrinsic connection between the feature combination and the pathogen type, thereby establishing a reliable mapping relationship, so that when the model faces a new unknown sample, it can accurately map the pathogen type to which it belongs, providing strong support for the precise detection of bovine mastitis pathogens, and improving the accuracy and efficiency of detection.

[0029] 2. In the present invention, by sending the analysis results to the interactive Web application module and performing intelligent sorting, users can quickly focus on key information, manual review ensures the accuracy and reliability of the results, generates test reports and provides a variety of report output methods, which is convenient for users to obtain detailed information for decision-making and archiving. At the same time, data is shared with third-party systems through API interfaces, which promotes collaborative cooperation in multiple fields and realizes efficient result processing and extensive information sharing.

[0030] 3. In the present invention, different components in the milk sample are stratified according to density differences through centrifugation, and impurities are effectively separated by removing the supernatant. Sterile physiological saline is added for resuspending and oscillation to evenly disperse the pathogens, thereby obtaining a relatively pure and evenly distributed pathogen sample. The effect and function of the present invention is to provide a high-quality sample basis for subsequent detection, high detection accuracy and stability, and ensure that the detection results can more truly and accurately reflect the situation of bovine mastitis pathogens. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a framework diagram of the AI ​​intelligent detection device for bovine mastitis pathogens proposed by the present invention;

[0032] Figure 2 This is a sample collection device architecture diagram of the AI ​​intelligent detection device for bovine mastitis pathogens proposed by the present invention;

[0033] Figure 3 This is a sample preprocessing module architecture diagram of the AI ​​intelligent detection device for bovine mastitis pathogens proposed by the present invention;

[0034] Figure 4 This is a data collection and feature extraction module architecture diagram of the AI ​​intelligent detection device for bovine mastitis pathogens proposed by the present invention;

[0035] Figure 5 This is a model training and mapping module architecture diagram of the AI ​​intelligent detection device for bovine mastitis pathogens proposed by the present invention;

[0036] Figure 6 This is a diagram of the pathogen detection and classification module architecture of the AI ​​intelligent detection device for bovine mastitis pathogens proposed by the present invention;

[0037] Figure 7 This is a result report and feedback module architecture diagram of the AI ​​intelligent detection device for bovine mastitis pathogens proposed by the present invention;

[0038] Figure 8 This is a flow chart of the AI ​​intelligent detection method for bovine mastitis pathogens proposed in the present invention. DETAILED DESCRIPTION

[0039] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] Embodiment 1

[0041] Reference Figure 1 In the first embodiment of the present invention, the present invention provides an AI intelligent detection device for bovine mastitis pathogens, comprising:

[0042] The sample collection device is used to collect milk samples after cleaning and disinfecting the cow's nipples, and mark them to provide the original data source for subsequent testing;

[0043] The sample pretreatment module is used to centrifuge the collected samples to remove the supernatant, add sterile physiological saline to resuspend and oscillate, and obtain pathogen samples;

[0044] The data acquisition and feature extraction module is used to inject the pre-processed samples into the detection device, use multiple sensors to detect the samples at the same time, and pre-process the electrical signals obtained by the detection and extract features;

[0045] The model training and mapping module is used to establish a convolutional neural network model and a training model, and learn the corresponding relationship between feature combinations and pathogen types during the training process to establish a pathogen type mapping;

[0046] The pathogen detection and classification module is used to predict the results of the data set after sample feature extraction through a convolutional neural network model, and map the prediction results to pathogen categories based on the sample data characteristics;

[0047] The result reporting and feedback module is used to send the analysis results to the interactive Web application, calculate the ranking value of the detected pathogens through the built-in intelligent ranking algorithm, allow manual review through the interactive interface, and publish the bovine mastitis pathogen detection report. Finally, it provides report download and API interface to send report data to the third-party system.

[0048] Specifically, the sample collection device collects milk samples from the cow's nipples, providing raw materials for subsequent analysis, cleaning and disinfecting the cow's nipples, reducing the possibility of external microorganisms mixing into the samples, ensuring the purity of the samples, and ensuring that the collected samples are representative and reliable, laying the foundation for the subsequent accurate detection of bovine mastitis pathogens; the centrifugal operation of the sample pretreatment module can separate the different components in the milk according to the density difference of the materials, remove the supernatant containing impurities such as fat and soluble protein, enrich the pathogens, and add sterile saline to resuspend the precipitate, so that the pathogens are evenly dispersed in the solution, and a relatively pure sample with evenly distributed pathogens is obtained, which reduces the interference of impurities on subsequent detection and improves the accuracy and stability of detection; through the multi-sensor detection in the data acquisition and feature extraction module, rich and comprehensive sample data is obtained, which can be more Accurately reflect the existence, type and status of pathogens; establish and train a convolutional neural network model suitable for processing bovine mastitis pathogen detection data through the model training and mapping module, accurately establish a mapping between feature combinations and pathogen types, and accurately predict the category of pathogens based on their characteristics; input the feature-extracted sample data set into the trained convolutional neural network model through the pathogen detection and classification module and map the prediction results to clear pathogen categories, providing key basis for the diagnosis and treatment of bovine mastitis and helping farmers to take targeted measures in a timely manner; the result report and feedback module can present the test results to users in an orderly and accurate manner through interactive Web applications, realize the sharing and collaboration of test data between different systems and users, and is conducive to integrating resources from all parties and improving the management level of dairy farming and disease prevention and control capabilities.

[0049] Reference Figure 2 The sample collection device includes a sterile sampling device and a sample marking device. The sterile sampling device is used to collect and store milk samples collected from the teats of cows. The sample marking device is used to mark the collected milk samples. The sample mark includes the cow number, sampling time and teat position.

[0050] Specifically, sterile sampling instruments can be used to prevent contamination from external microorganisms during the collection and storage of milk samples, so that microorganisms in the environment around the cow's nipples may not be misjudged as pathogens or cover up the true situation of pathogens; sample marking tools can facilitate the classification, storage and retrieval of samples.

[0051] Reference Figure 3 The sample pretreatment module includes a centrifugal unit, a supernatant removal unit and a resuspension unit. The centrifugal unit is used to generate centrifugal force through high-speed rotation to stratify different components in the milk sample according to density differences. The supernatant removal unit is used to remove the supernatant of the sample after centrifugation. The resuspension unit is used to disperse the pathogens in the resuspension through mechanical vibration.

[0052] Specifically, the centrifugal unit can be used to stratify the various components in the milk sample according to the density difference, so that the pathogens are relatively enriched at the bottom of the centrifuge tube, which helps to improve the sensitivity and accuracy of the detection; the supernatant removal unit can remove a large amount of impurities in the supernatant that are separated from the pathogens during the centrifugation process; the mechanical vibration of the resuspension unit can make the pathogens in the precipitate evenly dispersed in the resuspension, ensuring that the content of pathogens in each unit volume of the resuspension is relatively uniform, so that the sensor can more accurately reflect the true state of the pathogens.

[0053] Reference Figure 4 The data acquisition and feature extraction module includes a multi-sensor detection unit and a signal conversion and extraction unit. The sensor detection unit is used to synchronously detect samples using a variety of different types of sensors to obtain physical and chemical property data of samples in multiple dimensions. The multiple different types include optical sensors and electrochemical sensors. The signal conversion and extraction unit is used to convert various physical and chemical signals collected by the sensor into a unified electrical signal form, and extract valuable feature information related to pathogens from the signal data.

[0054] Specifically, rich information about the sample is obtained through the multi-sensor detection unit, including optical properties (absorbance, transmittance, color, fluorescence, etc.) and electrochemical properties (potential, current, conductivity, etc.), which can be combined to more comprehensively reflect the characteristics of the pathogens in the sample, avoiding the limitations that may exist in a single sensor; after the signal is converted into a unified electrical signal form by the signal conversion and extraction unit, feature extraction is performed, which can focus on key information related to the pathogens, remove irrelevant background noise and redundant data, and make subsequent analysis of the pathogens more accurate.

[0055] Reference Figure 5 The model training and mapping module includes a model construction unit and a model training unit. The model construction unit is used to construct a convolutional neural network. The model training unit is used to input the prepared training data into the constructed convolutional neural network model, perform forward propagation calculation to obtain the output result, and after multiple iterations, the model learns the features and mapping relationships in the data.

[0056] Specifically, the model building unit can be closely integrated with the characteristics of the data and the detection objectives, providing an infrastructure for subsequent training and accurate classification; the model parameters are gradually adjusted to the optimal values ​​through the model training unit, which enables the model to better fit the training data, accurately extract features and establish mapping relationships with various types of pathogens.

[0057] Reference Figure 6 The pathogen detection and classification module includes a model prediction unit and a classification decision unit. The model prediction unit is used to input the preprocessed sample feature data set into the trained convolutional neural network model, and obtain the predicted probability distribution of the sample belonging to different pathogen categories through the forward propagation calculation of the model. The classification decision unit is used to map the prediction result into a clear pathogen category based on the predicted probability distribution obtained by the model prediction unit.

[0058] Specifically, through in-depth analysis of the input sample features by the model prediction unit and the feature extraction and mapping capabilities of the CNN model, the probability distribution of samples belonging to different pathogens is generated to achieve prediction of unknown samples; the classification decision unit is used to judge the probability distribution predicted by the model according to the set classification threshold and category mapping rules, and the samples are mapped to the corresponding pathogen categories. The probability distribution predicted by the model is converted into a clear pathogen category, eliminating the ambiguity of the results and providing an accurate basis for the targeted treatment and prevention of bovine mastitis.

[0059] Reference Figure 7The result report and feedback module includes an interactive Web application module, a report generation unit and a report publishing and sharing unit. The interactive Web application module is used to sort and output the analysis results obtained by the pathogen detection and classification module. The report generation unit is used to generate a formal bovine mastitis pathogen detection report according to the audited pathogen detection results and related information. The report publishing and sharing unit is used to publish the generated bovine mastitis pathogen detection report to a designated platform or system, and provide multiple ways for users to obtain reports, and send report data to a third-party system through an API interface.

[0060] Specifically, the analysis results from pathogen detection are received through the interactive Web application module, and the data are collected and sorted. The sorted results are displayed through an intuitive and easy-to-understand user interface, so that users can further review the results on the interface; the reviewed pathogen detection results and related information are accurately filled into the report template through the report generation unit, which improves the professionalism and credibility of the report; the report publishing and sharing unit selects a suitable publishing platform and provides a convenient download function, so that the bovine mastitis pathogen detection report can be widely disseminated to users in need. Both internal farm personnel and external cooperative organizations can obtain reports in a timely manner, which promotes the circulation and sharing of information.

[0061] Reference Figure 8 In a second embodiment of the present invention, the present invention provides an AI intelligent detection method for bovine mastitis pathogens, comprising the following steps:

[0062] S1, sample collection: select the sampling cow's nipple through a special sterile sampling device, clean and disinfect it, collect the milk, quickly put it into a sterile container, and mark it;

[0063] S2, sample pretreatment: centrifuge the sample, discard the supernatant with a pipette, then add sterile saline resuspension solution and oscillate to obtain a pathogen sample;

[0064] S3, data acquisition and feature extraction: inject the pre-processed samples into the sample slot of the detection device, use multiple sensors to detect the samples at the same time, then pre-process the electrical signals obtained by the detection, and then use the data analysis algorithm to extract features from the pre-processed data to obtain a data set;

[0065] S4, model training and mapping: Establish a convolutional neural network model, learn a large amount of data from known pathogen samples, and use the back propagation algorithm to update the weights of the fully connected layer based on the difference between the predicted results and the real sample labels. During the training process, learn which pathogen type the feature combination corresponds to, so as to establish a stable mapping relationship;

[0066] S5, pathogen detection and classification: input the data set in S3 into the learned convolutional neural network model, calculate the prediction results through forward propagation, and then map the prediction results to pathogen categories based on the sample data characteristics;

[0067] S6, result report and feedback: the analysis results are sent to the interactive Web application module, and the ranking value of the detected pathogens is calculated by the built-in intelligent ranking algorithm. After manual review, the bovine mastitis pathogens are released and a test report is generated. The report is available for download or the report data is sent to a third-party system through the API interface.

[0068] Embodiment 3

[0069] The third embodiment of the present invention is based on the same inventive concept. The present invention proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the AI ​​intelligent detection method for bovine mastitis pathogens of the above embodiment are implemented.

[0070] Embodiment 4

[0071] The fourth embodiment of the present invention is based on the same inventive concept. The present invention proposes a computer, which includes: a processor and a memory; the processor and the memory communicate with each other; the memory is used to store instructions; the processor is used to execute the instructions in the memory to execute the AI ​​intelligent detection method for bovine mastitis pathogens of the above embodiment.

[0072] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0073] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. AI intelligent detection device for pathogens of mastitis in dairy cows, characterized in that: include: The sample collection device is used to collect milk samples after cleaning and disinfecting the cow's nipples, and mark them to provide the original data source for subsequent testing; The sample pretreatment module is used to centrifuge the collected samples to remove the supernatant, add sterile physiological saline to resuspend and oscillate, and obtain pathogen samples; The data acquisition and feature extraction module is used to inject the pre-processed samples into the detection device, use multiple sensors to detect the samples at the same time, and pre-process the electrical signals obtained by the detection and extract features; The model training and mapping module is used to establish a convolutional neural network model and a training model, and learn the corresponding relationship between feature combinations and pathogen types during the training process to establish a pathogen type mapping; The pathogen detection and classification module is used to predict the results of the data set after sample feature extraction through a convolutional neural network model, and map the prediction results to pathogen categories based on the sample data characteristics; The result reporting and feedback module is used to send the analysis results to the interactive Web application, calculate the ranking value of the detected pathogens through the built-in intelligent ranking algorithm, allow manual review through the interactive interface, and publish the bovine mastitis pathogen detection report. Finally, it provides report download and API interface to send report data to the third-party system.

2. The AI ​​intelligent detection device for bovine mastitis pathogens according to claim 1 is characterized in that: The sample collection device comprises a sterile sampling device and a sample marking device. The sterile sampling device is used to collect and store milk samples collected from the teats of cows. The sample marking device is used to mark the collected milk samples. The sample marking includes the cow number, sampling time and teat position.

3. The AI ​​intelligent detection device for bovine mastitis pathogens according to claim 1 is characterized in that: The sample pretreatment module includes a centrifugal unit, a supernatant removal unit and a resuspension unit. The centrifugal unit is used to generate centrifugal force through high-speed rotation to stratify different components in the milk sample according to density differences. The supernatant removal unit is used to remove the supernatant of the sample after centrifugation. The resuspension unit is used to disperse pathogens in the resuspension through mechanical vibration.

4. The AI ​​intelligent detection device for bovine mastitis pathogens according to claim 1 is characterized in that: The data acquisition and feature extraction module includes a multi-sensor detection unit and a signal conversion and extraction unit. The sensor detection unit is used to use a variety of different types of sensors to synchronously detect samples and obtain physical and chemical property data of samples in multiple dimensions. The multiple different types include optical sensors and electrochemical sensors. The signal conversion and extraction unit is used to convert various physical and chemical signals collected by the sensors into a unified electrical signal form, and extract valuable feature information related to pathogens from the signal data.

5. The AI ​​intelligent detection device for bovine mastitis pathogens according to claim 1, characterized in that: The model training and mapping module includes a model construction unit and a model training unit. The model construction unit is used to construct a convolutional neural network. The model training unit is used to input the prepared training data into the constructed convolutional neural network model, perform forward propagation calculations to obtain output results, and after multiple iterations, allow the model to learn the features and mapping relationships in the data.

6. The AI ​​intelligent detection device for bovine mastitis pathogens according to claim 1, characterized in that: The pathogen detection and classification module includes a model prediction unit and a classification decision unit. The model prediction unit is used to input the preprocessed sample feature data set into a trained convolutional neural network model, and obtain the predicted probability distribution of the sample belonging to different pathogen categories through the forward propagation calculation of the model. The classification decision unit is used to map the prediction result into a clear pathogen category based on the predicted probability distribution obtained by the model prediction unit.

7. The AI ​​intelligent detection device for bovine mastitis pathogens according to claim 1, characterized in that: The result report and feedback module includes an interactive Web application module, a report generation unit and a report publishing and sharing unit. The interactive Web application module is used to sort and output the analysis results obtained by the pathogen detection and classification module. The report generation unit is used to generate a formal bovine mastitis pathogen detection report based on the audited pathogen detection results and related information. The report publishing and sharing unit is used to publish the generated bovine mastitis pathogen detection report to a designated platform or system, and provide multiple ways for users to obtain reports, and send report data to a third-party system through an API interface.

8. AI intelligent detection method for pathogens of mastitis in dairy cows, characterized in that: The AI ​​intelligent detection device for bovine mastitis pathogens according to any one of claims 1 to 7 comprises the following steps: S1, sample collection: select the sampling cow's nipple through a special sterile sampling device, clean and disinfect it, collect the milk, quickly put it into a sterile container, and mark it; S2, sample pretreatment: centrifuge the sample, discard the supernatant with a pipette, then add sterile saline resuspension solution and oscillate to obtain a pathogen sample; S3, data acquisition and feature extraction: inject the pre-processed samples into the sample slot of the detection device, use multiple sensors to detect the samples at the same time, then pre-process the electrical signals obtained by the detection, and then use the data analysis algorithm to extract features from the pre-processed data to obtain a data set; S4, model training and mapping: Establish a convolutional neural network model, learn a large amount of data from known pathogen samples, and use the back propagation algorithm to update the weights of the fully connected layer based on the difference between the predicted results and the real sample labels. During the training process, learn which pathogen type the feature combination corresponds to, so as to establish a stable mapping relationship; S5, pathogen detection and classification: input the data set in S3 into the learned convolutional neural network model, calculate the prediction results through forward propagation, and then map the prediction results to pathogen categories based on the sample data characteristics; S6, result report and feedback: the analysis results are sent to the interactive Web application module, and the ranking value of the detected pathogens is calculated by the built-in intelligent ranking algorithm. After manual review, the bovine mastitis pathogens are released and a test report is generated. The report is available for download or the report data is sent to a third-party system through the API interface.

9. A computer 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, the AI ​​intelligent detection method for bovine mastitis pathogens as described in claim 8 is implemented.

10. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by the processor, the AI ​​intelligent detection method for bovine mastitis pathogens as claimed in claim 8 is implemented.

Images