Multi-dimensional correlation analysis method and system for pathogenic bacteria in aquatic products

Abstract

The invention provides a multi-dimensional correlation analysis method and system for pathogenic bacteria in aquatic products, and relates to the technical field of pathogenic bacteria detection.The method comprises the steps that aquatic product sample information and detection information are collected, an aquatic product pathogenic bacteria basic data set is constructed, structural arrangement is conducted, and a multi-dimensional correlation data table is formed; the method comprises the following steps: extracting co-occurrence records, occurrence frequency and concentration change corresponding relations of pathogenic bacteria in different sources, collection time intervals and storage and transportation conditions, constructing a pathogenic bacteria co-occurrence relation matrix, and forming a pathogenic bacteria multi-dimensional incidence relation graph; and according to the co-occurrence relation matrix and the multi-dimensional association relation graph, carrying out association determination on the distribution states of pathogenic bacteria in different batches of aquatic products, identifying a pathogenic bacteria propagation path and a high-risk association combination, generating a pathogenic bacteria multi-dimensional association analysis result, and outputting pollution source prompt information and risk level information. Therefore, identification and risk early warning of potential pollution sources of pathogenic bacteria in aquatic products are realized.

Description

Technical Field

[0001] This invention relates to the field of pathogen detection technology, and in particular to a multidimensional correlation analysis method and system for pathogenic bacteria in aquatic products. Background Technology

[0002] Currently, in the field of pathogen detection and analysis in aquatic products, common technical solutions are mainly based on microbial culture combined with molecular biological detection methods. Specifically, aquatic product samples are first pretreated and enriched to increase the number of potentially pathogenic bacteria. Then, they are isolated and cultured using selective media, followed by detection and confirmation of target pathogenic bacteria using techniques such as biochemical identification, PCR amplification, or real-time quantitative PCR. Simultaneously, some studies also combine the detection results with simple statistical analysis of different sample sources, detection times, and bacterial species to assess the distribution and contamination levels of common pathogenic bacteria in aquatic products, thereby achieving basic monitoring of the safety status of aquatic products.

[0003] However, in actual aquatic product circulation or batch quality testing scenarios, the aforementioned technologies typically only target and statistically analyze single indicators or single bacterial species. For example, during batch sampling of aquatic products, testing personnel often independently test for different pathogenic bacteria such as Vibrio parahaemolyticus and Salmonella, and make simple statistical judgments based on single test results. This approach makes it difficult to comprehensively analyze the correlation between multi-dimensional data such as "testing time - production environment - bacterial species - degree of contamination." When a batch of aquatic products is cross-contaminated at different times or from different sources, existing methods struggle to identify potential correlation patterns in a timely manner, easily leading to inaccurate determination of the source of contamination, thereby affecting subsequent risk assessments and regulatory decisions. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for multidimensional correlation analysis of pathogenic bacteria in aquatic products, aiming to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: Firstly, a multidimensional association analysis method for pathogenic bacteria in aquatic products, the method comprising: S1. Collect aquatic product sample information and testing information. The aquatic product sample information includes the aquatic product's origin, collection time, storage and transportation conditions, and batch number. The testing information includes the pathogenic bacteria detected for each aquatic product sample, the testing time, and the concentration value of the test results. The aquatic product sample information and testing information are then processed to standardize the fields and construct a basic dataset of aquatic product pathogenic bacteria. S2. The basic dataset of pathogenic bacteria in aquatic products is structured and organized, and the corresponding relationship between the source of aquatic products, collection time, storage and transportation conditions, batch number, pathogenic bacteria type and detection result concentration value is established. Multidimensional feature records are generated for each individual aquatic product sample, and a multidimensional association data table is formed. S3. Perform co-occurrence correlation analysis on the multidimensional correlation data table. By extracting the co-occurrence records, frequency of occurrence, and corresponding relationship of concentration value changes of each pathogenic bacteria under different aquatic product origins, collection time intervals, and storage and transportation conditions, construct a pathogenic bacteria co-occurrence relationship matrix. Based on the correlation relationship between each pathogenic bacteria species in the pathogenic bacteria co-occurrence relationship matrix, establish a node connection structure to form a multidimensional correlation diagram of pathogenic bacteria. S4. Based on the co-occurrence matrix of pathogenic bacteria and the multidimensional association diagram of pathogenic bacteria, the distribution status of pathogenic bacteria in different aquatic product origins, collection time intervals and storage and transportation conditions is determined. By identifying the association connection sequence and extension direction of pathogenic bacteria between different batches of aquatic products, the transmission path of pathogenic bacteria and high-risk association combinations are determined, and the multidimensional association analysis results of pathogenic bacteria in aquatic products are generated. S5. Output pollution source indication information and risk level information based on the results of multidimensional association analysis of pathogenic bacteria in aquatic products, so as to realize the association identification and risk warning of potential pollution sources of pathogenic bacteria in aquatic products.

[0006] Preferably, step S3 includes: The multidimensional correlation data table was hierarchically grouped according to the aquatic product's origin, collection time interval, storage and transportation conditions, and batch number. The pathogenic bacteria types, detection result concentration values, and corresponding detection times were extracted from each group. Establish paired records for pathogenic bacteria species that appear simultaneously within the same group within a preset time interval, and mark each paired record with a concentration level according to the position of the detection result concentration value within a preset concentration level range; For each pairing record, the number of repeated occurrences, the number of consecutive occurrences, and the number of occurrences across batches are counted to generate a pathogen pairing statistics table; Based on the pathogenic bacteria pairing statistics table, pairing records that appear only once and do not show synchronous changes in concentration level are removed, while pairing records that appear more than twice consecutively and have the same direction of concentration level change are retained, thus generating an effective co-occurrence table of pathogenic bacteria. Using the pathogenic bacteria species in the effective co-occurrence table of pathogenic bacteria as the horizontal and vertical axes, and using the number of repeated occurrences, consecutive occurrences, cross-batch occurrences, and synchronous changes in concentration levels among each pathogenic bacteria species as the corresponding relationship content, a pathogenic bacteria co-occurrence relationship matrix is ​​generated. Based on the correspondence between pathogenic bacteria species in the pathogenic bacteria co-occurrence matrix, each pathogenic bacteria node is hierarchically connected according to the association strength, and the aquatic product origin, collection time interval, and storage and transportation conditions are added to the corresponding connection relationship to form a multidimensional association diagram of pathogenic bacteria.

[0007] Preferably, step S4 includes: Extract the number of repeated occurrences, consecutive occurrences, cross-batch occurrences, and synchronous changes in concentration levels among the pathogenic bacteria in the co-occurrence relationship matrix. Compare the pairing relationships of each pathogenic bacteria in order according to the number of repeated occurrences, consecutive occurrences, cross-batch occurrences, and synchronous changes in concentration levels to determine the primary and secondary pathogenic bacteria pairings. Starting with the pairing of primary pathogenic bacteria, corresponding connections are extracted from the multidimensional association graph of pathogenic bacteria according to the order of collection time to construct the temporal transmission chain of pathogenic bacteria. The changes in the source of aquatic products, storage and transportation conditions and batch numbers of adjacent nodes in the time-series transmission chain of pathogens were compared item by item. The connection relationship where the source of aquatic products is consistent and the batch number changes in the order of collection time is determined to be the homologous transmission relationship. The connection relationship where the storage and transportation conditions are consistent and the collection time interval is continuously progressive is determined to be the circulation transmission relationship. The node positions, concentration level changes, and connection directions of the main associated pathogen pairings of the same pathogen in different batches of aquatic products in the time-series transmission chain of pathogens were merged and sorted to determine the transmission initiation node, transmission relay node, and transmission terminal node. When the same transmission initiation node corresponds to two or more main associated pathogen pairings, and the main associated pathogen pairings extend to different batch numbers, the transmission initiation node is identified as a high-risk pollution source node. Based on the initiation node, relay node, terminal node, homologous transmission relationship, and circulation transmission relationship, the transmission path of pathogens and high-risk association combinations are identified, and the results of multidimensional association analysis of pathogens in aquatic products are generated.

[0008] Preferably, the process of generating the pathogen co-occurrence matrix includes: Establish row and column coordinates for each pathogen species in the effective co-occurrence table of pathogens, and write the paired records of any two pathogens in the same group to the corresponding coordinate positions; The number of repetitions, consecutive occurrences, cross-batch occurrences, and synchronous changes in concentration levels in the paired records are written into the record field at the same coordinate position to form pathogenic bacteria paired relationship units; When the same pathogen pairing relationship occurs repeatedly in different aquatic product origins, different collection time intervals, or different storage and transportation conditions, the record fields at the corresponding coordinate positions are accumulated and merged, and the aquatic product origin, collection time interval, storage and transportation conditions, and batch number corresponding to each occurrence are retained. When no pathogenic bacteria pairing record exists at any coordinate position, that coordinate position is marked as a unit with no co-occurrence relationship. A co-occurrence matrix of pathogens is generated by constructing a complete matrix structure based on the row and column coordinates of all pathogenic bacteria species, the paired relationship units of pathogenic bacteria, and the non-co-occurrence relationship units.

[0009] Preferably, the process of hierarchically connecting each pathogenic node according to its association strength includes: Extract the number of repeated occurrences, consecutive occurrences, cross-batch occurrences, and synchronous changes in concentration levels for each pathogenic bacteria pairing unit in the pathogenic bacteria co-occurrence matrix; The pairing units of each pathogenic bacterium were screened item by item according to the order of comparison of the number of repeated occurrences, the number of consecutive occurrences, the number of occurrences across batches, and the synchronous changes in concentration levels. The pathogen pairing relationship unit that meets the preset retention conditions for the number of repetitions, the number of consecutive occurrences, the number of occurrences across batches, and the synchronous change of concentration level is determined as the first association strength unit; the pathogen pairing relationship unit that meets only three of the preset retention conditions is determined as the second association strength unit; and the pathogen pairing relationship unit that meets only two of the preset retention conditions is determined as the third association strength unit. Establish first-layer connection relationships, second-layer connection relationships, and third-layer connection relationships based on the first association strength unit, the second association strength unit, and the third association strength unit, respectively; A hierarchical connection structure is formed between pathogenic bacteria nodes according to the first-level connection relationship, the second-level connection relationship, and the third-level connection relationship.

[0010] Preferably, the process of identifying pathogenic transmission routes and high-risk associated combinations includes: Starting from the transmission initiation node, adjacent connections are continuously extracted along the chronological transmission chain of pathogens, and the source of aquatic products, collection time interval, storage and transportation conditions and batch number corresponding to each connection are extracted. When the types of pathogens in adjacent connections are consistent and the batch numbers change accordingly according to the collection time, the corresponding connections will be merged into the same pathogen transmission chain segment. When the transmission chain of the same pathogen extends and connects through pairing of the main associated pathogen, the end node of the previous transmission chain is spliced ​​with the beginning node of the next transmission chain to form a continuous transmission path. When any propagation chain segment corresponds to both source propagation and circulation propagation relationships, the propagation chain segment is divided into source propagation segment and circulation propagation segment, and written into the continuous propagation path in the order of source propagation segment first and circulation propagation segment second. All continuous propagation paths are summarized at the end, and continuous propagation paths with a common propagation start node are merged into the same propagation path group; When pathogenic bacteria transmission chains in the same transmission path group are combined and statistically analyzed, if there are two or more main associated pathogenic bacteria pairs in the same transmission path group and the main associated pathogenic bacteria pairs extend to different batches of aquatic products, the corresponding pathogenic bacteria combinations are identified as high-risk associated combinations.

[0011] Preferably, step S5 includes: The transmission initiation node, transmission relay node, transmission terminal node, pathogen transmission path and high-risk association combination were identified from the multidimensional association analysis results of pathogenic bacteria in aquatic products. The source of aquatic products, collection time interval, storage and transportation conditions and batch number corresponding to each transmission path were also extracted. Based on the aquatic product origin and batch number corresponding to the initiation node in each transmission path, different transmission paths are merged and organized to generate a set of pollution source nodes. The number of transmission paths, the number of high-risk association combinations, and the number of consecutive batch numbers corresponding to each pollution source node are counted. The number of transmission paths, the number of high-risk associated combinations, and the number of consecutively appearing batch numbers are compared with preset risk judgment thresholds. When the number of transmission paths is not less than the first threshold, the number of high-risk associated combinations is not less than the second threshold, and the number of consecutively appearing batch numbers is not less than the third threshold, the corresponding pollution source node is determined as a first-level risk node. When the number of transmission paths is not less than the fourth threshold, the number of high-risk associated combinations is not less than the fifth threshold, and the number of consecutively appearing batch numbers is not less than the sixth threshold, the corresponding pollution source node will be identified as a secondary risk node. When the number of transmission paths is not less than the seventh threshold, the number of high-risk associated combinations is not less than the eighth threshold, and the number of consecutively appearing batch numbers is not less than the ninth threshold, the corresponding pollution source node will be identified as a level three risk node. Pollution source alerts are generated based on the risk level of each pollution source node, the corresponding aquatic product origin, storage and transportation conditions, and transmission route information. The pollution source information is associated with the corresponding risk node level to form the risk level information of pathogenic bacteria contamination in aquatic products, and the pollution source information and risk level information are output.

[0012] Secondly, a multidimensional correlation analysis system for pathogenic bacteria in aquatic products, the system comprising: The data acquisition module is used to collect information on aquatic product samples and testing information. The aquatic product sample information includes the origin of the aquatic products, collection time, storage and transportation conditions and batch number. The testing information includes the types of pathogens detected for each aquatic product sample, the testing time and the concentration value of the test results. The module also performs field standardization processing on the aquatic product sample information and testing information to construct a basic dataset of pathogens in aquatic products. The data processing module is used to structure and organize the basic dataset of pathogenic bacteria in aquatic products. It establishes a correspondence between the source of aquatic products, collection time, storage and transportation conditions, batch number, pathogenic bacteria type and detection result concentration value, generates multidimensional feature records for individual aquatic product samples, and forms a multidimensional association data table. The co-occurrence analysis module is used to perform co-occurrence association analysis on the multidimensional association data table. By extracting the co-occurrence records, frequency of occurrence, and corresponding changes in concentration values ​​of pathogenic bacteria under different aquatic product origins, collection time intervals, and storage and transportation conditions, a pathogenic bacteria co-occurrence relationship matrix is ​​constructed. Based on the association relationship between each pathogenic bacteria species in the pathogenic bacteria co-occurrence relationship matrix, a node connection structure is established to form a multidimensional association relationship diagram of pathogenic bacteria. The transmission analysis module is used to determine the distribution status of pathogens in different aquatic product origins, collection time intervals, and storage and transportation conditions based on the pathogen co-occurrence relationship matrix and the pathogen multidimensional association relationship diagram. By identifying the association connection sequence and extension direction of pathogens between different batches of aquatic products, the transmission path of pathogens and high-risk association combinations are determined, and the multidimensional association analysis results of pathogens in aquatic products are generated. The risk output module is used to output pollution source indication information and risk level information based on the results of multidimensional correlation analysis of pathogenic bacteria in aquatic products, so as to realize the correlation identification and risk warning of potential pollution sources of pathogenic bacteria in aquatic products.

[0013] The above-described solution of the present invention has at least the following beneficial effects: First, by collecting information on aquatic product samples and testing, and standardizing fields such as the source of aquatic products, collection time, storage and transportation conditions, batch number, pathogenic bacteria species, and concentration values ​​of test results, a basic dataset of pathogenic bacteria in aquatic products is constructed. This enables the sample source information, testing time information, and pathogenic bacteria test results, which were originally scattered in different testing records, to be organized in a unified data structure, thereby providing a unified data foundation for the correlation analysis between multidimensional data.

[0014] Based on this, the basic dataset is structured and organized, and the correspondence between the aquatic product's origin, collection time, storage and transportation conditions, batch number, pathogenic bacteria type, and test result concentration value is established. This forms a multidimensional feature record and multidimensional association data table with a single aquatic product sample as the unit, enabling different test batches, samples from different sources, and test results of different pathogenic bacteria to be associated and expressed in the same data structure, thereby achieving multidimensional organization of aquatic product test data.

[0015] Furthermore, by extracting the co-occurrence records, frequency of occurrence, and corresponding changes in concentration values ​​of various pathogenic bacteria from different aquatic product origins, collection time intervals, and storage and transportation conditions, a pathogenic bacteria co-occurrence relationship matrix is ​​constructed, and a multidimensional association diagram of pathogenic bacteria is formed. This allows the co-occurrence relationships of different pathogenic bacteria in different source environments and circulation processes to be expressed in a structured form, thereby achieving a comprehensive analysis of the association relationships between multiple pathogenic bacteria.

[0016] After forming a pathogen co-occurrence matrix and a multidimensional association diagram, the distribution status of pathogens in different aquatic product origins, collection time intervals, and storage and transportation conditions is determined by association. The transmission paths of pathogens and high-risk association combinations are identified by combining the connection sequence and extension direction between different batches. This allows the transmission relationship of pathogens between different batches of aquatic products to be systematically identified, thereby enabling the analysis of potential cross-contamination relationships between multiple batches of samples.

[0017] After obtaining the results of the multidimensional correlation analysis of pathogens, the detection results can not only reflect the detection of pathogens in a single test, but also identify potential sources of pollution by combining the source environment, time sequence and circulation process. This provides data support for the supervision of aquatic product quality, pollution source tracing and risk assessment, and improves the comprehensive analysis capability of pathogen contamination in the circulation of aquatic products. Attached Figure Description

[0018] Figure 1 This is a flowchart of a method for multidimensional correlation analysis of pathogenic bacteria in aquatic products provided in an embodiment of the present invention. Detailed Implementation

[0019] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0020] like Figure 1As shown, embodiments of the present invention propose a multidimensional association analysis method for pathogenic bacteria in aquatic products, the method comprising: S1. Collect aquatic product sample information and testing information. The aquatic product sample information includes the aquatic product's origin, collection time, storage and transportation conditions, and batch number. The testing information includes the pathogenic bacteria detected for each aquatic product sample, the testing time, and the concentration value of the test results. The aquatic product sample information and testing information are then processed to standardize the fields and construct a basic dataset of aquatic product pathogenic bacteria. S2. The basic dataset of pathogenic bacteria in aquatic products is structured and organized, and the corresponding relationship between the source of aquatic products, collection time, storage and transportation conditions, batch number, pathogenic bacteria type and detection result concentration value is established. Multidimensional feature records are generated for each individual aquatic product sample, and a multidimensional association data table is formed. S3. Perform co-occurrence correlation analysis on the multidimensional correlation data table. By extracting the co-occurrence records, frequency of occurrence, and corresponding relationship of concentration value changes of each pathogenic bacteria under different aquatic product origins, collection time intervals, and storage and transportation conditions, construct a pathogenic bacteria co-occurrence relationship matrix. Based on the correlation relationship between each pathogenic bacteria species in the pathogenic bacteria co-occurrence relationship matrix, establish a node connection structure to form a multidimensional correlation diagram of pathogenic bacteria. S4. Based on the co-occurrence matrix of pathogenic bacteria and the multidimensional association diagram of pathogenic bacteria, the distribution status of pathogenic bacteria in different aquatic product origins, collection time intervals and storage and transportation conditions is determined. By identifying the association connection sequence and extension direction of pathogenic bacteria between different batches of aquatic products, the transmission path of pathogenic bacteria and high-risk association combinations are determined, and the multidimensional association analysis results of pathogenic bacteria in aquatic products are generated. S5. Output pollution source indication information and risk level information based on the results of multidimensional association analysis of pathogenic bacteria in aquatic products, so as to realize the association identification and risk warning of potential pollution sources of pathogenic bacteria in aquatic products.

[0021] In this embodiment of the invention, by collecting aquatic product sample information and testing information, and performing unified field processing and structured organization on the aquatic product origin, collection time, storage and transportation conditions, batch number, pathogenic bacteria type, and test result concentration value, data from different sources can be correlated within the same data structure, forming multidimensional feature records based on individual aquatic product samples, and further constructing a multidimensional correlation data table. This method transforms originally scattered testing data into an analytical data set with a unified structure, establishing a clear data correspondence between aquatic product origin information, circulation information, and pathogenic bacteria test results, thereby providing a complete and consistent data foundation for subsequent correlation analysis.

[0022] Based on a multidimensional correlation data table, this study extracts the co-occurrence records, frequencies, and concentration changes of various pathogenic bacteria species under different aquatic product origins, collection time intervals, and storage and transportation conditions. A pathogenic bacteria co-occurrence matrix is ​​constructed, and further, a node connection structure is established based on the correlation relationships in the matrix to form a multidimensional correlation graph of pathogenic bacteria. By incorporating the co-occurrence, concentration changes, and time and environmental conditions of pathogenic bacteria species into the matrix and graph structures, the overall distribution of different pathogenic bacteria across different origins and circulation processes can be expressed, providing a structured correlation basis for subsequent transmission relationship analysis. The collection time interval is a time analysis interval divided according to preset time units or continuous time periods based on the collection time.

[0023] After generating a pathogen co-occurrence matrix and a multidimensional association diagram of pathogens, the distribution of pathogens under different aquatic product origins, collection time intervals, and storage and transportation conditions is correlated. The transmission paths of pathogens among multiple aquatic product batches are identified by combining the connection sequence and extension direction between different batches. Simultaneously, high-risk association combinations formed by the associations between multiple pathogens are identified, thus generating multidimensional association analysis results of aquatic product pathogens. This process allows for the simultaneous reflection of the transmission relationships of pathogens across different batches, origins, and storage and transportation conditions within the same analytical framework, enabling the clear identification of the transmission chains and formation conditions of pathogens.

[0024] After obtaining the results of multidimensional correlation analysis, the transmission paths and high-risk association combinations are comprehensively organized to output corresponding pollution source alerts and risk level information. This allows different source nodes and their corresponding batches to be risk-identified according to transmission relationships, thereby achieving the correlation identification and risk warning of potential pathogenic bacteria contamination sources in aquatic products. By jointly analyzing transmission relationships and source information, the possible source location and transmission range of contamination can be further determined based on the detection data, providing data support for aquatic product quality supervision and traceability analysis.

[0025] For example, in a certain aquatic product quality testing scenario, batch sampling inspections are conducted on shrimp products from multiple aquaculture areas. The origin, collection time, transportation and storage conditions, and concentration values ​​of pathogens such as Vibrio parahaemolyticus and Salmonella are recorded for each batch. The collected data is uniformly organized to form a multidimensional correlation data table, and a pathogen co-occurrence relationship matrix and a pathogen multidimensional correlation relationship diagram are constructed. The co-occurrence and concentration change relationships of pathogens among different batches of samples are analyzed to identify the transmission paths of pathogens that appear continuously in multiple batches and the high-risk association combinations formed by the co-occurrence of multiple pathogens. This allows for the determination of the pollution source nodes in the corresponding aquaculture areas and circulation links, and the output of corresponding pollution source warning information and risk level information to assist regulatory personnel in tracing and controlling the risks of relevant batches of aquatic products.

[0026] In a preferred embodiment of the present invention, step S3 includes: The multidimensional correlation data table was hierarchically grouped according to the aquatic product's origin, collection time interval, storage and transportation conditions, and batch number. The pathogenic bacteria types, detection result concentration values, and corresponding detection times were extracted from each group. Establish paired records for pathogenic bacteria species that appear simultaneously within the same group within a preset time interval, and mark each paired record with a concentration level according to the position of the detection result concentration value within a preset concentration level range; For each pairing record, the number of repeated occurrences, the number of consecutive occurrences, and the number of occurrences across batches are counted to generate a pathogen pairing statistics table; Based on the pathogenic bacteria pairing statistics table, pairing records that appear only once and do not show synchronous changes in concentration level are removed, while pairing records that appear more than twice consecutively and have the same direction of concentration level change are retained, thus generating an effective co-occurrence table of pathogenic bacteria. Using the pathogenic bacteria species in the effective co-occurrence table of pathogenic bacteria as the horizontal and vertical axes, and using the number of repeated occurrences, consecutive occurrences, cross-batch occurrences, and synchronous changes in concentration levels among each pathogenic bacteria species as the corresponding relationship content, a pathogenic bacteria co-occurrence relationship matrix is ​​generated. Based on the correspondence between pathogenic bacteria species in the pathogenic bacteria co-occurrence matrix, each pathogenic bacteria node is hierarchically connected according to the association strength, and the aquatic product origin, collection time interval, and storage and transportation conditions are added to the corresponding connection relationship to form a multidimensional association diagram of pathogenic bacteria.

[0027] In this embodiment of the invention, by hierarchically grouping the multidimensional correlation data table according to the aquatic product's origin, collection time interval, storage and transportation conditions, and batch number, and extracting the pathogenic bacteria species, detection result concentration values, and corresponding detection times within each group, the original detection data can be systematically divided according to the source environment and time conditions, clearly distinguishing the environmental and temporal backgrounds between different test samples. Based on this, by establishing pathogenic bacteria pairing records and marking the detection result concentration values ​​within intervals, the relationships between different pathogenic bacteria appearing within the same time range can be accurately recorded, while retaining their corresponding concentration change information. Furthermore, by statistically analyzing the number of repeated occurrences, consecutive occurrences, and cross-batch occurrences, and by removing pairing records that appear only once and do not show synchronous changes in concentration level, while retaining consecutively occurring pairing records with consistent directions of concentration level changes, occasional detection results can be screened out, thus ensuring that the retained co-occurrence relationships reflect stable pathogenic bacteria co-occurrence characteristics. Finally, by constructing a co-occurrence relationship matrix of pathogenic bacteria and forming a multidimensional association diagram of pathogenic bacteria, the co-occurrence relationships between different pathogenic bacteria and their corresponding origins, time intervals, and storage and transportation conditions are uniformly expressed, thus providing a clear association structure basis for subsequent identification of transmission paths.

[0028] In a preferred embodiment of the present invention, the method for setting the preset time interval and the preset concentration level range includes: When determining the preset time interval, the detection time of the same batch of aquatic product samples within a continuous detection cycle is first obtained, and the interval between two adjacent detection times is organized according to the detection frequency. Based on the organization results, the basic time unit is determined. When the detection work is carried out once a day, 24 hours is determined as a basic time unit, and when the detection work is carried out twice a day, 12 hours is determined as a basic time unit. Then, the basic time unit is corrected by combining the actual residence time of aquatic products at the source, transportation and temporary storage ends, and a preset time interval is generated for establishing pathogen pairing records. Among them, when the same sample or the same source sample is tested within the preset time interval, the pathogenic bacteria species detected within that time range are regarded as being in the same pairing judgment window, so as to determine whether the two pathogenic bacteria constitute a simultaneous occurrence relationship.

[0029] When determining the preset concentration level range, the concentration values ​​of the test results in historical test records are first extracted and sorted according to the type of pathogen. Then, combined with the current testing standards, common contamination levels of the samples, and the distribution of historical test results, the concentration values ​​of the test results are divided into multiple continuous intervals. For example, test results below a first concentration threshold can be identified as low concentration level, test results between the first and second concentration thresholds can be identified as medium concentration level, and test results above the second concentration threshold can be identified as high concentration level. In specific implementation, the first and second concentration thresholds can be set separately according to different pathogen types and remain consistent within the same implementation period. Subsequently, the concentration value of each test record is compared with the corresponding interval item by item to obtain the concentration level label of the test record, and the concentration level label is written into the corresponding pathogen pairing record for subsequent comparison of the concentration changes of different pathogens within the same time window. The first and second concentration thresholds are example thresholds used to divide the concentration level intervals in the embodiment.

[0030] In a preferred embodiment of the present invention, according to the pathogenic bacteria pairing statistics table, pairing records that appear only once and do not show synchronous changes in concentration level are removed, while pairing records that appear more than twice consecutively and show the same direction of concentration level change are retained, thereby generating an effective co-occurrence table of pathogenic bacteria, including: First, the established pathogen pairing records are summarized according to the aquatic product's origin, collection time interval, storage and transportation conditions, and batch number, forming a corresponding pathogen pairing statistical table. In this statistical table, the number of occurrences, consecutive occurrences, cross-batch occurrences, and corresponding concentration level markings of each pathogen pairing in each collection time interval are recorded. Then, the statistical results of each pathogen pairing are extracted one by one to determine whether it only occurs once in a collection time interval and whether there is no synchronous change in concentration level during that occurrence. If both conditions are met, the pathogen pairing is identified as an occasional pairing record and deleted from the subsequent retained records. Among them, no synchronous change in concentration level means that although the two pathogens constituting the pathogen pairing are detected simultaneously in the corresponding time interval, their concentration levels do not show a state of common increase or common decrease.

[0031] For pathogen pairing records that need to be retained, first check whether the pathogen pairing occurs consecutively in two or more adjacent collection time intervals; if consecutive occurrences exist, further compare the order of concentration level changes corresponding to the pathogen pairing within the consecutive time intervals; when both pathogens change from low concentration level to medium concentration level, or both change from medium concentration level to high concentration level, or both change from high concentration level to medium concentration level, or both change from medium concentration level to low concentration level, the change state is determined to be consistent in the direction of concentration level change; when pathogen pairing simultaneously meets the conditions of occurring consecutively more than twice and consistent in the direction of concentration level change, the pathogen pairing is determined to be a valid co-occurrence pairing and written into the effective co-occurrence table of pathogens.

[0032] When generating the effective co-occurrence table of pathogenic bacteria, for each retained effective co-occurrence pair, the pathogenic species name, corresponding origin, corresponding collection time interval, corresponding storage and transportation conditions, corresponding batch number, number of repetitions, number of consecutive occurrences, number of occurrences across batches, and synchronous changes in concentration level are recorded. All effective co-occurrence pairs after screening are then summarized to form the effective co-occurrence table of pathogenic bacteria, which serves as input data for subsequently constructing the pathogenic bacteria co-occurrence relationship matrix. Those skilled in the art, based on the above-described time window settings, concentration level interval divisions, and pairing screening steps, can uniformly process the raw detection data and extract the effective co-occurrence relationships of pathogenic bacteria.

[0033] In a preferred embodiment of the present invention, step S4 includes: Extract the number of repeated occurrences, consecutive occurrences, cross-batch occurrences, and synchronous changes in concentration levels among the pathogenic bacteria in the co-occurrence relationship matrix. Compare the pairing relationships of each pathogenic bacteria in order according to the number of repeated occurrences, consecutive occurrences, cross-batch occurrences, and synchronous changes in concentration levels to determine the primary and secondary pathogenic bacteria pairings. Starting with the pairing of primary pathogenic bacteria, corresponding connections are extracted from the multidimensional association graph of pathogenic bacteria according to the order of collection time to construct the temporal transmission chain of pathogenic bacteria. The changes in the source of aquatic products, storage and transportation conditions and batch numbers of adjacent nodes in the time-series transmission chain of pathogens were compared item by item. The connection relationship where the source of aquatic products is consistent and the batch number changes in the order of collection time is determined to be the homologous transmission relationship. The connection relationship where the storage and transportation conditions are consistent and the collection time interval is continuously progressive is determined to be the circulation transmission relationship. The node positions, concentration level changes, and connection directions of the main associated pathogen pairings of the same pathogen in different batches of aquatic products in the time-series transmission chain of pathogens were merged and sorted to determine the transmission initiation node, transmission relay node, and transmission terminal node. When the same transmission initiation node corresponds to two or more main associated pathogen pairings, and the main associated pathogen pairings extend to different batch numbers, the transmission initiation node is identified as a high-risk pollution source node. Based on the initiation node, relay node, terminal node, homologous transmission relationship, and circulation transmission relationship, the transmission path of pathogens and high-risk association combinations are identified, and the results of multidimensional association analysis of pathogens in aquatic products are generated.

[0034] In this embodiment of the invention, by extracting the frequency of repetitions, consecutive occurrences, cross-batch occurrences, and synchronous changes in concentration levels among various pathogenic bacteria in the pathogenic bacteria co-occurrence relationship matrix, and by sequentially comparing the pairing relationships of each pathogenic bacteria, it is possible to identify primary associated pathogenic bacteria pairings with stable co-occurrence characteristics and relatively secondary associated pathogenic bacteria pairings among multiple pathogenic bacteria combinations. Based on this, starting with the primary associated pathogenic bacteria pairings, corresponding connection relationships are extracted from the multidimensional association relationship diagram of pathogenic bacteria according to the collection time sequence. This allows the associated nodes scattered in different detection batches to be linked together to form a temporal transmission chain of pathogenic bacteria. By comparing the changes in the aquatic product origin, storage and transportation conditions, and batch number of adjacent nodes in the temporal transmission chain, it is possible to identify homologous transmission relationships with consistent origins and continuous transmission according to the collection time sequence, as well as circulation transmission relationships formed under the same storage and transportation conditions. By further merging and organizing the node positions, concentration level changes, and connection directions of the main associated pathogens in different batches of aquatic products, the transmission initiation nodes, transmission relay nodes, and transmission terminal nodes can be determined, and a complete transmission structure can be formed among multiple batches, thereby realizing the identification of pathogen transmission paths and high-risk associated combinations.

[0035] In a preferred embodiment of the present invention, the pairing relationships of each pathogenic bacterium are compared sequentially according to the number of repetitions, the number of consecutive occurrences, the number of occurrences across batches, and the synchronous changes in concentration levels to determine the primary associated pathogenic bacterium pairings and the secondary associated pathogenic bacterium pairings, including: First, all pathogenic bacteria co-occurrence relationship units in the pathogenic bacteria co-occurrence relationship matrix are extracted, and the number of repetitions, consecutive occurrences, cross-batch occurrences, and concentration level synchronous change records corresponding to each pathogenic bacteria co-occurrence relationship unit are extracted. Then, each pathogenic bacteria co-occurrence relationship unit is compared item by item according to a pre-set comparison order, prioritizing the comparison of the number of repetitions. When the number of repetitions of two or more pathogenic bacteria co-occurrence relationship units is inconsistent, the pathogenic bacteria co-occurrence relationship unit with more repetitions is ranked first. When the number of repetitions is consistent, the consecutive occurrences are compared, and the pathogenic bacteria co-occurrence relationship unit with more consecutive occurrences is ranked first. When the consecutive occurrences are still consistent, the cross-batch occurrences are compared, and the pathogenic bacteria co-occurrence relationship unit with more cross-batch occurrences is ranked first. When the cross-batch occurrences are still consistent, the concentration level synchronous change records are compared, and the pathogenic bacteria co-occurrence relationship unit with more records of synchronous concentration level changes is ranked first.

[0036] After arranging all pathogenic bacteria pairing units sequentially, the pathogenic bacteria pairings that are at the beginning and simultaneously meet the following criteria—a preset number of repetitions, a preset number of consecutive occurrences, a preset number of occurrences across batches, and a record of synchronous changes in concentration levels—are identified as primary associated pathogenic bacteria pairings. Pathogenic bacteria pairings that do not meet the criteria for primary associated pathogenic bacteria pairings but are located after the primary associated pathogenic bacteria pairings in the sequential arrangement and meet at least some of the criteria are identified as secondary associated pathogenic bacteria pairings. Subsequently, the primary associated pathogenic bacteria pairings and secondary associated pathogenic bacteria pairings are written into their respective pairing classification result tables for use when constructing transmission relationships later.

[0037] In a preferred embodiment of the present invention, starting with the pairing of primary pathogenic bacteria, corresponding connections are extracted from the multidimensional association graph of pathogenic bacteria according to the chronological order of collection time to construct a temporal transmission chain of pathogenic bacteria, including: First, extract the pathogen species, origin, collection time interval, storage and transportation conditions, and batch number corresponding to each main associated pathogen pair from the main associated pathogen pair classification result table. Then, locate the initial connection relationship corresponding to the main associated pathogen pair in the pathogen multidimensional association relationship diagram. Next, using the initial connection relationship as the starting position, search for the next connection relationship that is adjacent to it and has temporal continuity according to the collection time from front to back. Here, temporal continuity means that the collection time corresponding to the next connection relationship is later than the collection time corresponding to the previous connection relationship, and the two are located in adjacent collection time intervals or in collection time intervals arranged continuously according to the detection order. When the next connection relationship is found, it is further determined whether the next connection relationship has the same pathogen species as the current connection relationship or forms a connection relationship through the main associated pathogen pair. If the connection condition is met, the next connection relationship is continued to the end of the current chain and used as the new current connection relationship to continue searching. If the connection condition is not met, the extension of the current chain is terminated.

[0038] Following the above method, starting from each primary pathogen pairing, multiple connections that are consecutive in collection time and interconnected in the relationship graph are extracted sequentially. The origin, collection time interval, storage and transportation conditions, and batch number corresponding to each node are recorded in the order of connection, forming a pathogenic bacteria time-series transmission chain. After repeating the above steps for all primary pathogen pairings, multiple pathogenic bacteria time-series transmission chains are obtained, and each pathogenic bacteria time-series transmission chain is written into the time-series transmission chain result table.

[0039] In a preferred embodiment of the present invention, the node positions, concentration level changes, and connection directions corresponding to the main associated pathogen pairings of the same pathogen in different batches of aquatic products in the time-series transmission chain of pathogens are merged and organized to determine the transmission initiation node, transmission relay node, and transmission termination node, including: First, the pathogenic bacteria time-series transmission chains in the result table are extracted one by one. From each chain, the pathogenic bacteria species, batch number, node order, concentration level marker, and the primary pathogenic bacteria pairing relationship between each node and the preceding and following nodes are extracted. Then, nodes of the same pathogenic bacteria in different batches of aquatic products are categorized. Multiple nodes with the same pathogenic bacteria species and located in the same transmission chain are numbered according to their order of appearance in the chain to determine their position. Next, the concentration level markers of adjacent nodes are compared one by one according to their order of appearance to form the sequence of concentration level changes of the same pathogenic bacteria in different batches. The specific changes from low to medium concentration, from medium to high concentration, or from high to medium concentration, or from medium to low concentration are recorded. Simultaneously, the primary pathogenic bacteria pairing relationships between adjacent nodes are extracted to determine the connection direction in the transmission chain. The connection direction is recorded in the order of the preceding node pointing to the following node.

[0040] After completing the above processing, the node types in each pathogenic bacteria transmission chain are determined as follows: the node at the beginning of the chain, with no other connections pointing to it and connecting to subsequent nodes, is identified as the propagation initiation node; the node in the middle of the chain, receiving both the previous connection and connecting to the next node, is identified as the propagation relay node; and the node at the end of the chain, receiving the previous connection and with no subsequent connections, is identified as the propagation termination node. In cases of branch connections, nodes connecting to multiple subsequent nodes are still identified as either propagation initiation nodes or propagation relay nodes, depending on whether there are any connections at their beginning. The identified propagation initiation nodes, propagation relay nodes, and propagation termination nodes are marked and written into the propagation node result table for subsequent identification of pathogenic bacteria transmission paths and high-risk association combinations.

[0041] In a preferred embodiment of the present invention, the process of generating a pathogen co-occurrence matrix includes: Establish row and column coordinates for each pathogen species in the effective co-occurrence table of pathogens, and write the paired records of any two pathogens in the same group to the corresponding coordinate positions; The number of repetitions, consecutive occurrences, cross-batch occurrences, and synchronous changes in concentration levels in the paired records are written into the record field at the same coordinate position to form pathogenic bacteria paired relationship units; When the same pathogen pairing relationship occurs repeatedly in different aquatic product origins, different collection time intervals, or different storage and transportation conditions, the record fields at the corresponding coordinate positions are accumulated and merged, and the aquatic product origin, collection time interval, storage and transportation conditions, and batch number corresponding to each occurrence are retained. When no pathogenic bacteria pairing record exists at any coordinate position, that coordinate position is marked as a unit with no co-occurrence relationship. A co-occurrence matrix of pathogens is generated by constructing a complete matrix structure based on the row and column coordinates of all pathogenic bacteria species, the paired relationship units of pathogenic bacteria, and the non-co-occurrence relationship units.

[0042] In this embodiment of the invention, by establishing row and column coordinates for each pathogen in the effective co-occurrence table of pathogens, and writing the pairing records of any two pathogens within the same group into the corresponding coordinate positions, the co-occurrence relationships between different pathogens can be uniformly organized in a matrix structure. Writing records of repeated occurrences, consecutive occurrences, cross-batch occurrences, and synchronous changes in concentration levels into the same coordinate position allows the association information between each pair of pathogens to be stored in a fixed structure, facilitating unified extraction and comparison in subsequent analyses. When the same pathogen pairing relationship occurs repeatedly in different aquatic product origins, different collection time intervals, or different storage and transportation conditions, by accumulating and merging the record fields at the corresponding coordinate positions and retaining the corresponding origin, time interval, storage and transportation conditions, and batch number, the association relationship and its formation conditions can be simultaneously recorded in the matrix structure. When there is no pathogen pairing record at a certain coordinate position, it is marked as a unit without co-occurrence relationship, keeping the matrix structure intact, thus forming a complete matrix structure containing all pathogen species and their co-occurrence relationships. The pathogen co-occurrence matrix constructed in this way can uniformly express the co-occurrence of pathogens and their corresponding environmental conditions, thus providing a structured data foundation for subsequent association strength analysis and transmission path identification.

[0043] In a preferred embodiment of the present invention, a pathogen co-occurrence relationship matrix is ​​generated by constructing a complete matrix structure according to the row and column coordinates of all pathogenic bacteria species, pathogenic bacteria pairing relationship units, and non-co-occurrence relationship units, including: First, all pathogenic bacteria species appearing in the effective co-occurrence table are extracted and deduplicated to form a set of pathogenic bacteria species. Then, the pathogenic bacteria species in the set are arranged according to a unified sorting rule, which can be based on the order of their first appearance in the detection records or the predetermined entry order of the pathogenic bacteria names. After sorting, all pathogenic bacteria species are simultaneously written into the row and column directions of a matrix, so that each row and column in the matrix corresponds to one pathogenic bacteria species, thus forming the row and column coordinates of all pathogenic bacteria species. Subsequently, the formed pathogenic bacteria pairing units are extracted one by one, and according to the two pathogenic bacteria names corresponding to the pairing unit, the corresponding intersection position is located in the row and column coordinates, and the pathogenic bacteria pairing unit is written into the corresponding intersection position. For pathogenic bacteria pairings with mutual correspondence, the same content can be written into the symmetrical intersection position to maintain the consistency of the pairing relationships in the matrix. Next, each row and column intersection is checked. When no pathogen pairing relationship unit is written at a certain intersection position, the intersection position is marked as a unit with no co-occurrence relationship. The unit with no co-occurrence relationship is used to indicate that the two corresponding pathogens have not formed a valid co-occurrence relationship within the current statistical range.

[0044] After writing and marking all intersection positions, the integrity of each row and column in the matrix is ​​verified to confirm that each pathogen is arranged in the corresponding row and column directions, and that each intersection position corresponds to a pathogen pairing unit or a unit with no co-occurrence relationship. Finally, all verified row and column coordinates, pathogen pairing units, and units with no co-occurrence relationship are summarized to form a complete matrix structure and generate a pathogen co-occurrence relationship matrix for subsequent use in extracting pathogen pairing relationships, comparing association strength, and establishing node connection relationships. The pathogen co-occurrence relationship matrix is ​​used to represent co-occurrence correspondences and belongs to an undirected correspondence structure.

[0045] In a preferred embodiment of the present invention, the process of hierarchically connecting each pathogenic bacteria node according to the association strength includes: Extract the number of repeated occurrences, consecutive occurrences, cross-batch occurrences, and synchronous changes in concentration levels for each pathogenic bacteria pairing unit in the pathogenic bacteria co-occurrence matrix; The pairing units of each pathogenic bacterium were screened item by item according to the order of comparison of the number of repeated occurrences, the number of consecutive occurrences, the number of occurrences across batches, and the synchronous changes in concentration levels. The pathogen pairing relationship unit that meets the preset retention conditions for the number of repetitions, the number of consecutive occurrences, the number of occurrences across batches, and the synchronous change of concentration level is determined as the first association strength unit; the pathogen pairing relationship unit that meets only three of the preset retention conditions is determined as the second association strength unit; and the pathogen pairing relationship unit that meets only two of the preset retention conditions is determined as the third association strength unit. Establish first-layer connection relationships, second-layer connection relationships, and third-layer connection relationships based on the first association strength unit, the second association strength unit, and the third association strength unit, respectively; A hierarchical connection structure is formed between pathogenic bacteria nodes according to the first-level connection relationship, the second-level connection relationship, and the third-level connection relationship.

[0046] In this embodiment of the invention, by extracting the repetition frequency, consecutive occurrence frequency, cross-batch occurrence frequency, and synchronous change records of concentration levels corresponding to each pathogenic bacteria pairing relationship unit in the pathogenic bacteria co-occurrence relationship matrix, and screening each pathogenic bacteria pairing relationship unit item by item according to a preset comparison order, the degree of association between different pathogenic bacteria can be uniformly judged. During the screening process, pairing relationship units that simultaneously meet preset retention conditions are divided into first association strength units, and pairing relationship units that meet some retention conditions are divided into second or third association strength units, so that the association relationship between different pathogenic bacteria pairings can be expressed hierarchically according to the degree of association. Subsequently, by establishing first-layer, second-layer, and third-layer connection relationships respectively, a hierarchical connection relationship network is formed between pathogenic bacteria nodes. Through this process, pathogenic bacteria combinations with high association degrees and pathogenic bacteria combinations with relatively low association degrees can be distinguished from various pathogenic bacteria combinations, thereby providing a hierarchical association structure basis for subsequent transmission path identification.

[0047] In a preferred embodiment of the present invention, the method for setting preset retention conditions includes: First, pathogenic bacteria pairing relationship units formed within multiple detection cycles are acquired, and the repetition frequency, consecutive occurrence frequency, cross-batch occurrence frequency, and synchronous change records of concentration level corresponding to each pathogenic bacteria pairing relationship unit are extracted. Then, the occurrence of different pathogenic bacteria pairing relationship units in actual detection scenarios is statistically organized to form the distribution of repetition, consecutive occurrence, cross-batch occurrence, and synchronous change of concentration level. Based on this, the lower limit requirements for screening stable co-occurrence relationships are set by combining the number of aquatic product origins, the number of collection time intervals, the number of storage and transportation conditions, and the length of batch transfer records.

[0048] Specifically, for the number of repeated occurrences, the repeated recording of the same pathogen pairing in two or more statistical groups can be used as a condition for repeated occurrence; for the number of consecutive occurrences, the consecutive occurrence of the same pathogen pairing in two or more adjacent collection time intervals can be used as a condition for consecutive occurrence; for the number of occurrences across batches, the recording of the same pathogen pairing in two or more different batch numbers can be used as a condition for cross-batch occurrence; for records of synchronous changes in concentration levels, the synchronous change condition can be defined as at least one common increase or decrease in concentration level of the two pathogens in continuous comparisons. After setting the above lower limits, each lower limit requirement is written into the retention condition configuration table to form the corresponding preset retention conditions; within the same implementation cycle, each preset retention condition remains consistent to ensure that different pathogen pairing relationship units are screened according to the same standard; when the detection object changes or the detection cycle range changes, the preset retention conditions can be updated again according to the statistical distribution of the corresponding detection data, and the updated conditions are used in the new association strength determination process.

[0049] In a preferred embodiment of the present invention, a first-layer connection relationship, a second-layer connection relationship, and a third-layer connection relationship are established based on a first association strength unit, a second association strength unit, and a third association strength unit, respectively, including: First, extract all pathogenic bacteria pairing relationship units after completing the association strength classification, and extract the pathogenic bacteria pairing relationship unit records that are determined to be the first association strength unit, the second association strength unit, and the third association strength unit respectively; then, take the two pathogenic bacteria in each pathogenic bacteria pairing relationship unit record as nodes to be connected, and extract the corresponding aquatic product origin, collection time interval, storage and transportation conditions, number of recurrences, number of consecutive occurrences, number of occurrences across batches, and synchronous change records of concentration level for that pathogenic bacteria pairing relationship unit record. For each pathogen pairing relationship record in the first association strength unit, a first-level connection is established between the corresponding two pathogen nodes, and the statistical content and condition information of the pathogen pairing relationship record are written into this connection to indicate that the two pathogens form the most stable association connection. For each pathogen pairing relationship record in the second association strength unit, a second-level connection is established between the corresponding two pathogen nodes, and the corresponding statistical content and condition information are written to indicate that the two pathogens form a secondary association connection. For each pathogen pairing relationship record in the third association strength unit, a third-level connection is established between the corresponding two pathogen nodes, and the corresponding statistical content and condition information are written to indicate that the two pathogens form a basic-level association connection.

[0050] After establishing all the connections, the connections at each level are numbered and organized, and each connection can be traced back to a specific pathogenic bacteria pairing unit record, thus forming a set of hierarchical connections that can be used for subsequent node relationship extraction and path identification.

[0051] In a preferred embodiment of the present invention, a hierarchical connection structure is formed between pathogenic bacteria nodes according to a first-layer connection relationship, a second-layer connection relationship, and a third-layer connection relationship, including: First, using all pathogenic bacteria species as the basic set of nodes, a unique node identifier is established for each pathogen, and all pathogenic bacteria nodes are uniformly written into the node list. Then, the first-level connection relationship, the second-level connection relationship, and the third-level connection relationship are extracted, and the connection records between the corresponding nodes are written in descending order of connection level. When writing the first-level connection relationship, pathogenic bacteria node pairs with first-level connection relationships are identified as core connection node pairs, and their connection direction, connection level, and corresponding condition information are recorded first. When writing the second-level connection relationship, pathogenic bacteria node pairs with second-level connection relationships are identified as extended connection node pairs, and they are attached to the established core connection node pairs to form an outward-extending connection structure. When writing the third-level connection relationship, pathogenic bacteria node pairs with third-level connection relationships are identified as peripheral connection node pairs, and they are written to the outside of the extended connection node pairs to form an outermost connection structure.

[0052] If the same pair of pathogen nodes simultaneously satisfies connection conditions at different levels, the higher-level connection is retained as the final connection level for that node pair. If the same node forms connection relationships at different levels with multiple other nodes, each connection relationship is recorded separately according to its connection level, allowing the node to simultaneously exhibit multi-level connection states. After writing all connection relationships hierarchically, the connection status between each pathogen node is uniformly organized, forming a hierarchical connection structure with the first-level connection relationship as the core layer, the second-level connection relationship as the extension layer, and the third-level connection relationship as the outer layer. Finally, the hierarchical connection structure is written into the pathogen node association structure table for subsequent use in constructing a multi-dimensional association graph of pathogens and identifying propagation paths.

[0053] In a preferred embodiment of the present invention, the process of identifying pathogenic bacteria transmission routes and high-risk association combinations includes: Starting from the transmission initiation node, adjacent connections are continuously extracted along the chronological transmission chain of pathogens, and the source of aquatic products, collection time interval, storage and transportation conditions and batch number corresponding to each connection are extracted. When the types of pathogens in adjacent connections are consistent and the batch numbers change accordingly according to the collection time, the corresponding connections will be merged into the same pathogen transmission chain segment. When the transmission chain of the same pathogen extends and connects through pairing of the main associated pathogen, the end node of the previous transmission chain is spliced ​​with the beginning node of the next transmission chain to form a continuous transmission path. When any propagation chain segment corresponds to both source propagation and circulation propagation relationships, the propagation chain segment is divided into source propagation segment and circulation propagation segment, and written into the continuous propagation path in the order of source propagation segment first and circulation propagation segment second. All continuous propagation paths are summarized at the end, and continuous propagation paths with a common propagation start node are merged into the same propagation path group; When pathogenic bacteria transmission chains in the same transmission path group are combined and statistically analyzed, if there are two or more main associated pathogenic bacteria pairs in the same transmission path group and the main associated pathogenic bacteria pairs extend to different batches of aquatic products, the corresponding pathogenic bacteria combinations are identified as high-risk associated combinations.

[0054] In this embodiment of the invention, by taking the transmission initiation node as the starting point and continuously extracting adjacent connections along the pathogenic bacteria temporal transmission chain according to the chronological order of collection time, and extracting the aquatic product origin, collection time interval, storage and transportation conditions, and batch number corresponding to each connection, the pathogenic bacteria association nodes originally scattered in different test batches can be continuously organized in chronological order, allowing the pathogenic bacteria association relationships between different batches to be expressed in the same time series structure. Furthermore, when the pathogenic bacteria species in adjacent connections remain consistent and the batch numbers change according to the collection time order, by merging the corresponding connections into the same pathogenic bacteria transmission chain segment, the continuous occurrence relationship of the same type of pathogenic bacteria across multiple aquatic product batches can be uniformly expressed, thereby forming a continuous transmission chain segment structure.

[0055] After a transmission chain is formed, when different transmission chains are connected through pairing with a primary pathogen, multiple transmission chains can be integrated into a continuous transmission path by splicing the terminal node of the previous transmission chain with the starting node of the next transmission chain. This allows the transmission process of pathogens between different batches to be expressed as a complete chain. When a transmission chain corresponds to both homologous transmission and circulation transmission relationships, by dividing the transmission chain into source transmission segments and circulation transmission segments, and writing the continuous transmission path with the source transmission segment first and the circulation transmission segment last, the transmission process of pathogens in the source and circulation stages can be distinguished and expressed, thus reflecting the transmission characteristics of pathogens in different stages.

[0056] Based on the formation of continuous transmission paths, transmission paths with a common initiation node are merged to form transmission path groups, which can unify and organize multiple transmission paths caused by the same pollution source. Furthermore, by combining and statistically analyzing the transmission chain segments of pathogens within the transmission path groups, and identifying the main associated pathogen pairs that appear simultaneously in the same transmission path group and extend to different batches of aquatic products, high-risk pathogen association combinations can be determined. Through this process, pathogen transmission paths can be identified simultaneously with pathogen combinations with a high degree of transmission association, creating a corresponding expression between the transmission path structure and pathogen combination relationships, thus providing a basis for subsequent pollution source analysis and risk level determination.

[0057] In a preferred embodiment of the present invention, when the same pathogen transmission chain segments extend and connect through pairing of primary associated pathogens, the terminal node of the preceding transmission chain segment is joined end-to-end with the starting node of the following transmission chain segment to form a continuous transmission path, including: First, extract all generated transmission chains of the same pathogen, and extract the starting node, ending node, pathogen type, collection time interval, batch number, and main pathogen pairing information corresponding to adjacent nodes for each transmission chain. Then, take a certain transmission chain as the previous transmission chain, extract the pathogen type, batch number, and collection time interval corresponding to its ending node, and then search for the subsequent transmission chains in which the starting node and ending node have a main pathogen pairing connection relationship, and take them as the next transmission chain. During the search process, prioritize retaining the next transmission chain whose collection time is later than the collection time corresponding to the ending node of the previous transmission chain and whose batch number changes continuously according to the collection time order, so as to ensure that the transmission direction is consistent with the time order.

[0058] Once a primary pathogen pairing relationship is confirmed between the terminal node of the previous transmission chain and the starting node of the next transmission chain, all existing nodes in the previous transmission chain are retained sequentially. Then, the remaining nodes in the next transmission chain, except for the starting node, are added to the end of the previous transmission chain in their original order. If the starting node of the next transmission chain and the terminal node of the previous transmission chain correspond to the same batch number and are in adjacent collection time intervals, the two nodes are considered as head-to-tail nodes. If the starting node of the next transmission chain and the terminal node of the previous transmission chain correspond to different batch numbers, but a primary pathogen pairing relationship has been established between them, this primary pathogen pairing relationship is used as the basis for chain splicing, and the next transmission chain is added to the end of the previous transmission chain.

[0059] After completing one end-to-end splicing, the newly formed chain is used as the chain to be extended to continue searching for other connectable transmission chains. When no new main associated pathogen pairing relationship exists, or the collection time sequence of subsequent transmission chains does not meet the progressive requirements, the splicing process is terminated. The above process is repeated for all possible transmission chains to form a continuous transmission path composed of multiple transmission chains connected end-to-end. The node sequence, connection sequence, source, collection time interval, storage and transportation conditions, and batch number corresponding to the continuous transmission path are written into the continuous transmission path record table for subsequent path merging and risk analysis.

[0060] In a preferred embodiment of the present invention, when any propagation chain segment corresponds to both a source propagation relationship and a flow propagation relationship, the propagation chain segment is divided into a source propagation segment and a flow propagation segment, and written into the continuous propagation path in the order of source propagation segment first and flow propagation segment second, including: First, the connection relationships between adjacent nodes in each transmission chain segment are extracted one by one, and the source of aquatic products, collection time interval, storage and transportation conditions, and batch number corresponding to each adjacent node are extracted respectively. Then, it is determined whether each adjacent connection relationship meets the conditions for determining the same source transmission relationship and the conditions for determining the circulation transmission relationship. Specifically, when the source of aquatic products of adjacent nodes is the same and the batch number changes in the order of collection time, the adjacent connection relationship is determined to be a same source transmission relationship. When the storage and transportation conditions of adjacent nodes are the same and the collection time interval is continuously progressive, the adjacent connection relationship is determined to be a circulation transmission relationship.

[0061] When the adjacent connections at the beginning of the same propagation chain meet the criteria for same-source propagation and the adjacent connections at the end meet the criteria for flow propagation, the corresponding preceding nodes and connections are divided into source propagation segments, and the corresponding following nodes and connections are divided into flow propagation segments. If an intermediate node serves as both the end node of the source propagation segment and the beginning node of the flow propagation segment, the intermediate node is retained as an inter-segment connection node, and its forward and backward connections are recorded respectively.

[0062] After dividing the transmission chain into segments, the nodes and connections in the source transmission segment are first written into the continuous transmission path in chronological order of collection time. Then, the nodes and connections in the circulation transmission segment are added sequentially after the source transmission segment in chronological order of collection time. If only source transmission relationships or only circulation transmission relationships exist within the same transmission chain segment, the entire transmission chain segment is written into the continuous transmission path as a single type of transmission segment. This method allows the transmission processes formed by the source link and the transmission processes formed by the circulation link within the same transmission chain to be expressed sequentially in the continuous transmission path, facilitating the subsequent differentiation between the pollution occurrence stage and the diffusion stage.

[0063] In a preferred embodiment of the present invention, step S5 includes: The transmission initiation node, transmission relay node, transmission terminal node, pathogen transmission path and high-risk association combination were identified from the multidimensional association analysis results of pathogenic bacteria in aquatic products. The source of aquatic products, collection time interval, storage and transportation conditions and batch number corresponding to each transmission path were also extracted. Based on the aquatic product origin and batch number corresponding to the initiation node in each transmission path, different transmission paths are merged and organized to generate a set of pollution source nodes. The number of transmission paths, the number of high-risk association combinations, and the number of consecutive batch numbers corresponding to each pollution source node are counted. The number of transmission paths, the number of high-risk associated combinations, and the number of consecutively appearing batch numbers are compared with preset risk judgment thresholds. When the number of transmission paths is not less than the first threshold, the number of high-risk associated combinations is not less than the second threshold, and the number of consecutively appearing batch numbers is not less than the third threshold, the corresponding pollution source node is determined as a first-level risk node. When the number of transmission paths is not less than the fourth threshold, the number of high-risk associated combinations is not less than the fifth threshold, and the number of consecutively appearing batch numbers is not less than the sixth threshold, the corresponding pollution source node will be identified as a secondary risk node. When the number of transmission paths is not less than the seventh threshold, the number of high-risk associated combinations is not less than the eighth threshold, and the number of consecutively appearing batch numbers is not less than the ninth threshold, the corresponding pollution source node will be identified as a level three risk node. Pollution source alerts are generated based on the risk level of each pollution source node, the corresponding aquatic product origin, storage and transportation conditions, and transmission route information. The pollution source information is associated with the corresponding risk node level to form the risk level information of pathogenic bacteria contamination in aquatic products, and the pollution source information and risk level information are output.

[0064] In this embodiment of the invention, by extracting the transmission initiation nodes, transmission relay nodes, transmission terminal nodes, pathogenic bacteria transmission paths, and high-risk association combinations identified in the multidimensional association analysis results of pathogenic bacteria in aquatic products, and extracting the aquatic product origin, collection time interval, storage and transportation conditions, and batch number corresponding to each transmission path, the transmission relationships and source information between different batches of samples can be uniformly organized, enabling the transmission structure information in the detection data to establish a correspondence with specific source locations. Based on this, by merging and organizing different transmission paths according to the aquatic product origin and batch number corresponding to the transmission initiation node, multiple transmission paths triggered by the same source node can be centrally expressed, generating a set of contamination source nodes. This allows the transmission relationships between multiple detection batches to be organized and analyzed around specific source nodes.

[0065] After forming a set of contamination source nodes, the transmission characteristics of these nodes can be comprehensively described from three aspects: the scope of transmission, the relationship between pathogenic bacteria, and the continuity of batches, by statistically analyzing the number of transmission paths, the number of high-risk associated combinations, and the number of consecutive batch numbers corresponding to each contamination source node. Based on this, by comparing the statistical results with preset risk assessment thresholds, the transmission intensity of different contamination source nodes can be graded, thus forming a risk node structure of different levels. By adopting a clear threshold determination method, the risk level determination process can have a unified standard, thereby improving the consistency of risk assessment among different tested batches.

[0066] After determining the risk level, by associating the risk level of the pollution source node with the corresponding aquatic product's origin, storage and transportation conditions, and transmission route information, pollution source alert information containing the pollution source location and risk level can be generated. By outputting pollution source alert information and risk level information, the transmission relationships, source information, and risk levels in aquatic product testing data can be uniformly expressed, thus providing a basis for aquatic product quality supervision, pollution tracing, and risk management, enabling the test results to further support the identification of potential pollution sources and risk warnings.

[0067] In a preferred embodiment of the present invention, the method for setting a preset risk assessment threshold includes: First, statistical results of pollution source nodes formed within multiple detection cycles are extracted, and the number of transmission paths, the number of high-risk association combinations, and the number of consecutively occurring batch numbers corresponding to each pollution source node are extracted. Then, the above three types of statistical results are sorted separately to form the distribution of the number of transmission paths, the distribution of the number of high-risk association combinations, and the distribution of the number of consecutively occurring batch numbers. Based on this, combined with actual regulatory needs and the number of sample batches, the lower limit of judgment corresponding to different risk levels is set in layers to form a preset risk judgment threshold for risk level determination.

[0068] The preset risk assessment thresholds include a set of thresholds for determining primary risk nodes, a set of thresholds for determining secondary risk nodes, and a set of thresholds for determining tertiary risk nodes. The set of thresholds for determining primary risk nodes includes a first threshold, a second threshold, and a third threshold, which correspond to the number of transmission paths, the number of high-risk association combinations, and the number of consecutively appearing batch numbers, respectively. The set of thresholds for determining secondary risk nodes includes a fourth threshold, a fifth threshold, and a sixth threshold, which correspond to the number of transmission paths, the number of high-risk association combinations, and the number of consecutively appearing batch numbers, respectively. The set of thresholds for determining tertiary risk nodes includes a seventh threshold, an eighth threshold, and a ninth threshold, which correspond to the number of transmission paths, the number of high-risk association combinations, and the number of consecutively appearing batch numbers, respectively.

[0069] When setting the thresholds for each group, first determine the highest lower limit for judgment corresponding to the first-level risk node, then determine the intermediate lower limit for judgment corresponding to the second-level risk node, and finally determine the basic lower limit for judgment corresponding to the third-level risk node; so that the first threshold is higher than the fourth threshold, the fourth threshold is higher than the seventh threshold, the second threshold is higher than the fifth threshold, the fifth threshold is higher than the eighth threshold, the third threshold is higher than the sixth threshold, and the sixth threshold is higher than the ninth threshold, thereby forming a decreasing threshold relationship corresponding to the risk level.

[0070] In practical implementation, pollution source nodes that meet high requirements for the number of transmission paths, high requirements for the number of high-risk associated combinations, and high requirements for the number of consecutively occurring batch numbers can be identified as Level 1 risk nodes; pollution source nodes that meet the intermediate requirements for all three statistical results can be identified as Level 2 risk nodes; and pollution source nodes that meet the basic requirements for all three statistical results can be identified as Level 3 risk nodes. After setting the thresholds, each threshold is written into the risk assessment configuration table and kept consistent within the same testing and analysis cycle. When the sample size, the number of circulating batches, or regulatory standards change, the corresponding thresholds can be readjusted based on the new statistical distribution, and the adjusted thresholds are used in the new risk level assessment process.

[0071] The risk level determination is carried out in the order of Level 1 risk nodes, Level 2 risk nodes, and Level 3 risk nodes. When the conditions for a higher level are met, the determination of a lower level will not be carried out.

[0072] Embodiments of the present invention also provide a multidimensional correlation analysis system for pathogenic bacteria in aquatic products, the system comprising: The data acquisition module is used to collect information on aquatic product samples and testing information. The aquatic product sample information includes the origin of the aquatic products, collection time, storage and transportation conditions and batch number. The testing information includes the types of pathogens detected for each aquatic product sample, the testing time and the concentration value of the test results. The module also performs field standardization processing on the aquatic product sample information and testing information to construct a basic dataset of pathogens in aquatic products. The data processing module is used to structure and organize the basic dataset of pathogenic bacteria in aquatic products. It establishes a correspondence between the source of aquatic products, collection time, storage and transportation conditions, batch number, pathogenic bacteria type and detection result concentration value, generates multidimensional feature records for individual aquatic product samples, and forms a multidimensional association data table. The co-occurrence analysis module is used to perform co-occurrence association analysis on the multidimensional association data table. By extracting the co-occurrence records, frequency of occurrence, and corresponding changes in concentration values ​​of pathogenic bacteria under different aquatic product origins, collection time intervals, and storage and transportation conditions, a pathogenic bacteria co-occurrence relationship matrix is ​​constructed. Based on the association relationship between each pathogenic bacteria species in the pathogenic bacteria co-occurrence relationship matrix, a node connection structure is established to form a multidimensional association relationship diagram of pathogenic bacteria. The transmission analysis module is used to determine the distribution status of pathogens in different aquatic product origins, collection time intervals, and storage and transportation conditions based on the pathogen co-occurrence relationship matrix and the pathogen multidimensional association relationship diagram. By identifying the association connection sequence and extension direction of pathogens between different batches of aquatic products, the transmission path of pathogens and high-risk association combinations are determined, and the multidimensional association analysis results of pathogens in aquatic products are generated. The risk output module is used to output pollution source indication information and risk level information based on the results of multidimensional correlation analysis of pathogenic bacteria in aquatic products, so as to realize the correlation identification and risk warning of potential pollution sources of pathogenic bacteria in aquatic products.

[0073] It should be noted that this system is a system corresponding to the above method. All implementation methods in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0074] Embodiments of the present invention also provide a computing device, including: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0075] Embodiments of the present invention also provide a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0076] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A multidimensional correlation analysis method for pathogenic bacteria in aquatic products, characterized in that, The method includes: S1. Collect aquatic product sample information and testing information. The aquatic product sample information includes the aquatic product's origin, collection time, storage and transportation conditions, and batch number. The testing information includes the pathogenic bacteria detected for each aquatic product sample, the testing time, and the concentration value of the test results. The aquatic product sample information and testing information are then processed to standardize the fields and construct a basic dataset of aquatic product pathogenic bacteria. S2. The basic dataset of pathogenic bacteria in aquatic products is structured and organized, and the corresponding relationship between the source of aquatic products, collection time, storage and transportation conditions, batch number, pathogenic bacteria type and detection result concentration value is established. Multidimensional feature records are generated for each individual aquatic product sample, and a multidimensional association data table is formed. S3. Perform co-occurrence correlation analysis on the multidimensional correlation data table. By extracting the co-occurrence records, frequency of occurrence, and corresponding relationship of concentration value changes of each pathogenic bacteria under different aquatic product origins, collection time intervals, and storage and transportation conditions, construct a pathogenic bacteria co-occurrence relationship matrix. Based on the correlation relationship between each pathogenic bacteria species in the pathogenic bacteria co-occurrence relationship matrix, establish a node connection structure to form a multidimensional correlation diagram of pathogenic bacteria. S4. Based on the co-occurrence matrix of pathogenic bacteria and the multidimensional association diagram of pathogenic bacteria, the distribution status of pathogenic bacteria in different aquatic product origins, collection time intervals and storage and transportation conditions is determined. By identifying the association connection sequence and extension direction of pathogenic bacteria between different batches of aquatic products, the transmission path of pathogenic bacteria and high-risk association combinations are determined, and the multidimensional association analysis results of pathogenic bacteria in aquatic products are generated. S5. Output pollution source indication information and risk level information based on the results of multidimensional association analysis of pathogenic bacteria in aquatic products, so as to realize the association identification and risk warning of potential pollution sources of pathogenic bacteria in aquatic products.

2. The multidimensional correlation analysis method for pathogenic bacteria in aquatic products according to claim 1, characterized in that, Step S3 includes: The multidimensional correlation data table was hierarchically grouped according to the aquatic product's origin, collection time interval, storage and transportation conditions, and batch number. The pathogenic bacteria types, detection result concentration values, and corresponding detection times were extracted from each group. Establish paired records for pathogenic bacteria species that appear simultaneously within the same group within a preset time interval, and mark each paired record with a concentration level according to the position of the detection result concentration value within a preset concentration level range; For each pairing record, the number of repeated occurrences, the number of consecutive occurrences, and the number of occurrences across batches are counted to generate a pathogen pairing statistics table; Based on the pathogenic bacteria pairing statistics table, pairing records that appear only once and do not show synchronous changes in concentration level are removed, while pairing records that appear more than twice consecutively and have the same direction of concentration level change are retained, thus generating an effective co-occurrence table of pathogenic bacteria. Using the pathogenic bacteria species in the effective co-occurrence table of pathogenic bacteria as the horizontal and vertical axes, and using the number of repeated occurrences, consecutive occurrences, cross-batch occurrences, and synchronous changes in concentration levels among each pathogenic bacteria species as the corresponding relationship content, a pathogenic bacteria co-occurrence relationship matrix is ​​generated. Based on the correspondence between pathogenic bacteria species in the pathogenic bacteria co-occurrence matrix, each pathogenic bacteria node is hierarchically connected according to the association strength, and the aquatic product origin, collection time interval, and storage and transportation conditions are added to the corresponding connection relationship to form a multidimensional association diagram of pathogenic bacteria.

3. The multidimensional correlation analysis method for pathogenic bacteria in aquatic products according to claim 2, characterized in that, Step S4 includes: Extract the number of repeated occurrences, consecutive occurrences, cross-batch occurrences, and synchronous changes in concentration levels among the pathogenic bacteria in the co-occurrence relationship matrix. Compare the pairing relationships of each pathogenic bacteria in order according to the number of repeated occurrences, consecutive occurrences, cross-batch occurrences, and synchronous changes in concentration levels to determine the primary and secondary pathogenic bacteria pairings. Starting with the pairing of primary pathogenic bacteria, corresponding connections are extracted from the multidimensional association graph of pathogenic bacteria according to the order of collection time to construct the temporal transmission chain of pathogenic bacteria. The changes in the source of aquatic products, storage and transportation conditions and batch numbers of adjacent nodes in the time-series transmission chain of pathogens were compared item by item. The connection relationship where the source of aquatic products is consistent and the batch number changes in the order of collection time is determined to be the homologous transmission relationship. The connection relationship where the storage and transportation conditions are consistent and the collection time interval is continuously progressive is determined to be the circulation transmission relationship. The node positions, concentration level changes, and connection directions of the main associated pathogen pairings of the same pathogen in different batches of aquatic products in the time-series transmission chain of pathogens were merged and sorted to determine the transmission initiation node, transmission relay node, and transmission terminal node. When the same transmission initiation node corresponds to two or more main associated pathogen pairings, and the main associated pathogen pairings extend to different batch numbers, the transmission initiation node is identified as a high-risk pollution source node. Based on the initiation node, relay node, terminal node, homologous transmission relationship, and circulation transmission relationship, the transmission path of pathogens and high-risk association combinations are identified, and the results of multidimensional association analysis of pathogens in aquatic products are generated.

4. The multidimensional correlation analysis method for pathogenic bacteria in aquatic products according to claim 2, characterized in that, The process of generating a pathogen co-occurrence matrix includes: Establish row and column coordinates for each pathogen species in the effective co-occurrence table of pathogens, and write the paired records of any two pathogens in the same group to the corresponding coordinate positions; The number of repetitions, consecutive occurrences, cross-batch occurrences, and synchronous changes in concentration levels in the paired records are written into the record field at the same coordinate position to form pathogenic bacteria paired relationship units; When the same pathogen pairing relationship occurs repeatedly in different aquatic product origins, different collection time intervals, or different storage and transportation conditions, the record fields at the corresponding coordinate positions are accumulated and merged, and the aquatic product origin, collection time interval, storage and transportation conditions, and batch number corresponding to each occurrence are retained. When no pathogenic bacteria pairing record exists at any coordinate position, that coordinate position is marked as a unit with no co-occurrence relationship. A co-occurrence matrix of pathogens is generated by constructing a complete matrix structure based on the row and column coordinates of all pathogenic bacteria species, the paired relationship units of pathogenic bacteria, and the non-co-occurrence relationship units.

5. The multidimensional correlation analysis method for pathogenic bacteria in aquatic products according to claim 2, characterized in that, The process of hierarchically connecting each pathogenic node according to its association strength includes: Extract the number of repeated occurrences, consecutive occurrences, cross-batch occurrences, and synchronous changes in concentration levels for each pathogenic bacteria pairing unit in the pathogenic bacteria co-occurrence matrix; The pairing units of each pathogenic bacterium were screened item by item according to the order of comparison of the number of repeated occurrences, the number of consecutive occurrences, the number of occurrences across batches, and the synchronous changes in concentration levels. The pathogen pairing relationship unit that meets the preset retention conditions for the number of repetitions, the number of consecutive occurrences, the number of occurrences across batches, and the synchronous change of concentration level is determined as the first association strength unit; the pathogen pairing relationship unit that meets only three of the preset retention conditions is determined as the second association strength unit; and the pathogen pairing relationship unit that meets only two of the preset retention conditions is determined as the third association strength unit. Establish first-layer connection relationships, second-layer connection relationships, and third-layer connection relationships based on the first association strength unit, the second association strength unit, and the third association strength unit, respectively; A hierarchical connection structure is formed between pathogenic bacteria nodes according to the first-level connection relationship, the second-level connection relationship, and the third-level connection relationship.

6. The multidimensional correlation analysis method for pathogenic bacteria in aquatic products according to claim 3, characterized in that, The process of identifying pathogen transmission routes and high-risk associated combinations includes: Starting from the transmission initiation node, adjacent connections are continuously extracted along the chronological transmission chain of pathogens, and the source of aquatic products, collection time interval, storage and transportation conditions and batch number corresponding to each connection are extracted. When the types of pathogens in adjacent connections are consistent and the batch numbers change accordingly according to the collection time, the corresponding connections will be merged into the same pathogen transmission chain segment. When the transmission chain of the same pathogen extends and connects through pairing of the main associated pathogen, the end node of the previous transmission chain is spliced ​​with the beginning node of the next transmission chain to form a continuous transmission path. When any propagation chain segment corresponds to both source propagation and circulation propagation relationships, the propagation chain segment is divided into source propagation segment and circulation propagation segment, and written into the continuous propagation path in the order of source propagation segment first and circulation propagation segment second. All continuous propagation paths are summarized at the end, and continuous propagation paths with a common propagation start node are merged into the same propagation path group; When pathogenic bacteria transmission chains in the same transmission path group are combined and statistically analyzed, if there are two or more main associated pathogenic bacteria pairs in the same transmission path group and the main associated pathogenic bacteria pairs extend to different batches of aquatic products, the corresponding pathogenic bacteria combinations are identified as high-risk associated combinations.

7. The method for multidimensional correlation analysis of pathogenic bacteria in aquatic products according to claim 1, characterized in that, Step S5 includes: The transmission initiation node, transmission relay node, transmission terminal node, pathogen transmission path and high-risk association combination were identified from the multidimensional association analysis results of pathogenic bacteria in aquatic products. The source of aquatic products, collection time interval, storage and transportation conditions and batch number corresponding to each transmission path were also extracted. Based on the aquatic product origin and batch number corresponding to the initiation node in each transmission path, different transmission paths are merged and organized to generate a set of pollution source nodes. The number of transmission paths, the number of high-risk association combinations, and the number of consecutive batch numbers corresponding to each pollution source node are counted. The number of transmission paths, the number of high-risk associated combinations, and the number of consecutively appearing batch numbers are compared with preset risk judgment thresholds. When the number of transmission paths is not less than the first threshold, the number of high-risk associated combinations is not less than the second threshold, and the number of consecutively appearing batch numbers is not less than the third threshold, the corresponding pollution source node is determined as a first-level risk node. When the number of transmission paths is not less than the fourth threshold, the number of high-risk associated combinations is not less than the fifth threshold, and the number of consecutively appearing batch numbers is not less than the sixth threshold, the corresponding pollution source node will be identified as a secondary risk node. When the number of transmission paths is not less than the seventh threshold, the number of high-risk associated combinations is not less than the eighth threshold, and the number of consecutively appearing batch numbers is not less than the ninth threshold, the corresponding pollution source node will be identified as a level three risk node. Pollution source alerts are generated based on the risk level of each pollution source node, the corresponding aquatic product origin, storage and transportation conditions, and transmission route information. The pollution source information is associated with the corresponding risk node level to form the risk level information of pathogenic bacteria contamination in aquatic products, and the pollution source information and risk level information are output. Among them, the first threshold is higher than the fourth threshold, the fourth threshold is higher than the seventh threshold, the second threshold is higher than the fifth threshold, the fifth threshold is higher than the eighth threshold, so that the third threshold is higher than the sixth threshold, and the sixth threshold is higher than the ninth threshold.

8. A multidimensional correlation analysis system for pathogenic bacteria in aquatic products, characterized in that, The system, used in any one of claims 1 to 7, comprises: The data acquisition module is used to collect information on aquatic product samples and testing information. The aquatic product sample information includes the origin of the aquatic products, collection time, storage and transportation conditions and batch number. The testing information includes the types of pathogens detected for each aquatic product sample, the testing time and the concentration value of the test results. The module also performs field standardization processing on the aquatic product sample information and testing information to construct a basic dataset of pathogens in aquatic products. The data processing module is used to structure and organize the basic dataset of pathogenic bacteria in aquatic products. It establishes a correspondence between the source of aquatic products, collection time, storage and transportation conditions, batch number, pathogenic bacteria type and detection result concentration value, generates multidimensional feature records for individual aquatic product samples, and forms a multidimensional association data table. The co-occurrence analysis module is used to perform co-occurrence association analysis on the multidimensional association data table. By extracting the co-occurrence records, frequency of occurrence, and corresponding changes in concentration values ​​of pathogenic bacteria under different aquatic product origins, collection time intervals, and storage and transportation conditions, a pathogenic bacteria co-occurrence relationship matrix is ​​constructed. Based on the association relationship between each pathogenic bacteria species in the pathogenic bacteria co-occurrence relationship matrix, a node connection structure is established to form a multidimensional association relationship diagram of pathogenic bacteria. The transmission analysis module is used to determine the distribution status of pathogens in different aquatic product origins, collection time intervals, and storage and transportation conditions based on the pathogen co-occurrence relationship matrix and the pathogen multidimensional association relationship diagram. By identifying the association connection sequence and extension direction of pathogens between different batches of aquatic products, the transmission path of pathogens and high-risk association combinations are determined, and the multidimensional association analysis results of pathogens in aquatic products are generated. The risk output module is used to output pollution source indication information and risk level information based on the results of multidimensional correlation analysis of pathogenic bacteria in aquatic products, so as to realize the correlation identification and risk warning of potential pollution sources of pathogenic bacteria in aquatic products.

9. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images