Search

CN-122020598-A - Multidimensional correlation analysis method and system for pathogenic bacteria in aquatic products

CN122020598ACN 122020598 ACN122020598 ACN 122020598ACN-122020598-A

Abstract

The invention provides a multidimensional association analysis method and a multidimensional association analysis system for pathogenic bacteria in aquatic products, and relates to the technical field of pathogen detection. The method comprises the steps of extracting the corresponding relation among the co-occurrence records, occurrence frequency and concentration change of pathogenic bacteria under different sources, collecting time intervals and storage and transportation conditions, constructing a pathogenic bacteria co-occurrence relation matrix and forming a pathogenic bacteria multidimensional association relation graph, carrying out association judgment on the pathogenic bacteria distribution states in different batches of aquatic products according to the co-occurrence relation matrix and the multidimensional association relation graph, identifying pathogenic bacteria propagation paths and high risk association combinations, generating pathogenic bacteria multidimensional association analysis results, and outputting pollution source prompt information and risk grade information so as to realize identification and risk early warning of potential pollution sources of pathogenic bacteria in the aquatic products.

Inventors

  • JIANG XIAOYING
  • LIN WEIYAN
  • SHEN XIAOFEN
  • CAO SHUANG
  • JIANG FENG

Assignees

  • 厦门中集信检测技术有限公司

Dates

Publication Date
20260512
Application Date
20260407

Claims (10)

  1. 1. The multidimensional correlation analysis method for pathogenic bacteria in aquatic products is characterized by comprising the following steps: S1, acquiring aquatic product sample information and detection information, wherein the aquatic product sample information comprises an aquatic product source place, acquisition time, storage and transportation conditions and batch numbers, the detection information comprises pathogenic bacteria detection types, detection time and detection result concentration values corresponding to each aquatic product sample, and field unified processing is carried out on the aquatic product sample information and the detection information to construct an aquatic product pathogenic bacteria basic data set; s2, carrying out structural arrangement on a basic data set of aquatic product pathogenic bacteria, establishing a corresponding relation among the source place, the collection time, the storage and transportation conditions, the batch number, the pathogenic bacteria types and the concentration value of a detection result, generating a multidimensional characteristic record taking a single aquatic product sample as a unit, and forming a multidimensional associated data table; S3, performing co-occurrence association analysis processing on the multi-dimensional association data table, constructing a pathogenic bacteria co-occurrence relationship matrix by extracting the corresponding relationship among the co-occurrence records, the occurrence frequency and the detection result concentration value change of all pathogenic bacteria types under different aquatic product sources, collection time intervals and storage and transportation conditions, and establishing a node connection structure according to the association among all pathogenic bacteria types in the pathogenic bacteria co-occurrence relationship matrix to form a pathogenic bacteria multi-dimensional association relationship graph; S4, carrying out association judgment on pathogen distribution states under different aquatic product sources, collection time intervals and storage and transportation conditions according to a pathogen co-occurrence relation matrix and a pathogen multidimensional association relation diagram, and determining pathogen propagation paths and high risk association combinations by identifying association connection sequences and extension directions of pathogens among different aquatic product batches to generate an aquatic product pathogen multidimensional association analysis result; And S5, outputting pollution source prompt information and risk level information according to the multidimensional correlation analysis result of the aquatic product pathogenic bacteria so as to realize correlation identification and risk early warning of potential pollution sources of the aquatic product pathogenic bacteria.
  2. 2. The method for multidimensional scaling of pathogenic bacteria in aquatic products according to claim 1, the method is characterized in that the step S3 comprises the following steps: layering and grouping the multidimensional associated data table according to the source places of the aquatic products, the collection time interval, the storage and transportation conditions and the batch numbers, and extracting the pathogen types, the concentration values of detection results and the corresponding detection time in each group; Establishing pairing records of pathogenic bacteria types which simultaneously appear in the same group in a preset time interval, and marking the concentration level of each pairing record according to the position of the concentration value of the detection result in a preset concentration level interval; counting the repeated occurrence times, the continuous occurrence times and the cross-batch occurrence times of each pairing record respectively to generate a pathogenic bacteria pairing statistical table; Rejecting pairing records which only appear once and have no synchronous change of concentration level according to a pathogenic bacteria pairing statistical table, and reserving pairing records which appear more than two times continuously and have consistent change direction of concentration level to generate a pathogenic bacteria effective co-occurrence table; The pathogenic bacteria types in the pathogenic bacteria effective co-occurrence table are taken as the abscissa and the ordinate, and the synchronous change records of the repeated occurrence times, the continuous occurrence times, the cross-batch occurrence times and the concentration level among all the pathogenic bacteria types are taken as the corresponding relation content, so that a pathogenic bacteria co-occurrence relation matrix is generated; And according to the corresponding relation content among all pathogenic bacteria types in the pathogenic bacteria co-occurrence relation matrix, layering and connecting all pathogenic bacteria nodes according to the association strength, and attaching the aquatic product source place, the collection time interval and the storage and transportation condition to the corresponding connection relation to form a pathogenic bacteria multidimensional association relation diagram.
  3. 3. The method for multidimensional scaling of pathogenic bacteria in aquatic products according to claim 2, the method is characterized in that the step S4 comprises the following steps: Extracting the synchronous change records of the repeated occurrence times, the continuous occurrence times, the cross-batch occurrence times and the concentration level among all pathogenic bacteria types in the pathogenic bacteria co-occurrence relation matrix, sequentially comparing the synchronous change records of the repeated occurrence times, the continuous occurrence times, the cross-batch occurrence times and the concentration level of all pathogenic bacteria pairing relations, and determining main associated pathogenic bacteria pairing and secondary associated pathogenic bacteria pairing; the method comprises the steps of taking main associated pathogenic bacteria pairing as a starting point, extracting corresponding connection relations in a pathogenic bacteria multidimensional association relation graph according to acquisition time sequence, and constructing a pathogenic bacteria time sequence transfer chain; The method comprises the steps of comparing the source places of aquatic products, storage and transportation conditions and the change conditions of batch numbers of adjacent nodes in a pathogenic bacteria time sequence transfer chain item by item, determining the connection relation of the source places of the aquatic products which are consistent and the batch numbers which correspondingly change according to the acquisition time sequence as a homologous propagation relation, and determining the connection relation of the storage and transportation conditions which are consistent and the acquisition time interval which continuously advances as a circulation propagation relation; Combining and sorting the node positions, concentration level change sequences and connection directions corresponding to main associated pathogenic bacteria pairing of the same pathogenic bacteria in different aquatic product batches in a pathogenic bacteria time sequence transmission chain, and determining a transmission starting node, a transmission relay node and a transmission tail end node; When the same transmission starting node corresponds to more than two main associated pathogenic bacteria pairs, and the main associated pathogenic bacteria pairs are respectively extended to different batch numbers, determining the transmission starting node as a high-risk pollution source node; And identifying a pathogenic bacteria propagation path and a high risk association combination according to the propagation starting node, the propagation relay node, the propagation end node, the homologous propagation relationship and the circulation propagation relationship, and generating a multidimensional association analysis result of the pathogenic bacteria of the aquatic product.
  4. 4. The method of claim 2, wherein the generating a pathogenic co-occurrence relationship matrix comprises: Establishing row-column corresponding coordinates for each consistent pathogen type in the pathogen effective co-occurrence table, and writing pairing records of any two pathogens in the same group into corresponding coordinate positions; The repeated occurrence times, the continuous occurrence times, the cross-batch occurrence times and the concentration level synchronous change records in the pairing records are respectively and correspondingly written into the record fields of the same coordinate position to form a pathogenic bacteria pairing relation unit; when the same pathogenic bacteria pairing relationship repeatedly appears in different aquatic product sources, different collection time intervals or different storage and transportation conditions, accumulating and merging the record fields of the corresponding coordinate positions, and reserving the aquatic product sources, the collection time intervals, the storage and transportation conditions and the batch numbers which appear corresponding each time; when any coordinate position does not have a pathogen pairing record, marking the coordinate position as a co-occurrence relation unit; And forming a complete matrix structure according to row-column corresponding coordinates of all pathogenic bacteria types, pathogenic bacteria pairing relation units and non-co-occurrence relation units, and generating a pathogenic bacteria co-occurrence relation matrix.
  5. 5. The method of claim 2, wherein the step of hierarchically connecting pathogenic nodes according to the association strength comprises: Extracting the synchronous change records of the repeated occurrence times, the continuous occurrence times, the cross-batch occurrence times and the concentration level corresponding to each pathogenic bacteria pairing relation unit in the pathogenic bacteria co-occurrence relation matrix; Screening each pathogenic bacteria pairing relation unit one by one according to the sequence of the repeated occurrence times, the continuous occurrence times, the cross-batch occurrence times and the sequence of the synchronous change records of the concentration levels; Determining pathogenic bacteria pairing relation units which meet preset reservation conditions in the synchronous change records of the repeated occurrence times, the continuous occurrence times, the cross-batch occurrence times and the concentration level as first association strength units, determining pathogenic bacteria pairing relation units which meet the preset reservation conditions as second association strength units, and determining pathogenic bacteria pairing relation units which meet the preset reservation conditions as third association strength units; respectively establishing a first layer connection relation, a second layer connection relation and a third layer connection relation according to the first association strength unit, the second association strength unit and the third association strength unit; And forming a layered connection structure among pathogen nodes according to the first layer connection relation, the second layer connection relation and the third layer connection relation.
  6. 6. A method of multidimensional association analysis of pathogenic bacteria in an aquatic product as claimed in claim 3 wherein the process of identifying a combination of pathogen propagation paths and high risk associations comprises: Continuously extracting adjacent connection relations along the acquisition time sequence in the pathogenic bacteria time sequence transmission chain by taking a transmission starting node as a starting point, and extracting aquatic product source places, acquisition time intervals, storage and transportation conditions and batch numbers corresponding to the connection relations; when the types of pathogenic bacteria in adjacent connection relations are kept consistent and the batch numbers correspondingly change according to the acquisition time sequence, merging the corresponding connection relations into the same pathogenic bacteria transmission chain segment; when the same pathogenic bacteria propagation chain segments are in extension connection through the main associated pathogenic bacteria pairing, end nodes of the former propagation chain segments and initial nodes of the latter propagation chain segments are spliced end to form a continuous propagation path; when any propagation chain segment corresponds to the homologous propagation relationship and the circulation propagation relationship, dividing the propagation chain segment into a source propagation segment and a circulation propagation segment, and writing the source propagation segment and the circulation propagation segment into a continuous propagation path according to the sequence of the source propagation segment and the circulation propagation segment; terminal summarizing is carried out on all continuous propagation paths, and the continuous propagation paths with common propagation starting nodes are merged into the same propagation path group; and carrying out combination statistics on pathogenic bacteria propagation chain segments in the same propagation path group, and determining corresponding pathogenic bacteria combinations as high-risk associated combinations when more than two main associated pathogenic bacteria pairs exist in the same propagation path group and the main associated pathogenic bacteria pairs are respectively extended to different aquatic product batches.
  7. 7. The method for multidimensional scaling of pathogenic bacteria in aquatic products according to claim 1, the method is characterized in that the step S5 comprises the following steps: Extracting a propagation starting node, a propagation relay node, a propagation tail end node, a pathogenic bacteria propagation path and a high risk association combination which are identified in a multidimensional association analysis result of the aquatic product pathogenic bacteria, and extracting an aquatic product source place, an acquisition time interval, storage and transportation conditions and a batch number which correspond to each propagation path; Merging and sorting different propagation paths according to the source places and batch numbers of aquatic products corresponding to the propagation starting nodes in each propagation path to generate a pollution source node set, and counting the number of propagation paths, the number of high-risk association combinations and the number of batch numbers continuously appearing corresponding to each pollution source node; Respectively comparing the number of the propagation paths, the number of the high-risk association combinations and the number of the batch numbers which continuously appear with a preset risk judgment threshold, and determining the corresponding pollution source node as a primary risk node when the number of the propagation paths is not smaller than a first threshold, the number of the high-risk association combinations is not smaller than a second threshold and the number of the batch numbers which continuously appear is not smaller than a third threshold; When the number of the propagation paths is not less than a fourth threshold value, the number of the high-risk association combinations is not less than a fifth threshold value, and the number of the batch numbers which continuously appear is not less than a sixth threshold value, determining the corresponding pollution source node as a secondary risk node; When the number of propagation paths is not less than a seventh threshold, the number of high-risk association combinations is not less than an eighth threshold, and the number of batch numbers appearing continuously is not less than a ninth threshold, determining the corresponding pollution source node as a three-level risk node; Generating pollution source prompt information according to the risk node grade corresponding to each pollution source node, the corresponding aquatic product source place, the storage and transportation conditions and the propagation path information; carrying out association labeling on the pollution source prompt information and the corresponding risk node grade to form aquatic product pathogenic bacteria pollution risk grade information, and outputting the pollution source prompt information and the risk grade information; Wherein 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.
  8. 8. A system for multidimensional association analysis of pathogenic bacteria in an aquatic product, for use in a method as claimed in any one of claims 1 to 7, the system comprising: The data acquisition module is used for acquiring aquatic product sample information and detection information, wherein the aquatic product sample information comprises an aquatic product source place, acquisition time, storage and transportation conditions and batch numbers, the detection information comprises pathogenic bacteria detection types, detection time and detection result concentration values corresponding to each aquatic product sample, and field unified processing is carried out on the aquatic product sample information and the detection information to construct an aquatic product pathogenic bacteria basic data set; the data arrangement module is used for carrying out structural arrangement on the basic data set of aquatic product pathogenic bacteria, establishing a corresponding relation among the source place, the collection time, the storage and transportation conditions, the batch number, the pathogenic bacteria type and the concentration value of the detection result, generating a multidimensional characteristic record taking a single aquatic product sample as a unit, and forming a multidimensional associated data table; The co-occurrence analysis module is used for carrying out co-occurrence association analysis processing on the multi-dimensional association data table, constructing a pathogenic bacteria co-occurrence relation matrix by extracting the corresponding relation of the co-occurrence record, the occurrence frequency and the detection result concentration value change of each pathogenic bacteria type under the condition of different aquatic product sources, collection time intervals and storage and transportation, and establishing a node connection structure according to the association relation among all pathogenic bacteria types in the pathogenic bacteria co-occurrence relation matrix to form a pathogenic bacteria multi-dimensional association relation graph; The transmission analysis module is used for carrying out association judgment on pathogen distribution states under different aquatic product sources, acquisition time intervals and storage and transportation conditions according to the pathogen co-occurrence relation matrix and the pathogen multidimensional association relation graph, and determining pathogen transmission paths and high risk association combinations by identifying the association connection sequence and the extension direction of the pathogens among different aquatic product batches to generate an aquatic product pathogen multidimensional association analysis result; The risk output module is used for outputting pollution source prompt information and risk grade information according to the multidimensional association analysis result of the aquatic product pathogenic bacteria so as to realize association identification and risk early warning of the potential pollution sources of the aquatic product pathogenic bacteria.
  9. 9. A computing device, comprising: one or more processors; Storage means for storing one or more programs which when executed by the one or more processors cause the one or more processors to implement the method of any of claims 1 to 7.
  10. 10. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a program which, when executed by a processor, implements the method according to any of claims 1 to 7.

Description

Multidimensional correlation analysis method and system for pathogenic bacteria in aquatic products Technical Field The invention relates to the technical field of pathogen detection, in particular to a multidimensional correlation analysis method and a multidimensional correlation analysis system for pathogenic bacteria in aquatic products. Background At present, in the field of detection and analysis of pathogenic bacteria in aquatic products, a common technical scheme is mainly realized based on a microorganism culture combined with molecular biology detection method. Specifically, pretreatment and enrichment culture are carried out on aquatic product samples to increase the number of possible pathogenic bacteria, then separation culture is carried out through a selective culture medium, and then detection and confirmation are carried out on target pathogenic bacteria by using technologies such as biochemical identification, PCR amplification or real-time fluorescence quantitative PCR. Meanwhile, some researches can also combine detection results to carry out simple statistical analysis on different sample sources, detection time and strain types, so as to evaluate the distribution condition and pollution level of common pathogenic bacteria in the aquatic products and realize basic monitoring on the safety condition of the aquatic products. However, in the actual aquatic product circulation or batch quality detection scenario, the above technology usually performs detection and statistics only for a single index or a single strain. For example, in the batch sampling process of aquatic products, a detector often detects different pathogenic bacteria such as vibrio parahaemolyticus and salmonella separately, and makes simple statistical judgment according to a single detection result. The method is difficult to comprehensively analyze the association relation among multidimensional data such as detection time, production place environment, strain type and pollution degree, and when cross pollution occurs to a batch of aquatic products under different time or different source conditions, the existing method is difficult to identify potential association modes in time, and inaccurate pollution source judgment is easily caused, so that subsequent risk assessment and supervision decision are influenced. Disclosure of Invention The invention aims to provide a multidimensional correlation analysis method and a multidimensional correlation analysis system for pathogenic bacteria in aquatic products, and aims to solve the problems in the background technology. In order to solve the technical problems, the technical scheme of the invention is as follows: in a first aspect, a method for multidimensional association analysis of pathogenic bacteria in an aquatic product, the method comprising: S1, acquiring aquatic product sample information and detection information, wherein the aquatic product sample information comprises an aquatic product source place, acquisition time, storage and transportation conditions and batch numbers, the detection information comprises pathogenic bacteria detection types, detection time and detection result concentration values corresponding to each aquatic product sample, and field unified processing is carried out on the aquatic product sample information and the detection information to construct an aquatic product pathogenic bacteria basic data set; s2, carrying out structural arrangement on a basic data set of aquatic product pathogenic bacteria, establishing a corresponding relation among the source place, the collection time, the storage and transportation conditions, the batch number, the pathogenic bacteria types and the concentration value of a detection result, generating a multidimensional characteristic record taking a single aquatic product sample as a unit, and forming a multidimensional associated data table; S3, performing co-occurrence association analysis processing on the multi-dimensional association data table, constructing a pathogenic bacteria co-occurrence relationship matrix by extracting the corresponding relationship among the co-occurrence records, the occurrence frequency and the detection result concentration value change of all pathogenic bacteria types under different aquatic product sources, collection time intervals and storage and transportation conditions, and establishing a node connection structure according to the association among all pathogenic bacteria types in the pathogenic bacteria co-occurrence relationship matrix to form a pathogenic bacteria multi-dimensional association relationship graph; S4, carrying out association judgment on pathogen distribution states under different aquatic product sources, collection time intervals and storage and transportation conditions according to a pathogen co-occurrence relation matrix and a pathogen multidimensional association relation diagram, and determining pathogen propagation paths and high risk asso