Biological detection data analysis method

Abstract

The invention relates to a biological detection data analysis method, which belongs to the technical field of data analysis, and comprises the following steps: obtaining inspection data, generating a unique sample number, calling a standard template and matching a standard detection item, generating a sample preparation guide, receiving a triggered preparation confirmation instruction, obtaining a sample preparation completion instruction, and triggering a sample detection instruction. The method comprises the following steps: acquiring original data of equipment, judging according to a judgment standard to obtain a detection result, calling a multi-dimensional data analysis engine to perform statistical analysis on the detection result based on the detection result, and calculating an epidemic disease positive rate and comparing the epidemic disease positive rate with a preset positive rate reference range to obtain an analysis result, so that intelligent upgrading in the field of biological detection is realized; the traceability and the processing normalization of sample information are ensured, the problems of large personal error and data analysis lagging in the traditional detection process are effectively solved, and the comprehensive purpose of improving the detection accuracy, the detection efficiency and the risk early warning capability is achieved.

Description

Technical Field

[0001] This invention relates to the field of data analysis technology, and more specifically to a method for analyzing biological detection data. Background Technology

[0002] In the field of biological testing, the real-time, accurate analysis and visualization of testing data are of paramount importance for improving the risk management level of the agricultural industry. With the in-depth advancement of the Digital China strategy, the industry has placed higher demands on the depth of data mining, response speed and decision support capabilities of testing data, making it imperative to promote the development of testing technology towards intelligence and digitalization.

[0003] Currently, existing technologies mainly rely on process management backends, such as Laboratory Information Management Systems (LIMS), which can realize basic business such as sample registration and task allocation; or lightweight query applications based on platforms such as WeChat official accounts, which support single-dimensional result display.

[0004] However, existing technologies still rely on manual data entry for equipment, which is inefficient and prone to errors, and lacks data authenticity and end-to-end traceability capabilities. Furthermore, the data analysis dimensions are limited, making it difficult to achieve multi-dimensional integrated analysis of time, space, and attributes, thus hindering the acquisition of in-depth insights. At the same time, there are also issues with poor data visualization, with most data consisting of static tables and lacking interactive exploration functions. This makes it impossible to support users' independent drill-down and real-time decision-making, resulting in prominent data silos and making it difficult to form a business closed loop of detection, analysis, and application. Summary of the Invention

[0005] In view of the shortcomings of related technologies, the purpose of this invention is to provide a biological detection data analysis method to solve the technical problems of existing technologies, which rely on manual data entry, have a single analysis dimension and poor visualization interactivity, resulting in difficulties in data traceability, superficial insights and inability to support real-time decision-making, thus forming information silos between detection and application.

[0006] This invention provides a method for analyzing biological detection data, comprising the following steps: Data acquisition steps: Acquire data based on the original sample to be tested, and generate a unique sample number for the data; wherein, the data includes basic information, sample information, and testing requirements; Sample preparation confirmation steps: Retrieve the standard template from the preset sample preparation standard library based on the sample number; Match the standard test item from the preset test item library based on the test requirements; Generate a sample preparation guide based on the standard test item and the standard template; Receive a triggered preparation confirmation instruction based on the sample preparation guide; Obtain a sample preparation completion instruction based on the preparation confirmation instruction. Sample testing and confirmation steps: Based on the sample preparation completion instruction, a sample testing instruction is triggered. Based on the sample testing instruction, the raw equipment data generated by the test equipment during the testing process is obtained. According to the testing items in the testing requirements, the corresponding testing judgment criteria are retrieved from the preset testing result judgment rule library. The raw equipment data is judged according to the judgment criteria to obtain the testing result. Detection result analysis steps: Call the preset multi-dimensional data analysis engine to perform statistical analysis on the detection results, calculate the disease positivity rate and compare it with the preset positivity rate benchmark range to obtain the analysis results.

[0007] By establishing a fully digitalized process from data acquisition to test result analysis, and utilizing the unique identifier of the sample number and standard template matching technology, the traceability of sample information and the standardization of processing have been achieved. By integrating the scattered testing processes into a coherent automated workflow, the problems of data silos and inconsistent operations in traditional methods are solved, ultimately improving testing efficiency and result reliability.

[0008] In some embodiments of the present invention, the step of acquiring the submitted data specifically includes: Based on a preset classification rule base, classification labels are assigned to the original test samples corresponding to the submitted test data, and the sample information is compared with the test requirements to determine if there are any mismatches. If there is no mismatch between the sample information and the detection requirements, the sample status of the original sample to be tested will be updated to "to be prepared". Otherwise, generate a warning message.

[0009] By automating the classification and matching of submitted data in the backend, intelligent auditing of the consistency between sample information and testing requirements is achieved. It can automatically identify unreasonable submission requests and generate warnings in the early stages of the process, thereby effectively avoiding the waste of subsequent testing resources and result deviations caused by the mismatch between samples and projects. This achieves the purpose of proactive risk prevention and control and ensuring the standardization of testing operations.

[0010] In some embodiments of the present invention, the sample preparation and verification step specifically includes: Based on the classification label on the original test sample corresponding to the sample number, the standard template in the preset sample preparation standard library is retrieved; wherein, the standard template includes the preparation process, preparation environment requirements and sample number calculation rules corresponding to the original test samples with different classification labels. Sample preparation guidelines are generated based on the standard testing items and the standard template; Based on the sample preparation guidelines, a preparation confirmation instruction triggered after each preparation operation is completed is received, and a sample preparation completion instruction is obtained according to the preparation confirmation instruction. Based on the sample preparation completion instruction, a preparation record is generated, and the sample status of the prepared sample is updated to be tested.

[0011] By dynamically generating standardized preparation guidelines based on classification labels and driving the process forward with a step-by-step confirmation mechanism, refined management of the sample preparation process has been achieved. Transforming manual operations into standardized instruction flows effectively reduces operational arbitrariness and ensures the controllability and traceability of the preparation process.

[0012] In some embodiments of the present invention, the sample detection and confirmation step further includes: Based on the aforementioned testing requirements, the weights are set for the matching degree of departmental expertise, the current workload of the department, the skill proficiency of testing personnel, and the availability of testing equipment. The task allocation score is obtained by weighting the department's professional matching degree, the department's current task load, the testing personnel's skill proficiency, and the availability of the testing equipment using a multi-factor weighting allocation method. Based on the task allocation score, target departments and target testing personnel are assigned to the prepared samples.

[0013] By comprehensively evaluating the department's professional matching degree, task load, personnel skills and equipment status through a multi-factor weight allocation algorithm, intelligent scheduling of testing tasks has been achieved. By quantifying the status of multidimensional resources as the basis for allocation, the subjectivity and inefficiency of manual allocation are solved, thereby achieving the goal of detecting optimized resource allocation and improving task execution efficiency.

[0014] In some embodiments of the present invention, the sample detection and confirmation step specifically includes: Acquire the equipment data generated by the test equipment during the testing process, and encapsulate the equipment data in a unified format to obtain the original equipment data; The original data of the equipment is cleaned, verified and standardized to obtain test data. The test data is then associated and bound with the sample number and the test items in the test requirements to generate a data traceability chain for the equipment. Record the start-up and stop times of the test equipment during the testing process and generate an instrument usage log.

[0015] By automatically collecting raw data from equipment through standardized interfaces and performing cleaning, verification, and standardization processing, a data traceability chain is generated. This achieves fully automated data collection and accurate association from the equipment end to the information system, eliminating manual input errors and ensuring the authenticity and integrity of the data. This solves the problems of easy tampering and difficulty in tracing test data, and achieves the goal of providing a data foundation for the credibility of test results.

[0016] In some embodiments of the present invention, the sample detection and confirmation step further includes: The detection data is input into a pre-trained machine learning decision model to obtain the model detection result, and the model detection result is compared with the detection result. If the model detection result is consistent with the detection result, then the output model detection result is the final detection result; Otherwise, the test data will be marked as pending manual review, and the final test result will be obtained through manual judgment by the testing personnel.

[0017] By introducing machine learning models and comparing and verifying them with the judgment results of the rule base, auxiliary judgments can be provided in complex situations that cannot be covered by the rules, and manual intervention can be triggered when the results conflict. This improves the accuracy and reliability of judgments for complex detection items, and achieves the goal of making up for the limitations of single rule judgments and ensuring the scientific nature of the results.

[0018] In some embodiments of the present invention, the method further includes a consumables management step: In the sample preparation verification step and / or sample testing verification step, the consumable requirements are determined based on the standard template and / or the standard testing items, and the consumable type and quantity are selected and registered according to the consumable requirements. Based on the type and quantity of consumables, a consumable usage record associated with the sample number is generated, and the consumable inventory data is updated according to the consumable usage record.

[0019] By linking the use of consumables with specific sample numbers and standard procedures and automatically updating the inventory, refined and traceable management of consumables usage is achieved. It can accurately record the consumption of consumables for each test and link with the inventory system, thereby solving the problems of extensive consumable management and difficulty in traceability, and achieving the goal of reducing costs and making consumables fully controllable and traceable.

[0020] In some embodiments of the present invention, the method further includes a report generation step: Based on the final test results, an initial test report is generated; Based on the initial test report, the test items, the final test results, the equipment data traceability chain, the instrument usage log, and the consumable usage record are subject to multi-level review. After multi-level review and approval, a final test report is generated based on the preset report template library and the initial test report.

[0021] By conducting multi-level reviews of the test results and related traceability data chains, the rigor and authority of the generated test reports are ensured. Through multi-level and multi-element cross-reviews, a complete quality assurance system is formed, thereby solving the problem of report review becoming a mere formality and achieving the goal of maximizing the quality and legal validity of the final test reports.

[0022] In some embodiments of the present invention, the detection result analysis step further includes: The test results or the final test results are statistically analyzed to calculate the overall positive rate of the epidemic, the proportion of each epidemic disease detected, and the positive rate of the main epidemic diseases within a preset time period, and a statistical result set is obtained. Based on the pre-selected dimension combination and the statistical result set, the detection results or the final detection results and their associated sample information and basic information are subjected to joint queries and data aggregation through SQL multi-table join and data aggregation algorithms to obtain cross-analysis results. If the overall positive rate of the disease within the preset statistical period exceeds the preset positive rate benchmark range, the data of the overall positive rate of the disease within the preset statistical period will be marked as abnormal data, and an abnormal analysis report will be generated. The analysis results are obtained based on the statistical result set, the cross-analysis results, and the anomaly analysis report.

[0023] By conducting multi-dimensional statistical analysis, cross-analysis, and in-depth mining of anomaly warnings on the final test results, the transformation from single test data to macro-level decision support has been achieved. It can proactively reveal hidden trends, correlations, and risks in the data, thereby solving the problem of low utilization rate of massive test data and achieving the goal of providing users with forward-looking insights and precise prevention and control basis.

[0024] In some embodiments of the present invention, the method further includes a data display step: Using the ECharts chart component, different data types in the statistical results set are matched with corresponding charts for display, resulting in multiple types of charts. Based on the overall positive rate of the disease and the place of origin in the sample information, the national and regional distribution of the target disease type is displayed using map software SDK and PostGIS spatial database, resulting in a national distribution map and a regional distribution map. A multi-level drill-down system is constructed based on the multiple types of charts, the national distribution map, and the regional distribution map. The multiple types of charts, the national distribution map, and the regional distribution map are dynamically associated according to the multi-level drill-down system.

[0025] By integrating diverse display methods such as charts, maps, and interactive drill-down, complex data analysis results are transformed into intuitive and explorable visual information, lowering the threshold for data interpretation and solving the problems of traditional reports being unintuitive and lacking in depth. This greatly enhances the user's data analysis experience and decision-making efficiency. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Figure 1 A flowchart of a biological detection data analysis method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of a structured data collection form provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of another structured data collection form provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of another structured data collection form provided in an embodiment of the present invention; Figure 5 A schematic diagram illustrating the trend of positive rates for major infectious diseases over the past three years, provided as an embodiment of the present invention; Figure 6 This is a schematic diagram illustrating the detection of different strains of an epidemic disease, provided by an embodiment of the present invention. Figure 7 This is a schematic diagram of the overall positive rate of epidemic diseases nationwide in a multi-level data drill-down system provided in an embodiment of the present invention; Figure 8 This is a schematic diagram illustrating the overall positive rate of disease in a certain region within a multi-level data drill-down system provided in an embodiment of the present invention. Figure 9 Provided for embodiments of the present invention Figure 8 A diagram showing the detailed list of tested samples from the corresponding region. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this application clearer, the application is described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. All other embodiments obtained by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application. It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations according to this application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. Furthermore, it should be understood that the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such process, method, product, or apparatus. Currently, there are two main types of information system solutions in the field of biological testing: one is the process management system represented by the laboratory information management system, which can realize basic functions such as sample registration, task allocation, and report generation. However, such systems are mostly PC-based, focusing on back-end business control and lacking front-end data statistical analysis capabilities for users. Users need to export data through the back-end and then use third-party tools such as Excel and SPSS for analysis, which is cumbersome and has poor timeliness. Another type is lightweight applications that focus on data display, such as the data query function on WeChat official accounts developed by some testing institutions. These functions can only display single-dimensional test results, such as whether a single sample is qualified. They cannot achieve multi-dimensional statistics such as the trend of disease positivity rate and variety distribution, and they do not support data drill-down analysis such as drilling down from the national distribution to the city-level data.

[0028] From a technical perspective, existing technologies have the following prominent problems: First, the equipment has weak data interaction capabilities. Most systems still rely on manual input of test equipment data, which is not only inefficient but also prone to data errors and cannot guarantee data authenticity. Manual input of test equipment data is prone to data tampering and errors, and lacks an automatic correlation mechanism for equipment usage time and consumable usage records, making it impossible to achieve full-chain traceability of data, equipment, and consumables. Once a dispute arises over the test results, it is difficult to trace the root cause of the problem, which does not meet the data traceability requirements in the "Accreditation Review Criteria for Inspection and Testing Institutions". Secondly, the data analysis is limited to a single dimension, only capable of basic numerical statistics, lacking multi-dimensional integrated analysis of time, space, and attribute dimensions; The existing system requires manual data export for secondary analysis, which cannot achieve real-time statistical analysis of the detection data. Furthermore, the analysis dimension is limited to a single indicator and cannot integrate multi-dimensional data such as time, space, and attributes for in-depth mining. This makes it impossible for users to quickly grasp the epidemic pattern and formulate precise prevention and control strategies. Third, the visualization is poor, with data mostly displayed in tabular form, lacking intuitive charts and not supporting interactive operations. Users cannot filter or drill down data according to their needs, making it difficult to quickly obtain useful information. The tabular display method cannot intuitively present data trends such as the changes in the positive rate of diseases over the past three years, or the spatial distribution of a certain disease across provinces and cities nationwide. Furthermore, it does not support data drill-down operations. If users need to view the city-level distribution data of a certain disease, they need to re-filter and query, which is cumbersome and does not conform to the digital product design concept of prioritizing user experience. It also cannot meet the management needs of regulatory authorities to keep abreast of regional disease dynamics in real time.

[0029] Furthermore, existing technologies mostly involve one-way data exchange between the front-end and back-end. The front-end cannot synchronize dynamic data such as the detection progress and review results of the back-end in real time, which makes it impossible for users to obtain the detection status in a timely manner. Moreover, the detection data of the back-end cannot be quickly fed back to the front-end for statistical analysis, forming data silos and failing to achieve closed-loop management of detection, analysis, and application.

[0030] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other. The technical solution of the present invention will be described in detail below with reference to specific embodiments and accompanying drawings.

[0031] like Figure 1 As shown, the present invention provides a method for analyzing biological detection data, comprising the following steps: Step S1 for obtaining test data: Obtain test data based on the original sample to be tested, and generate a unique sample number for the test data; wherein, the test data includes basic information, sample information and testing requirements; In some embodiments, step S1 of acquiring test data specifically includes: Based on a preset classification rule base, classification labels are assigned to the original samples to be tested corresponding to the submitted data, and the sample information is compared with the testing requirements to see if there are any mismatches. If there is no mismatch between the sample information and the detection requirements, the sample status of the original sample to be tested will be updated to "to be prepared". Otherwise, generate a warning message.

[0032] By automating the classification and matching of submitted data in the backend, intelligent auditing of the consistency between sample information and testing requirements is achieved. It can automatically identify unreasonable submission requests and generate warnings in the early stages of the process, thereby effectively avoiding the waste of subsequent testing resources and result deviations caused by the mismatch between samples and projects. This achieves the purpose of proactive risk prevention and control and ensuring the standardization of testing operations.

[0033] Furthermore, such as Figures 2 to 4 As shown, a structured data collection form is designed on the front end of the mini-program. Based on the original sample to be tested, the submitted data is obtained through guided filling and intelligent verification mechanisms to ensure the completeness and accuracy of the user's submitted data. The submitted data includes basic information, sample information, and testing requirements; The basic information includes the user name, contact information, and submission date; specifically, the place of origin can be selected through a three-level drop-down menu linking "province, city, and county". Sample information includes sample name, variety, origin, quantity, and storage conditions; Testing requirements include the target disease type, testing items, and report delivery method; specifically, the testing items offer multiple selection options, allowing users to add custom testing requirements.

[0034] Before submitting the acquired test data, it is first intelligently verified by preset intelligent inspection rules. If the inspection passes, it can be submitted to the backend of the mini program; if the inspection fails, the error reason is displayed in real time to ensure that the data meets the backend processing standards. Optionally, the preset intelligent inspection rules are: the number of samples must be greater than 0, at least one test item must be selected, and the contact information format must conform to the mobile phone number or email address specifications.

[0035] After the data is submitted for testing, a unique sample number is generated using an event-driven model. The sample number consists of the user ID, the submission date, and a random 6-digit number. The sample number and its corresponding test data are encapsulated in JSON format, encrypted, and then pushed to the backend of the mini-program. The backend of the mini-program will verify the data integrity and signature validity of the submitted data. After verification, the submitted data will be synchronized to the MySQL database. Automatically assign classification labels to the original test samples corresponding to the submitted test data based on a preset classification rule base; Optionally, the preset classification rule base includes classification by target disease type into viral diseases, bacterial diseases, and parasitic diseases, and classification by species into blood samples, tissue samples, and secretion samples; At the same time, compare the sample information with the testing requirements to see if there are any mismatches; for example, whether the poultry samples should be tested for avian influenza. If there is a mismatch between the sample information and the testing requirements, an early warning message will be generated and automatically pushed to the back-end staff; If there is no mismatch between the sample information and the testing requirements, the sample status of the original sample to be tested will be updated to "to be prepared". Back-end staff can view the list of samples to be verified and the data of the original samples to be tested. After verifying that everything is correct, they can click to confirm the transfer.

[0036] Sample preparation confirmation step S2: Retrieve the standard template from the preset sample preparation standard library based on the sample number; match the standard test items from the preset test item library based on the test requirements; generate sample preparation guidelines based on the standard test items and the standard template; receive the triggered preparation confirmation instruction based on the sample preparation guidelines; and obtain the sample preparation completion instruction based on the preparation confirmation instruction. In some embodiments, the sample preparation verification step S2 specifically includes: Based on the classification label on the original test sample corresponding to the sample number, the standard template in the preset sample preparation standard library is retrieved; the standard template includes the preparation process, preparation environment requirements and sample number calculation rules corresponding to the original test samples with different classification labels. Sample preparation guidelines are generated based on standard testing items and standard templates; Based on the sample preparation guide, receive the preparation confirmation instruction triggered after each preparation operation is completed, and obtain the sample preparation completion instruction according to the preparation confirmation instruction; Based on the sample preparation completion instruction, a preparation record is generated, and the sample status of the prepared sample is updated to be tested.

[0037] By dynamically generating standardized preparation guidelines based on classification labels and driving the process forward with a step-by-step confirmation mechanism, refined management of the sample preparation process has been achieved. Transforming manual operations into standardized instruction flows effectively reduces operational arbitrariness and ensures the controllability and traceability of the preparation process.

[0038] Furthermore, the standard templates in the preset sample preparation standard library include the preparation process, preparation environment requirements, and sample number calculation rules corresponding to the original test samples with different classification labels; Optionally, the preparation process may involve centrifuging to separate serum from blood samples and grinding and homogenizing tissue samples. The preparation environment requirements may include temperature and humidity; The formula for calculating the number of samples is: number of test items × 2 + 1 spare sample; After selecting a sample number, the system can automatically retrieve the corresponding standard template based on the classification label on the original sample to be tested, and display the preparation steps and parameters. Based on the testing requirements in the submitted data, the corresponding standard testing items are matched from the preset testing item library; for example, if the testing requirement is swine fever testing, the two basic items of swine fever virus nucleic acid testing and swine fever antibody testing are automatically bound. Back-end staff can add or delete testing items according to the actual situation, and the required number of sample backups will be automatically updated after modification; Sample preparation guidelines are generated based on standard templates and standard testing items. These guidelines are displayed on the mini-program interface, allowing users to prepare samples according to the guidelines and confirm each preparation step on the interface. After the user clicks the confirmation button, they will receive a preparation confirmation instruction, and finally receive a sample preparation completion instruction based on the preparation confirmation instruction. After the sample preparation is completed, the back-end staff clicks the preparation confirmation on the back-end management terminal to generate a preparation record and update the sample status of the prepared sample to be tested. Optionally, the preparation record includes the personnel involved in the preparation, the preparation time, the type of consumables, and the quantity of consumables.

[0039] Sample testing and confirmation step S3: Based on the sample preparation completion instruction, trigger the sample testing instruction, obtain the original equipment data generated by the test equipment during the testing process based on the sample testing instruction, retrieve the corresponding testing judgment standard from the preset test result judgment rule library according to the test items in the test requirements, judge the original equipment data according to the judgment standard, and obtain the test result.

[0040] In some embodiments, the sample detection and verification step S4 further includes: Weights are set based on the matching degree of department specialties to testing needs, the current workload of the department, the skill proficiency of testing personnel, and the availability of testing equipment. The multi-factor weighting method was used to assign weights to the department's professional matching degree, the department's current workload, the skill proficiency of the testing personnel, and the availability of the testing equipment. The task allocation score was obtained by weighted calculation. Based on the task allocation score, target departments and target testing personnel are assigned to the prepared samples.

[0041] By comprehensively evaluating the department's professional matching degree, task load, personnel skills and equipment status through a multi-factor weight allocation algorithm, intelligent scheduling of testing tasks has been achieved. By quantifying the status of multidimensional resources as the basis for allocation, the subjectivity and inefficiency of manual allocation are solved, thereby achieving the goal of detecting optimized resource allocation and improving task execution efficiency.

[0042] Furthermore, the input parameters of the multi-factor weight allocation algorithm include the department's professional matching degree, the department's current task matching, the skill proficiency of the testing personnel, and the availability of the testing equipment; The calculation method for departmental professional matching degree is as follows: Based on the preset departmental professional mapping library, classification labels are assigned to departments, and the departmental professional matching degree is calculated according to the matching of the target disease type in the testing needs with the department's classification label; for example, viral disease testing is preferentially assigned to molecular biology testing departments; The current workload of a department is calculated as follows: (Number of tasks received / Department's maximum capacity) × 100%; The method for calculating the skill proficiency of testing personnel is as follows: obtain the historical testing data of testing personnel, which includes the historical testing accuracy rate and completion efficiency score of testing personnel, and calculate the skill proficiency of testing personnel based on the historical testing accuracy rate and completion efficiency score; A multi-factor weighting algorithm was used to assign weights to the department's professional matching degree, the department's current workload, the skill proficiency of the testing personnel, and the availability of the testing equipment. Optionally, the weight of departmental professional matching degree is set to 40%; the weight of the current workload of the department is set to 30%; the weight of the skill proficiency of the testing personnel is set to 20%; and the availability of the testing equipment is set to 10%. The task allocation score is obtained by weighted calculation. The department and testing personnel with the highest task allocation scores were selected as the target department and target testing personnel; The prepared samples to be tested are allocated to the target departments and target testing personnel. Task reminders are pushed to the target testing personnel through the service notification of the mini program. The task reminders include the sample number, the test items, and the completion deadline. Meanwhile, the task status can be updated in the backend management terminal, and department heads can manually adjust task assignments; for example, cross-department scheduling can be carried out in special circumstances.

[0043] In some embodiments, the sample detection and verification step S3 further includes: Acquire the equipment data generated by the test equipment during the testing process, and encapsulate the equipment data in a unified format to obtain the raw equipment data; The raw equipment data is cleaned, verified and standardized to obtain test data. The test data is then associated with the sample number and the test items in the test requirements to generate a data traceability chain for the equipment. Record the start-up and stop times of the testing equipment during the testing process and generate an instrument usage log.

[0044] By automatically collecting raw data from equipment through standardized interfaces and performing cleaning, verification, and standardization processing, a data traceability chain is generated. This achieves fully automated data collection and accurate association from the equipment end to the information system, eliminating manual input errors and ensuring the authenticity and integrity of the data. This solves the problems of easy tampering and difficulty in tracing test data, and achieves the goal of providing a data foundation for the credibility of test results.

[0045] Furthermore, a unified interface protocol should be developed for different types of testing equipment; For example, the balance uses the RS485 protocol to transmit weight data, and the real-time PCR instrument transmits detection data such as fluorescence intensity value and Ct value through an interface; The raw data from the equipment is packaged in the format of equipment number, collection time, data value, and data status. Data is collected every 100ms to ensure the real-time nature of the data. After obtaining the raw equipment data, the raw equipment data is parsed and verified to obtain the test data. The test data is stored in the database and associated with its corresponding sample number and the test items in the test requirements to generate the equipment data traceability chain. The equipment data traceability chain includes the sample number, test item, equipment number, collection time and data value. Optionally, the operation of parsing and verifying the original data of the equipment includes removing outliers that exceed a reasonable range and using linear interpolation to complete missing data; among which, outliers include negative balance weights; During the testing process, the instrument usage time is automatically recorded, starting from the start of the equipment and stopping when the testing task is completed, generating an instrument usage log for instrument maintenance reminders and usage efficiency analysis. The instrument usage log includes the user, usage duration, and number of samples tested, and is used for instrument maintenance reminders and usage efficiency analysis.

[0046] In some embodiments, the sample detection and verification step S3 further includes: The detection data is input into a pre-trained machine learning decision model to obtain the model detection result, and the model detection result is compared with the detection result. If the model detection result is consistent with the detection result, then the output model detection result is the final detection result; Otherwise, the test data will be marked as pending manual review, and the final test result will be obtained through manual judgment by the testing personnel.

[0047] By introducing machine learning models and comparing and verifying them with the judgment results of the rule base, auxiliary judgments can be provided in complex situations that cannot be covered by the rules, and manual intervention can be triggered when the results conflict. This improves the accuracy and reliability of judgments for complex detection items, and achieves the goal of making up for the limitations of single rule judgments and ensuring the scientific nature of the results.

[0048] Furthermore, different judgment criteria are preset for different test items in the preset test result judgment rule library; For example, in PCR testing, if the Ct value is less than 35, it is considered positive; if the Ct value is greater than or equal to 35 and less than or equal to 40, it is considered suspicious; if the Ct value is greater than 40, it is considered negative. In antibody testing, if the antibody titer is greater than or equal to 1:16, it is considered qualified. The pre-defined test result judgment rule library allows administrators to add and modify judgment criteria through a backend visual interface; For complex detection projects, a machine learning model is introduced. By obtaining historical detection data, the machine learning model is used to obtain a machine learning judgment model. The detection data is then input into the machine learning judgment model to obtain the model detection results. Alternatively, the machine learning model can be a random forest classification model; For example, in a genotype detection project, a random forest classification model is trained by obtaining the sample detection feature values ​​of known genotypes to obtain a machine learning judgment model; After inputting the detection data into the machine learning judgment model to obtain the model's detection results, the results are compared with the detection results. If the model detection result is consistent with the detection result, then the output model detection result is the final detection result; Otherwise, the test data will be marked as awaiting manual review, and the final test result will be obtained through manual judgment by the testing personnel. Inspectors can view the rule base for judging test results, model test results, and the judgment criteria in the preset test result judgment rule base. They can manually adjust the test results to be manually reviewed, record the review record, and ensure that the results are traceable. The review record includes the reviewer, review comments, and reasons for adjustment. Step S4 for analyzing test results: Call the preset multi-dimensional data analysis engine to perform statistical analysis on the test results, calculate the positive rate of the disease and compare it with the preset positive rate benchmark range to obtain the analysis results.

[0049] Based on the above biological detection data analysis methods, by establishing a digital link for the entire process from data acquisition to test result analysis, and by using the unique identifier of sample number and standard template matching technology, the traceability of sample information and the standardization of processing are achieved. By integrating the scattered testing processes into a coherent automated workflow, the problems of data silos and inconsistent operations in traditional methods are solved, ultimately improving testing efficiency and result reliability.

[0050] In some embodiments, the biological detection data analysis method further includes a consumables management step: In the sample preparation confirmation step S2 and / or sample testing confirmation step S3, the consumable requirements are determined based on the standard template and / or standard testing items, and the consumable type and quantity are selected and registered according to the consumable requirements. Based on the type and quantity of consumables, generate consumable usage records associated with the sample number, and update inventory data according to the consumable usage records.

[0051] By linking the use of consumables with specific sample numbers and standard procedures and automatically updating the inventory, refined and traceable management of consumables usage is achieved. It can accurately record the consumption of consumables for each test and link with the inventory system, thereby solving the problems of extensive consumable management and difficulty in traceability, and achieving the goal of reducing costs and making consumables fully controllable and traceable.

[0052] Furthermore, the types of consumables include centrifuge tubes, reagent bottles, and PCR reaction solutions; Based on the type and quantity of consumables, the corresponding quantity is deducted from the predefined inventory table in the database, and a consumable usage record is generated; the consumable usage record includes the sample number, the testing task number, and the time of receipt; When the inventory of a certain type of consumables falls below a preset threshold, an alert message is sent to the inventory administrator via mini-program notification and SMS reminder, and a list of consumables with insufficient inventory is displayed on the backend management terminal, allowing the administrator to generate purchase orders with one click; Optionally, the preset threshold can be set to 1.5 times the average monthly consumption. Establish a traceability file for consumables, recording the batch number, manufacturer, expiration date, and inspection report number for each batch of consumables upon entry into the warehouse. When consumables are used, they are linked to sample test data. If abnormal test results occur later, the batch information of the consumables used can be traced through the sample number, making it easier to investigate the cause of the problem.

[0053] In some embodiments, the biometric data analysis method further includes a report generation step: An initial test report is generated based on the final test results; Based on the initial test report, the test items, final test results, equipment data traceability chain, instrument usage logs and consumable usage records are subject to multi-level review. After multi-level review and approval, the final test report is generated based on the preset report template library and the initial test report.

[0054] By conducting multi-level reviews of the test results and related traceability data chains, the rigor and authority of the generated test reports are ensured. Through multi-level and multi-element cross-reviews, a complete quality assurance system is formed, thereby solving the problem of report review becoming a mere formality and achieving the goal of maximizing the quality and legal validity of the final test reports.

[0055] Furthermore, after the testing personnel complete the manual judgment, they submit the initial test report to the department head for first-level review. The department head reviews the completeness of the test items and the accuracy of the final test results. After the first-level review is passed, it is sent to the quality control department for the second-level review. The QC department reviews the equipment data traceability chain, consumable usage records, and instrument usage logs. After the second-level review is passed, the initial test report is sent to the person in charge of the institution for a third-level review, which includes reviewing the report format and signing authority. After the Level 3 review is approved, based on the preset report template library, the system automatically extracts basic information, sample information, test data, and final judgment results from the database, fills them into the corresponding fields of the template, and generates a final test report in PDF format. It also supports adding electronic seals of the institution and electronic signatures of the testers. The preset report template library is divided into templates for disease testing reports and quality testing reports according to the type of testing. Once the final test report is generated, a notification will be sent to the user via a mini-program. The user can view and download the report online, or choose to have a paper report mailed to them. The mini-program can automatically synchronize the report information to a third-party logistics system, generate a tracking number, and send it back to the user.

[0056] In some embodiments, the detection result analysis step S4 further includes: Statistical analysis is performed on the test results or final test results to calculate the overall positive rate of the epidemic, the proportion of each epidemic disease tested, and the positive rate of the main epidemic diseases within a preset time period, and a statistical result set is obtained. Based on the pre-selected dimension combination and the statistical result set, the detection results or the final detection results and their associated sample information and basic information are coupled and aggregated using SQL multi-table join and data aggregation algorithms to obtain cross-analysis results. If the overall positive rate of the disease within the preset statistical period exceeds the preset positive rate benchmark range, the data of the overall positive rate of the disease within the preset statistical period will be marked as abnormal data and an abnormal analysis report will be generated. The analysis results are obtained based on the statistical result set, cross-analysis results, and anomaly analysis report.

[0057] By conducting multi-dimensional statistical analysis, cross-analysis, and in-depth mining of anomaly warnings on the final test results, the transformation from single test data to macro-level decision support has been achieved. It can proactively reveal hidden trends, correlations, and risks in the data, thereby solving the problem of low utilization rate of massive test data and achieving the goal of providing users with forward-looking insights and precise prevention and control basis.

[0058] Furthermore, the mini-program's front end has a built-in multi-dimensional data analysis engine to enable real-time calculation and in-depth mining of detection data; Based on the test results or final test results, the overall positive rate of the epidemic, the proportion of each epidemic test, and the positive rate of the main epidemic within a preset time period are calculated through a multi-dimensional data analysis engine. The overall positive rate of the epidemic is calculated as the ratio of the number of positive samples to the total number of samples tested. The percentage of each disease tested is calculated as the ratio of the number of samples tested for a certain disease to the total number of samples tested. The positive rate of major diseases within the preset time period can be the positive rate of major diseases within the past three years, calculated by year, and supports users to select time range to filter data; for example, users can select the past month, the past three months or a custom date; Based on the test results or final test results, multi-dimensional cross-analysis is performed through a multi-dimensional data analysis engine. It supports the combination of dimensions such as target disease type, variety, and age, or the combination of dimensions such as target disease type, origin, and test time. The analysis is performed through SQL multi-table join and data aggregation algorithms to obtain cross-analysis results. For example, analyzing the positivity rate of swine fever in piglets aged 30-60 days or the positivity rate of avian influenza in East China in the third quarter of 2024; Based on the test results or final test results, an abnormal data early warning analysis is performed through a multi-dimensional data analysis engine, and a preset positive rate benchmark range is established based on historical test data in a predefined time series database. Optionally, the preset positive rate benchmark range can be ±5% of the average positive rate of a certain disease over the past 6 months; If the overall positive rate of the epidemic within the preset statistical period exceeds the preset positive rate benchmark range, the data of the overall positive rate of the epidemic within the preset statistical period will be marked as abnormal data, highlighted in red in the chart on the front end of the mini program, and an abnormal analysis report will be generated. The anomaly analysis report includes the anomaly time period, the regions involved, and possible causes. In some embodiments, the biodetection data analysis method further includes a data visualization step: By using ECharts chart components, different data types in the statistical results set can be matched with corresponding charts for display, resulting in multiple types of charts; Based on the overall positive rate of the disease and the place of origin in the sample information, the national and regional distribution of the target disease type is displayed by using map software SDK and PostGIS spatial database, resulting in national and regional distribution maps; A multi-level drill-down system is constructed based on multiple types of charts, national distribution maps, and regional distribution maps. The multi-level drill-down system is used to dynamically link multiple types of charts, national distribution maps, and regional distribution maps.

[0059] By integrating diverse display methods such as charts, maps, and interactive drill-down, complex data analysis results are transformed into intuitive and explorable visual information, lowering the threshold for data interpretation and solving the problems of traditional reports being unintuitive and lacking in depth. This greatly enhances the user's data analysis experience and decision-making efficiency.

[0060] Furthermore, the analysis results are displayed in charts, spatial distribution diagrams, and multi-level drill-down displays through the mini-program front-end, enhancing data readability and user interaction experience. The chart display is as follows: the ECharts component is integrated to match the corresponding charts for different data types in the statistical results set. For example, the overall positivity rate of the epidemic is displayed using a bar chart, where the X-axis represents the target disease type and the Y-axis represents the positivity rate; like Figure 5 As shown, the trend of the positive rate of major diseases in the past three years is analyzed based on the positive rate of major diseases in the past three years. The trend is displayed in a line chart, where the X-axis represents the year and the Y-axis represents the positive rate. Each line represents a disease. like Figure 6 As shown, the detection of different varieties of diseases is analyzed based on the statistical results set and cross-analysis results, and the results are displayed in a pie chart, where each sector represents the detection percentage of different varieties. Charts allow users to view specific values ​​by hovering the mouse over them, and to hide or show the corresponding data series by clicking the legend. Integrating Tencent Map SDK and PostGIS spatial database, based on the overall positive rate of the epidemic and the place of origin in the sample information, the national distribution of the target epidemic type is displayed on the front end of the mini program, using a combination of heat map and marker points; The heatmap uses different shades of color to indicate the positivity rate of a region, with red representing a high positivity rate and blue representing a low positivity rate. Markers indicate the number of positive samples in each province and city. Users can click on a province on the map to view the distribution data of the municipal-level administrative regions under that province, enabling spatial drilling down to the national, provincial, and municipal levels. like Figures 7 to 9 As shown, the design employs a three-level drill-down system: overview, subdivision, and details. When a user clicks on a specific disease in the overall positive rate bar chart, they can drill down to a national distribution map of that disease, a line graph showing the trend of the positive rate over the past three years, and a pie chart showing the detection status of different varieties. Clicking on a specific city-level region on the national distribution map allows users to drill down to the histogram of susceptible age distribution, radar chart of genotype distribution, and a detailed list of tested samples for that region. The data context remains coherent during the drilling process, and users can return to the previous level of operation, making it easy for users to trace back the data analysis path. The sample details list includes the sample number, testing time, and final test result.

[0061] It should be noted that the above is a reference method for biological detection data analysis, and the present invention is not limited thereto.

[0062] The embodiments of the present invention achieve traceability of sample information and standardization of processing, thereby improving detection efficiency and result reliability. It solves the technical problem that existing technologies rely on manual data entry, have a single analysis dimension, and poor visualization interactivity, resulting in difficulties in data traceability, superficial insights, and an inability to support real-time decision-making, thus forming information silos between detection and application.

[0063] Finally, it should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other. The above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them; although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications can still be made to the specific implementation of the present invention or equivalent substitutions can be made to some technical features without departing from the spirit of the technical solutions of the present invention, and all such modifications and substitutions should be covered within the scope of the technical solutions claimed in the present invention.

Claims

1. A method of biological detection data analysis, characterized by, The method comprises the following steps: A submission data acquisition step: obtaining submission data based on an original sample to be tested, and generating a unique sample number for the submission data; wherein the submission data includes basic information, sample information, and detection requirements; A sample preparation confirmation step: based on the sample number, a standard template in a preset sample preparation standard library is retrieved; based on the detection requirements, a standard detection item is matched from a preset detection item library, a sample preparation guide is generated based on the standard detection item and the standard template, a preparation confirmation instruction triggered after each preparation operation is completed is received based on the sample preparation guide, and a sample preparation completion instruction is obtained according to the preparation confirmation instruction; A sample detection confirmation step: based on the sample preparation completion instruction, a sample detection instruction is triggered, device raw data generated by a test equipment during detection is obtained based on the sample detection instruction, a corresponding detection judgment standard is retrieved from a preset detection result judgment rule library according to the detection item in the detection requirements, the device raw data is judged according to the judgment standard, and a detection result is obtained; A detection result analysis step: a preset multi-dimensional data analysis engine is called to statistically analyze the detection result, an epidemic disease positive rate is calculated and compared with a preset positive rate benchmark range, and an analysis result is obtained.

2. The biological detection data analysis method of claim 1, wherein, The submission data acquisition step specifically comprises: Based on a preset classification rule library, a classification label is assigned to the original sample to be tested corresponding to the submission data, and the sample information is compared with the detection requirements to determine whether there are mismatched items; If there are no mismatched items between the sample information and the detection requirements, the sample state of the original sample to be tested is updated to be prepared; Otherwise, a warning information is generated.

3. The biological detection data analysis method of claim 2, wherein, The sample preparation confirmation step specifically comprises: Based on the classification label of the original sample to be tested corresponding to the sample number, a standard template in the preset sample preparation standard library is retrieved; wherein the standard template includes preparation processes, preparation environment requirements, and sample portion calculation rules corresponding to the original sample to be tested of different classification labels; Based on the standard detection item and the standard template, a sample preparation guide is generated; Based on the sample preparation guide, a preparation confirmation instruction triggered after each preparation operation is completed is received, and a sample preparation completion instruction is obtained according to the preparation confirmation instruction; Based on the sample preparation completion instruction, a preparation record is generated, and the sample state of the prepared sample is updated to be detected.

4. The biological detection data analysis method of claim 3, wherein, The sample detection confirmation step further comprises: Based on the detection requirements, the department professional matching degree, the department current task load, the detection personnel skill proficiency, and the available state of the test equipment are set weights; The department professional matching degree, the department current task load, the detection personnel skill proficiency, and the available state of the test equipment are set weights by a multi-factor weight distribution method, and the task allocation score is obtained by weighted calculation; Based on the task allocation score, the target department and the target detection personnel are allocated for the prepared sample.

5. The biological detection data analysis method of claim 1, wherein, The sample detection confirmation step further comprises: Device data generated by the test equipment during the detection process is obtained, and the device data is packaged in a unified format to obtain device raw data; The detection data is obtained by cleaning, verifying and standardizing the original data of the equipment, and the detection data is associated and bound with the sample number corresponding to the detection data and the detection items in the detection requirement, and a device data traceability chain is generated; The starting time and stopping time of the test equipment during the detection process are recorded to generate an instrument use log.

6. The biological detection data analysis method of claim 5, wherein, The sample detection confirmation step further comprises: The detection data is input into a pre-trained machine learning judgment model to obtain a model detection result, and the model detection result is compared with the detection result; If the model detection result is consistent with the detection result, the model detection result is output as the final detection result; Otherwise, the detection data is marked as to be manually reviewed, and the final detection result is obtained by manual judgment by the detection personnel.

7. The biological detection data analysis method of claim 6, wherein, The method further comprises a consumable management step: In the sample preparation confirmation step and / or sample detection confirmation step, the consumable requirement is determined based on the standard template and / or the standard detection item, and the consumable type and consumable quantity are selected and registered according to the consumable requirement; Based on the consumable type and consumable quantity, a consumable use record associated with the sample number is generated, and the inventory data of the consumables is updated according to the consumable use record.

8. The biological detection data analysis method of claim 7, wherein, The method further comprises a report generation step: An initial detection report is generated according to the final detection result; Based on the initial detection report, the detection items, the final detection result, the device data traceability chain, the instrument use log and the consumable use record are subjected to multi-level audit; After passing the multi-level audit, a final detection report is generated based on a pre-set report template library and the initial detection report.

9. The biological detection data analysis method of claim 5, wherein, The detection result analysis step further comprises: Data statistics are performed on the detection result or the final detection result, the overall positive rate of the epidemic disease, the detection proportion of each epidemic disease, and the positive rate of the main epidemic disease within a preset time are calculated to obtain a statistical result set; Based on the pre-selected dimension combination and the statistical result set, the detection result or the final detection result and its associated sample information and basic information are simultaneously queried and data aggregated by SQL multi-table inquiry and data aggregation algorithm to obtain cross-analysis results; If the overall positive rate of the epidemic disease in a preset statistical period exceeds a preset positive rate benchmark range, the data of the overall positive rate of the epidemic disease in the preset statistical period is marked as abnormal data, and an abnormal analysis report is generated; The analysis result is obtained based on the statistical result set, the cross-analysis result and the abnormal analysis report.

10. The biological detection data analysis method of claim 9, wherein, The method further comprises a data display step: Different data types in the statistical result set are matched with corresponding charts for display through ECharts chart components to obtain multiple types of charts; Based on the overall positive rate of the epidemic disease and the source of the sample information, the national distribution and regional distribution of the epidemic disease of the target epidemic disease type are displayed through a map software SDK and a PostGIS spatial database to obtain a national distribution map and a regional distribution map; Based on the multiple types of charts, the national distribution chart and the regional distribution chart, a multi-level drilling system is constructed, and the multiple types of icons, the national distribution chart and the regional distribution chart are dynamically associated according to the multi-level drilling system.

Images