An intelligent vaccination management system and method based on data analysis
Through the intelligent vaccination management system based on data analysis, the problem of integrating and matching vaccination information for migrant populations is solved, the scientificity and accuracy of vaccination strategies are improved, and public health safety and the health interests of vaccinators are ensured.
Patent Information
- Application Number
- CN202411217405.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-02
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-09-02
AI Technical Summary
When handling vaccination of different mobile types of populations, it is difficult for the prior art to effectively integrate and match vaccination information, resulting in insufficient accuracy and adaptability of vaccination decisions, and the inability to accurately meet individual immunity needs, increasing health risks.
Adopt a data analysis-based vaccination intelligent management system to dynamically analyze the flow type, geographical location information and epidemiological data of vaccinators, build a priority scoring model, and combine the multi-dimensional system adaptation index and time series data consistency index to ensure vaccination data and generate highly compatible vaccination protocols through feature vector matching scores.
It significantly improves the reliability and accuracy of the vaccination strategy, ensures public health safety, maximizes the health interests of vaccinated people, and ensures that mobile vaccinated people get the most suitable immune protection in the new environment.
Smart Images

Figure CN118762798B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of vaccination management, and more specifically, to an intelligent vaccination management system and method based on data analysis. Background Art
[0002] In modern vaccination management, with the frequent cross-regional mobility of the population, the health status and vaccination needs of mobile vaccine recipients have become increasingly complex. Especially for inflow and outflow populations, their vaccination information is scattered in management systems in different regions, and the vaccination management systems in various regions differ in data format, field type, and timestamp records. When the historical vaccination information of mobile vaccine recipients is transferred to new areas, it often faces compatibility issues, resulting in the vaccine recipients being unable to receive timely and accurate immunization protection in the new environment. Therefore, a dynamic analysis mechanism is needed to combine the vaccine recipient's mobility type, geographic location information, and vaccination history to build an effective vaccination decision model to ensure the health and safety of the vaccine recipients in the new area.
[0003] However, existing technologies have difficulty in effectively solving the problem of integrating and matching vaccination information when dealing with people of different mobile types. For inflowing populations, the standardization and adaptability of vaccination records are easily affected by changes in geographical location, and vaccination information may not match the vaccination management system in the new area, affecting the accuracy of vaccination decisions. For outflowing populations, unfinished vaccination records may not be effectively tracked in the new environment, resulting in interruptions or omissions in vaccination plans. In addition, the vaccination needs of mobile populations are dynamically adjusted as the environment changes, and existing technical means cannot flexibly adapt to such changes, often resulting in vaccination strategies that cannot accurately meet individual immune needs, thereby increasing the health risks of mobile populations. These problems show that existing technologies lack effective dynamic adjustment and refined management methods when dealing with vaccinations for people of different mobile types.
[0004] In order to solve the above problems, a technical solution is now provided. Summary of the invention
[0005] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide an intelligent vaccination management system and method based on data analysis, which effectively improves the scientificity and accuracy of vaccination management through systematic dynamic analysis and precise matching mechanism. First, through the dynamic analysis of the flow type of the vaccine recipient, combined with geographic location information, vaccination history and epidemiological data, a priority scoring model is constructed to ensure that vaccination decisions can adapt to the health risks and needs of the current environment of the vaccine recipient. Secondly, through the comprehensive calculation of the multidimensional system adaptation index and the time series data consistency index, the compatibility and synchronization of historical vaccination information in cross-system transfer are effectively evaluated, avoiding decision-making errors caused by inconsistent data, and ensuring the accuracy and completeness of vaccination data. Finally, by quantizing the feature vectors of the concerned items of the drugs to be vaccinated and the corresponding items of the mobile vaccine recipients, a highly compatible comprehensive matching score is generated by using nonlinear combination and dynamic threshold judgment to ensure that the vaccine recipients obtain the most suitable vaccination plan in the new environment. Overall, the present invention significantly improves the reliability and accuracy of the vaccination strategy, ensures public health safety, and maximizes the health benefits of the vaccine recipients to solve the problems raised in the above background technology.
[0006] To achieve the above object, the present invention provides the following technical solutions:
[0007] S1, obtain the historical vaccination information of mobile people to be vaccinated;
[0008] S2, divide the mobile recipients into inflow and outflow types according to historical vaccination information, determine the priority of the drugs to be vaccinated according to the mobile type, and obtain the focus items of the drugs to be vaccinated;
[0009] S3, for the inflow type, identify the degree of standardization of historical vaccination information relative to the inflow area and determine whether it is an entry-clear type;
[0010] S4. If it is determined that a clear type is entered, the corresponding items of the mobile persons to be vaccinated are determined based on the items of interest of the drugs to be vaccinated, and the matching results are obtained through matching analysis of the items of interest and the corresponding items.
[0011] In a preferred embodiment, step S2 includes the following contents:
[0012] S2.1, obtain the geographic location information and the timestamp of each vaccination in the historical vaccination information of the mobile vaccine recipients; arrange the geographic location information in chronological order to form a time series trajectory data set; standardize the geographic location information from different sources to ensure that the location data is processed under a unified geographic coordinate system or regional coding system; then, determine the current residence of the mobile vaccine recipients and mark it as the current node;
[0013] For each location in the historical trajectory, the spatial distance from the current residence is calculated, and the distances are arranged in chronological order to form a distance sequence; then, the distance sequence is differentially analyzed to calculate the distance change rate between adjacent historical locations, thereby forming distance change data;
[0014] In the distance change data, attention is paid to the distance change rate of the last section of positions; if the distance gradually decreases in the last few positions and the distance between the end position and the current residence is less than the set threshold, and the time point is within the limited time, it is judged as an inflow type; if the distance change rate of the historical trajectory shows a gradual increase in the last few positions, and the distance of the last position is greater than the distance of the initial position, it is judged as an outflow type.
[0015] In a preferred embodiment, S2.2, when determining the priority score of the drug to be vaccinated, the mobility type of the mobile vaccinee is combined, and the vaccination history, geographic location information and current vaccination strategy are analyzed; if the mobile vaccinee is an inflow type, then according to the epidemiological data and vaccination policy of the inflow area, the vaccination history record is analyzed to identify whether there is a lack of vaccines related to infectious diseases in the corresponding area or the vaccination interval exceeds the recommended time; if it is found that the vaccinee is at risk of exposure in the inflow area, and his vaccination history fails to cover the main infectious diseases in the corresponding area, the relevant vaccine is set to a high priority; further analyze the vaccination time interval, if the time interval between the last dose and the recommended time has passed, or the vaccination time is insufficient, resulting in the immune effect not meeting the requirements, then the priority of the relevant vaccine is increased to ensure that the vaccinee obtains continuous protection; if the vaccinee is at risk of environmental exposure in the new area, the relevant vaccine is set as a priority vaccination;
[0016] Construct a priority assessment model for preferential vaccination drugs based on inflow types:
[0017] Data input and feature selection:
[0018] A1. Select epidemiological data of the inflow area;
[0019] A2. Includes historical vaccination information of mobile recipients, specific vaccination time, dose, and vaccination interval;
[0020] A3. Analyze the environmental changes of the mobile vaccinated people from their original areas to the areas they move to;
[0021] A4. Consider the interval between the last vaccination and the current recommended vaccination, especially whether it exceeds the recommended protection period of the vaccine;
[0022] If the mobile vaccinated person is of the outflow type, the epidemiological data of the target area will be analyzed to assess the exposure risk in the new environment; if the infectious disease prevention and control requirements of the target area are different from those of the current area, priority will be given to vaccines that can provide effective protection in the new area; if the target area has high prevention and control requirements for certain infectious diseases, and the vaccine recipient has not yet received the relevant vaccine, the priority of such vaccines will be increased to ensure that they are immune protected before arriving in the new area; if the vaccination record shows that some vaccines have not been fully administered and prevention and control are still required in the target area, all doses will be completed before leaving to ensure the continuity of the immune effect;
[0023] Construct a priority assessment model for preferential vaccination drugs based on outflow types:
[0024] B1. Analyze epidemiological data of target areas where outflows are heading;
[0025] B2. Consider whether the vaccines related to the diseases controlled in the target area have been administered in the vaccination records of the mobile recipients, and whether the vaccination doses are complete;
[0026] B3. Evaluate whether the vaccination history of mobile recipients meets the public health requirements of the target area;
[0027] B4. Analyze the current vaccination records to see if there are any unfinished vaccine doses;
[0028] The Logistic regression model was used for classification discussion: for the inflow type, the A1-4 features were input; for the outflow type, the B1-4 features were input, and the priority score for the vaccine to be administered was output; the basic formula of the model established a linear combination by taking each input feature as an independent variable, and converted it into a priority score through the Logistic function;
[0029] The output priority score is compared with the corresponding threshold, and the drugs to be vaccinated whose priority scores are greater than or equal to the corresponding threshold are collected and sorted according to the priority score.
[0030] In a preferred embodiment, S2.3, obtaining of items of interest:
[0031] First, select the top-ranked drugs to be vaccinated and define preliminary items of concern from the characteristics of the drugs to be vaccinated; for each vaccine, conduct preliminary screening based on the characteristics of the drugs to determine which characteristics are particularly important in the inflow or outflow type scenario; after the preliminary screening, organize these characteristics into a list of items of concern to be evaluated;
[0032] Each candidate focus item is vectorized to make it a feature vector that is convenient for subsequent calculations; specifically, each focus item is converted into a binary variable to indicate whether the relevant focus item is important in the current context.
[0033] In a preferred embodiment, step S3 includes the following contents:
[0034] In the process of generating the multidimensional system adaptation index, first, all data fields are extracted from the historical vaccination information of the mobile vaccinated persons, and mapped one by one with the field types of the vaccination management system of the inflow area; according to the mapping results, it is determined whether the data format of each field is compatible with the inflow system and marked as compatible or incompatible;
[0035] Next, for each field type, the matching degree between the historical vaccination information and the field type of the vaccination management system in the inflow area is calculated; the matching degree includes the type, structure, and length attributes of the data field; a comprehensive calculation is performed based on the matching of the data field in these three aspects; the matching result is converted through the Logit function to obtain an exponential score;
[0036] After completing the format conversion and type matching of all fields, evaluate the integrity of the data mapping between systems; combine the format mapping result of each field with its corresponding type matching degree, and multiply it by the weight value of the corresponding field in the system; calculate the scores of all fields comprehensively to obtain the final multi-dimensional system adaptation index.
[0037] In a preferred embodiment, when generating the time series data consistency index, first, the timestamp of each record is extracted from the historical vaccination information of the mobile vaccinated person, and compared with the corresponding record in the vaccination management system of the inflow area; for each pair of records, the time difference between the two is calculated, and the time difference is processed using logarithmic transformation;
[0038] Next, check whether the time sequence of historical vaccination information is consistent with the vaccination records in the inflow system; compare the order of each group of adjacent vaccination events to determine whether the order of records in the two systems is the same; if the order is consistent, the score is full, if not, the score is zero, forming a sequential consistency score set;
[0039] The timestamp difference and sequential consistency score are combined to calculate the synchronization score of each record between the two systems; the timestamp difference and sequential consistency score are combined into a single score value through the Logit function to reflect the synchronization of the two systems; finally, based on the synchronization score of all records, the time series data consistency index of the entire data set is calculated.
[0040] In a preferred embodiment, when generating a comprehensive judgment of a clear entry type, it is first necessary to normalize the multidimensional system adaptation index and the time series data consistency index, and convert the two indexes into values within the same scale range; through logarithmic and radical operations, the indexes are converted into numerical values that fluctuate within a similar interval; then, the interaction between the two indexes is analyzed to obtain an interaction score, and the combined effect of the two is evaluated through a multiplication relationship;
[0041] Finally, the interaction score and dynamic threshold are combined to comprehensively judge whether the historical vaccination information is of the clearly entered type; when the interaction score is higher than the dynamic threshold, it is judged as the clearly entered type; if it is lower than the dynamic threshold, it is judged as the unclearly entered type.
[0042] In a preferred embodiment, step S4 includes the following contents:
[0043] S4.1, obtaining a list of items of interest for each drug to be vaccinated from step S2;
[0044] S4.2, matching and calculating the feature vectors of the items of interest of the drugs to be vaccinated and the feature vectors of the corresponding items of the mobile persons to be vaccinated; the matching calculation uses the vector inner product calculation method to evaluate the fit between each item of interest and the corresponding item; and a list of matching scores is generated, each score reflecting the degree of match between the item of interest and the corresponding item in a certain dimension;
[0045] All matching scores are nonlinearly combined to generate a comprehensive matching score;
[0046] S4.3, S4.3, compare the comprehensive matching score with the matching threshold to determine whether the comprehensive matching score meets the vaccination standard; if the comprehensive matching score is higher than the matching threshold, it is judged that the relevant vaccine is highly consistent with the needs of the mobile recipients and is given priority for vaccination.
[0047] An intelligent vaccination management system based on data analysis, comprising: a historical vaccination information collection module, a flow type determination module, a normative identification module and a corresponding item matching module;
[0048] The historical vaccination information collection module obtains the historical vaccination information of the mobile recipients and transmits the historical vaccination information data to the mobile type determination module;
[0049] The flow type determination module divides the flow recipients into inflow and outflow types according to the historical vaccination information, determines the priority of the drugs to be vaccinated according to the flow type, obtains the focus items of the drugs to be vaccinated, and transmits the flow type and the focus item data of the drugs to be vaccinated to the input normative identification module;
[0050] The standardization identification module identifies the standardization degree of the historical vaccination information relative to the inflow area according to the inflow type, determines whether it is an input clear type, and transmits the judgment result of the input clear type to the corresponding item matching module;
[0051] If the corresponding item matching module determines that a clear type of data has been entered, it will determine the corresponding items of the mobile persons to be vaccinated based on the items of interest of the drugs to be vaccinated, and obtain the matching results through matching analysis between the items of interest and the corresponding items.
[0052] Technical effects and advantages of an intelligent vaccination management system and method based on data analysis of the present invention:
[0053] 1. Step S2 dynamically analyzes the mobility type of the vaccine recipients, combines geographic location information, vaccination history, epidemiological data and other factors, accurately identifies and classifies the vaccination needs of the vaccine recipients in different regions, and builds a priority scoring model to ensure that the vaccination decision for each vaccine can fully adapt to the health risks and needs of the vaccine recipient's current environment. Through feature vector quantization and data-driven optimization, the focus items are highly matched with the actual situation of the vaccine recipients, which effectively improves the accuracy and scientificity of vaccination. In addition, through real-time adjustment and adaptive update mechanisms, the system can continuously optimize vaccination strategies to ensure that vaccine recipients always receive the most appropriate immune protection under different mobility types, thereby improving vaccination effects and ensuring public health safety.
[0054] 2. The present invention effectively evaluates the compatibility and synchronization of historical vaccination information in cross-system transfers through comprehensive calculations of the multi-dimensional system adaptation index and the time series data consistency index, ensuring the accuracy and completeness of the data in the vaccination management system of the inflow area. Through a detailed analysis of the data format, field type, timestamp, and sequence, possible inconsistencies between systems are identified and corrected, thereby avoiding vaccination decision errors caused by data loss, misreading, or time confusion. Ultimately, a comprehensive judgment of whether the historical vaccination information is of a clear input type provides a solid basis for vaccination decisions in the inflow area, improves the reliability and accuracy of the vaccination strategy, ensures the vaccination effect of the vaccinee in the new area, and ensures public health safety.
[0055] 3. The present invention accurately matches the items of interest of the drugs to be vaccinated with the corresponding items of the mobile vaccine recipients, and uses advanced methods such as feature vector quantization, nonlinear combination, and dynamic threshold judgment to generate a comprehensive matching score, thereby effectively improving the fit between the vaccine and the vaccine recipient. Through this matching mechanism, it can ensure that each mobile vaccine recipient obtains the vaccine that best suits his or her health status and immune needs in the new environment, thereby optimizing the vaccination effect, reducing vaccination risks, and improving the accuracy and reliability of vaccination strategies. Ultimately, it provides a scientific decision-making basis for the vaccination management system, ensures public health safety, and maximizes the health benefits of the vaccine recipients. BRIEF DESCRIPTION OF THE DRAWINGS
[0056] Figure 1 A schematic diagram of a process of an intelligent vaccination management method based on data analysis according to the present invention;
[0057] Figure 2 The present invention is a structural schematic diagram of an intelligent vaccination management system based on data analysis. DETAILED DESCRIPTION
[0058] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0059] Example: Figure 1 The present invention provides an intelligent vaccination management method based on data analysis, comprising:
[0060] S1, obtain the historical vaccination information of mobile people to be vaccinated;
[0061] S2, divide the mobile recipients into inflow and outflow types according to historical vaccination information, determine the priority of the drugs to be vaccinated according to the mobile type, and obtain the focus items of the drugs to be vaccinated;
[0062] S3, for the inflow type, identify the degree of standardization of historical vaccination information relative to the inflow area and determine whether it is an entry-clear type;
[0063] S4. If it is determined that a clear type is entered, the corresponding items of the mobile persons to be vaccinated are determined based on the items of interest of the drugs to be vaccinated, and the matching results are obtained through matching analysis of the items of interest and the corresponding items.
[0064] In step S2, it is crucial to distinguish the mobility types of mobile vaccine recipients during the vaccination process, because different mobility types may directly affect the vaccine recipients' immunization needs and vaccine selection priorities. The mobility types are mainly divided into two categories: "inflow" and "outflow", which represent the vaccine recipients flowing into the current area from other places, or flowing out of the current area to other areas. For inflow mobile vaccine recipients, their historical vaccination information may come from different regions or systems, and these data may differ from the standards and requirements of the current area. Therefore, it is necessary to specifically identify the degree of standardization of their information entry to ensure that their vaccination history is complete, accurate and meets local requirements in the new area. This is crucial to determine their immune protection needs in the new area. The vaccine selection of inflow mobile vaccine recipients often needs to consider their exposure risk in the new environment and the matching with local epidemiological characteristics, so as to give priority to those vaccines that can provide the best protection in the current area. By analyzing their vaccination history and combining the public health strategy of the new area, their vaccine priority can be accurately determined, thereby avoiding the risk of unnecessary repeated vaccination or insufficient protection. In contrast, when choosing vaccines, outflow-type mobile recipients need to ensure that they still have sufficient immunity after leaving the current area. This requires evaluating whether their vaccination information meets the vaccination requirements of the future destination to ensure that their health and safety can be continuously protected in the new environment. Therefore, distinguishing the types of mobility can not only improve the accuracy of vaccination, but also effectively adapt to the changes in health needs of recipients due to geographical changes, thereby achieving more personalized and scientific immunization management.
[0065] In step S3, for the "inflow" type of mobile vaccine recipients, it is necessary to identify the degree of standardization of their historical vaccination information relative to the inflow area, because the standardization of vaccination information directly affects the accuracy and effectiveness of vaccination decisions. The historical vaccination information of inflow vaccine recipients usually comes from different regions or systems, and the entry standards, formats and completeness of this information may differ from the requirements of the inflow area. If these differences are not identified, information mismatches, omissions or misunderstandings may occur in the vaccine recipient's vaccination history in the new area, thus affecting vaccination decisions. For example, different regions may use different coding methods, data entry formats or vaccination record standards, and the inflow area may not be able to directly interpret this information or misinterpret the vaccination status, making it impossible to accurately determine which vaccines need priority or booster shots.
[0066] Furthermore, identifying the degree of standardization of historical vaccination information can also help ensure the accuracy and completeness of the data, which is particularly important for formulating scientific vaccination plans. If the vaccination information of the influx is not standardized, the vaccines that should have been protected may not be correctly identified, thereby missing key vaccination needs, or unnecessary repeated vaccinations due to misinterpretation of information, which not only wastes resources but may also increase the health risks of the vaccine recipients. Therefore, by identifying and standardizing the historical vaccination information of the influx, it can be ensured that this information is correctly interpreted and applied in the new area, providing a reliable basis for subsequent vaccination plans.
[0067] Finally, the standardization of influx vaccination information is also related to the effective implementation of local public health strategies. Vaccination strategies in influx areas are usually based on local epidemiological characteristics and public health goals. If the influx vaccination information cannot be correctly identified, the implementation of these strategies may be affected. For example, influx areas may require certain vaccines to achieve herd immunity. If historical vaccination information is not standardized, influxes may fail to obtain the required vaccine protection in a timely manner, which in turn affects the health and safety of the entire community. Therefore, identifying the standardization of historical vaccination information is not only a guarantee of individual health, but also a necessary measure for the public health safety of influx areas.
[0068] In step S4, even if it has been confirmed that the historical vaccination information belongs to the clear entry type, it is still necessary to determine the corresponding items of the mobile vaccine recipients based on the focus items of the drugs to be vaccinated, and obtain the final matching results through the matching results of the focus items and the corresponding items. This is because the clear entry type only ensures the standardization and integrity of the data, but does not specifically analyze and match the actual immunization needs of the vaccine recipients with the requirements of the current drugs to be vaccinated.
[0069] First, the entry of historical vaccination information has a clear type. Although it indicates the standardization and reliability of the data, this information needs to be further matched with the items of concern for the drugs to be vaccinated. The items of concern for the drugs to be vaccinated usually include vaccine-specific requirements, such as vaccination intervals, types of vaccines previously vaccinated, vaccination reactions, etc. These items of concern are crucial to determining whether the vaccine recipient needs additional vaccinations, booster shots, or avoidance of certain vaccines. For example, some vaccines may require multiple vaccinations within a specific time interval. If historical records show that the vaccine recipient does not meet this requirement, the system needs to discover this through matching and recommend booster shots. Therefore, determining the corresponding items of the mobile vaccine recipients and matching them can accurately assess the gap between their existing immune status and the demand for vaccines to be vaccinated, thereby making scientific vaccination decisions.
[0070] Secondly, through the matching results of the focus items and the corresponding items, the special needs or potential risks of the vaccine recipients in the new environment can also be identified. Even if the vaccination information is entered in a clear type, this information may be standardized in the original area, but there may be differences or deficiencies in the new area. For example, a vaccine may not be necessary in the original area, but it must be vaccinated in the inflow area due to different epidemiological characteristics. In this case, these differences can be discovered through matching analysis to ensure that the vaccine recipients are properly protected in the new environment. In addition, some focus items may also involve personalized needs such as the vaccine recipient's health status, allergy history, etc. This information must be matched with the safety requirements of the vaccine to ensure the safety and effectiveness of the vaccination.
[0071] Finally, the matching of focus items with corresponding items also provides a more detailed and accurate basis for vaccination decisions. Even if the data entry is clear, relying directly on these data to make decisions may not capture all details or special circumstances. Through a systematic matching process, it can be ensured that all relevant factors are taken into account, so that more comprehensive and personalized vaccination recommendations can be made. This process not only improves the accuracy of decision-making, but also reduces vaccination risks and optimizes the allocation of vaccination resources. Therefore, in the key link of vaccination decision-making, obtaining the final matching result through the matching results of focus items and corresponding items is a necessary step to ensure that vaccine recipients are fully protected in the new environment.
[0072] Step S1 includes the following contents:
[0073] S1.1, first identify possible data sources for mobile people to be vaccinated. Common data sources include vaccination databases, local health systems, community health record systems, personal health device data (such as smart bracelets or health applications), and vaccination record systems in the original place of residence.
[0074] According to the mobility history of the mobile vaccine recipients, the different regions and systems where they may have been vaccinated are classified and sorted to ensure that all possible vaccination information data sources are identified and recorded. This step aims to comprehensively cover all vaccination histories and avoid missing key data.
[0075] S1.2, after identifying all possible data sources, obtain access to these data through appropriate channels and permission management mechanisms. Usually it is necessary to collaborate with local health departments or data management agencies to obtain legal data access authorization.
[0076] Verify the identity information of the mobile vaccine recipients, such as ID number, medical card number, etc., and ensure the security and legitimacy of the data through identity verification procedures. This step also includes the specific operation of downloading vaccination data from relevant systems after obtaining authorization.
[0077] S1.3: Integrate the vaccination information obtained from different sources and convert all data into a unified format for subsequent analysis and processing. This includes dealing with data format differences that may exist in different systems, such as date format, vaccine code, and the representation of vaccination locations.
[0078] Through data conversion tools or custom scripts, all data are normalized to conform to a unified structure and standard. For example, the date format of different systems is unified into the "YYYY-MM-DD" format, and the vaccine name or code is unified into an internationally accepted standard.
[0079] S1.4, clean the integrated data, remove redundant records, and correct obvious errors (such as input errors, missing data, inconsistent records, etc.). The cleaning step also includes a standard check on the data, such as whether there are blank fields or duplicate records.
[0080] Apply data cleaning algorithms or manual inspection methods to identify and remove duplicate vaccination records and repair or complete missing data. For example, if the same vaccine vaccination record is recorded repeatedly in different systems, only the most complete and accurate version is retained and inconsistent data items are corrected.
[0081] S1.5, organize all vaccination information into time series in chronological order to ensure that the historical vaccination information is continuous and complete in time dimension. Time series organization will help to analyze whether the vaccination interval meets the requirements in the future.
[0082] Arrange the vaccination events in chronological order to form a complete vaccination history timeline. For unclear time records, try to restore the complete time sequence by inferring the context or comparing with other related records.
[0083] S1.6, combined with the geographic location information in the vaccination history, clearly mark the location of each vaccination to facilitate subsequent determination of the inflow or outflow type and perform regional matching.
[0084] Geographical location information is attached to each vaccination record, such as the region and city where the vaccination took place. If geographic location information is missing in some records, it can be supplemented by cross-checking with other information or through logical inference.
[0085] S1.7, after completing data acquisition and collation, store the processed historical vaccination information in an integrated database or data warehouse to prepare for subsequent analysis and processing.
[0086] Processing content: Ensure the security and accessibility of data storage, and use a reliable storage system to save the cleaned and sorted inoculation information. Through data backup and access control measures, ensure the integrity and security of the data, while making it efficiently available in subsequent analysis steps.
[0087] In modern vaccination management systems, with the increasing global population mobility, traditional static information analysis methods can no longer effectively cope with complex population mobility situations, especially in groups that move frequently across regions, where the complexity of historical information, geographic location changes, and vaccine needs of vaccine recipients has increased significantly. In order to ensure the accuracy and scientificity of vaccination decisions, it is necessary to introduce dynamic analysis methods to accurately identify mobility types and conduct comprehensive evaluations based on vaccine characteristics, epidemiological data, environmental adaptability, and other factors. Step S2 aims to construct an accurate vaccine priority scoring model by acquiring and analyzing data such as the geographic location information, historical vaccination records, and mobility types of vaccine recipients, and on this basis, scientifically determine the focus items for each vaccine to ensure that when flowing into or out of different regions, the vaccination strategy can adapt to the changing environment and health needs, thereby providing the most appropriate immune protection for vaccine recipients. Through data-driven optimization and real-time adjustment, the system can continuously update and optimize vaccination decisions, improve the overall vaccination effect, and ensure public health safety.
[0088] Step S2 includes the following contents:
[0089] In modern vaccination management, with the increase in population mobility, accurately identifying the mobility type of mobile vaccine recipients has become an important part of improving the accuracy of vaccination decisions. Mobility not only affects the exposure risk and immunization needs of vaccine recipients, but also has a profound impact on the accurate interpretation of their vaccination history and the selection of subsequent vaccines. Especially in the case of movement between different regions, vaccination records may be scattered in multiple databases, and changes in geographical location are directly related to the effectiveness of vaccination decisions. Therefore, in the vaccination management system, it is first necessary to obtain and analyze the geographical location information of mobile vaccine recipients, and combine time series data to construct their movement trajectory. This trajectory analysis not only helps to determine whether the vaccine recipients have flowed into their current residence from other regions, or whether they plan to flow out, but also provides a scientific basis for subsequent vaccination decisions. Through precise geographical location and time series analysis, the mobility type of vaccine recipients can be effectively determined, thus laying a solid foundation for the formulation of vaccination strategies.
[0090] S2.1, obtain the geographic location information and the timestamp of each vaccination in the historical vaccination information of the mobile vaccine recipients. Arrange the geographic location information in chronological order to form a time series trajectory data set. Standardize the geographic location information from different sources to ensure that the location data is processed under a unified geographic coordinate system or regional coding system. Next, determine the current residence of the mobile vaccine recipients and mark it as the current node.
[0091] For each location in the historical trajectory, the spatial distance to the current residence is calculated, and the distances are arranged in chronological order to form a distance sequence. Subsequently, the distance sequence is differentially analyzed to calculate the distance change rate between adjacent historical locations, thereby forming distance change data.
[0092] In the distance change data, attention is paid to the distance change rate of the last section of positions; if the distance gradually decreases in the last few positions and the distance between the end position and the current residence is less than the set threshold, and the time point is within the limited time, it is judged as an inflow type; if the distance change rate of the historical trajectory shows a gradual increase in the last few positions, and the distance of the last position is greater than the distance of the initial position, it is judged as an outflow type.
[0093] For example: if the distances of the last three consecutive positions decrease successively, and the distance between the terminal position and the current residence is less than the set threshold (for example, 100 meters), and the timestamp of the position is within the last 30 days, it is judged as an inflow type; on the contrary, if the distances of the last three consecutive positions in the historical trajectory increase successively, and the distance between the terminal position and the current residence is more than twice the distance of the initial position, it is judged as an outflow type.
[0094] Traditional methods for classifying mobile populations usually rely on static data such as residential information, registered addresses, or simple entry and exit records. These methods have obvious shortcomings in dealing with the increasingly complex population mobility in modern society. First, static information lacks real-time performance and cannot reflect the actual residential dynamics of individuals. Especially among people who move frequently, traditional methods may not be able to accurately capture their mobility status. Secondly, methods that rely on registered addresses or entry and exit records are susceptible to data lags and cannot update the latest location of individuals in a timely manner, resulting in inaccurate classification results. In addition, traditional methods usually ignore the continuity and time dimensions of geographic location information and cannot identify subtle changes in mobility patterns and trajectories, which is particularly insufficient for judging whether a person is actually flowing in or out of an area, or the risk of being exposed to different vaccination environments.
[0095] Therefore, the traditional classification method of mobile population has obvious limitations in accuracy and timeliness, and it is difficult to meet the needs of accurate identification of mobile population in modern vaccination management. In order to solve these problems, step S2.1 introduces a trajectory analysis method based on geographic location information and time series. This method can more accurately classify mobile populations into "inflow" or "outflow" types by dynamically capturing the movement trajectory of individuals and combining the changes in geographic space and time dimensions, thereby providing a more reliable basis for vaccination decisions.
[0096] S2.2, when determining the priority score of the drugs to be vaccinated, the mobility type of the mobile vaccine recipients is taken into consideration, and their vaccination history, geographic location information, and current vaccination strategies are analyzed. If the mobile vaccine recipients are of the inflow type, their vaccination history records are analyzed based on the epidemiological data and vaccination policies of the inflow area to identify whether there are any missing vaccines related to infectious diseases in the area or the vaccination interval exceeds the recommended time. If it is found that the vaccine recipient is at risk of exposure in the inflow area and his vaccination history does not cover the major infectious diseases in the area, the relevant vaccines will be set as high priority. Further analyze the vaccination time interval. If the time interval between the last dose and the recommended time has passed, or the vaccination time is insufficient, resulting in the immunization effect not meeting the requirements, the priority of the relevant vaccines will be increased to ensure that the vaccine recipients receive continuous protection. If the vaccine recipient is at risk of environmental exposure in the new area, such as the risk of waterborne diseases, the relevant vaccines will be set as priority vaccination.
[0097] Construct a priority assessment model for preferential vaccination drugs based on inflow types:
[0098] Data input and feature selection:
[0099] A1. Select epidemiological data of the inflow area, such as the prevalence of major infectious diseases and vaccination coverage.
[0100] A2. Includes historical vaccination information of mobile people waiting to be vaccinated, including the specific vaccination time, dose, and vaccination interval of the vaccine.
[0101] A3. Analyze the environmental changes of the mobile recipients from their original areas to the areas they move into, paying special attention to possible new exposure risks, such as climate differences, natural disaster risks, etc.
[0102] A4. Consider the interval between the last vaccination and the current recommended vaccination time, especially whether it exceeds the recommended protection period of the vaccine.
[0103] If the mobile vaccinated person is of the outflow type, the epidemiological data of the target area will be analyzed to assess the risk of exposure in the new environment. If the infectious disease prevention and control requirements of the target area are different from those of the current area, priority will be given to vaccines that can provide effective protection in the new area. If the target area has high prevention and control requirements for certain infectious diseases, and the vaccine recipient has not yet received the relevant vaccine, the priority of such vaccines will be increased to ensure that they are immune before arriving in the new area. If the vaccination record shows that some vaccines have not been fully administered and prevention and control are still required in the target area, all doses will be completed before leaving to ensure the continuity of the immunization effect.
[0104] Construct a priority assessment model for preferential vaccination drugs based on outflow types:
[0105] B1. Analyze the epidemiological data of the target areas where the outflows are about to go, and pay attention to the epidemic situation of infectious diseases that is different from the current area.
[0106] B2. Consider whether the vaccines related to the disease prevention and control in the target area in the vaccination records of the mobile recipients have been administered and whether the vaccination doses are complete.
[0107] B3. Evaluate whether the vaccination history of mobile recipients meets the public health requirements of the target area, especially whether certain vaccines are still effective after crossing regions.
[0108] B4. Analyze the current vaccination records to see if there are any unfinished vaccine doses, especially those that are still needed for prevention and control in the target areas.
[0109] The Logistic regression model is used for classification discussion: for the inflow type, input A1-A4 features; for the outflow type, input B1-B4 features, and output the priority score of the vaccine to be vaccinated. The basic formula of the model takes each input feature as an independent variable, establishes a linear combination, and converts it into a priority score through the Logistic function.
[0110] The output priority score is compared with the corresponding threshold, and the drugs to be vaccinated whose priority scores are greater than or equal to the corresponding threshold are collected and sorted according to the priority score.
[0111] S2.3, Acquisition of Focus Items:
[0112] First, select the drugs to be vaccinated that are ranked first, and define preliminary concerns from the characteristics of the drugs to be vaccinated, including key factors such as drug effectiveness, adaptability, time requirements, regional applicability, and vaccination safety. For each vaccine, conduct a preliminary screening based on the characteristics of the drug to determine which characteristics are particularly important in the inflow or outflow type scenarios. For example, for the inflow type, focus on the effectiveness and adaptability of the drug in the new area, while for the outflow type, focus on the cross-regional compatibility and vaccination completeness of the vaccine. After the preliminary screening, organize these characteristics into a list of concerns to be evaluated, laying the foundation for subsequent in-depth analysis.
[0113] Each candidate item of interest is vectorized to make it a feature vector that is convenient for subsequent calculations. Specifically, each item of interest is converted into a binary variable to indicate whether the relevant item of interest is important in the current situation. For example, if a vaccine is of high importance to the current region in the inflow type, the corresponding item in the feature vector is set to "1"; if it is not important, it is set to "0". This representation of feature vectors is not only concise and clear, but also convenient for subsequent calculation and matching with the corresponding items of mobile vaccinated persons. In the process of vectorization, the representation of different features can be further refined, such as using Boolean vectors to distinguish the applicability of vaccines in different regions.
[0114] After the initial definition and feature vectorization, data-driven optimization is performed using historical data and actual vaccination results. By analyzing large-scale vaccination data, it is possible to identify which items of interest have the most significant impact on vaccination results in actual applications. Machine learning algorithms can be used to mine patterns in historical data and adjust the feature vector representation of the items of interest. For example, by applying the random forest algorithm to analyze historical vaccination data, the correlation pattern between the effectiveness of a specific vaccine in different regions and the vaccination success rate is identified. Based on these patterns, the feature vectors of the items of interest for the vaccine are adjusted, for example, by increasing the weight representation of geographical adaptability features, or introducing more sophisticated sub-items to capture more complex adaptability features. This data-driven optimization process ensures that the selection of items of interest not only conforms to theoretical analysis, but also reflects higher accuracy in practical applications.
[0115] In actual application, as new data continues to accumulate, the focus items and matching calculation mechanisms are regularly adjusted and adaptively updated in real time. By introducing a dynamic update algorithm, the definition of focus items can be adjusted as the epidemiological environment changes, ensuring the long-term effectiveness of the model. For example, when the epidemic trend of an infectious disease in a certain area changes, the system automatically re-evaluates the focus items of the relevant vaccines, incorporates the newly emerging important features into the feature vector, and improves the matching score of the vaccine. This real-time adjustment mechanism not only improves the flexibility of the system, but also continuously optimizes the vaccination effect, ensuring that the vaccine recipients receive the most appropriate immune protection in different environments.
[0116] Step S2 dynamically analyzes the mobility type of the vaccine recipients, combines geographic location information, vaccination history, epidemiological data and other factors, accurately identifies and classifies the vaccination needs of the vaccine recipients in different areas, and builds a priority scoring model to ensure that the vaccination decision for each vaccine can fully adapt to the health risks and needs of the vaccine recipient's current environment. Through feature vector quantization and data-driven optimization, the focus items are highly matched with the actual situation of the vaccine recipients, which effectively improves the accuracy and scientificity of vaccination. In addition, through real-time adjustment and adaptive update mechanisms, the system can continuously optimize the vaccination strategy to ensure that the vaccine recipients always receive the most appropriate immune protection under different mobility types, thereby improving the vaccination effect and ensuring public health safety.
[0117] In the inflow type of vaccination management, the historical vaccination information of mobile inoculated persons usually comes from different regions and systems. Due to the differences in data format, field type, timestamp record and other aspects of the vaccination management systems in various regions, historical vaccination information may face system incompatibility and insufficient data synchronization when transferred to the inflow area. This inconsistency may lead to misinterpretation of vaccination records, data loss or confusion in time sequence, thus affecting the accuracy and effectiveness of vaccination decisions. Therefore, in order to ensure that vaccination information can be seamlessly connected in the new system and maintain the integrity and timeliness of the data, it is necessary to conduct a detailed evaluation of historical vaccination information through the "multidimensional system adaptation index" and "time series data consistency index" respectively. The multidimensional system adaptation index is used to analyze the compatibility of vaccination information in data format and field type to ensure that information can be correctly transmitted between different systems. The time series data consistency index evaluates the time sequence and synchronization of timestamps of vaccination information to ensure that there will be no time confusion or data lag when vaccination records are transferred across systems. Through the comprehensive calculation in these two directions, it is possible to effectively judge whether the vaccination information is accurate and standardized, providing a solid guarantee for the health management of inflow type inoculated persons.
[0118] Step S3 includes the following contents:
[0119] In modern vaccination management, with the frequent flow of population, the vaccination information of mobile recipients often comes from different regions and systems. Vaccination management systems in different regions differ in data format, field type, and data standard, which may lead to compatibility issues when historical vaccination information is transferred across systems. If this information cannot be accurately connected to the vaccination management system of the inflow area, it will directly affect the accuracy of vaccination decisions and may even cause important vaccination history to be ignored. Therefore, it is necessary to evaluate the degree of adaptation of vaccination information between different systems to ensure seamless integration and application of data. The multidimensional system adaptation index provides a quantitative evaluation index by comprehensively analyzing the compatibility between historical vaccination information and the inflow system, ensuring that vaccination information can be smoothly and accurately transferred and applied, and providing a solid foundation for vaccination decisions in the inflow area.
[0120] In the process of generating the multidimensional system adaptation index, first of all, it is necessary to extract all data fields from the historical vaccination information of the mobile vaccinated persons and map them one by one with the field types of the vaccination management system in the inflow area. Establish data format conversion rules to ensure the compatibility of field types and data formats. According to the mapping results, determine whether the data format of each field is compatible with the inflow system and mark it as compatible or incompatible.
[0121] Next, for each field type, calculate the matching degree between the historical vaccination information and the field type of the vaccination management system in the inflow area. The matching degree includes the type, structure, length and other attributes of the data field. Considering the matching of the data field in these three aspects, a comprehensive calculation is performed. The matching result needs to be converted through the Logit function to obtain an exponential score.
[0122] After completing the format conversion and type matching of all fields, evaluate the integrity of the data mapping between systems. Combine the format mapping results of each field with its corresponding type matching degree and multiply it by the weight value of the field in the system. Calculate the scores of all fields together to obtain the final multi-dimensional system adaptation index. This index reflects the overall compatibility between historical vaccination information and the vaccination management system of the inflow area, ensuring that data can be seamlessly connected and highly compatible when applied across systems.
[0123] The multidimensional system adaptation index effectively evaluates the compatibility between historical vaccination information and the vaccination management system of the inflow area through refined format conversion rules and type matching analysis, ensuring the accuracy and completeness of data in cross-system applications. This index provides a scientific basis for the cross-regional transfer of vaccination information, reduces the risk of data loss or misinterpretation due to system incompatibility, and thus improves the reliability and accuracy of vaccination decisions. Through the application of the multidimensional system adaptation index, the vaccination management system can better integrate vaccination data from different regions, provide more comprehensive and effective information support for the vaccination strategies of the inflow areas, and ultimately ensure public health safety.
[0124] In modern vaccination management systems, with the frequent cross-regional population mobility, historical vaccination information of vaccine recipients is often scattered in management systems in different regions. Since vaccination management systems in different regions may differ in data entry time, record order, etc., historical vaccination information is prone to time lag or disordered order when transferred across systems. This inconsistency may lead to inaccurate vaccination decisions and affect the immune protection effect of vaccine recipients in new areas. Therefore, it is necessary to evaluate and ensure the synchronization and consistency of vaccination information between different systems through the time series data consistency index. The index accurately analyzes the timestamps and order of vaccination records, identifies possible asynchronous entry and data lag problems, and provides a scientific basis for the accurate integration of vaccination information, ensuring that vaccine recipients can obtain timely and correct vaccination decision support after flowing into new areas.
[0125] When generating the time series data consistency index, first, extract the timestamp of each record from the historical vaccination information of the mobile vaccinated person and compare it with the corresponding record in the vaccination management system of the inflow area. For each pair of records, calculate the time difference between the two, and use logarithmic transformation to process the time difference to ensure that large time differences do not excessively affect subsequent calculations. The time difference data set is thus formed, which is used to determine the consistency of the timestamps between the two systems.
[0126] Next, check whether the historical vaccination information is consistent with the time sequence of the vaccination records in the inflow system. Compare each group of adjacent vaccination events in order to determine whether the order of the records in the two systems is the same. If the order is consistent, the score is full, if not, the score is zero, forming a sequential consistency score set.
[0127] The timestamp difference and the sequential consistency score are combined to calculate the synchronization score of each record between the two systems. The timestamp difference and the sequential consistency score are combined into a single score value through the Logit function to reflect the synchronization of the two systems. Finally, based on the synchronization score of all records, the temporal data consistency index of the entire data set is calculated to reflect the overall data synchronization and consistency between the historical vaccination information and the vaccination management system of the inflow area, ensuring that the data maintains temporal accuracy and sequential consistency between different systems.
[0128] The time series data consistency index conducts a comprehensive analysis of the timestamp differences and sequence consistency of vaccination records, effectively identifying and correcting the time and sequence inconsistencies of historical vaccination information between different systems, ensuring the timeliness and accuracy of the data. The index provides a reliable guarantee for the time synchronization and sequence consistency of vaccination information in cross-system applications, avoiding vaccination decision errors caused by data lags or sequence confusion. Through the application of the time series data consistency index, the vaccination management system can better integrate and utilize vaccination data from different regions, improve the vaccination effect of vaccine recipients in new areas, and ensure public health safety.
[0129] When generating a comprehensive judgment of a clear entry type, it is first necessary to normalize the multidimensional system adaptation index and the time series data consistency index, and convert the two indexes into values within the same scale range. Through logarithmic and radical operations, the indexes are converted into values that fluctuate within a similar interval. Next, the interaction between the two indexes is analyzed to obtain the interaction score, and the combined effect of the two is evaluated through the multiplication relationship. The multiplication operation amplifies the difference between the indexes, so that when one index is low, the overall evaluation score will drop rapidly, thus reflecting the potential uncertainty of the data.
[0130] Then, a dynamic threshold adjustment mechanism is applied to calculate a comprehensive threshold suitable for the current situation based on the historical data and actual vaccination cases in the inflow area. Different upper and lower threshold ranges are set to flexibly judge whether the entry type is met under different conditions. The threshold is dynamically adjusted according to the difference between the two indexes to ensure flexibility and accuracy of judgment.
[0131] Finally, the interaction score and dynamic threshold are combined to comprehensively judge whether the historical vaccination information is of the clear entry type. When the interaction score is higher than the dynamic threshold, it is judged as the clear entry type; if it is lower than the dynamic threshold, it is judged as the unclear entry type. This method comprehensively considers system compatibility and data synchronization, and through normalization processing, interaction analysis and dynamic threshold adjustment, it provides a flexible and accurate evaluation tool to ensure the accuracy and reliability of vaccination information in cross-system applications.
[0132] The present invention effectively evaluates the compatibility and synchronization of historical vaccination information in cross-system transfers through comprehensive calculations of the multi-dimensional system adaptation index and the time series data consistency index, ensuring the accuracy and completeness of the data in the vaccination management system of the inflow area. Through a detailed analysis of the data format, field type, timestamp, and sequence, possible inconsistencies between systems are identified and corrected, thereby avoiding vaccination decision errors caused by data loss, misreading, or time confusion. Ultimately, a comprehensive judgment of whether the historical vaccination information is of a clear input type provides a solid basis for vaccination decisions in the inflow area, improves the reliability and accuracy of the vaccination strategy, ensures the vaccination effect of the vaccine recipient in the new area, and ensures public health safety.
[0133] In modern vaccination management, against the background of increasingly frequent mobility, there is a complex matching relationship between the health status of the vaccine recipient, vaccination history and the public health needs of the inflow area. In order to ensure that each mobile vaccine recipient can obtain the most appropriate vaccination plan, it is necessary to accurately match the items of concern of the drug to be vaccinated with the specific situation of the vaccine recipient based on a scientific matching mechanism. Step S4 can effectively integrate information from different sources through systematic matching calculations and nonlinear combinations, generate a comprehensive matching score, and judge the compatibility of the vaccine with the vaccine recipient based on this. This process not only ensures the scientificity and personalization of the vaccination plan, but also improves the effectiveness and safety of vaccination, providing strong support for vaccination decisions.
[0134] Step S4 includes the following contents:
[0135] S4.1, obtain a list of items of concern for each drug to be vaccinated from step S2. The items of concern include key factors such as effectiveness, adaptability, vaccination time requirements, regional applicability, vaccination safety, etc. The items of concern have been processed by feature vectorization for subsequent calculations.
[0136] According to the historical vaccination information of the mobile inoculated people, relevant information is extracted and converted into feature vectors corresponding to the items of interest for the drugs to be vaccinated. For example, if a certain item of interest is "adaptability", the corresponding item may be information such as the health status, exposure risk or environmental adaptability of the mobile inoculated people in the inflow area. This information is also feature vectorized to ensure that it has the same dimension and structure as the vector of the item of interest.
[0137] S4.2, match the feature vector of the item of interest of the drug to be vaccinated with the feature vector of the corresponding item of the mobile person to be vaccinated. The matching calculation uses vector inner product or similarity calculation to evaluate the fit between each item of interest and the corresponding item. This process generates a list of matching scores, each score reflecting the degree of match between the item of interest and the corresponding item in a certain dimension.
[0138] All matching scores are nonlinearly combined to generate a comprehensive matching score. The matching results are converted into the final matching score through exponential functions or other nonlinear transformations. The nonlinear combination amplifies the impact of high matching degree on the final result, ensuring that the matching results can accurately reflect the fit between the mobile vaccine recipients and the drugs to be vaccinated.
[0139] S4.3, S4.3, compare the comprehensive matching score with the matching threshold to determine whether the comprehensive matching score meets the vaccination standard. Combined with the type of mobility, the public health strategy of the inflow area, and historical vaccination information, adjust the threshold to ensure the accuracy of the matching result and the actual vaccination needs. If the comprehensive matching score is higher than the matching threshold, it is judged that the relevant vaccine is highly consistent with the needs of the mobile recipients, and priority vaccination is recommended.
[0140] Generate a final matching result report, including the matching score of each drug to be vaccinated, recommended vaccination priority and related instructions. This report is used to guide medical institutions or vaccination management systems to provide the most suitable vaccination plan for mobile vaccinated people.
[0141] The present invention accurately matches the items of interest of the drugs to be vaccinated with the corresponding items of the mobile vaccine recipients, and uses advanced methods such as feature vector quantization, nonlinear combination, and dynamic threshold determination to generate a comprehensive matching score, thereby effectively improving the fit between the vaccine and the vaccine recipient. Through this matching mechanism, it can ensure that each mobile vaccine recipient obtains the vaccine that best suits their health status and immune needs in the new environment, thereby optimizing the vaccination effect, reducing vaccination risks, and improving the accuracy and reliability of the vaccination strategy. Ultimately, it provides a scientific decision-making basis for the vaccination management system, ensures public health safety, and maximizes the health benefits of the vaccine recipients.
[0142] Figure 2 The present invention provides an intelligent vaccination management system based on data analysis, comprising: a historical vaccination information collection module, a flow type determination module, a normative identification module and a corresponding item matching module;
[0143] The historical vaccination information collection module obtains the historical vaccination information of the mobile recipients and transmits the historical vaccination information data to the mobile type determination module;
[0144] The flow type determination module divides the flow recipients into inflow and outflow types according to the historical vaccination information, determines the priority of the drugs to be vaccinated according to the flow type, obtains the focus items of the drugs to be vaccinated, and transmits the flow type and the focus item data of the drugs to be vaccinated to the input normative identification module;
[0145] The standardization identification module identifies the standardization degree of the historical vaccination information relative to the inflow area according to the inflow type, determines whether it is an input clear type, and transmits the judgment result of the input clear type to the corresponding item matching module;
[0146] If the corresponding item matching module determines that a clear type of data has been entered, it will determine the corresponding items of the mobile persons to be vaccinated based on the items of interest of the drugs to be vaccinated, and obtain the matching results through matching analysis between the items of interest and the corresponding items.
[0147] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.
[0148] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.
[0149] It should be noted that, in this article, if there are relational terms such as first and second, etc., they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "including a..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0150] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A vaccination intelligent management method based on data analysis, characterized in that: Includes steps: S1, obtain the historical vaccination information of mobile people to be vaccinated; S2, according to the historical vaccination information, the mobile recipients are divided into inflow and outflow types, the priority of the drugs to be vaccinated is determined according to the flow type, and the concerned items of the drugs to be vaccinated are obtained; S3, for the inflow type, identify the degree of standardization of historical vaccination information relative to the inflow area and determine whether it is an entry-clear type; S4, if it is determined that the input type is clear, the corresponding item of the mobile person to be vaccinated is determined according to the concerned item of the drug to be vaccinated, and the matching result is obtained through matching analysis between the concerned item and the corresponding item; Items of concern refer to the characteristics or key factors of the drug to be vaccinated, including effectiveness, adaptability, vaccination time requirements, regional applicability, and vaccination safety, which are used to guide vaccination decisions; The corresponding items refer to the features extracted from the historical vaccination information of the mobile recipients, which are matched with the items of interest of the drugs to be vaccinated, and are used to evaluate the fit between the vaccine and the recipients; Step S2 includes the following contents: S2.1, obtain the geographic location information and the timestamp of each vaccination in the historical vaccination information of the mobile vaccine recipients; arrange the geographic location information in chronological order to form a time series trajectory data set; standardize the geographic location information from different sources to ensure that the location data is processed under a unified geographic coordinate system or regional coding system; then, determine the current residence of the mobile vaccine recipients and mark it as the current node; For each location in the historical trajectory, the spatial distance from the current residence is calculated, and the distances are arranged in chronological order to form a distance sequence; then, the distance sequence is differentially analyzed to calculate the distance change rate between adjacent historical locations, thereby forming distance change data; In the distance change data, pay attention to the distance change rate of the last position; if the distance gradually decreases in the last few positions and the distance between the terminal position and the current residence is less than the set threshold, and the time point is within the specified time, it is determined to be an inflow type; if the distance change rate of the historical trajectory shows a gradual increase in the last few positions, and the distance of the last position is greater than the distance of the initial position, it is determined to be an outflow type; S2.2, when determining the priority score of the drugs to be vaccinated, the mobility type of the mobile vaccine recipients shall be taken into consideration, and their vaccination history, geographic location information and current vaccination strategy shall be analyzed; if the mobile vaccine recipients are of the inflow type, their vaccination history records shall be analyzed according to the epidemiological data and vaccination policies of the inflow area to identify whether there are any missing vaccines related to infectious diseases in the corresponding area or the vaccination interval exceeds the recommended time; if it is found that the vaccine recipients are at risk of exposure in the inflow area and their vaccination history does not cover the major infectious diseases in the corresponding area, the relevant vaccines shall be set as high priority; then the vaccination time interval shall be analyzed, if the time interval between the last dose and the recommended time has passed, or the vaccination time is insufficient, resulting in the immunization effect not meeting the requirements, the priority of the relevant vaccines shall be increased to ensure that the vaccine recipients receive continuous protection; if the vaccine recipients are at risk of environmental exposure in the new area, the relevant vaccines shall be vaccinated; Construct a priority assessment model for preferential vaccination drugs based on inflow types: Data input and feature selection: A1. Select epidemiological data of the inflow area; A2. Includes historical vaccination information of mobile recipients, specific vaccination time, dose, and vaccination interval; A3. Analyze the environmental changes of the mobile vaccinated people from their original areas to the areas they move to; A4. Consider whether the interval between the last vaccination and the current recommended vaccination time exceeds the recommended protection period of the vaccine; If the mobile vaccinated person is of the outflow type, the epidemiological data of the target area shall be analyzed to assess the exposure risk in the new environment; if the infectious disease prevention and control requirements of the target area are different from those of the current area, the vaccine that can provide effective protection in the new area shall be administered; if the target area has high prevention and control requirements for certain infectious diseases, and the vaccine recipient has not yet received the relevant vaccine, the priority of such vaccine shall be increased to ensure that the person is immune protected before arriving in the new area; if the vaccination record shows that some vaccines have not been fully administered and prevention and control are still required in the target area, all doses shall be completed before leaving to ensure the continuity of the immune effect; Construct a priority assessment model for preferential vaccination drugs based on outflow types: B1. Analyze epidemiological data of target areas where outflows are heading; B2. Consider whether the vaccines related to the diseases controlled in the target area have been administered in the vaccination records of the mobile recipients, and whether the vaccination doses are complete; B3. Evaluate whether the vaccination history of mobile recipients meets the public health requirements of the target area; B4. Analyze the current vaccination records to see if there are any unfinished vaccine doses; The Logistic regression model was used for classification discussion: for the inflow type, the A1-A4 features were input; for the outflow type, the B1-B4 features were input, and the priority score for the vaccine to be administered was output; the basic formula of the model established a linear combination by taking each input feature as an independent variable, and converted it into a priority score through the Logistic function; Compare the output priority score with the corresponding threshold, collect the drugs to be vaccinated whose priority scores are greater than or equal to the corresponding threshold, and sort them according to the priority scores; S2.3, Acquisition of Focus Items: First, select the top-ranked drugs to be vaccinated and define preliminary items of concern from the characteristics of the drugs to be vaccinated; for each vaccine, perform preliminary screening based on the characteristics of the drugs; after the preliminary screening, organize these characteristics into a list of items of concern to be evaluated; Each candidate focus item is vectorized to make it a feature vector that is convenient for subsequent calculations. Specifically, each focus item is converted into a binary variable to indicate whether the relevant focus item is important in the current context. Step S3 includes the following contents: The process of obtaining the multi-dimensional system adaptation index: First, all data fields are extracted from the historical vaccination information of mobile vaccinated persons, and mapped one by one with the field types of the vaccination management system in the inflow area; based on the mapping results, it is determined whether the data format of each field is compatible with the inflow system and marked as compatible or incompatible; Next, for each field type, the matching degree between the historical vaccination information and the field type of the vaccination management system in the inflow area is calculated; the matching degree includes the type, structure, and length attributes of the data field; a comprehensive calculation is performed based on the matching of the data field in these three aspects; the matching result is converted through the Logit function to obtain an exponential score; After completing the format conversion and type matching of all fields, evaluate the integrity of the data mapping between systems; combine the format mapping result of each field with its corresponding type matching degree, and multiply it by the weight value of the corresponding field in the system; calculate the scores of all fields comprehensively to obtain the final multi-dimensional system adaptation index; The process of obtaining the time series data consistency index is as follows: First, the timestamp of each record was extracted from the historical vaccination information of the mobile vaccinated persons, and compared with the corresponding record in the vaccination management system of the inflow area; for each pair of records, the time difference between the two was calculated, and the time difference was processed using logarithmic transformation; Next, check whether the time sequence of historical vaccination information is consistent with the vaccination records in the inflow system; compare the order of each group of adjacent vaccination events to determine whether the order of records in the two systems is the same; if the order is consistent, the score is full, if not, the score is zero, forming a sequential consistency score set; Combine the timestamp difference and the sequential consistency score to calculate the synchronization score of each record between the two systems; use the Logit function to combine the timestamp difference and the sequential consistency score into a single score value to reflect the synchronization of the two systems; finally, calculate the time series data consistency index of the entire data set based on the synchronization scores of all records; When generating a comprehensive judgment of a clear entry type, it is first necessary to normalize the multidimensional system adaptation index and the time series data consistency index, converting the two indexes into values within the same scale range; through logarithmic and radical operations, convert the indexes into values that fluctuate within a similar interval; then, analyze the interaction between the two indexes to obtain the interaction score, and evaluate the combined effect of the two through the multiplication relationship; Finally, the interaction score and dynamic threshold are combined to comprehensively judge whether the historical vaccination information is of the clearly entered type; when the interaction score is higher than the dynamic threshold, it is judged as the clearly entered type; if it is lower than the dynamic threshold, it is judged as the unclearly entered type.
2. The method for intelligent management of vaccination based on data analysis according to claim 1, characterized in that: Step S4 includes the following contents: S4.1, obtaining a list of items of interest for each drug to be vaccinated from step S2; S4.2, matching and calculating the feature vectors of the items of interest of the drugs to be vaccinated and the feature vectors of the corresponding items of the mobile persons to be vaccinated; the matching calculation uses the vector inner product calculation method to evaluate the fit between each item of interest and the corresponding item; and a list of matching scores is generated, each score reflecting the degree of match between the item of interest and the corresponding item in a certain dimension; All matching scores are nonlinearly combined to generate a comprehensive matching score; S4.3, compare the comprehensive matching score and the matching threshold to determine whether the comprehensive matching score meets the vaccination standard; if the comprehensive matching score is higher than the matching threshold, it is judged that the relevant vaccine is highly consistent with the needs of the mobile people to be vaccinated.
3. An intelligent vaccination management system based on data analysis, used to implement an intelligent vaccination management method based on data analysis as described in any one of claims 1-2, characterized in that: include: Historical vaccination information collection module, flow type determination module, normative identification module and corresponding item matching module; The historical vaccination information collection module obtains the historical vaccination information of the mobile recipients and transmits the historical vaccination information data to the mobile type determination module; The flow type determination module divides the flow recipients into inflow and outflow types according to the historical vaccination information, determines the priority of the drugs to be vaccinated according to the flow type, obtains the focus items of the drugs to be vaccinated, and transmits the flow type and the focus item data of the drugs to be vaccinated to the input normative identification module; The standardization identification module identifies the standardization degree of the historical vaccination information relative to the inflow area according to the inflow type, determines whether it is an input clear type, and transmits the judgment result of the input clear type to the corresponding item matching module; If the corresponding item matching module determines that a clear type of data has been entered, it will determine the corresponding items of the mobile persons to be vaccinated based on the items of interest of the drugs to be vaccinated, and obtain the matching results through matching analysis between the items of interest and the corresponding items.
