Intelligent intelligent medical treatment management system and method

Abstract

The present application belongs to the technical field of medical and health information processing, and provides a smart intelligent medical management system and method. The fixed management relationship, dynamic activity relationship and outpatient disease information of unit personnel are acquired to form case correlation mapping data; the aggregation difference of cases in different relationships is analyzed according to disease types to determine disease effective correlation relationship and case correlation groups, and cross-group bridging relationship is identified; the case correlation groups are taken as the initial analysis range, the disease analysis range is expanded along the cross-group bridging relationship, and the disease effective analysis boundary is formed; the disease aggregation state is determined according to the boundary case aggregation degree, case time evolution degree and personnel correlation migration degree to generate unit disease aggregation medical warning results. The present application can reduce the case correlation omission across fixed management units and across dynamic activity groups and irrelevant personnel relationship interference, and improve the accuracy of unit disease aggregation identification and early warning.

Description

Technical Field

[0001] This invention relates to the field of medical and health information processing technology, and in particular to a smart intelligent medical treatment management system and method. Background Technology

[0002] With the advancement of information technology in healthcare management, medical records, organizational affiliation and accommodation management information, as well as various organizational activity records, are gradually being collected. During their time under organizational management, employees maintain relatively stable organizational affiliations and accommodation management relationships. Simultaneously, they develop dynamic activity relationships through shared participation in meetings, training sessions, shifts, tasks, and group activities. Therefore, a single case of illness may be associated with multiple fixed management groups and dynamic activity groups, and the actual scope of association for similar cases may extend beyond the original organizational or accommodation management units.

[0003] Compared to fixed management relationships, the temporal organization of different dynamic activities differs significantly. Some activities are repeated according to a predetermined schedule within a certain period, with personnel associations manifesting as multiple effective activity periods occurring at intervals; other activities are conducted continuously within a relatively concentrated time frame, with personnel associations primarily concentrated within the corresponding activity periods. The effective association time varies for different diseases; when cases occur within a certain timeframe after an activity ends, activity relationships may still explain the distribution of cases. Therefore, establishing case associations solely based on whether unit personnel participated in a particular activity, or uniformly based on the overall start and end times of the activity, may introduce personnel relationships where shared activity periods did not actually occur.

[0004] Traditional cluster analysis of disease among employees typically categorizes cases by fixed management units such as departments, work groups, or accommodation units, or by pre-defined activity ranges. When cases are linked due to shared activities across organizational or accommodation units, separately analyzing cases by fixed management units can easily fragment related cases into different analytical scopes. Conversely, including all activity relationships among employees in the same unit can lead to numerous duplicate or irrelevant relationships. Furthermore, the same case may be repeatedly covered by multiple fixed management relationships and multiple dynamic activity relationships, resulting in varying explanatory power of different relationships for the actual distribution of current disease cases.

[0005] Furthermore, the scope of existing disease analysis is usually determined based on fixed management units or pre-established personnel contact networks, making it difficult to further screen personnel relationships that have disease explanatory value based on the actual distribution of cases, and to dynamically adjust the scope of disease analysis by utilizing common personnel and time continuity relationships between different case-related groups. This can easily lead to the fragmentation or inappropriate expansion of the scope of disease clusters.

[0006] Therefore, there is an urgent need for an intelligent medical treatment management system and method that can combine the fixed management relationships of personnel in an organization with the dynamic activity relationships of different time periods, screen effective personnel association relationships based on disease type and actual case distribution, and dynamically determine the scope of disease cluster analysis. Summary of the Invention

[0007] In order to overcome the defects and shortcomings of the existing technology, the present invention provides a smart medical treatment management system and method.

[0008] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a smart medical treatment management system, comprising the following modules: Personnel Association Module: Obtains the fixed management relationships, dynamic activity relationships, and outpatient disease information of personnel in the unit, and maps the outpatient disease information to the corresponding fixed management relationships and dynamic activity relationships with the personnel in the unit as the association objects, forming case association mapping data; Association Analysis Module: Based on case association mapping data, it analyzes the clustering differences of cases in fixed management relationships and dynamic activity relationships according to disease type, determines the effective disease association relationship according to the association distribution of similar cases, and identifies cross-group bridging relationships connecting different case association groups; Boundary Evolution Module: Starting with the case-associated groups corresponding to the effective disease associations as the initial analysis scope, the module gradually acquires associated case-associated groups along cross-group bridging relationships and expands the disease analysis scope to form the effective disease analysis boundary; Clustering identification module: Within the effective analysis boundary of the disease, the disease clustering status is quantified based on the degree of clustering of boundary cases, the degree of case time evolution, and the degree of personnel association migration, and the unit disease clustering visit warning results are generated.

[0009] According to the above technical solution, the steps for obtaining fixed management relationships, dynamic activity relationships, and outpatient disease information of unit personnel, and mapping outpatient disease information to corresponding fixed management relationships and dynamic activity relationships using unit personnel as associated objects to form case association mapping data include: S110. Obtain the fixed management relationships, dynamic activity relationships, and outpatient disease information of the unit's personnel.

[0010] S120. Using the unit personnel as the association index, outpatient cases are associated with the fixed management relationship and the dynamic activity relationship corresponding to the case, and the dynamic activity relationship is matched with the effective time period of the activity corresponding to the case to form case association mapping data.

[0011] According to the above technical solution, the steps for analyzing the clustering differences of cases in fixed management relationships and dynamic activity relationships based on case association mapping data according to disease type, determining effective disease association relationships based on the association distribution of similar cases, and identifying cross-group bridging relationships connecting different case association groups include: S210. Group the case association mapping data according to disease type, and collect them separately according to fixed management relationship and dynamic activity relationship. Statistically analyze the number of covered cases, the number of covered personnel, and the time distribution characteristics of the association relationship corresponding to fixed management relationship and dynamic activity relationship, and calculate the corresponding case density to form disease relationship feature data.

[0012] S220. Based on the disease relationship characteristic data, compare the overlap between the case sets corresponding to fixed management relationships and dynamic activity relationships to determine the new explanatory power of fixed management relationships and dynamic activity relationships for disease types; combine the temporal consistency of cases covered by fixed management relationships and dynamic activity relationships, the effectiveness of fixed management relationships and dynamic activity relationships at the stage of case occurrence, and the degree of fit between disease types and fixed management relationship types and dynamic activity relationship types to determine the association explanatory power of fixed management relationships and dynamic activity relationships for the current disease type case distribution, and screen the effective disease association relationship set based on the association explanatory power.

[0013] S230. Based on the set of effective disease associations, the cases and individuals associated with the effective disease associations are grouped into case association groups. Individuals in different case association groups are compared to identify common individuals who simultaneously belong to different case association groups. Based on the time sequence of case occurrences and the effective association period corresponding to the dynamic activity relationship in different case association groups, the temporal continuity of the group connection relationship formed by the common individuals is verified, and a cross-group bridging relationship connecting different case association groups is generated.

[0014] According to the above technical solution, the steps of taking the case-related groups corresponding to the effective disease association as the initial analysis scope, gradually obtaining the associated case-related groups along the cross-group bridging relationship and expanding the disease analysis scope to form the effective disease analysis boundary include: S310. Based on cross-group bridging relationships, connect the case-related groups to form case-related group connectivity groups; determine the boundary evolution starting group based on the connection direction of cross-group bridging relationships within the case-related group connectivity groups, and aggregate the unit personnel range corresponding to the boundary evolution starting group to form the initial disease analysis scope; S320. Taking the case-related groups corresponding to the initial disease analysis scope as the starting point of boundary evolution, along the connection direction of cross-group bridging relationships, screen case-related groups that are connected to the case-related groups within the disease analysis scope but have not yet been included in the disease analysis scope. Include the screened case-related groups into the disease analysis scope and continue to evolve the boundary until there are no cross-group bridging relationships that can further expand the disease analysis scope, thus forming an effective disease analysis boundary.

[0015] According to the above technical solution, the steps for quantifying the disease clustering status within the effective disease analysis boundary based on the degree of clustering of boundary cases, the degree of case temporal evolution, and the degree of personnel association migration, and generating unit disease clustering medical visit early warning results include: S410. Determine the degree of clustering of boundary cases based on the number of cases and the number of people corresponding to the effective disease analysis boundary, determine the degree of case temporal evolution based on the change in the number of cases corresponding to the time of case occurrence, and determine the degree of human association migration based on the newly included case-related groups and corresponding cross-group bridging relationships during the formation of the effective disease analysis boundary, thereby forming the disease clustering status characteristics.

[0016] S420. Determine the disease cluster risk level based on the characteristics of disease clustering status, and generate unit disease clustering medical visit early warning results in combination with the effective disease analysis boundary.

[0017] Secondly, this application provides an electronic device, including a processor and a memory, wherein the memory stores a computer program that can be called by the processor, and the processor executes a smart medical treatment management system by calling the computer program stored in the memory.

[0018] Thirdly, this application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform a smart medical management system.

[0019] Compared with the prior art, this application has the following advantages and beneficial effects: By associating and matching the medical records of employees with fixed management relationships and dynamic activity relationships over time, and based on the case association distribution corresponding to different disease types, we can identify personnel relationships that have practical explanatory power for the current disease. This avoids interference from irrelevant relationships, dilution of local clustering characteristics, and omission of case associations across fixed management units caused by using fixed management scope or uniform personnel relationships for case aggregation. By distinguishing dynamic activity relationships with different organizational characteristics over time, and combining the case occurrence time with the corresponding effective activity period for relationship screening, invalid personnel associations introduced by treating the overall time span of periodic activities as continuous association time are reduced; the scope of disease analysis is dynamically determined through cross-group bridging relationships, retaining effective case associations formed across organizational management units and accommodation management units, reducing the possibility of disease clusters being fragmented or inappropriately expanded, and improving the accuracy and pertinence of disease cluster identification and early warning for medical treatment among personnel in the unit. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the structure of a smart medical treatment management system provided in an embodiment of this application; Figure 2 This is a flowchart illustrating the association resolution module provided in an embodiment of this application; Figure 3 This is a flowchart illustrating the boundary evolution module provided in an embodiment of this application; Figure 4 This is a flowchart illustrating the aggregation identification module provided in an embodiment of this application; Figure 5 This is a schematic diagram of a smart medical treatment management method provided in an embodiment of this application. Detailed Implementation

[0021] The technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments and specific features in the embodiments are detailed descriptions of the technical solution of this application, rather than limitations thereof. In the absence of conflict, the embodiments and technical features in the embodiments can be combined with each other.

[0022] Example 1

[0023] In the field of healthcare information processing, disease clustering identification is used to identify the concentrated distribution range of similar disease cases from outpatient case data, providing a basis for disease monitoring and early warning of medical visits. In the context of workplace health management, employees simultaneously have relatively stable fixed management relationships such as organizational affiliation and accommodation management, and can also form dynamic activity relationships due to joint participation in different activities. Employees of the same workplace may be in multiple relationships at the same time.

[0024] Traditional methods for identifying disease clusters typically group cases according to pre-defined organizational units, population groups, or contact relationships, and determine whether a cluster has formed based on the number of cases, the proportion of occurrence, and changes over time. However, when multiple interpersonal relationships coexist within an organization, existing methods still have the following limitations: First, existing disease cluster analyses typically use fixed management units such as organizational management units or accommodation management units, or a certain type of personnel activity relationship, as the scope of case aggregation. When personnel in a unit have multiple personnel relationships, such as organizational affiliation, accommodation management, and cross-management unit activities, using a single relationship for analysis may easily miss cases that are actually related between different management units, while directly merging multiple personnel relationships may easily expand the scope of case analysis and introduce personnel unrelated to the current disease.

[0025] Secondly, existing methods for case association based on shared activity records usually determine the association between cases and activities based on the start and end times of activities or preset time windows. However, different units have different frequencies and durations of activities. Using a uniform continuous time range or fixed time window can easily include time periods in which no actual shared activities occurred within the association scope, affecting the accuracy of matching between personnel activity relationships and case occurrences.

[0026] Third, the scope of existing disease cluster analysis is usually determined based on pre-set organizational management units, accommodation management units, activity groups, or contact networks. When similar cases gradually appear in multiple groups with different relationships between people, the existing analysis scope is difficult to adjust according to the actual correlation of cases, which can easily lead to the segmentation of related cases for analysis or the continuous expansion of the scope of irrelevant personnel.

[0027] In view of the above-mentioned technical limitations, such as Figure 1 As shown, this embodiment provides a smart medical treatment management system, which specifically includes the following modules: Personnel Association Module: Obtains the fixed management relationships, dynamic activity relationships, and outpatient disease information of personnel in the unit, and maps the outpatient disease information to the corresponding fixed management relationships and dynamic activity relationships with the personnel in the unit as the association objects, forming case association mapping data; Association Analysis Module: Based on case association mapping data, it analyzes the clustering differences of cases in fixed management relationships and dynamic activity relationships according to disease type, determines the effective disease association relationship according to the association distribution of similar cases, and identifies cross-group bridging relationships connecting different case association groups; Boundary Evolution Module: Starting with the case-associated groups corresponding to the effective disease associations as the initial analysis scope, the module gradually acquires associated case-associated groups along cross-group bridging relationships and expands the disease analysis scope to form the effective disease analysis boundary; Clustering identification module: Within the effective analysis boundary of the disease, the disease clustering status is quantified based on the degree of clustering of boundary cases, the degree of case time evolution, and the degree of personnel association migration, and the unit disease clustering visit warning results are generated.

[0028] In this embodiment, the fixed management relationships, dynamic activity relationships, and outpatient disease information of unit personnel are obtained. Using unit personnel as the associated objects, the outpatient disease information is mapped to the corresponding fixed management relationships and dynamic activity relationships, forming case association mapping data, which includes the following specific contents: S110. Obtain the fixed management relationships, dynamic activity relationships, and outpatient disease information of the unit's personnel.

[0029] The "unit" mentioned in this embodiment refers to an organizational entity that implements relatively stable organizational affiliation management, unified accommodation management, and phased collective activity management for its personnel. The personnel of the unit simultaneously have fixed management relationships based on organizational affiliation and accommodation affiliation, and can form phased dynamic activity relationships due to joint participation in different activities.

[0030] In this embodiment, fixed management relationships, dynamic activity relationships, and outpatient disease information of unit personnel are acquired to provide a data foundation for subsequent disease type-based relationship analysis and disease analysis boundary construction. Fixed management relationships characterize the relatively stable management affiliation of unit personnel within a certain timeframe; dynamic activity relationships characterize the personnel relationships formed due to joint participation in phased activities; and outpatient disease information records the occurrence of illnesses among unit personnel and their corresponding time information.

[0031] Specifically, the fixed management relationship originates from the unit's personnel management data and includes at least personnel identifiers, fixed management relationship identifiers, and fixed management relationship types. The fixed management relationship types include organizational affiliation and accommodation management, which are used to characterize the organizational affiliation and accommodation management relationships of the unit's personnel, respectively. The same unit's personnel can simultaneously correspond to both organizational affiliation fixed management relationships and accommodation management fixed management relationships. Dynamic activity relationships are derived from the unit's activity management data and include at least the dynamic activity identifier, dynamic activity type, participant identifier, and activity validity time information; for periodic dynamic activity relationships, the activity participation record records the corresponding participant identifiers according to each actual activity period.

[0032] Outpatient disease information comes from historical outpatient cases in the unit's medical management system. It includes at least the personnel ID corresponding to the case, the disease type, and the time of the case occurrence, and is used to determine the case set corresponding to different disease types.

[0033] Because personnel within a unit simultaneously possess both relatively stable fixed management relationships and dynamic activity relationships formed through shared activities, different relationships may overlap in coverage of the same case, and the time organization methods for different dynamic activities differ. Therefore, this embodiment simultaneously acquires fixed management relationships, dynamic activity relationships, and outpatient disease information, and records the corresponding relationship types and the effective time information of activities corresponding to dynamic activity relationships, providing a data foundation for subsequent case association mapping, identification of effective disease association relationships, and construction of disease analysis boundaries.

[0034] S120. Using the unit personnel as the association index, outpatient cases are associated with the fixed management relationship and the dynamic activity relationship corresponding to the case, and the dynamic activity relationship is matched with the effective time period of the activity corresponding to the case to form case association mapping data.

[0035] Using the personnel identifier corresponding to the case in the outpatient disease information as the association index, it is associated with the personnel identifier in the fixed management relationship and the participant identifier in the dynamic activity relationship, respectively, to determine the set of fixed management relationships to which the personnel of the unit corresponding to each case belong, and the set of dynamic activity relationships to which the personnel of the unit corresponding to the case participate.

[0036] The fixed management relationship type is used to distinguish relatively stable management affiliations among personnel within a unit, including fixed management relationships based on organizational affiliation and fixed management relationships based on accommodation. The dynamic activity type is used to distinguish phased personnel relationships formed through different implementation methods, including periodic dynamic activity relationships and concentrated dynamic activity relationships. Periodic dynamic activity relationships are those that are repeatedly carried out multiple times according to a predetermined schedule within a certain period. The corresponding effective activity time information includes the start and end times of each actual dynamic activity, consisting of multiple intermittent actual activity periods. Concentrated dynamic activity relationships are those that are carried out within a single continuous time frame. The corresponding effective activity time information includes the start and end times of the dynamic activity.

[0037] Historical outpatient cases in the outpatient disease information are grouped according to disease type. For historical outpatient cases corresponding to each disease type, the effective time period information of each dynamic activity relationship involving the personnel of the corresponding unit is obtained. For a dynamic activity relationship corresponding to a case, the actual activity time period in which the personnel of the unit corresponding to the case actually participated is obtained; when the case occurrence time falls within any actual activity time period in which the personnel of the unit corresponding to the case actually participated, the case activity time interval is set to 0; when the case occurrence time is later than one or more actual activity time periods in which the personnel of the unit corresponding to the case actually participated, the actual activity time period with the closest end time to the case occurrence time is selected from the actual activity time periods in which the personnel of the unit corresponding to the case actually participated and which are earlier than the case occurrence time, the difference between the case occurrence time and the end time of the actual activity time period is calculated, and the difference is determined as the case activity time interval; when the case occurrence time is earlier than the first actual activity time period in which the personnel of the unit corresponding to the case actually participated, no case activity time interval sample is formed between the case and the corresponding dynamic activity relationship.

[0038] Each case activity time interval obtained in the above manner is taken as a case activity time interval sample, forming case activity time interval sample data corresponding to each disease type; when the same case corresponds to multiple dynamic activity relationships, the corresponding case activity time interval samples are obtained respectively.

[0039] Based on the case activity time interval samples corresponding to each case activity time interval sample, the case activity time interval samples are divided into 0-day time interval group, 1-day time interval group, 2-day time interval group, and other time interval groups in ascending order. The number of case activity time interval samples corresponding to each time interval group is counted to form the time interval distribution data corresponding to the current disease type.

[0040] In this embodiment, the length of the preset time interval is set to 1 day. The basis for this setting is that the unit's activity arrangements and the outpatient cases of the unit's personnel can be recorded on a daily time scale. Therefore, 1 day is used as the statistical granularity of the time interval to characterize the distribution of the number of cases occurring in different time ranges after the actual activity period ends.

[0041] Based on the time interval distribution data, starting from the 0-day time interval group, the number of case activity time interval samples corresponding to each time interval group is accumulated sequentially in ascending order of the number of days in the time interval. For each time interval group, the sum of the number of case activity time interval samples from the 0-day time interval group to that time interval group is divided by the sum of the number of case activity time interval samples corresponding to all time interval groups for the current disease type to obtain the cumulative sample coverage ratio for that time interval group.

[0042] When the cumulative sample coverage ratio corresponding to a certain time interval group reaches the preset coverage ratio threshold for the first time, the maximum time interval corresponding to that time interval group is determined as the maximum associated time interval corresponding to the current disease type.

[0043] Based on the maximum association time interval corresponding to the current disease type, the case activity time interval between each case and the dynamic activity relationship is screened. When the case activity time interval is less than or equal to the maximum association time interval corresponding to the current disease type, the association between the current case and the corresponding dynamic activity relationship is retained, and the actual activity period used to determine the case activity time interval is determined as the effective association period corresponding to the current case. When the case activity time interval is greater than the maximum association time interval corresponding to the current disease type, or when no case activity time interval sample is formed, the association between the current case and the corresponding dynamic activity relationship is removed.

[0044] After completing the dynamic activity relationship screening, the case-related disease information, fixed management association information, and effective dynamic activity association information are linked and stored to form case association mapping data. Among them, the case-related disease information includes the disease type and the time of case occurrence, the fixed management association information includes the fixed management relationship identifier and the fixed management relationship type, and the effective dynamic activity association information includes the dynamic activity identifier, the dynamic activity type, and the effective association time period corresponding to the current case.

[0045] In this embodiment, as Figure 2 As shown, based on case association mapping data, the clustering differences of cases in fixed management relationships and dynamic activity relationships are analyzed according to disease type. Effective disease association relationships are determined based on the association distribution of similar cases, and cross-group bridging relationships connecting different case association groups are identified. The specific steps include: S210. Group the case association mapping data according to disease type, and collect them separately according to fixed management relationship and dynamic activity relationship. Statistically analyze the number of covered cases, the number of covered personnel, and the time distribution characteristics of the association relationship corresponding to fixed management relationship and dynamic activity relationship, and calculate the corresponding case density to form disease relationship feature data.

[0046] The case association mapping data is grouped and aggregated according to the disease type in the case association mapping data to obtain the case association mapping set corresponding to each disease type; For each disease type, the case association mapping set is independently grouped and collected according to fixed management relationship type and fixed management relationship identifier, and dynamic activity type and dynamic activity identifier. Case records with the same fixed management relationship type and fixed management relationship identifier are assigned to the corresponding fixed management relationship case set, and case records with the same dynamic activity type and dynamic activity identifier are assigned to the corresponding dynamic activity relationship case set. The same case record can belong to one or more fixed management relationship case sets and one or more dynamic activity relationship case sets. For each fixed management relationship case set and each dynamic activity relationship case set, the case records are deduplicated according to the personnel identifier corresponding to the case. When there are multiple case records for the same unit personnel corresponding to the current disease type, the case record with the earliest occurrence time is retained as the deduplicated case record. The deduplicated case records are used as the corresponding case set to participate in the subsequent case set calculation, and the number of deduplicated cases is counted to obtain the number of covered cases corresponding to each fixed management relationship and each dynamic activity relationship, so as to avoid the repeated calculation of the number of covered cases due to multiple visits by the same unit personnel for the same disease.

[0047] In subsequent comparisons involving unions, differences, and overlaps between case sets with different relationships, the personnel identifier corresponding to each case is used as the unique identifier to determine whether the elements of the case sets are the same. When the same personnel identifier corresponds to different case occurrence times in the case set union operation, the case record with the earliest occurrence time is retained as the case record corresponding to that personnel in the union result.

[0048] Furthermore, based on the personnel management data of the unit corresponding to the fixed management relationship identifier and the activity participation record corresponding to the dynamic activity identifier, the covered personnel set corresponding to each fixed management relationship and each dynamic activity relationship is obtained, and the number of personnel in each covered personnel set is counted to obtain the number of covered personnel corresponding to each fixed management relationship and each dynamic activity relationship. Among them, the set of personnel covered by the fixed management relationship is the set of all personnel from units that belong to the corresponding fixed management relationship during the current disease cluster analysis period (including personnel from units that are not infected), and the set of personnel covered by the dynamic activity relationship is the set of all personnel from units that actually participated in the corresponding dynamic activity (including personnel from units that are not infected).

[0049] The disease cluster analysis period is the global time boundary for this unit's disease cluster analysis, determined according to the preset statistical period; if no preset statistical period is set, the period for this disease cluster analysis is the earliest occurrence time to the latest occurrence time of all cases of the current disease type.

[0050] Divide the number of covered cases corresponding to each fixed management relationship and each dynamic activity relationship by the number of covered personnel to obtain the case density corresponding to each fixed management relationship and each dynamic activity relationship. This density is used to characterize the concentration of similar cases within the covered personnel of different relationships.

[0051] Simultaneously, based on the aforementioned fixed management relationship case set and dynamic activity relationship case set, and according to the occurrence time of cases in each case set, the temporal distribution characteristics of the corresponding association relationships for each fixed management relationship and each dynamic activity relationship are statistically analyzed: For a set of cases with fixed management relationships, the daily occurrence time distribution of cases corresponding to each fixed management relationship is obtained by statistically analyzing the occurrence time distribution of cases corresponding to fixed management relationships on a calendar day basis. For a set of cases with dynamic activity relationships, the activity time intervals of each case determined by S120 are called, and the number of cases corresponding to each activity time interval is counted according to a preset interval of 1 day, so as to obtain the time interval distribution of cases corresponding to dynamic activity relationships.

[0052] The temporal distribution characteristics of association relationships include the daily occurrence time distribution of cases corresponding to fixed management relationships and the time interval distribution of cases corresponding to dynamic activity relationships; among them, the statistical granularity of the temporal distribution characteristics of association relationships is consistent with S120, using 1 day as the minimum time interval; Based on the number of cases covered, the number of people covered, the case density, and the time distribution characteristics of the correlation relationships for each disease type, the data are collected according to disease type, relationship type, and corresponding relationship identifier to form disease relationship feature data.

[0053] S220. Based on the disease relationship characteristic data, compare the overlap between the case sets corresponding to fixed management relationships and dynamic activity relationships to determine the new explanatory power of fixed management relationships and dynamic activity relationships for disease types; combine the temporal consistency of cases covered by fixed management relationships and dynamic activity relationships, the effectiveness of fixed management relationships and dynamic activity relationships at the stage of case occurrence, and the degree of fit between disease types and fixed management relationship types and dynamic activity relationship types to determine the association explanatory power of fixed management relationships and dynamic activity relationships for the current disease type case distribution, and screen the effective disease association relationship set based on the association explanatory power.

[0054] For the disease relationship feature data corresponding to the current disease type, fixed management relationships and dynamic activity relationships are selected as candidate associations, and then sorted from high to low according to the case density corresponding to each candidate association.

[0055] Fixed management relationships include fixed management relationships based on organizational affiliation and fixed management relationships based on accommodation management; dynamic activity relationships include periodic dynamic activity relationships and centralized dynamic activity relationships. Each candidate relationship is distinguished by a corresponding relationship identifier, and different organizational affiliations, accommodation management relationships, or dynamic activities under the same relationship type are treated as independent candidate relationships for subsequent analysis.

[0056] Let the current disease type be d, and the i-th candidate association be d. Candidate relationships include fixed management relationships based on the relatively stable management affiliation of personnel within an organization, as well as dynamic activity relationships.

[0057] The i-th candidate association The corresponding case set is denoted as Let the set of all cases corresponding to the current disease type be denoted as .

[0058] Candidate associations are processed sequentially from high to low according to the case density ranking. For the i-th candidate association, the case sets corresponding to the effective disease associations ranked before the i-th candidate association and whose association explanatory power reaches the preset explanatory power threshold are subjected to a union operation to obtain the included case set. Candidate associations that have not yet reached the preset explanatory power threshold will not be included in the already included case set.

[0059] Perform a difference operation between the set of cases corresponding to the i-th candidate association and the set of cases already included to obtain the set of newly explained cases. .

[0060] The newly added explanatory case set is used to characterize cases that are not yet covered by other candidate associations ranked higher in the current candidate association coverage, that is, cases that the current candidate association can newly explain relative to the existing case explanation range.

[0061] Furthermore, based on the number of cases corresponding to the newly added explanatory case set. Number of new personnel Number of cases covered by candidate associations and the total number of cases of the current disease type Calculate the explanatory contribution of new cases: ; in: The contribution to the new case explanation corresponding to the i-th candidate association is dimensionless and is used to characterize the new case explanation value of the i-th candidate association after excluding duplicate explanations from other candidate associations; The larger the value, the more independent case clusters formed by the corresponding fixed management relationship or dynamic activity relationship can be found in the candidate association; The number of newly explained cases corresponding to the i-th candidate association is obtained by performing a difference operation between the current candidate association's case set and the already included case set. The number of new personnel corresponding to the i-th candidate association is obtained by performing a difference operation between the set of personnel covered by the current candidate association (output by calling S210) and the set of personnel covered by the already included associations. It is used to represent the scale of new associated personnel in the current candidate association. Let be the number of cases corresponding to the i-th candidate association, used to represent the size of the cases corresponding to the current candidate association; The total number of cases corresponding to the current disease type is obtained by deduplicating the cases belonging to the current disease type in the case association mapping data according to the case deduplication rules described in S210.

[0062] A non-zero smoothing parameter is preset to limit the minimum denominator of the first term corresponding to the intensity of local case clustering; in this embodiment, it is set to 1. When the number of newly added personnel... When the value is 0, as a special calculation rule, the first term corresponding to the local case clustering intensity in the above formula is set to 0, while the second term corresponding to the independent case explanatory power and the third term corresponding to the global case coverage power are still calculated according to the above formula; when the number of new personnel When the value is greater than 0, the contribution of newly added cases is calculated according to the above formula. This avoids incorrectly assigning the maximum local case cluster strength when no new covered persons are added to the current candidate association.

[0063] in, , , The fusion weights represent the local case clustering intensity, the explanatory power of independent cases, and the global case coverage, respectively. These are all adjustable parameters ranging from [0,1] and satisfying the following conditions: This makes the range of the newly added explanatory contribution of the case [0,1]; This is used to characterize the concentration of cases within the scope of newly added personnel in the current candidate association, and to avoid the dilution of local high-cluster case characteristics due to the large scale of personnel covered by fixed management relationships; Used to characterize the independent case explanation ability provided by the current candidate association, reducing the contribution of associations that highly overlap with other candidate associations but cannot provide new explanations; Used to characterize the extent to which newly added explanatory cases of the current candidate association cover the overall case range of the current disease type.

[0064] By calculating the contribution of newly added cases, the evaluation results of associations that cover a large number of cases but mainly repeat the scope of existing relationships are reduced, while highlighting fixed management relationships or dynamic activity relationships that can reflect the true scope of disease clustering.

[0065] Furthermore, for each candidate association, corresponding case time variables are determined based on the association type: For fixed management relationships, the occurrence date of each case is converted into a relative day number based on the start date of this disease cluster analysis period, serving as the corresponding case time variable to characterize the time distribution of cases covered by the fixed management relationship within the analysis period; for dynamic activity relationships, the case activity time intervals corresponding to each case determined in S120 are used as the corresponding case time variables to characterize the time distribution of case occurrence time relative to the actual association period of the corresponding dynamic activity. Specifically, for periodic dynamic activity relationships, the case activity time interval is the case activity time interval determined in S120 based on the effective association period corresponding to the current case.

[0066] For each candidate association, calculate the difference between the time variable of each case and the average time variable of all cases in the corresponding candidate association. Square all differences and sum them to obtain the sum of squared time biases. Divide the sum of squared time biases by the number of cases in the corresponding candidate association to obtain the average squared time bias. Take the square root of the average squared time bias to obtain the standard deviation of the time variable of the case corresponding to the i-th candidate association. This refers to the degree of temporal dispersion of cases.

[0067] The maximum association time interval corresponding to the current disease type Using the normalized benchmark of temporal dispersion, the case-time consistency index corresponding to the i-th candidate association is calculated according to the following formula: ; in, is the case-time consistency index corresponding to the i-th candidate association, with a value range of [0,1]. The larger the value, the more concentrated the time distribution of the cases covered by the current candidate association is under the corresponding time reference. Among them, the fixed management relationship uses the relative day number corresponding to the date of occurrence of the case as the time reference, and the dynamic activity relationship uses the time interval of the case activity as the time reference.

[0068] The maximum association time interval corresponding to the current disease type is used to characterize the time scale in which the current disease cases have statistical association significance under the personnel association relationship. Therefore, it serves as a unified normalization benchmark for the dispersion of the occurrence time of cases with fixed management relationship and the dispersion of the activity time interval of cases with dynamic activity relationship. When the maximum association time interval corresponding to the current disease type is 0, 1 day is used as the minimum normalization benchmark, which is consistent with the minimum time statistical granularity of 1 day used in S120.

[0069] Furthermore, based on the corresponding status between fixed management relationships and dynamic activity relationships and the time of case occurrence, the relationship time validity index corresponding to each candidate association relationship is calculated.

[0070] For fixed management relationships, since the fixed management relationships of organizational affiliation and accommodation management are relatively stable management affiliations during this disease cluster analysis period, the relationship time validity index corresponding to the fixed management relationship is set to 1.

[0071] For dynamic activity relationships, calculate the case activity time matching score for each case. When the case activity time interval is 0, i.e., the case occurrence time falls within the valid association time period corresponding to the current case determined in S120, the case activity time matching score is set to 1; when the case activity time interval is greater than 0, the case activity time matching score is calculated according to the following formula: ; in, Let j be the time interval between the activity of the j-th case; The score is the match score for the activity time of the j-th case.

[0072] The average of the case activity time matching scores for all cases covered by the current dynamic activity relationship is used to obtain the relationship time validity index corresponding to that dynamic activity relationship. : ; in, This represents the number of covered cases corresponding to the i-th dynamic activity relationship; The value range is [0,1], which is used to characterize the degree of matching between the time of occurrence of the case and the effective associated time period corresponding to the current case.

[0073] The correlation between dynamic activity relationships and case occurrence time is evaluated by using the relationship time validity index. This reduces the evaluation results of dynamic activity relationships that are only based on the participation of personnel in the corresponding dynamic activities, but whose case occurrence time is far from the actual activity period.

[0074] Furthermore, based on the historical case association characteristics corresponding to the current disease type, the relationship type fit coefficients corresponding to different relationship types are determined. This is used to characterize the degree of fit between different relationship types and the current disease case distribution.

[0075] The values ​​for case-specific contribution, case-time consistency index, and relation-time validity index are all within the range of [0,1]. The association explanatory power is calculated using the following formula, combined with the relation type fit coefficient corresponding to the current disease type: ; in: The explanatory power of the i-th association for the current disease type; The matching coefficient for the current disease type is used to reflect the degree of matching between the current disease type and the relationship type to which the i-th candidate association belongs. The relationship type is a fixed management relationship of organization affiliation, a fixed management relationship of accommodation management, a periodic dynamic activity relationship, or a centralized dynamic activity relationship. Contributed to the explanation of new cases; As an indicator of case-time consistency; This is a time-based validity indicator for the relationship.

[0076] Candidate associations are processed in descending order of case density. For each candidate association, the contribution of new cases to the explanation of the current case set is calculated, and the explanatory power of the association is calculated by combining the case temporal consistency index, the relationship temporal validity index, and the relationship type fit coefficient.

[0077] When the explanatory power of a candidate association reaches a preset explanatory power threshold, the candidate association is determined as a valid disease association, and its corresponding case set is merged into the already included case set, and its corresponding covered personnel set is merged into the already included association covered personnel set; when the explanatory power of the association does not reach the preset explanatory power threshold, the already included case set and the already included association covered personnel set are not updated.

[0078] After processing the current candidate association, continue processing the next candidate association in the sorting until all candidate associations are processed, forming a set of valid disease associations corresponding to the current disease type.

[0079] S230. Based on the set of effective disease associations, the cases and individuals associated with the effective disease associations are grouped into case association groups. Individuals in different case association groups are compared to identify common individuals who simultaneously belong to different case association groups. Based on the time sequence of case occurrences and the effective association period corresponding to the dynamic activity relationship in different case association groups, the temporal continuity of the group connection relationship formed by the common individuals is verified, and a cross-group bridging relationship connecting different case association groups is generated.

[0080] Based on the set of valid disease associations determined in S220, the set of cases and the set of covered persons corresponding to each valid disease association are grouped into a case association group.

[0081] Intersection operations are performed on the sets of individuals corresponding to different case-associated groups. Individuals belonging to two or more case-associated groups are identified as common individuals. Common individuals are associated with both case-associated groups through their respective valid disease associations, thus forming candidate cross-group bridging relationships. It is not required that the two case-associated groups correspond to the same fixed management relationship identifier or dynamic activity identifier.

[0082] For candidate cross-group bridging relationships, the occurrence time of the first case in two case-associated groups is compared. The case-associated group with the earlier occurrence time of the first case is identified as the source case-associated group, and the case-associated group with the later occurrence time of the first case is identified as the target case-associated group.

[0083] When the first case of two case-related groups occurs at the same time, the case-related groups corresponding to the effective disease associations of the two case-related groups are determined as the source case-related groups according to the order of processing of the effective disease associations in S220. The case-related groups corresponding to the effective disease associations processed first are determined as the target case-related groups. The source and target are only used to determine the traversal direction of the subsequent disease analysis boundary and are not used to indicate the actual order of disease transmission.

[0084] When the target case is associated with a dynamic activity relationship, the common personnel corresponding to the candidate cross-group bridging relationship are verified based on the activity participation record of the effective association period corresponding to the first case of the target case. The common personnel who actually participated in the effective association period are retained, and the retained common personnel are used as the common personnel corresponding to the current candidate cross-group bridging relationship. When there are no common personnel who actually participated in the effective association period, the corresponding candidate cross-group bridging relationship is removed.

[0085] When the target case is associated with a dynamic activity relationship, and the first case in the target case-associated group occurs later than the first case in the source case-associated group, the effective association period determined when establishing the case association mapping between the first case in the target case and the corresponding dynamic activity relationship in S120 is called, and the time range between the first case in the source case-associated group and the first case in the target case-associated group is determined as the group connection time range.

[0086] When there is no overlap between the effective association period and the group connection time range, the candidate cross-group bridging relationship is determined not to meet the bridging temporal continuity condition; when there is an overlap between the effective association period and the group connection time range, and the occurrence time of the first case in the target case-related group is within the effective association period, the bridging time interval is set to 0; when there is an overlap, and the occurrence time of the first case in the target case-related group is later than the end time of the effective association period, the difference between the occurrence time of the first case in the target case-related group and the end time of the effective association period is determined as the bridging time interval.

[0087] If the bridging time interval is less than or equal to the maximum association time interval corresponding to the current disease type, the candidate cross-population bridging relationship is determined to meet the bridging temporal continuity condition; otherwise, the candidate cross-population bridging relationship is determined not to meet the bridging temporal continuity condition.

[0088] When the target case-associated group corresponds to a fixed management relationship, and the occurrence time of the first case in the target case-associated group is later than the occurrence time of the first case in the source case-associated group, the time interval between the occurrence times of the first cases in the two case-associated groups is calculated; when the time interval is less than or equal to the maximum association time interval corresponding to the current disease type, the candidate cross-group bridging relationship is determined to meet the bridging temporal continuity condition; otherwise, it is determined not to meet the bridging temporal continuity condition.

[0089] Candidate cross-population bridging relationships with the same occurrence time of the first case or that meet the bridging sequence continuity condition are retained, while other candidate cross-population bridging relationships are eliminated.

[0090] Based on the retained candidate cross-group bridging relationships corresponding to common personnel, source case-related groups, target case-related groups, and traversal directions, directed cross-group bridging relationships are formed.

[0091] When forming a directed cross-group bridging relationship, the system saves the effective disease association types corresponding to the source case association group and the target case association group respectively; when the source case association group or the target case association group corresponds to a dynamic activity relationship, the system saves the corresponding dynamic activity identifier and dynamic activity type; when the target case association group corresponds to a dynamic activity relationship, the system also saves the effective association time period used for bridging temporal continuity verification.

[0092] Specifically, when the target case-associated group corresponds to a fixed management relationship, the current disease cluster analysis period is used as the effective bridging time range for the corresponding cross-group bridging relationship; when the target case-associated group corresponds to a dynamic activity relationship, the effective association period used for the continuity verification of the bridging sequence of the first case in the target case-associated group is used as the effective bridging time range for the corresponding cross-group bridging relationship.

[0093] In this embodiment, as Figure 3 As shown, the initial analysis scope is the case-associated population corresponding to the effective disease association. The analysis then expands the disease analysis scope by progressively obtaining associated case-associated populations along cross-population bridging relationships, forming the effective disease analysis boundary. This includes the following specific steps: S310. Based on cross-group bridging relationships, connect the case-related groups to form case-related group connectivity groups; determine the boundary evolution starting group based on the connection direction of cross-group bridging relationships within the case-related group connectivity groups, and aggregate the personnel range of the units corresponding to the boundary evolution starting group to form the initial disease analysis scope.

[0094] Based on the cross-population bridging relationship determined in S230, without ignoring the bridging direction, all case-related groups corresponding to the current disease type are grouped into connectivity groups, and case-related groups directly or indirectly connected through one or more cross-population bridging relationships are assigned to the same connectivity group.

[0095] For each connected group, traverse the directed cross-group bridging relationships within the connected group, count the number of times each case-associated group appears as the target case-associated group, and obtain the number of incoming bridging relationships corresponding to each case-associated group; determine the case-associated groups with 0 incoming bridging relationships as the boundary evolution starting groups, and form the boundary evolution starting group set of the corresponding connected group.

[0096] The case sets corresponding to each case-associated group in the initial set of boundary evolution are combined to form the initial case set of the corresponding connected group; the personnel sets corresponding to each case-associated group are combined to form the initial personnel set of the corresponding connected group, and the initial personnel set is determined as the initial disease analysis scope corresponding to the connected group.

[0097] S320. Taking the case-related groups corresponding to the initial disease analysis scope as the starting point of boundary evolution, along the connection direction of cross-group bridging relationships, screen case-related groups that are connected to the case-related groups within the disease analysis scope but have not yet been included in the disease analysis scope. Include the screened case-related groups into the disease analysis scope and continue to evolve the boundary until there are no cross-group bridging relationships that can further expand the disease analysis scope, thus forming an effective disease analysis boundary.

[0098] For each connected group, the boundary evolution starting group determined in S310 is taken as the case-associated group that has been included in the disease analysis at the beginning of the first round of expansion, and the initial case set and initial personnel set formed in S310 are taken as the case set and personnel set that have been gathered at the beginning of the first round of expansion, respectively; the case-associated groups that have been included in the disease analysis at the beginning of subsequent rounds of expansion are the boundary evolution starting group and the target end case-associated groups newly included in previous rounds.

[0099] In each round of expansion, the directed cross-population bridging relationship formed by S230 is invoked, and the source-end case-associated population and target-end case-associated population of each directed cross-population bridging relationship are compared with the case-associated populations already included in the disease analysis scope up to the start of this round of expansion. When the source-end case-associated population belongs to the case-associated populations already included in the disease analysis scope and the target-end case-associated population does not belong to the case-associated populations already included in the disease analysis scope, the corresponding directed cross-population bridging relationship is determined as a candidate expansion bridging relationship, and the corresponding target-end case-associated population is determined as a candidate expanded case-associated population.

[0100] For each candidate extended case association group, the corresponding case set is compared with the case set already collected up to the start of this round of expansion to obtain the corresponding boundary new case set; the difference operation between the case sets is performed according to the case set element comparison rules determined in S210. When the boundary new case set is not empty, the corresponding candidate extended case association group is retained as the case association group to be included in this round; when the boundary new case set is empty, the scope of disease analysis is not further expanded through the corresponding candidate extended bridging relationship.

[0101] When there are duplicate case-related groups to be included in this round, duplicate case-related groups to be included are deduplicated; within the same round, the deduplicated case-related groups to be included are simultaneously included in the disease analysis scope, and their corresponding case sets and personnel sets are respectively joined with the case sets and personnel sets already collected in the previous round.

[0102] Simultaneously, the current expansion round, the newly included case-related groups in the current round, and the directed cross-group bridging relationships used to connect the corresponding source-end case-related groups and target-end case-related groups are recorded to form a boundary evolution record.

[0103] After completing the current round of expansion, the next round of expansion will continue based on the case-related groups already included in the expansion; the iteration will stop when there are no candidate expanded case-related groups with newly added cases with boundaries.

[0104] The case-related groups that have been included after the iteration stops, along with their corresponding case sets, personnel sets, and boundary evolution records, are identified as the effective disease analysis boundaries for the corresponding connected groups.

[0105] In this embodiment, as Figure 4 As shown, within the effective disease analysis boundary, the disease clustering status is quantified based on the degree of clustering of boundary cases, the degree of case temporal evolution, and the degree of personnel association and migration, generating unit disease clustering medical visit early warning results, including the following specific steps: S410. Determine the degree of clustering of boundary cases based on the number of cases and the number of people corresponding to the effective disease analysis boundary, determine the degree of case temporal evolution based on the change in the number of cases corresponding to the time of case occurrence, and determine the degree of human association migration based on the newly included case-related groups and corresponding cross-group bridging relationships during the formation of the effective disease analysis boundary, thereby forming the disease clustering status characteristics.

[0106] Based on the effective disease analysis boundary determined in S320, obtain the case association groups that have been included after the iteration stops, along with their corresponding case sets, personnel sets, and boundary evolution records. From the boundary evolution records, obtain the target-end case association groups newly included in each expansion round and their corresponding directed cross-group bridging relationships. Then, call the effective disease association types and bridging effective time ranges corresponding to the source-end case association groups and target-end case association groups in S230 to form the boundary evolution association information corresponding to the effective disease analysis boundary.

[0107] Based on the personnel identifiers corresponding to each case in the case set corresponding to the effective disease analysis boundary, a boundary disease personnel set is formed; the number of personnel in the boundary disease personnel set and the total number of personnel in the case set corresponding to the effective disease analysis boundary are counted, and the degree of boundary case clustering is obtained by dividing the number of personnel in the boundary disease personnel set by the total number of personnel.

[0108] Furthermore, according to the 1-day statistical granularity determined by S120, the number of cases corresponding to the current time interval and the previous time interval within the effective analysis boundary of the disease is counted. When the number of cases in the previous time interval is greater than 0, the difference between the number of cases in the current time interval and the number of cases in the previous time interval is divided by the number of cases in the previous time interval to obtain the change ratio of the number of cases. When the change in the number of cases is greater than 1, the change in the number of cases is corrected to 1; when the change in the number of cases is less than 0, the change in the number of cases is corrected to 0. The corrected change in the number of cases is then determined as the degree of case evolution over time.

[0109] If the number of cases in the current time interval is 0, and the number of cases in the current time interval is greater than 0, then the case time evolution degree is set to 1; otherwise, it is set to 0.

[0110] The value of the case time evolution degree ranges from [0,1]. The larger the value of the case time evolution degree, the more obvious the growth trend of the number of cases. Among them, the current time interval is the most recent natural day that has ended before the time point of this disease cluster analysis, and the previous time interval is the natural day before the current time interval.

[0111] Furthermore, the difference between the personnel set obtained after stopping iteration S320 and the initial personnel set formed in S310 is performed to obtain the newly included personnel set during the expansion of the effective disease analysis boundary. The number of personnel in the newly included personnel set is counted and divided by the total number of personnel corresponding to the effective disease analysis boundary to obtain the personnel association migration degree. When no new personnel are included, the personnel association migration degree is set to 0. The greater the personnel association migration degree, the higher the extent of expansion of the effective disease analysis boundary along the personnel range formed by the cross-population bridging relationship.

[0112] Simultaneously, according to the expansion rounds in the S320 boundary evolution record, the target end case-related groups newly included in each round and their corresponding directed cross-group bridging relationships are obtained. The effective disease association types corresponding to the source end case-related groups and target end case-related groups in S230 are called to determine the relationship types involved in the personnel scope expansion in each round. Among them, the relationship types include fixed management relationship of organizational affiliation, fixed management relationship of accommodation management, periodic dynamic activity relationship, and centralized dynamic activity relationship.

[0113] The degree of clustering of boundary cases, the degree of case temporal evolution, and the degree of migration of personnel associations are associated and stored according to the current disease type and the corresponding effective analysis boundary of the disease to form disease clustering status characteristics, which are used to characterize the concentration level of patients, the growth trend of the number of cases, and the degree of expansion of the scope of personnel association within the effective analysis boundary of the disease.

[0114] S420. Determine the disease cluster risk level based on the characteristics of disease clustering status, and generate unit disease clustering medical visit early warning results in combination with the effective disease analysis boundary.

[0115] For the effective analysis boundary of the current disease type, the disease clustering status features formed by S410 are invoked to obtain the degree of clustering of boundary cases, the degree of case time evolution, and the degree of personnel association migration.

[0116] The disease clustering risk value is obtained by multiplying the degree of clustering of border cases, the degree of case temporal evolution, and the degree of personnel association migration by the clustering weight, temporal evolution weight, and association migration weight corresponding to the current disease type, and then adding the three products together.

[0117] Among them, the clustering degree weight, time evolution weight, and association migration weight all range from [0,1], and the sum of the three is 1. They are used to characterize the degree of influence of the concentration level of patients within the boundary, the growth trend of the number of cases, and the expansion of the association range of personnel in the unit on the current disease type clustering risk.

[0118] For different disease types, corresponding risk assessment thresholds are set, with the value range being [0,1]. The risk assessment thresholds are pre-calibrated based on the disease cluster risk values ​​and corresponding risk confirmation results obtained from historical disease cluster events of the corresponding disease type, and stored in association according to disease type.

[0119] The disease cluster risk value corresponding to the current disease type is compared with the risk assessment threshold corresponding to the disease type; when the disease cluster risk value is greater than or equal to the risk assessment threshold, the disease cluster risk level is determined to be high risk; when the disease cluster risk value is less than the risk assessment threshold, the disease cluster risk level is determined to be low risk.

[0120] Furthermore, based on the effective disease analysis boundary determined in S320, the case-associated groups already included after the iteration stops are obtained, along with their corresponding personnel sets, case sets, and boundary evolution records. The scope of personnel involved in the current disease cluster is determined based on the personnel sets; the types of effective disease associations corresponding to each included case-associated group are used to determine the organizational affiliation fixed management relationship, accommodation fixed management relationship, periodic dynamic activity relationship, and centralized dynamic activity relationship involved in the current disease cluster; and the formation path of the effective disease analysis boundary is determined based on the expansion rounds in the boundary evolution record, the newly included target-end case-associated groups in each round, and the corresponding directed cross-group bridging relationships.

[0121] When the directed cross-population bridging relationship in the boundary evolution record involves dynamic activity relationships, the dynamic activities that need to be focused on are determined based on the corresponding dynamic activity identifier and dynamic activity type recorded in S230. When the target case-associated group corresponds to a dynamic activity relationship, the effective bridging time range recorded in S230 is further invoked to determine the effective activity period associated with the current disease cluster in the corresponding dynamic activity, and the unit personnel that need to be focused on are determined based on the personnel set of the corresponding target case-associated group. When the directed cross-population bridging relationship involves a fixed management relationship, the fixed management relationship and corresponding unit personnel that need to be focused on are determined based on the personnel set of the corresponding case-associated group.

[0122] The system associates and stores disease type, degree of clustering of boundary cases, degree of case evolution over time, degree of migration of related personnel, disease clustering risk value, disease clustering risk level, scope of personnel in the unit corresponding to the effective analysis boundary of the disease, the related groups of cases that have been included, boundary evolution records, and number of cases to generate early warning results for disease clustering in the unit.

[0123] When the disease cluster risk level is high, the personnel of the unit corresponding to the effective disease analysis boundary, as well as the fixed management relationships and dynamic activity relationships involved in the boundary evolution record, are identified as key early warning targets; when the disease cluster risk level is low, the personnel of the unit corresponding to the effective disease analysis boundary, as well as the relevant fixed management relationships and dynamic activity relationships, are continuously monitored.

[0124] Among the above parameters, the preset coverage ratio threshold in S120, and the threshold in S220... , , The fusion weight, relation type adaptation coefficient, and preset explanatory power threshold, as well as the clustering weight, time evolution weight, association migration weight, and risk judgment threshold in S420, are all parameters that can be calibrated offline.

[0125] Before formal deployment, historical outpatient case data from different time periods, seasons, and disease types within the target unit, as well as historical clustered disease event data confirmed by the unit's health management personnel, were selected and divided into non-overlapping parameter calibration datasets and independent validation datasets. Historical clustered disease event data should include at least the disease type, cases involved in the event, personnel from the involved units, relevant fixed management relationships and dynamic activity relationships, and risk confirmation results.

[0126] For the relationship type fitting coefficient, for each disease type, the total number of candidate associations corresponding to fixed management relationships of organizational affiliation, fixed management relationships of accommodation, periodic dynamic activity relationships, and centralized dynamic activity relationships are counted, as well as the number of candidate associations confirmed to be related to historical clustered disease events. The number of confirmed related candidate associations is divided by the total number of candidate associations of the corresponding relationship type to obtain the historical effective association ratio of the corresponding relationship type, and the historical effective association ratio is used as the initial value of the relationship type fitting coefficient.

[0127] Based on the initial value of the relation type adaptation coefficient, adjust the relation type adaptation coefficient corresponding to each relation type within the range of [0,1], and compare it with the preset coverage ratio threshold of S120 and the threshold of S220. , , Candidate parameter combinations are obtained by combining fusion weights and preset explanatory thresholds. Each candidate parameter combination is used to process the parameter calibration dataset, and the dataset is validated using an independent validation dataset. For example, when the recall rate of effective disease association identification is greater than or equal to 90%, the misjudgment rate of irrelevant associations is less than or equal to 10%, the proportion of effective disease analysis boundary covering real clustered cases is greater than or equal to 85%, and the proportion of irrelevant personnel included is less than or equal to 15%, the corresponding candidate parameter combination is determined to have passed the validation.

[0128] When there are multiple validated candidate parameter combinations, select the candidate parameter combination with the smallest sum of the misjudgment ratio of irrelevant relationships and the inclusion ratio of irrelevant personnel. Store the relationship type adaptation coefficients in the combination according to the disease type and the relationship type, and store the remaining parameters according to the disease type.

[0129] After completing the calibration of the parameters corresponding to S120 and S220, the calibrated parameters are used to process historical clustered disease events to obtain the degree of clustering of boundary cases, the degree of case time evolution, and the degree of personnel association migration corresponding to each historical event.

[0130] Different candidate combinations were set for the clustering degree weight, time evolution weight, and association migration weight in S420. The disease clustering risk value corresponding to each historical clustered disease event was calculated, and different disease clustering risk values ​​were selected as candidate risk judgment thresholds.

[0131] Independent validation datasets were processed using different candidate weight combinations and candidate risk thresholds. When the false alarm rate was less than or equal to 20%, the corresponding candidate weight combination and candidate risk threshold were deemed to have passed validation. When multiple valid risk parameter combinations existed, the combination with the lowest sum of the high-risk event false negative rate and the low-risk event false positive rate was selected and stored as the system default risk parameter according to disease type. These requirements can be adjusted based on the organization's personnel size, the organization of personnel's dynamic activities, and the organization's health management requirements.

[0132] Non-zero smoothing parameters in S220 As a fixed parameter, it is set to 1 in this embodiment and does not participate in offline calibration. The 1-day statistical granularity used in S120, S210 and S410 and the 1-day minimum calculation benchmark used in S220 are all fixed parameters. The basis for their setting is that the unit's personnel management, activity arrangement and outpatient case records can usually be statistically processed on a daily time scale, so there is no need to perform offline calibration.

[0133] The preset statistical period in S210 is a system operation configuration parameter, which is set by the unit's health management personnel according to the actual needs of medical visit statistics. When no preset statistical period is set, the period for this disease cluster analysis is the earliest case occurrence time to the latest case occurrence time of all cases of the current disease type.

[0134] Example 2

[0135] like Figure 5 As shown, this embodiment provides a smart medical treatment management method; specifically, it includes the following steps: S1. Obtain the fixed management relationships, dynamic activity relationships, and outpatient disease information of the unit's personnel, and map the outpatient disease information to the corresponding fixed management relationships and dynamic activity relationships with the unit's personnel as the associated objects, forming case association mapping data; S2. Based on case association mapping data, analyze the clustering differences of cases in fixed management relationships and dynamic activity relationships according to disease type, determine the effective disease association relationship according to the association distribution of similar cases, and identify cross-group bridging relationships connecting different case association groups. S3. Using the case-related groups corresponding to the effective disease association as the initial analysis scope, the associated case-related groups are gradually obtained along the cross-group bridging relationship, and the scope of disease analysis is expanded to form the effective disease analysis boundary; S4. Within the effective analysis boundary of the disease, quantify the disease clustering status based on the degree of clustering of boundary cases, the degree of case time evolution, and the degree of personnel association migration, and generate unit disease clustering medical treatment early warning results.

[0136] Embodiments of the present invention also provide an electronic device, including a memory, a processor, and a communication bus; the memory and the processor are connected via the communication bus. The memory stores an intelligent medical treatment management system as provided in the above embodiments, which can be loaded and executed by the processor.

[0137] The memory can be used to store instructions, programs, code, code sets, or instruction sets. The memory may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for at least one function, and instructions for implementing the intelligent medical treatment management system provided in the above embodiments. The data storage area may store data involved in the intelligent medical treatment management system provided in the above embodiments.

[0138] A processor may include one or more processing cores. The processor executes instructions, programs, code sets, or instruction sets stored in memory, and calls data stored in memory to perform various functions and process data as described in this application. The processor may be at least one of a specific application-specific integrated circuit, a digital signal processor, a digital signal processing device, a programmable logic device, a field-programmable gate array, a central processing unit, a controller, a microcontroller, and a microprocessor. It is understood that, for different devices, the electronic devices used to implement the above-described processor functions may also be other types, and the embodiments of this application do not specifically limit the specific implementation.

[0139] A communication bus may include a pathway for transmitting information between the aforementioned components. The communication bus can be a PCI bus or an EISA bus, etc. Communication buses can be categorized into address buses, data buses, control buses, etc.

[0140] This application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed as described in the above embodiments, representing a smart medical management system.

[0141] In this embodiment, a computer-readable storage medium can be a tangible device that holds and stores instructions used by an instruction execution device. A computer-readable storage medium can be, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination thereof. Specifically, a computer-readable storage medium can be a portable computer disk, a hard disk, a USB flash drive, a random access memory, a read-only memory, an erasable programmable read-only memory, a static random access memory, a portable compressed disk read-only memory, a digital multifunction disk, a memory stick, a floppy disk, an optical disk, a magnetic disk, a mechanical encoding device, or any combination thereof.

[0142] The terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0143] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the foregoing application concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions claimed in this application.

Claims

1. A smart medical treatment management system, characterized in that, Includes the following modules: Personnel Association Module: Obtains the fixed management relationships, dynamic activity relationships, and outpatient disease information of personnel in the unit, and maps the outpatient disease information to the corresponding fixed management relationships and dynamic activity relationships with the personnel in the unit as the association objects, forming case association mapping data; Association Analysis Module: Based on case association mapping data, it analyzes the clustering differences of cases in fixed management relationships and dynamic activity relationships according to disease type, determines the effective disease association relationship according to the association distribution of similar cases, and identifies cross-group bridging relationships connecting different case association groups; Boundary Evolution Module: Starting with the case-associated groups corresponding to the effective disease associations as the initial analysis scope, the module gradually acquires associated case-associated groups along cross-group bridging relationships and expands the disease analysis scope to form the effective disease analysis boundary; Clustering identification module: Within the effective analysis boundary of the disease, the disease clustering status is quantified based on the degree of clustering of boundary cases, the degree of case time evolution, and the degree of personnel association migration, and the unit disease clustering visit warning results are generated.

2. The intelligent medical treatment management system according to claim 1, characterized in that, Obtain the fixed management relationships, dynamic activity relationships, and outpatient disease information of personnel within the organization, and map the outpatient disease information to the corresponding fixed management relationships and dynamic activity relationships, using the personnel as the associated objects, to form case association mapping data. This includes the following steps: Obtain information on the fixed management relationships, dynamic activity relationships, and outpatient disease information of the unit's personnel; Using personnel within the organization as the associated index, outpatient cases are linked to the fixed management relationships and dynamic activity relationships corresponding to the cases. The dynamic activity relationships are then time-matched based on the time of the case occurrence and the effective time period of the corresponding dynamic activity, forming case association mapping data.

3. The intelligent medical treatment management system according to claim 2, characterized in that, Based on case association mapping data, the clustering differences of cases in fixed management relationships and dynamic activity relationships are analyzed according to disease type. Effective disease association relationships are determined based on the association distribution of similar cases, including the following steps: The case association mapping data is grouped according to disease type and then aggregated according to fixed management relationship and dynamic activity relationship. The number of covered cases, the number of covered personnel, and the time distribution characteristics of the association relationship are statistically analyzed for fixed management relationship and dynamic activity relationship, and the corresponding case density is calculated to form disease relationship feature data. Based on disease relationship characteristic data, the overlap between the case sets corresponding to fixed management relationships and dynamic activity relationships is compared to determine the new explanatory power of fixed management relationships and dynamic activity relationships for disease types. Combining the temporal consistency of cases covered by fixed management relationships and dynamic activity relationships, the effectiveness of fixed management relationships and dynamic activity relationships at the stage of case occurrence, and the degree of fit between disease types and fixed management relationship types and dynamic activity relationship types, the association explanatory power of fixed management relationships and dynamic activity relationships for the current disease type case distribution is determined, and the effective disease association relationship set is screened based on the association explanatory power.

4. The intelligent medical treatment management system according to claim 3, characterized in that, Identifying cross-group bridging relationships connecting different case-related groups includes the following steps: Based on the set of effective disease associations, cases and individuals associated with these associations are grouped into case association groups. Individuals in different case association groups are compared to identify common individuals who simultaneously belong to different case association groups. Based on the chronological order of case occurrences and the effective association period corresponding to the dynamic activity relationships in different case association groups, the temporal continuity of the group connection relationships formed by these common individuals is verified, generating cross-group bridging relationships that connect different case association groups.

5. The intelligent medical treatment management system according to claim 4, characterized in that, The initial analysis scope is the case-associated population corresponding to the effective disease association. The analysis then expands the scope of disease analysis by progressively acquiring associated case-associated populations along cross-population bridging relationships, thus forming the effective disease analysis boundary. This includes the following steps: Based on cross-population bridging relationships, case-related groups are connected and divided into case-related group connectivity groups; based on the connection direction of cross-population bridging relationships within the case-related group connectivity groups, the boundary evolution starting group is determined, and the scope of personnel in the units corresponding to the boundary evolution starting group is aggregated to form the initial disease analysis scope; Starting with the case-associated groups corresponding to the initial disease analysis scope as the boundary evolution starting point, along the connection direction of cross-group bridging relationships, we screen out case-associated groups that are connected to the case-associated groups within the disease analysis scope but have not yet been included in the disease analysis scope. We then include the screened case-associated groups into the disease analysis scope and continue to evolve the boundary until there are no more cross-group bridging relationships that can further expand the disease analysis scope, thus forming an effective disease analysis boundary.

6. The intelligent medical treatment management system according to claim 5, characterized in that, Within the effective analysis boundary of the disease, the disease clustering status is quantified based on the degree of clustering of boundary cases, the degree of case temporal evolution, and the degree of personnel association and migration, generating a unit-wide disease clustering visitation early warning result, including the following steps: The degree of clustering of cases at the effective disease analysis boundary is determined by the number of cases and the number of people corresponding to the effective disease analysis boundary. The degree of case temporal evolution is determined by the change in the number of cases corresponding to the time of case occurrence. The degree of human association migration is determined by the newly included case-related groups and corresponding cross-group bridging relationships during the formation of the effective disease analysis boundary, thus forming the characteristics of disease clustering status. The risk level of disease clustering is determined based on the characteristics of disease clustering status, and the unit disease clustering medical visit early warning result is generated by combining the effective disease analysis boundary.

7. A smart medical treatment management method, implemented based on the smart medical treatment management system according to any one of claims 1-6, characterized in that, The method includes: S1. Obtain the fixed management relationships, dynamic activity relationships, and outpatient disease information of the unit's personnel, and map the outpatient disease information to the corresponding fixed management relationships and dynamic activity relationships with the unit's personnel as the associated objects, forming case association mapping data; S2. Based on case association mapping data, analyze the clustering differences of cases in fixed management relationships and dynamic activity relationships according to disease type, determine the effective disease association relationship according to the association distribution of similar cases, and identify cross-group bridging relationships connecting different case association groups. S3. Using the case-related groups corresponding to the effective disease association as the initial analysis scope, the associated case-related groups are gradually obtained along the cross-group bridging relationship, and the scope of disease analysis is expanded to form the effective disease analysis boundary; S4. Within the effective analysis boundary of the disease, quantify the disease clustering status based on the degree of clustering of boundary cases, the degree of case time evolution, and the degree of personnel association migration, and generate unit disease clustering medical treatment early warning results.

8. A computer-readable storage medium, characterized in that, The system stores a computer program that can be loaded by a processor and executed as described in any one of claims 1 to 6, which is a smart medical treatment management system.

Images