A medical data fusion method and system based on semantic mapping

Abstract

The application relates to the technical field of medical data fusion, and relates to a medical data fusion method and system based on semantic mapping, which comprises the following steps: receiving a medical data fusion request, obtaining a business data node corresponding to the anonymous user number, confirming a set of participating end nodes by using the authorization node and the business data node, confirming an effective data end group in the set of participating end nodes based on the semantic mapping node set, obtaining a mapping field node set based on the effective data end group and a pre-constructed semantic mapping rule library, performing conflict checking and fusion on the mapping field node set, obtaining a set of fusion data nodes, generating a flow record node based on the authorization node, the semantic mapping node set, the mapping field node set and the set of fusion data nodes, and realizing medical data fusion. The application can realize intelligent screening, fusion and credible tracing of cross-institutional multi-source heterogeneous medical data.

Description

Technical Field

[0001] This invention relates to the field of medical data fusion technology, and in particular to a medical data fusion method and system based on semantic mapping. Background Technology

[0002] With the deepening of healthcare informatization, user medical data is scattered across multiple heterogeneous systems such as electronic medical records and laboratory tests, forming data silos that are difficult to effectively integrate in collaborative diagnosis and treatment scenarios. Healthcare data fusion technology aims to collect, standardize, and converge multi-source heterogeneous data into a unified view, providing a data foundation for precision medicine and clinical decision-making.

[0003] Currently, traditional medical data fusion methods mostly employ a pre-fixed data source and manually configured field mapping relationships. This approach lacks an effective assessment and automated screening mechanism for the actual data provision capabilities (such as field availability and data completeness) of the data source before fusion. This results in a large number of invalid requests being sent to data sources that do not possess the required data. Furthermore, traditional methods lack adequate protection for user privacy during the fusion process, posing a risk of data leakage during cross-platform transmission and processing, and failing to meet increasingly stringent data security and privacy compliance requirements. Therefore, how to achieve intelligent screening, fusion, and reliable traceability of multi-source heterogeneous medical data across institutions has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a medical data fusion method based on semantic mapping and a computer-readable storage medium. Its main purpose is to realize intelligent screening, fusion and reliable traceability of multi-source heterogeneous medical data across institutions.

[0005] To achieve the above objectives, the present invention provides a medical data fusion method based on semantic mapping, comprising:

[0006] Receive a medical data fusion request, parse the medical data fusion request to obtain the anonymous user ID, request source, target service stage and fusion data item, and obtain the authorized node based on the anonymous user ID, request source, target service stage and fusion data item;

[0007] Obtain the business data node corresponding to the anonymous user ID, wherein the business data node includes appointment information, communication records, service information index and follow-up records;

[0008] The participating node set is identified using the authorized node and the business data node;

[0009] A standard data node set is obtained based on the participating end node set, a semantic mapping node set is obtained using the standard data node set, and a valid data end group is identified in the participating end node set based on the semantic mapping node set.

[0010] Based on the effective data terminal group and the pre-built semantic mapping rule base, a mapping field node set is obtained, and the mapping field node set is subjected to conflict verification and fusion to obtain a fused data node set.

[0011] Based on the authorized node, semantic mapping node set, mapping field node set, and fusion data node set, a flow record node is generated, and the fusion data node set and flow record node are sent to the request source end to realize medical data fusion.

[0012] Optionally, obtaining the authorization node based on the anonymous user ID, the request source, the target service stage, and the fused data item includes:

[0013] Obtain the set of user authorization records, and use the anonymous user ID to retrieve the set of candidate authorization records from the set of user authorization records;

[0014] Using the request source, target service stage, and fused data items, target authorization records are filtered from the candidate authorization record set. The target authorization records include: a list of permitted fields, sensitive values ​​of the fields, and authorization validity period.

[0015] Obtain the current time and the fusion random code. If the current time is within the authorization validity period, perform one-way hashing on the anonymous user ID, request source, target service stage, current time and fusion random code to obtain the fusion identifier code.

[0016] By associating the anonymous user ID, the list of permitted fields, the sensitive values ​​of the fields, the authorization validity period, and the fusion identifier code, the authorization node is obtained.

[0017] Optionally, the step of identifying the participating node set using the authorized node and the business data node includes:

[0018] Based on the list of permitted fields, perform field matching on the business data nodes to filter out the list of permitted data items;

[0019] Using the list of permitted data items, retrieve multiple candidate data endpoints from the pre-defined data endpoint registration table, and perform the following operations on each of the multiple candidate data endpoints:

[0020] Obtain a list of available fields and historical fusion pass rates from the candidate data source;

[0021] The list of available fields is matched with the list of permitted data items to obtain a list of fields that can be merged.

[0022] If the list of fields that can be merged is not empty, then the candidate data end, the list of fields that can be merged, and the historical merging pass rate are associated to obtain the participating end nodes. The participating end nodes are then summarized to obtain the set of participating end nodes.

[0023] Optionally, obtaining the standard data node set based on the participating end node set includes:

[0024] Perform the following operation on each participating node in the participating node set:

[0025] Using the list of fusionable fields, edge data nodes are extracted from the candidate data nodes, wherein the edge data nodes include: identity identification field and medical content field;

[0026] One-way hashing is performed on the anonymous user ID, candidate data endpoint, and fusion identifier to obtain the intra-end replacement ID;

[0027] The identity identifier field is replaced with an in-device replacement number, and the medical content field is masked based on the field sensitivity value to obtain the de-identified data node;

[0028] The unidentified data nodes are standardized to obtain standard data nodes.

[0029] By aggregating the standard data nodes, a set of standard data nodes is obtained, in which each standard data node corresponds to a different candidate data endpoint.

[0030] Optionally, obtaining the semantic mapping node set using the standard data node set includes:

[0031] Extract standard data nodes sequentially from the standard data node set, obtain the number of filled fields corresponding to the standard data nodes, obtain the total number of fields corresponding to the list of fields that can be merged, calculate the ratio of the number of filled fields to the total number of fields, and obtain the field completeness.

[0032] The semantic mapping rule base is used to perform semantic matching on standard data nodes to obtain semantic matching degree and stage consistency degree;

[0033] By considering the completeness of associated fields, semantic matching degree, stage consistency degree, and historical fusion pass rate, semantic mapping nodes are obtained. By summarizing the semantic mapping nodes, a semantic mapping node set is obtained.

[0034] Optionally, the step of identifying a valid data endpoint group in the participating endpoint node set based on the semantic mapping node set includes:

[0035] Semantic mapping nodes are extracted sequentially from the set of semantic mapping nodes, and the semantic usability value corresponding to the semantic mapping node is calculated. The calculation formula is as follows:

[0036]

[0037] in, Indicates the available values ​​of the semantics. This indicates the completeness of the field. Indicates the semantic matching degree. This indicates the consistency of the stated stages. This indicates the historical fusion pass rate. This indicates the preset zero-prevention value;

[0038] If the semantic availability value is not less than the preset availability threshold, the candidate data ends corresponding to the semantic mapping node are confirmed as valid data ends, and the valid data ends are aggregated to obtain a valid data end group.

[0039] Optionally, obtaining the mapping field node set based on the valid data endpoint group and the pre-built semantic mapping rule base includes:

[0040] The valid data endpoints are initially grouped to obtain candidate domain groups and non-candidate domain groups;

[0041] Calculate the interoperability of each valid data endpoint within the candidate domain group to obtain multiple interoperability scores within the group;

[0042] Valid data endpoints with interoperability below a preset interoperability threshold among multiple groups are removed from the candidate domain groups to obtain the adjustment domain groups. The number of valid data endpoints in the adjustment domain groups is counted to obtain the number of valid data endpoints.

[0043] If the number of valid data endpoints is less than the preset endpoint threshold, then the interoperability between each valid data endpoint in the non-candidate domain group and the candidate domain group is calculated to obtain multiple inter-group interoperability.

[0044] The out-of-group interoperability is sorted in descending order to obtain the out-of-group interoperability sequence;

[0045] Extract an out-of-group interoperability subsequence from the out-of-group interoperability sequence, wherein the number of out-of-group interoperability in the out-of-group interoperability subsequence is the difference between the end number threshold and the number of valid data ends;

[0046] The effective data ends corresponding to each out-of-group interoperability in the out-of-group interoperability subsequence are moved from the non-candidate domain group to the adjustment domain group to obtain the interoperability domain group.

[0047] Otherwise, the adjusted domain group will be identified as an interconnected domain group;

[0048] The semantic mapping rule base is used to perform rule mapping on the standard data nodes corresponding to each valid data terminal in the interconnected domain group, resulting in a set of mapped field nodes.

[0049] Optionally, the step of performing conflict verification and fusion on the mapped field node set to obtain a fused data node set includes:

[0050] The mapping field nodes in the mapping field node set are grouped to obtain a set of fields to be merged, wherein each set of fields to be merged includes one or more mapping field nodes;

[0051] Extract the fields to be merged sequentially from the set of fields to be merged, count the number of mapped field nodes in the fields to be merged, and obtain the number of nodes in the group;

[0052] If there is only one node in the group, then the mapping field node in the field group to be merged will be identified as the merged data node.

[0053] If the number of nodes in a group is greater than one, then the conflict detection results between the mapping field nodes in the field group to be merged are checked. The conflict detection result is either that there is a conflict or that there is no conflict.

[0054] When the conflict detection result is that there is no conflict, all mapped field nodes in the field group to be merged are merged and processed to obtain the merged data node.

[0055] When the conflict detection result indicates a conflict, the mapped field nodes are extracted sequentially from the field group to be fused, and the field matching degree and time validity corresponding to the mapped field node are obtained. The fusion weight of the mapped field node is calculated based on the field matching degree and time validity, as shown in the following formula:

[0056]

[0057] in, Indicates the fusion weights, Indicates the matching degree of the field. Indicates the validity of the time period. This indicates that there are a total of [number] fields in the group to be merged. One mapping field node;

[0058] The fusion weights are aggregated to obtain n fusion weights, where each fusion weight corresponds one-to-one with a mapping field node. The mapping field node corresponding to the largest fusion weight among the n fusion weights is taken as the fusion data node.

[0059] The data nodes are aggregated and merged to obtain the merged data node set.

[0060] Optionally, generating a flow record node based on the authorized node, semantic mapping node set, mapping field node set, and fused data node set includes:

[0061] Perform one-way hashing on the authorized nodes to obtain the authorization digest;

[0062] A one-way hashing process is performed on the semantic mapping node set and the authorization digest to obtain the evaluation digest;

[0063] A one-way hash is performed on the mapping field node set and the evaluation summary to obtain the mapping summary;

[0064] Perform one-way hashing on the fused data node set and mapping digest to obtain the fused digest, and record the generation time of the fused digest to obtain the transfer time;

[0065] By associating the fusion identifier, request source, target service stage, interconnected domain group, authorization digest, evaluation digest, mapping digest, fusion digest, and transfer time, the transfer record node is obtained.

[0066] To achieve the above objectives, the present invention also provides a medical data fusion system based on semantic mapping, comprising:

[0067] The authorized node construction module is used to receive medical data fusion requests, parse the medical data fusion requests, obtain the anonymous user ID, the request source, the target service stage, and the fused data items, and obtain the authorized node based on the anonymous user ID, the request source, the target service stage, and the fused data items;

[0068] The valid data terminal confirmation module is used to obtain the business data node corresponding to the anonymous user ID, wherein the business data node includes appointment information, communication records, service information index and follow-up records;

[0069] The participating node set is identified using the authorized node and the business data node;

[0070] A standard data node set is obtained based on the participating end node set, a semantic mapping node set is obtained using the standard data node set, and a valid data end group is identified in the participating end node set based on the semantic mapping node set.

[0071] The data fusion module is used to obtain a set of mapping field nodes based on the effective data terminal group and the pre-built semantic mapping rule library, and to perform conflict verification and fusion on the set of mapping field nodes to obtain a fused data node set.

[0072] The data flow traceability module is used to generate flow record nodes based on the authorized nodes, semantic mapping node set, mapping field node set, and fusion data node set, and send the fusion data node set and flow record nodes to the request source end to realize medical data fusion.

[0073] To address the above problems, the present invention also provides an electronic device, the electronic device comprising:

[0074] Memory, storing at least one instruction;

[0075] The processor executes the instructions stored in the memory to implement the semantic mapping-based medical data fusion method described above.

[0076] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the semantic mapping-based medical data fusion method described above.

[0077] To address the problems described in the background section, this invention receives a medical data fusion request, parses the request to obtain an anonymous user ID, the request source, the target service stage, and the fused data items. Based on these information, it obtains the authorized node. This invention confirms the fusion permission range through multi-dimensional screening of user authorization records and then performs one-way hashing using time and random codes, ensuring that each fusion action has a unique traceable credential. This achieves precise control over the list of permitted fields and sensitive values ​​without exposing the user's true identity, thus improving the security of medical data fusion and obtaining the data corresponding to the anonymous user ID. The business data nodes include appointment information, communication records, service data indexes, and follow-up records. The participating end node set is identified using the authorized nodes and business data nodes. This invention achieves a single-end usability and cross-end non-association medical data information protection method through in-end replacement numbering. Combined with hierarchical shielding and standardized processing, subsequent fusion protects user privacy while facilitating cross-end field alignment. By matching the field authorization scope, the actual capabilities of candidate data ends, and historical fusion performance, data ends lacking data provision capabilities or not overlapping with the authorization scope are eliminated in advance, making the participating end node set more focused and improving the effectiveness of subsequent medical data fusion. The participating node set obtains a standard data node set, which is then used to obtain a semantic mapping node set. Based on this semantic mapping node set, valid data node groups are identified within the participating node set. This invention utilizes a combination of four indicators—field completeness, semantic matching degree, stage consistency, and historical fusion pass rate—to form a quantifiable multi-dimensional evaluation, providing comparable numerical criteria for determining the validity of candidate data nodes. Based on the valid data node groups and a pre-built semantic mapping rule base, a mapping field node set is obtained. This mapping field node set undergoes conflict verification and fusion to obtain a fused data node set. This invention, through secondary screening based on interoperability and on-demand completion, effectively eliminates... In addition to valid data sources with poor interoperability, the number of data sources is dynamically supplemented according to their interoperability from high to low, thereby achieving a balance between the breadth of data coverage and the quality of data alignment. This provides dual assurance for the quality of the fusion results. Based on the authorized node, semantic mapping node set, mapping field node set, and fusion data node set, a flow record node is generated. The fusion data node set and flow record node are sent to the requesting source to achieve medical data fusion. It can be seen that this invention cryptographically binds the authorization, evaluation, mapping, and fusion links in a causal order through multi-level nested digests. This ensures that any tampering with medical data in any link will lead to inconsistencies in all subsequent digests, thus making the entire medical data fusion process traceable. Therefore, this invention can realize intelligent screening, fusion, and reliable traceability of multi-source heterogeneous medical data across institutions. Attached Figure Description

[0078] Figure 1 This is a flowchart illustrating a semantic mapping-based medical data fusion method according to an embodiment of the present invention.

[0079] Figure 2 A functional block diagram of a medical data fusion system based on semantic mapping provided in an embodiment of the present invention;

[0080] Figure 3 This is a schematic diagram of the structure of an electronic device that implements the semantic mapping-based medical data fusion method according to an embodiment of the present invention.

[0081] Explanation of reference numerals in the attached figures:

[0082] 10. Electronic device; 11. Processor; 12. Memory; 13. Bus.

[0083] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0084] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0085] This application provides a medical data fusion method based on semantic mapping. The executing entity of the semantic mapping-based medical data fusion method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the semantic mapping-based medical data fusion method can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0086] Reference Figure 1 The diagram shown is a flowchart illustrating a semantic mapping-based medical data fusion method according to an embodiment of the present invention. In this embodiment, the semantic mapping-based medical data fusion method includes:

[0087] S1. Receive a medical data fusion request, parse the medical data fusion request to obtain the anonymous user ID, request source, target service stage and fusion data item, and obtain the authorized node based on the anonymous user ID, request source, target service stage and fusion data item.

[0088] It should be explained that the medical data fusion request is a request message submitted by the requesting party to trigger the merging of multi-terminal medical data. The anonymous user ID is a pre-assigned string number for the user, serving as a unique identification credential. The requesting party is the platform or system that initiates the medical data fusion request. The target service stage is the business stage served by this fusion. The fused data item is the specific text that the requesting party needs to obtain, such as an examination report. The authorization node is a data structure containing the allowed field range and identity credentials for this fusion operation, serving as the basis for subsequent data filtering.

[0089] For example, suppose the follow-up visit management system submits a medical data fusion request to the medical data fusion platform. The medical data fusion platform parses the medical data fusion request and obtains the anonymous user ID U2026X001, the request source as the follow-up visit management system, the target service stage as the follow-up visit stage, and the fused data item as the examination report.

[0090] Understandably, the authorized node needs to confirm that the current fusion is within the scope of user authorization, and also needs to provide traceable identity credentials for each subsequent operation. Therefore, obtaining the authorized node based on the anonymous user ID, the request source, the target service stage, and the fusion data item includes:

[0091] Obtain the set of user authorization records, and use the anonymous user ID to retrieve the set of candidate authorization records from the set of user authorization records;

[0092] Using the request source, target service stage, and fused data items, target authorization records are filtered from the candidate authorization record set. The target authorization records include: a list of permitted fields, sensitive values ​​of the fields, and authorization validity period.

[0093] Obtain the current time and the fusion random code. If the current time is within the authorization validity period, perform one-way hashing on the anonymous user ID, request source, target service stage, current time and fusion random code to obtain the fusion identifier code.

[0094] By associating the anonymous user ID, the list of permitted fields, the sensitive values ​​of the fields, the authorization validity period, and the fusion identifier code, the authorization node is obtained.

[0095] It should be understood that the user authorization record set is a structured authorization data set archived after being signed by the user. The candidate authorization record set is a set of authorization records selected from the user authorization record set that correspond to the anonymous user ID. The target authorization record is an authorization record further selected from the candidate authorization record set that matches the request source, target service stage, and fused data items. The permitted field list is a set of field names in the target authorization record that are allowed to be fused. The field sensitivity value is the sensitivity level value corresponding to each allowed fusion field (the larger the value, the higher the sensitivity level). The authorization validity period is the effective period of the authorization in the target authorization record. The current time is the time when this medical data fusion request was initiated. The fusion random code is a random string generated when the medical data fusion request is issued, used to avoid the fusion identifier code being predicted. The one-way hash processing is the process of calculating the fusion identifier code using a one-way hash algorithm on the input string and outputting the fusion identifier code. Optionally, the SHA-256 hash algorithm can be used to achieve this purpose. The input string is: anonymous user ID, request source, target service stage, current time, and fusion random code. The fusion identifier is a string obtained by performing a one-way hash on the input string, serving as the credential for this fusion operation. It should be clarified that the purpose of performing a one-way hash on the input string in this embodiment of the invention is to generate an irreversible and unique fusion identifier by performing a one-way hash on the anonymous user ID, request source, target service stage, current time, and fusion random code. This not only prevents the original information from being retrieved to protect user privacy, but also serves as a unique credential for each fusion operation, tracing back to the corresponding target authorization record. This enables precise control over the list of permitted fields and sensitive values, improving the security of medical data fusion.

[0096] For example, suppose a search is performed from the user authorization record set using the anonymous user ID U2026X001, yielding a candidate authorization record set (containing 3 candidate authorization records). Filtering is performed using the request source (follow-up visit management system), the target service stage (follow-up visit stage), and the fused data item (examination report), resulting in one target authorization record. The permitted fields are examination results, medication records, and allergy history, with sensitivity values ​​of 2, 2, and 3 respectively. The authorization validity period is from January 1, 2026 to December 31, 2026. The current time is March 27, 2026, within the authorization validity period, and the fused random code is R8F3K2. The anonymous user ID U2026X001, the request source (follow-up visit management system), the target service stage (follow-up visit stage), the current time of the fused data item (examination report) March 27, 2026, and the fusion random code R8F3K2 are concatenated and then processed using SHA-256 one-way hashing to obtain the fusion identifier code a1b2c3d4e5f60a8b. This embodiment of the invention confirms the scope of fusion permissions through multi-dimensional screening of user authorization records, and then combines time and random code for one-way hashing, ensuring that each fusion action has a unique traceable credential. This achieves precise control over the list of permitted fields and sensitive values ​​of fields without exposing the user's true identity, thus improving the security of medical data fusion.

[0097] S2. Obtain the business data node corresponding to the anonymous user ID, wherein the business data node includes appointment information, communication records, service information index and follow-up records.

[0098] It should be explained that the business data node is a collection of user's full-process business data associated with the anonymous user ID, which forms the basis for subsequent determination of which fields to integrate and which data terminals to integrate with. The appointment information is the appointment record generated when a user schedules an examination, such as the appointment time. The communication record is the conversation record generated during the user's consultation. The service data index is the index information of the storage location of medical data (such as imaging reports and laboratory reports) generated from each of the user's previous medical visits on the medical data fusion platform. The follow-up record is the health status record formed by medical staff periodically tracking the user after their medical visit.

[0099] For example, by searching the medical data fusion platform using the anonymous user ID U2026X001, the business data nodes corresponding to the anonymous user ID U2026X001 are obtained, including appointment information: cardiology appointment on March 20, 2026; communication records: consultation conversation with a cardiologist; service data index: electrocardiogram; and follow-up records: medication follow-up records in April 2026.

[0100] S3. Use the authorized nodes and business data nodes to identify the set of participating nodes.

[0101] It should be explained that the participating node set is a collection of multiple data endpoint nodes that, after authorization filtering, can participate in this medical data fusion, and is used to define the scope of data sources for this fusion. The data endpoints are data sources within the medical data fusion platform that can provide the medical data required for this fusion, such as electronic medical record systems, laboratory information systems, and image archiving systems.

[0102] Understandably, the selection of participating data terminals requires considering both the fields authorized by the user and the fields actually provided by the candidate data terminals, as well as their historical fusion performance. Therefore, the process of identifying the participating terminal node set using the authorized nodes and business data nodes includes:

[0103] Based on the list of permitted fields, perform field matching on the business data nodes to filter out the list of permitted data items;

[0104] Using the list of permitted data items, retrieve multiple candidate data endpoints from the pre-defined data endpoint registration table, and perform the following operations on each of the multiple candidate data endpoints:

[0105] Obtain a list of available fields and historical fusion pass rates from the candidate data source;

[0106] The list of available fields is matched with the list of permitted data items to obtain a list of fields that can be merged.

[0107] If the list of fields that can be merged is not empty, then the candidate data end, the list of fields that can be merged, and the historical merging pass rate are associated to obtain the participating end nodes. The participating end nodes are then summarized to obtain the set of participating end nodes.

[0108] It should be understood that the permitted data item list is a list of fields obtained by intersecting the field names actually involved in the business data nodes with the permitted field list. The data terminal registration table is a pre-built list containing basic information of each data terminal. Specifically, the data terminal registration table is constructed as follows: basic information of each data terminal, such as data terminal identifier, data terminal type, and list of available fields, is collected from all data terminals registered in the medical data fusion platform in advance. This information is then summarized to form the data terminal registration table, which is used to quickly retrieve and filter data terminals that meet the conditions during the execution of the fusion task.

[0109] Understandably, the candidate data endpoints are those whose available fields and permitted data item lists overlap in the data endpoint registration table, serving as alternatives for this fusion. The available field list is the set of fields that the candidate data endpoint can provide. The historical fusion pass rate is the proportion of fields in fusion tasks participated in by the candidate data endpoint over a past period that were ultimately considered as fusion data nodes; the historical fusion pass rate can be obtained from the historical fusion logs of the medical data fusion platform. The fusionable field list is the set of fields obtained by intersecting the available field list and permitted data item list of the candidate data endpoint. The participating endpoint node is a data structure obtained by associating the candidate data endpoint's identifier, fusionable field list, and historical fusion pass rate.

[0110] For example, suppose the permitted field list is: {Examination Report, Medication Record, Allergy History}. The actual business data node involves two types of fields: Examination Report and Medication Record. Therefore, the permitted data item list is: {Examination Report, Medication Record}. Retrieving from the data end registration table yields four candidate data ends: Electronic Medical Record System, Laboratory Information System, Imaging Archive System, and Physical Examination Center System. For instance, matching the Electronic Medical Record System, its available field list is: {Medication Record, Allergy History, Diagnosis Conclusion}, with a historical fusion pass rate of 0.92. Matching the available field list {Medication Record, Allergy History, Diagnosis Conclusion} with the permitted data item list {Examination Report, Medication Record} yields the fusionable field list for the Electronic Medical Record System as {Medication Record}. Since the fusionable field list is not empty, the corresponding participating end node is {Electronic Medical Record System - (Medication Record) - 0.92}. Similarly, the fusionable field lists and participating end nodes for the Laboratory Information System, Imaging Archive System, and Physical Examination Center System can be obtained. This invention, through a three-way matching of the authorized scope of fields, the actual capabilities of candidate data terminals, and historical fusion performance, preemptively eliminates data terminals that lack data provision capabilities or have no overlap with the authorized scope, making the participating terminal node set more focused and improving the effectiveness of subsequent medical data fusion.

[0111] S4. Obtain a standard data node set based on the participating end node set, obtain a semantic mapping node set using the standard data node set, and identify a valid data end group in the participating end node set based on the semantic mapping node set.

[0112] It should be explained that the standard data node set is a collection of nodes formed after the medical data provided by each data terminal in the participating terminal node set has been standardized and de-identified. Each standard data node corresponds to a candidate data terminal. The semantic mapping node set is a set of nodes used to evaluate the usability of each standard data node for this fusion. Each semantic mapping node includes four indicators: field completeness, semantic matching degree, stage consistency, and historical fusion success rate. The effective data terminal group is a set of multiple candidate data terminals that have been confirmed as effective for this fusion after being filtered by semantic usability values.

[0113] Understandably, due to differences in field naming, units, and formats among different data endpoints, and the involvement of user privacy fields, de-identification and standardization processing are necessary. Therefore, obtaining a standard data node set based on the participating endpoint node set includes:

[0114] Perform the following operation on each participating node in the participating node set:

[0115] Using the list of fusionable fields, edge data nodes are extracted from the candidate data nodes, wherein the edge data nodes include: identity identification field and medical content field;

[0116] One-way hashing is performed on the anonymous user ID, candidate data endpoint, and fusion identifier to obtain the intra-end replacement ID;

[0117] The identity identifier field is replaced with an in-device replacement number, and the medical content field is masked based on the field sensitivity value to obtain the de-identified data node;

[0118] The unidentified data nodes are standardized to obtain standard data nodes.

[0119] By aggregating the standard data nodes, a set of standard data nodes is obtained, in which each standard data node corresponds to a different candidate data endpoint.

[0120] It should be understood that the aforementioned edge data node is composed of two parts: an identity identifier field and a medical content field, extracted from the original records within the candidate data terminal according to the list of fusionable fields. The identity identifier field is used within the candidate data terminal to identify the user's true identity, such as ID number and name. The medical content field is the field in the edge data node that records actual medical data, such as laboratory result values, medication names, and dosages. The edge replacement number is a string obtained by concatenating the anonymous user number, data terminal identifier, and fusion identifier code, and then performing a one-way hash process. It is only available within this fusion task, ensuring that the identifier of the same user is different across different candidate data terminals, avoiding cross-terminal correlation tracking. The masking process involves masking the medical content field based on its sensitivity value. The higher the sensitivity value, the stronger the masking. For example, when the ID number field has a high sensitivity value, the ID number is completely masked; when the name field has a low sensitivity value, only the surname is retained, and the given name is masked, such as Wang**. The de-identified data node is the data node obtained after the identity identifier field has been replaced and the medical content field has been masked. The data standardization process involves performing conversion operations on medical content fields from different candidate data sources according to a pre-defined standard field dictionary. This includes normalizing field names, standardizing units of measurement, and ensuring the compliance of value ranges, making them uniformly compliant with the data specifications of the medical data fusion platform. For example, mg / dL and mmol / L are uniformly converted to mmol / L. The standard field dictionary is a predefined structured data dictionary containing all fusionable field data information. This embodiment of the invention achieves a medical data information protection method that is available only on one end and not linked across ends by replacing numbers within the end. Combined with hierarchical masking and standardization processing, subsequent fusion can both protect user privacy and facilitate cross-end field alignment.

[0121] Furthermore, in order to quantitatively evaluate the usability of each standard data node for this fusion, the step of obtaining the semantic mapping node set using the standard data node set includes:

[0122] Extract standard data nodes sequentially from the standard data node set, obtain the number of filled fields corresponding to the standard data nodes, obtain the total number of fields corresponding to the list of fields that can be merged, calculate the ratio of the number of filled fields to the total number of fields, and obtain the field completeness.

[0123] The semantic mapping rule base is used to perform semantic matching on standard data nodes to obtain semantic matching degree and stage consistency degree;

[0124] By considering the completeness of associated fields, semantic matching degree, stage consistency degree, and historical fusion pass rate, semantic mapping nodes are obtained. By summarizing the semantic mapping nodes, a semantic mapping node set is obtained.

[0125] It should be understood that the number of filled fields refers to the number of fields in the standard data node that have actually been assigned valid values ​​(i.e., non-empty). For example, each standard data node corresponds to a participating end node, which contains a list of fusionable fields. After data standardization, each field in the fusionable field list is checked to see if it has a non-empty value in the standard data node. The number of fields with non-empty values ​​is the number of filled fields. For example, if a participating end node's fusionable field list contains 5 fields (e.g., {medication records, allergy history, diagnosis conclusion, examination report, surgical history}), and after standardization, only 3 fields actually have valid values ​​(medication records, allergy history, diagnosis conclusion), while examination report and surgical history are empty, then the number of filled fields for this standard data node is 3. The total number of fields refers to the total number of fields defined in the participating end node's fusionable field list (5 in this example). The field completeness is 3 / 5 = 0.6.

[0126] For example, the process of using the semantic mapping rule base to perform semantic matching on standard data nodes to obtain semantic matching degree and stage consistency degree is as follows: Assume the target service stage is: follow-up visit stage, and the standard data nodes are {examination report, medication record, surgical record}. By querying the semantic mapping rule base, the semantic similarity weights of each field with the follow-up visit stage are obtained: examination report is 0.95, medication record is 0.85, and surgical record is 0.40. Taking the average of the three, the semantic matching degree of the standard data node is approximately 0.73. Then, the business stage labels associated with each field are queried: examination report and medication record are both marked as follow-up period, consistent with the target service (each scored 1.0 point), while surgical record is marked as surgical period, unrelated to the target service stage (scored 0 points). Taking the average, the stage consistency degree is approximately 0.67. Specifically, the semantic mapping rule base is a pre-constructed set of rules used to measure the strength of association between standard fields and target service stages in the medical data fusion platform. The semantic mapping rule base can be constructed based on statistical analysis of historical fusion logs and terminology mapping from a standard medical terminology database. The semantic matching degree is a quantitative indicator measuring the degree of semantic association between each field within a standard data node in the semantic mapping rule base and the target service stage. The stage consistency degree is the proportion of service stages belonging to fields already filled in a standard data node that are consistent with the current target service stage. The semantic mapping node is formed after considering the completeness of associated fields, semantic matching degree, stage consistency degree, and historical fusion pass rate. This embodiment of the invention combines four indicators—field completeness, semantic matching degree, stage consistency degree, and historical fusion pass rate—to form a quantifiable multi-dimensional evaluation, providing a comparable numerical basis for subsequently determining the effectiveness of candidate data.

[0127] Furthermore, in order to filter out truly suitable and effective data endpoints for this fusion, the effective data endpoint group is identified from the participating endpoint node set based on the semantic mapping node set, including:

[0128] Semantic mapping nodes are extracted sequentially from the set of semantic mapping nodes, and the semantic usability value corresponding to the semantic mapping node is calculated. The calculation formula is as follows:

[0129]

[0130] in, Indicates the available values ​​of the semantics. This indicates the completeness of the field. Indicates the semantic matching degree. This indicates the consistency of the stated stages. This indicates the historical fusion pass rate. This indicates the preset zero-prevention value;

[0131] If the semantic availability value is not less than the preset availability threshold, the candidate data ends corresponding to the semantic mapping node are confirmed as valid data ends, and the valid data ends are aggregated to obtain a valid data end group.

[0132] It should be understood that the zero-prevention value is an extremely small positive number introduced to prevent the denominator from being zero. The available threshold is a preset threshold used to determine whether candidate data is valid, and can be configured according to the data quality requirements of this fusion. The item takes a larger value when both field completeness and semantic matching are high. The closer the two indicators are, the larger their values ​​become. Therefore, the semantic usability value reaches its maximum when both indicators are high and their values ​​are similar. This embodiment of the invention integrates four indicators into a single semantic usability value expression, giving higher priority to candidate data ends that are both high in completeness and matching degree and are close to each other, thus reducing the bias caused by considering only a single indicator.

[0133] S5. Based on the effective data terminal group and the pre-built semantic mapping rule base, obtain the mapping field node set, perform conflict verification and fusion on the mapping field node set, and obtain the fused data node set.

[0134] It should be explained that the mapped field node set is the set of field nodes formed by mapping the standard data nodes of each data end in the effective data end group through a semantic mapping rule base. The fused data node set is the field-level fusion result obtained after conflict verification and fusion of the mapped field node set.

[0135] Understandably, in a valid data set confirmed solely by semantically usable values, some data endpoints may have weak interoperability, which can lower the quality of subsequent fusion. Therefore, the step of obtaining the mapping field node set based on the valid data set and the pre-built semantic mapping rule base includes:

[0136] The valid data endpoints are initially grouped to obtain candidate domain groups and non-candidate domain groups;

[0137] Calculate the interoperability of each valid data endpoint within the candidate domain group to obtain multiple interoperability scores within the group;

[0138] Valid data endpoints with interoperability below a preset interoperability threshold among multiple groups are removed from the candidate domain groups to obtain the adjustment domain groups. The number of valid data endpoints in the adjustment domain groups is counted to obtain the number of valid data endpoints.

[0139] If the number of valid data endpoints is less than the preset endpoint threshold, then the interoperability between each valid data endpoint in the non-candidate domain group and the candidate domain group is calculated to obtain multiple inter-group interoperability.

[0140] The out-of-group interoperability is sorted in descending order to obtain the out-of-group interoperability sequence;

[0141] Extract an out-of-group interoperability subsequence from the out-of-group interoperability sequence, wherein the number of out-of-group interoperability in the out-of-group interoperability subsequence is the difference between the end number threshold and the number of valid data ends;

[0142] The effective data ends corresponding to each out-of-group interoperability in the out-of-group interoperability subsequence are moved from the non-candidate domain group to the adjustment domain group to obtain the interoperability domain group.

[0143] Otherwise, the adjusted domain group will be identified as an interconnected domain group;

[0144] The semantic mapping rule base is used to perform rule mapping on the standard data nodes corresponding to each valid data terminal in the interconnected domain group, resulting in a set of mapped field nodes.

[0145] It should be understood that the initial grouping is an operation of dividing the valid data endpoint groups according to the type of the data endpoint. Specifically, valid data endpoints with the same data endpoint type label are grouped into the same candidate domain group. The data endpoint type label is a category identifier pre-labeled for each registered data endpoint in the medical data fusion platform, including clinical business, medical technology support, management support, and other categories. For example, assuming the valid data endpoint group is: {Electronic Medical Record System, Medical Order System, Laboratory Information System, Image Archiving System}, this grouping according to the data endpoint type label will group the Electronic Medical Record System and Medical Order System, which belong to the clinical business category, into the candidate domain group, i.e., the candidate domain group is: {Electronic Medical Record System, Medical Order System}, and the Laboratory Information System and Image Archiving System, which belong to the medical technology support category, into the non-candidate domain group, i.e., the non-candidate domain group is: {Laboratory Information System, Image Archiving System}.

[0146] Understandably, the interoperability score is a numerical value that measures the degree of data alignment between two valid data endpoints. It can be calculated by weighting one or more indicators that reflect the data matching quality in historical fusion tasks. The higher the interoperability score, the better the data interoperability between the two valid data endpoints. The specific indicator items and their weights used in the calculation can be dynamically configured according to actual business needs and the historical performance of the valid data endpoints. For example, one or more indicators such as field hit rate, unit consistency rate, time series alignment rate, field completeness rate, and data accuracy rate can be selected and calculated by weighted summation. This flexible indicator configuration method allows the interoperability score to adapt to the differentiated requirements for medical data alignment quality in different scenarios.

[0147] Further, the intra-group interoperability is the average interoperability between a valid data endpoint in a candidate domain group and all valid data endpoints in the candidate domain group except itself. The inter-group interoperability is the average interoperability between a valid data endpoint in a non-candidate domain group and all valid data endpoints in the candidate domain group. The interoperability threshold is the threshold for determining whether a valid data endpoint needs to be removed from the candidate domain group, and can be configured according to the fusion quality requirements. The endpoint number threshold is the minimum number of valid data endpoints required to be included in the interoperable domain group. The adjusted domain group is the group obtained after removing valid data endpoints from the candidate domain group that do not meet the interoperability standard. The interoperable domain group is the group finally used for rule mapping after supplementation or confirmation.

[0148] Understandably, the rule mapping is a process of mapping fields in standard data nodes to platform standard fields and filling in values ​​based on the semantic mapping rule library. Each filling operation generates a mapped field node. Specifically, assuming a standard data node contains the source field: Collection Time - Lab Report - Blood Glucose Value (90 mg / dL), and the semantic mapping rule library defines this field as corresponding to the platform standard field: Fasting Blood Glucose (mmol / L), the rule mapping process unifies the field name to Fasting Blood Glucose and converts the value 90 mg / dL to 5.0 mmol / L according to the conversion relationship (1 mmol / L = 18 mg / dL) before filling it in. The generated mapped field node is: {Platform Standard Field: Fasting Blood Glucose, Standardized Value: 5.0, Unit: mmol / L, Source Field: Lab Report - Blood Glucose Value, Collection Time}. The platform standard fields are a standardized set of field names, data types, units of measurement, and value ranges uniformly predefined in the standard field dictionary by the medical data fusion platform, serving as the unified target format for all valid data returned when performing rule mapping. This invention employs a two-stage interoperability screening and on-demand supplementation method. By eliminating valid data sources with poor interoperability, it dynamically supplements the number of data sources from high to low interoperability, thereby achieving a balance between the breadth of data coverage and the quality of data alignment, and providing dual assurance for the quality of the fusion results.

[0149] Further, the conflict verification and fusion of the mapped field node set to obtain a fused data node set includes:

[0150] The mapping field nodes in the mapping field node set are grouped to obtain a set of fields to be merged, wherein each set of fields to be merged includes one or more mapping field nodes;

[0151] Extract the fields to be merged sequentially from the set of fields to be merged, count the number of mapped field nodes in the fields to be merged, and obtain the number of nodes in the group;

[0152] If there is only one node in the group, then the mapping field node in the field group to be merged will be identified as the merged data node.

[0153] If the number of nodes in a group is greater than one, then the conflict detection results between the mapping field nodes in the field group to be merged are checked. The conflict detection result is either that there is a conflict or that there is no conflict.

[0154] When the conflict detection result is that there is no conflict, all mapped field nodes in the field group to be merged are merged and processed to obtain the merged data node.

[0155] When the conflict detection result indicates a conflict, the mapped field nodes are extracted sequentially from the field group to be fused, and the field matching degree and time validity corresponding to the mapped field node are obtained. The fusion weight of the mapped field node is calculated based on the field matching degree and time validity, as shown in the following formula:

[0156]

[0157] in, Indicates the fusion weights, Indicates the matching degree of the field. Indicates the validity of the time period. This indicates that there are a total of [number] fields in the group to be merged. One mapping field node;

[0158] The fusion weights are aggregated to obtain n fusion weights, where each fusion weight corresponds one-to-one with a mapping field node. The mapping field node corresponding to the largest fusion weight among the n fusion weights is taken as the fusion data node.

[0159] The data nodes are aggregated and merged to obtain the merged data node set.

[0160] It should be understood that the field group to be merged is a group formed by aggregating the mapping field nodes with the same field name in the mapping field node set according to the platform standard field name. For example, all mapping field nodes mapped to the platform standard field (medication record) are grouped together.

[0161] Understandably, the result of the conflict detection between the mapping field nodes in the field group to be merged is as follows: the consistency of the mapping field nodes returned by multiple valid data terminals under the same platform standard field is checked to determine whether there are any contradictions between the nodes in terms of value, unit, timestamp or reference range. When all mapping field nodes are consistent in the above dimensions or the differences are within the preset tolerance range, it is determined that there is no conflict; otherwise, it is determined that there is a conflict.

[0162] For example, suppose a set of fields to be fused is a platform standard field: fasting blood glucose. This field includes mapped field nodes from three valid data sources (electronic medical record system, laboratory information system, and image archiving system), with values ​​of 5.0 mmol / L, 5.1 mmol / L, and 5.0 mmol / L respectively. If the units are consistent and the numerical differences are within a preset tolerance range, the conflict detection result is no conflict. If the values ​​are 5.0 mmol / L, 5.0 mmol / L, and 10.2 mmol / L, and the 10.2 mmol / L value significantly deviates from the previous two (5.0 mmol / L and 5.0 mmol / L), then the conflict detection result is that a conflict exists. It is important to clarify that conflict detection for specific test values ​​is only performed on mapped field nodes within the same acquisition time window. Test values ​​at different time points reflect different physiological states of the human body and are not directly comparable. Before conflict detection, the mapped field nodes must be time-aligned according to the acquisition time, and only mapped field nodes with time differences within the preset tolerance range are subject to conflict detection.

[0163] Furthermore, the merging and fusion process involves aggregating the values ​​of each mapped field node in the same group to obtain the fused data node when there are no conflicts. For example, for numeric fields, the average of multiple fields is taken; for non-numeric fields (such as text or set fields), the union of multiple fields is taken. The field matching degree is the degree of semantic fit between the mapped field node and the platform standard field under the semantic mapping rule base. The semantic similarity is calculated by converting the mapped field node into a first semantic vector, converting the platform standard field into a second semantic vector, calculating the cosine similarity between the first and second semantic vectors, and using the cosine similarity as the semantic similarity. The Doc2Vec algorithm can be used to convert the mapped field node into a first semantic vector and the platform standard field into a second semantic vector; this is existing technology and will not be elaborated further here.

[0164] The time validity is used to characterize the proximity between the collection time and the current time in the mapping field node. Within the validity period, the closer the time validity is to 1, the higher the timeliness of the mapping field node. The calculation method is: Time validity = max[0,1-(t1-t0) / (t2-t0)], where t0 represents the collection time, t1 represents the current time, and t2 represents the valid deadline.

[0165] It should be clarified that the reason for selecting the mapping field node corresponding to the largest fusion weight among n fusion weights as the fusion data node is that the fusion weight comprehensively reflects the overall quality of the mapping field node in three dimensions: data source credibility, semantic matching accuracy, and data timeliness. The larger the fusion weight, the more trustworthy the data of that mapping field node is. When multi-source data conflict and cannot be directly aggregated, selecting the mapping field node with the largest fusion weight as the fusion data node allows for optimal selection in a quantifiable way, thereby ensuring the accuracy, reliability, and timeliness of the fused medical data. This embodiment of the invention, through conflict detection and diversion processing and weight normalization selection based on multi-dimensional indicators, retains the merging method in conflict-free scenarios while selecting the optimal field value in a quantifiable way in conflict scenarios, thus improving the accuracy of the fusion results.

[0166] S6. Generate a flow record node based on the authorized node, semantic mapping node set, mapping field node set, and fusion data node set, and send the fusion data node set and flow record node to the request source end to realize medical data fusion.

[0167] It should be explained that the flow record node is a data structure used to record summary information of each stage of the current fusion operation, such as authorization, evaluation, mapping, and fusion, and can be used for subsequent traceability.

[0168] Understandably, in order to ensure independent traceability of each step without exposing the original medical data, it is necessary to generate summaries for the four steps of authorization, evaluation, mapping, and fusion, and nest them hierarchically in causal order. Therefore, the generation of flow record nodes based on the authorization node, semantic mapping node set, mapping field node set, and fusion data node set includes:

[0169] Perform one-way hashing on the authorized nodes to obtain the authorization digest;

[0170] A one-way hashing process is performed on the semantic mapping node set and the authorization digest to obtain the evaluation digest;

[0171] A one-way hash is performed on the mapping field node set and the evaluation summary to obtain the mapping summary;

[0172] Perform one-way hashing on the fused data node set and mapping digest to obtain the fused digest, and record the generation time of the fused digest to obtain the transfer time;

[0173] By associating the fusion identifier, request source, target service stage, interconnected domain group, authorization digest, evaluation digest, mapping digest, fusion digest, and transfer time, the transfer record node is obtained.

[0174] It should be understood that the authorization digest is a string obtained by performing a one-way hash on only the authorization nodes. The evaluation digest is a string obtained by concatenating the semantic mapping node set with the authorization digest and then performing a one-way hash. Since the authorization digest of the previous level is appended to the end of the input during hash calculation, the evaluation digest and the authorization digest form a cryptographic dependency. The mapping digest is a string obtained by concatenating the mapping field node set with the evaluation digest and then performing a one-way hash. The fusion digest is a string obtained by concatenating the fused data node set with the mapping digest and then performing a one-way hash. The flow time is the moment the fusion digest is generated, marking the time point when this fusion is completed. This embodiment of the invention uses multi-level nested digests to cryptographically bind the authorization, evaluation, mapping, and fusion stages in a causal order, so that if medical data is tampered with at any stage, all subsequent digests will be inconsistent, thus making the entire medical data fusion process traceable.

[0175] To address the problems described in the background section, this invention receives a medical data fusion request, parses the request to obtain an anonymous user ID, the request source, the target service stage, and the fused data items. Based on these information, it obtains the authorized node. This invention confirms the fusion permission range through multi-dimensional screening of user authorization records and then performs one-way hashing using time and random codes, ensuring that each fusion action has a unique traceable credential. This achieves precise control over the list of permitted fields and sensitive values ​​without exposing the user's true identity, thus improving the security of medical data fusion and obtaining the data corresponding to the anonymous user ID. The business data nodes include appointment information, communication records, service data indexes, and follow-up records. The participating end node set is identified using the authorized nodes and business data nodes. This invention achieves a single-end usability and cross-end non-association medical data information protection method through in-end replacement numbering. Combined with hierarchical shielding and standardized processing, subsequent fusion protects user privacy while facilitating cross-end field alignment. By matching the field authorization scope, the actual capabilities of candidate data ends, and historical fusion performance, data ends lacking data provision capabilities or not overlapping with the authorization scope are eliminated in advance, making the participating end node set more focused and improving the effectiveness of subsequent medical data fusion. The participating node set obtains a standard data node set, which is then used to obtain a semantic mapping node set. Based on this semantic mapping node set, valid data node groups are identified within the participating node set. This invention utilizes a combination of four indicators—field completeness, semantic matching degree, stage consistency, and historical fusion pass rate—to form a quantifiable multi-dimensional evaluation, providing comparable numerical criteria for determining the validity of candidate data nodes. Based on the valid data node groups and a pre-built semantic mapping rule base, a mapping field node set is obtained. This mapping field node set undergoes conflict verification and fusion to obtain a fused data node set. This invention, through secondary screening based on interoperability and on-demand completion, effectively eliminates... In addition to valid data sources with poor interoperability, the number of data sources is dynamically supplemented according to their interoperability from high to low, thereby achieving a balance between the breadth of data coverage and the quality of data alignment. This provides dual assurance for the quality of the fusion results. Based on the authorized node, semantic mapping node set, mapping field node set, and fusion data node set, a flow record node is generated. The fusion data node set and flow record node are sent to the requesting source to achieve medical data fusion. It can be seen that this invention cryptographically binds the authorization, evaluation, mapping, and fusion links in a causal order through multi-level nested digests. This ensures that any tampering with medical data in any link will lead to inconsistencies in all subsequent digests, thus making the entire medical data fusion process traceable. Therefore, this invention can realize intelligent screening, fusion, and reliable traceability of multi-source heterogeneous medical data across institutions.

[0176] like Figure 2 The diagram shown is a functional block diagram of a medical data fusion system based on semantic mapping provided in an embodiment of the present invention.

[0177] The semantic mapping-based medical data fusion system 100 of this invention can be installed in an electronic device. Depending on the functions implemented, the semantic mapping-based medical data fusion system 100 may include an authorized node construction module 101, a valid data endpoint confirmation module 102, a data fusion module 103, and a data flow traceability module 104. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and which are stored in the memory of the electronic device.

[0178] The authorized node construction module 101 is used to receive medical data fusion requests, parse the medical data fusion requests, obtain the anonymous user number, the request source, the target service stage and the fusion data items, and obtain the authorized node based on the anonymous user number, the request source, the target service stage and the fusion data items;

[0179] The valid data terminal confirmation module 102 is used to obtain the business data node corresponding to the anonymous user number, wherein the business data node includes appointment information, communication records, service information index and follow-up records.

[0180] The participating node set is identified using the authorized node and the business data node;

[0181] A standard data node set is obtained based on the participating end node set, a semantic mapping node set is obtained using the standard data node set, and a valid data end group is identified in the participating end node set based on the semantic mapping node set.

[0182] The data fusion module 103 is used to obtain a set of mapping field nodes based on the effective data end group and the pre-built semantic mapping rule base, and to perform conflict verification and fusion on the set of mapping field nodes to obtain a fused data node set.

[0183] The circulation traceability module 104 is used to generate circulation record nodes based on the authorized node, semantic mapping node set, mapping field node set and fusion data node set, and send the fusion data node set and circulation record nodes to the request source end to realize medical data fusion.

[0184] In detail, the modules in the semantic mapping-based medical data fusion system 100 described in this embodiment of the invention employ the same methods as described above. Figure 1 The method uses the same techniques as the semantic mapping-based medical data fusion method described in the article and can produce the same technical effects, so it will not be repeated here.

[0185] like Figure 3 The diagram shown is a schematic representation of an electronic device that implements a semantic mapping-based medical data fusion method according to an embodiment of the present invention.

[0186] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as a medical data fusion method program based on semantic mapping.

[0187] The memory 11 includes at least one type of readable storage medium, such as flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as a portable hard drive. In other embodiments, the memory 11 can be an external storage device of the electronic device 1, such as a plug-in portable hard drive, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device 1. Furthermore, the memory 11 includes both internal storage units and external storage devices of the electronic device 1. The memory 11 can be used not only to store application software and various types of data installed on the electronic device 1, such as the code of a medical data fusion method program based on semantic mapping, but also to temporarily store data that has been output or will be output.

[0188] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., a semantic mapping-based medical data fusion method program) and calls data stored in the memory 11 to perform various functions of the electronic device 1 and process data.

[0189] The bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 can be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize the connection and communication between the memory 11 and at least one processor 10, etc.

[0190] Figure 3 Only electronic devices with components are shown; it will be understood by those skilled in the art that... Figure 3 The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0191] For example, although not shown, the electronic device 1 may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0192] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), which is typically used to establish communication connections between the electronic device 1 and other electronic devices.

[0193] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), and optionally, a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device 1 and to display a visual user interface.

[0194] The semantic mapping-based medical data fusion method program stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When run in the processor 10, it can achieve the following:

[0195] Receive a medical data fusion request, parse the medical data fusion request to obtain the anonymous user ID, request source, target service stage and fusion data item, and obtain the authorized node based on the anonymous user ID, request source, target service stage and fusion data item;

[0196] Obtain the business data node corresponding to the anonymous user ID, wherein the business data node includes appointment information, communication records, service information index and follow-up records;

[0197] The participating node set is identified using the authorized node and the business data node;

[0198] A standard data node set is obtained based on the participating end node set, a semantic mapping node set is obtained using the standard data node set, and a valid data end group is identified in the participating end node set based on the semantic mapping node set.

[0199] Based on the effective data terminal group and the pre-built semantic mapping rule base, a mapping field node set is obtained, and the mapping field node set is subjected to conflict verification and fusion to obtain a fused data node set.

[0200] Based on the authorized node, semantic mapping node set, mapping field node set, and fusion data node set, a flow record node is generated, and the fusion data node set and flow record node are sent to the request source end to realize medical data fusion.

[0201] Specifically, the processor 10's implementation method for the above instructions can be found in [reference needed]. Figures 1 to 3 The descriptions of the relevant steps in the corresponding embodiments are not repeated here.

[0202] Furthermore, if the modules / units integrated in the electronic device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0203] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device, can perform the following:

[0204] Receive a medical data fusion request, parse the medical data fusion request to obtain the anonymous user ID, request source, target service stage and fusion data item, and obtain the authorized node based on the anonymous user ID, request source, target service stage and fusion data item;

[0205] Obtain the business data node corresponding to the anonymous user ID, wherein the business data node includes appointment information, communication records, service information index and follow-up records;

[0206] The participating node set is identified using the authorized node and the business data node;

[0207] A standard data node set is obtained based on the participating end node set, a semantic mapping node set is obtained using the standard data node set, and a valid data end group is identified in the participating end node set based on the semantic mapping node set.

[0208] Based on the effective data terminal group and the pre-built semantic mapping rule base, a mapping field node set is obtained, and the mapping field node set is subjected to conflict verification and fusion to obtain a fused data node set.

[0209] Based on the authorized node, semantic mapping node set, mapping field node set, and fusion data node set, a flow record node is generated, and the fusion data node set and flow record node are sent to the request source end to realize medical data fusion.

[0210] In the embodiments provided by this invention, it should be understood that the disclosed devices, systems, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative, and actual implementations may have other classification methods.

[0211] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0212] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0213] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0214] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A medical data fusion method based on semantic mapping, characterized in that, The method includes: Receive a medical data fusion request, parse the medical data fusion request to obtain the anonymous user ID, request source, target service stage and fusion data item, and obtain the authorized node based on the anonymous user ID, request source, target service stage and fusion data item; The process of obtaining the authorization node based on the anonymous user ID, the request source, the target service stage, and the fused data items includes: Obtain the set of user authorization records, and use the anonymous user ID to retrieve the set of candidate authorization records from the set of user authorization records; Using the request source, target service stage, and fused data items, target authorization records are filtered from the candidate authorization record set. The target authorization records include: a list of permitted fields, sensitive values ​​of the fields, and authorization validity period. Obtain the current time and the fusion random code. If the current time is within the authorized validity period, perform one-way hashing on the anonymous user ID, request source, target service stage, current time and fusion random code to obtain the fusion identifier code. By associating the anonymous user ID, the list of permitted fields, the sensitive values ​​of the fields, the authorization validity period, and the fusion identifier code, the authorization node can be obtained. Obtain the business data node corresponding to the anonymous user ID, wherein the business data node includes appointment information, communication records, service information index and follow-up records; The participating node set is identified using the authorized node and the business data node; The step of identifying the participating node set using the authorized node and business data node includes: Based on the list of permitted fields, perform field matching on the business data nodes to filter out the list of permitted data items; Using the list of permitted data items, retrieve multiple candidate data endpoints from the pre-defined data endpoint registration table, and perform the following operations on each of the multiple candidate data endpoints: Obtain a list of available fields and historical fusion pass rates from the candidate data source; The list of available fields is matched with the list of permitted data items to obtain a list of fields that can be merged. If the list of fields that can be merged is not empty, then associate the candidate data end, the list of fields that can be merged, and the historical fusion pass rate to obtain the participating end nodes, and summarize the participating end nodes to obtain the set of participating end nodes; A standard data node set is obtained based on the participating end node set, a semantic mapping node set is obtained using the standard data node set, and a valid data end group is identified in the participating end node set based on the semantic mapping node set. Based on the effective data terminal group and the pre-built semantic mapping rule base, a mapping field node set is obtained, and the mapping field node set is subjected to conflict verification and fusion to obtain a fused data node set; Based on the authorized node, semantic mapping node set, mapping field node set, and fusion data node set, a flow record node is generated, and the fusion data node set and flow record node are sent to the request source end to realize medical data fusion.

2. The medical data fusion method based on semantic mapping as described in claim 1, characterized in that, The process of obtaining the standard data node set based on the participating end node set includes: Perform the following operation on each participating node in the participating node set: Using the list of fusionable fields, edge data nodes are extracted from the candidate data nodes, wherein the edge data nodes include: identity identification field and medical content field; One-way hashing is performed on the anonymous user ID, candidate data endpoint, and fusion identifier to obtain the intra-end replacement ID; The identity identifier field is replaced with an in-device replacement number, and the medical content field is masked based on the field sensitivity value to obtain the de-identified data node; The unidentified data nodes are standardized to obtain standard data nodes. By aggregating the standard data nodes, a set of standard data nodes is obtained, in which each standard data node corresponds to a different candidate data endpoint.

3. The medical data fusion method based on semantic mapping as described in claim 2, characterized in that, The step of obtaining the semantic mapping node set using the standard data node set includes: Extract standard data nodes sequentially from the standard data node set, obtain the number of filled fields corresponding to the standard data nodes, obtain the total number of fields corresponding to the list of fields that can be merged, calculate the ratio of the number of filled fields to the total number of fields, and obtain the field completeness. The semantic mapping rule base is used to perform semantic matching on standard data nodes to obtain semantic matching degree and stage consistency degree; By considering the completeness of associated fields, semantic matching degree, stage consistency degree, and historical fusion pass rate, semantic mapping nodes are obtained. By summarizing the semantic mapping nodes, a semantic mapping node set is obtained.

4. The medical data fusion method based on semantic mapping as described in claim 3, characterized in that, The process of identifying valid data endpoint groups from the participating endpoint node set based on the semantic mapping node set includes: Semantic mapping nodes are extracted sequentially from the set of semantic mapping nodes, and the semantic usability value corresponding to the semantic mapping node is calculated. The calculation formula is as follows: ； in, Indicates the available values ​​of the semantics. This indicates the completeness of the field. Indicates the semantic matching degree. This indicates the consistency of the stated stages. This indicates the historical fusion pass rate. This indicates the preset zero-prevention value; If the semantic availability value is not less than the preset availability threshold, the candidate data ends corresponding to the semantic mapping node are confirmed as valid data ends, and the valid data ends are aggregated to obtain a valid data end group.

5. The medical data fusion method based on semantic mapping as described in claim 4, characterized in that, The step of obtaining the mapping field node set based on the effective data endpoint group and the pre-built semantic mapping rule base includes: The valid data endpoints are initially grouped to obtain candidate domain groups and non-candidate domain groups; Calculate the interoperability of each valid data endpoint within the candidate domain group to obtain multiple interoperability scores within the group; Valid data endpoints with interoperability below a preset interoperability threshold among multiple groups are removed from the candidate domain groups to obtain the adjustment domain groups. The number of valid data endpoints in the adjustment domain groups is counted to obtain the number of valid data endpoints. If the number of valid data endpoints is less than the preset endpoint threshold, then the interoperability between each valid data endpoint in the non-candidate domain group and the candidate domain group is calculated to obtain multiple inter-group interoperability. The out-of-group interoperability is sorted in descending order to obtain the out-of-group interoperability sequence; Extract an out-of-group interoperability subsequence from the out-of-group interoperability sequence, wherein the number of out-of-group interoperability in the out-of-group interoperability subsequence is the difference between the end number threshold and the number of valid data ends; The effective data ends corresponding to each out-of-group interoperability in the out-of-group interoperability subsequence are moved from the non-candidate domain group to the adjustment domain group to obtain the interoperability domain group. Otherwise, the adjusted domain group will be identified as an interconnected domain group; The semantic mapping rule base is used to perform rule mapping on the standard data nodes corresponding to each valid data terminal in the interconnected domain group, resulting in a set of mapped field nodes.

6. The medical data fusion method based on semantic mapping as described in claim 5, characterized in that, The process of performing conflict verification and fusion on the mapped field node set to obtain a fused data node set includes: The mapping field nodes in the mapping field node set are grouped to obtain a set of fields to be merged, wherein each set of fields to be merged includes one or more mapping field nodes; Extract the fields to be merged sequentially from the set of fields to be merged, count the number of mapped field nodes in the fields to be merged, and obtain the number of nodes in the group; If there is only one node in the group, then the mapping field node in the field group to be merged will be identified as the merged data node. If the number of nodes in a group is greater than one, then the conflict detection results between the mapping field nodes in the field group to be merged are checked. The conflict detection result is either that there is a conflict or that there is no conflict. When the conflict detection result is that there is no conflict, all mapped field nodes in the field group to be merged are merged and processed to obtain the merged data node. When the conflict detection result indicates a conflict, the mapped field nodes are extracted sequentially from the field group to be fused, and the field matching degree and time validity corresponding to the mapped field node are obtained. The fusion weight of the mapped field node is calculated based on the field matching degree and time validity, as shown in the following formula: ； in, Indicates the fusion weights, Indicates the matching degree of the field. Indicates the validity of the time period. This indicates that there are a total of [number] fields in the group to be merged. One mapping field node; The fusion weights are aggregated to obtain n fusion weights, where each fusion weight corresponds one-to-one with a mapping field node. The mapping field node corresponding to the largest fusion weight among the n fusion weights is taken as the fusion data node. The data nodes are aggregated and merged to obtain the merged data node set.

7. The medical data fusion method based on semantic mapping as described in claim 6, characterized in that, The generation of flow record nodes based on the authorized nodes, semantic mapping node set, mapping field node set, and fused data node set includes: Perform one-way hashing on the authorized nodes to obtain the authorization digest; A one-way hashing process is performed on the semantic mapping node set and the authorization digest to obtain the evaluation digest; A one-way hash is performed on the mapping field node set and the evaluation summary to obtain the mapping summary; Perform one-way hashing on the fused data node set and mapping digest to obtain the fused digest, and record the generation time of the fused digest to obtain the transfer time; By associating the fusion identifier, request source, target service stage, interconnected domain group, authorization digest, evaluation digest, mapping digest, fusion digest, and transfer time, the transfer record node is obtained.

8. A medical data fusion system based on semantic mapping, characterized in that, The system includes: The authorized node construction module is used to receive medical data fusion requests, parse the medical data fusion requests, obtain the anonymous user ID, request source, target service stage and fusion data item, and obtain the authorized node based on the anonymous user ID, request source, target service stage and fusion data item; The process of obtaining the authorization node based on the anonymous user ID, the request source, the target service stage, and the fused data items includes: Obtain the set of user authorization records, and use the anonymous user ID to retrieve the set of candidate authorization records from the set of user authorization records; Using the request source, target service stage, and fused data items, target authorization records are filtered from the candidate authorization record set. The target authorization records include: a list of permitted fields, sensitive values ​​of the fields, and authorization validity period. Obtain the current time and the fusion random code. If the current time is within the authorized validity period, perform one-way hashing on the anonymous user ID, request source, target service stage, current time and fusion random code to obtain the fusion identifier code. By associating the anonymous user ID, the list of permitted fields, the sensitive values ​​of the fields, the authorization validity period, and the fusion identifier code, the authorization node can be obtained. The valid data terminal confirmation module is used to obtain the business data node corresponding to the anonymous user ID, wherein the business data node includes appointment information, communication records, service information index and follow-up records; The participating node set is identified using the authorized node and the business data node; The step of identifying the participating node set using the authorized node and business data node includes: Based on the list of permitted fields, perform field matching on the business data nodes to filter out the list of permitted data items; Using the list of permitted data items, retrieve multiple candidate data endpoints from the pre-defined data endpoint registration table, and perform the following operations on each of the multiple candidate data endpoints: Obtain a list of available fields and historical fusion pass rates from the candidate data source; The list of available fields is matched with the list of permitted data items to obtain a list of fields that can be merged. If the list of fields that can be merged is not empty, then associate the candidate data end, the list of fields that can be merged, and the historical fusion pass rate to obtain the participating end nodes, and summarize the participating end nodes to obtain the set of participating end nodes; A standard data node set is obtained based on the participating end node set, a semantic mapping node set is obtained using the standard data node set, and a valid data end group is identified in the participating end node set based on the semantic mapping node set. The data fusion module is used to obtain a set of mapping field nodes based on the effective data terminal group and the pre-built semantic mapping rule base, and to perform conflict verification and fusion on the set of mapping field nodes to obtain a fused data node set. The data flow traceability module is used to generate flow record nodes based on the authorized node, semantic mapping node set, mapping field node set, and fusion data node set, and send the fusion data node set and flow record nodes to the request source end to realize medical data fusion.

Images