CN-122022850-A - Medicine whole-flow circulation tracing system and method

CN122022850ACN 122022850 ACN122022850 ACN 122022850ACN-122022850-A

Abstract

The invention relates to the technical field of medicine data management, in particular to a medicine whole-flow circulation traceability system and method, comprising the steps of obtaining multi-source traceability data of a plurality of medicine units of a target medicine; the method comprises the steps of obtaining a plurality of time slices through time division, comprehensively analyzing the association condition of the medicine units among different links under each information dimension based on multi-source traceability data of all the medicine units to obtain flow association coefficients among the medicine units, constructing local circulation networks in the corresponding time slices by combining the flow association coefficients, and splicing graph structures of the local circulation networks of all the time slices by utilizing the flow association coefficients of the medicine units to obtain a full-flow circulation model of a target medicine, so that a full-flow circulation track visual view of the medicine units is obtained. The invention improves the integrity and accuracy of the whole process tracing of the medicine and realizes the automatic adaptation and association of heterogeneous data.

Inventors

YANG JING
ZHONG WENQING

Assignees

澜如(天津)科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260312

Claims (10)

1. The whole-flow circulation tracing method of the medicine is characterized by comprising the following steps: Acquiring multi-source traceability data of a plurality of medicine units of a target medicine, wherein the multi-source traceability data comprises traceability coding information, date information, main body information and position information of each link of the corresponding medicine unit in the circulation process; Acquiring a plurality of time slices through time division, comprehensively analyzing the association condition of the medicine units in each information dimension among different links based on the traceability coding information, the date information and the position information of all the medicine units, and obtaining flow association coefficients among the medicine units; carrying out graph structure splicing on local circulation networks of all time slices by utilizing flow association coefficients of the medicine units and combining main body information of the medicine units in the circulation process to obtain a full-flow circulation model of the target medicine; And generating a full-flow circulation track visible view of each medicine unit through the full-flow circulation model.
2. The method for tracing the whole drug flow according to claim 1, wherein the specific obtaining method of the flow association coefficient is as follows: clustering all the medicine units in any link according to the traceability coding information, the date information and the position information of all the medicine units in any link to obtain a plurality of clusters; Based on the association conditions of the drug units in the clusters contained in the adjacent links in the circulation process in the traceable coding information, the date information and the position information respectively, and combining preset weights, calculating the flow association coefficients among the drug units.
3. The method according to claim 2, wherein the calculating the process association coefficient between the medicine units based on the association condition of the medicine units in the cluster included in the adjacent links in the circulation process in the traceable code information, the date information and the position information respectively and by combining with the preset weight comprises the following specific steps: For the medicine units in any adjacent links, calculating the time association degree between the medicine units according to the date information distribution and the difference condition in the clustering clusters to which the medicine units belong; calculating Euclidean distance between the medicine units based on the position information of the medicine units, thereby obtaining the space association degree between the medicine units; weights are preset for the time association degree, the coding logic matching degree and the space association degree among the medicine units respectively, and flow association coefficients among the medicine units are obtained in a weighted fusion mode.
4. The method for tracing the whole flow of medicine according to claim 3, wherein the specific method for obtaining the time association degree between the medicine units is as follows: Calculating the time evolution coefficients among the drug units according to the discrete condition and the distribution level of the time information of the drug units in the clustering clusters to which the drug units in the adjacent links belong respectively; and combining the time evolution coefficients among the medicine units and the differences of the corresponding time information to obtain the time association degree among the medicine units.
5. The method for tracing the whole flow of medicine according to claim 3, wherein the specific method for obtaining the matching degree of the coding logic between the medicine units is as follows: and processing the corresponding traceable coding information of the medicine units in the adjacent links respectively through a preset rule analyzer to obtain corresponding coding data, and taking Jaccard coefficients of the corresponding coding data among the medicine units as the coding logic matching degree among the medicine units.
6. The method for tracing the whole drug flow according to claim 1, wherein the specific acquisition method of the local flow network is as follows: determining an effective association threshold according to the statistical distribution of the flow association coefficients, and screening drug unit pairs formed by corresponding drug units when the outflow Cheng Guan association coefficients are greater than or equal to the effective association threshold; taking a combination formed by a main body and links corresponding to each medicine unit as a network node, taking an association relation formed by the medicine units to the contained medicine units as a directed edge, and taking a flow association coefficient between the medicine units to the contained medicine units as an initial weight of the edge, so that a local circulation sub-network between adjacent links in any time slice is formed; And splicing all adjacent links in the time slice to the corresponding local circulation sub-network, de-overlapping and merging the nodes according to the main body identification, and keeping directivity and weight information at the edges so as to construct the corresponding local circulation network.
7. The method for tracing the whole-process circulation of the medicine according to claim 1, wherein the specific acquisition method of the whole-process circulation model is as follows: The main body information at least comprises a medicine enterprise, a distribution company, a hospital and a pharmacy in the circulation process; acquiring a circulation mode of a corresponding time slice based on the distribution condition of time information of the medicine units in any time slice and the distribution condition of the quantity of medicines circulated by a distribution company; and combining the circulation modes of all the time slices, and splicing the local circulation networks corresponding to all the time slices to form a full-flow circulation model.
8. The method for tracing the circulation of the whole medicine process according to claim 7, wherein the specific acquisition method of the circulation mode of the time slice is as follows: For any time slice, taking the average time interval from production to use of the medicine units in the time slice as the circulation speed characteristic of the corresponding time slice; Calculating standard deviation of the quantity of the medicine units circulated by all the distribution companies in the distribution link, and calculating concentration characteristics of the corresponding time slices based on the standard deviation, wherein the standard deviation and the concentration characteristics are in negative correlation; Obtaining the distribution companies with the quantity of the circulated medicine units larger than the average quantity of the medicine units circulated by all distribution companies from all distribution companies, marking the distribution companies as target distribution companies, and taking the quantity proportion of the target distribution companies in all distribution companies as main body liveness characteristics of corresponding time slices; And taking the circulation speed characteristic, the concentration characteristic and the main body liveness characteristic of the time slice as inputs, and identifying a circulation mode corresponding to the time slice through a pre-trained neural network, wherein the circulation mode comprises a conventional distribution mode, an emergency direct distribution mode and a mixed mode.
9. The full-flow circulation traceability method of the medicine according to claim 1, wherein the specific acquisition method of the time slice is as follows: Based on date information of all the medicine units, a circulation time sequence is constructed, the circulation time sequence of the target medicine is divided according to preset time granularity, a plurality of sequence segments are formed, and a section formed by the maximum value and the minimum value of elements in each sequence segment is used as a corresponding time slice.
10. A drug whole-flow circulation tracing system comprising a memory, a processor and a computer program stored in the memory and operable on the processor, wherein the processor implements the steps of a drug whole-flow circulation tracing method according to any one of claims 1 to 9 when executing the computer program.

Description

Medicine whole-flow circulation tracing system and method Technical Field The invention relates to the technical field of medicine data management, in particular to a medicine whole-flow circulation traceability system and method. Background The current medicine whole-flow traceability system mainly depends on a one-object one-code technology, but still has the condition that each link system adopts different coding rules and data formats, so that data are scattered in different systems such as production, circulation, use and the like, and a whole-chain association cannot be effectively established. The production links mostly adopt NMPA electronic supervision codes, the circulation links mostly use custom batch numbers, and the use links adopt different traceability code formats. The problems of non-uniform data format, non-open system interface and non-automatic data acquisition lead to the insufficient accuracy of the whole chain association of the drug traceability system, which seriously affects the drug safety supervision and the drug safety of patients. In the prior art, a system capable of automatically adapting to different system coding rules, intelligently associating multi-link data and outputting full-chain traceability results in real time is lacking, so that a medicine traceability target with 'source traceability, overtaking and responsibility traceability' is difficult to realize. Disclosure of Invention The invention provides a medicine full-flow circulation tracing system and method, which aim to solve the existing problems, namely the lack of a system which can automatically adapt to different system coding rules, intelligently correlate multi-link data and output full-chain tracing results in real time in the prior art, so that medicine tracing targets of 'source is detectable, destination is traceable and responsibility is accountable' are difficult to realize. The invention relates to a drug whole-flow circulation traceability system and a drug whole-flow circulation traceability method, which adopt the following technical scheme: The embodiment of the invention provides a medicine whole-flow circulation tracing method, which comprises the following steps: Acquiring multi-source traceability data of a plurality of medicine units of a target medicine, wherein the multi-source traceability data comprises traceability coding information, date information, main body information and position information of each link of the corresponding medicine unit in the circulation process; Acquiring a plurality of time slices through time division, comprehensively analyzing the association condition of the medicine units in each information dimension among different links based on the traceability coding information, the date information and the position information of all the medicine units, and obtaining flow association coefficients among the medicine units; carrying out graph structure splicing on local circulation networks of all time slices by utilizing flow association coefficients of the medicine units and combining main body information of the medicine units in the circulation process to obtain a full-flow circulation model of the target medicine; And generating a full-flow circulation track visible view of each medicine unit through the full-flow circulation model. Optionally, the specific method for obtaining the flow association coefficient includes: clustering all the medicine units in any link according to the traceability coding information, the date information and the position information of all the medicine units in any link to obtain a plurality of clusters; Based on the association conditions of the drug units in the clusters contained in the adjacent links in the circulation process in the traceable coding information, the date information and the position information respectively, and combining preset weights, calculating the flow association coefficients among the drug units. Optionally, the calculating the process association coefficient between the medicine units based on association conditions of the medicine units in the cluster clusters included in the adjacent links in the circulation process in the traceable coding information, the date information and the position information respectively and by combining with preset weights comprises the following specific methods: For the medicine units in any adjacent links, calculating the time association degree between the medicine units according to the date information distribution and the difference condition in the clustering clusters to which the medicine units belong; calculating Euclidean distance between the medicine units based on the position information of the medicine units, thereby obtaining the space association degree between the medicine units; weights are preset for the time association degree, the coding logic matching degree and the space association degree among the medicine units respectively, and flow associa