CN114360742A - Medicine warning data updating system and method based on data analysis - Google Patents
Medicine warning data updating system and method based on data analysis Download PDFInfo
- Publication number
- CN114360742A CN114360742A CN202210027905.2A CN202210027905A CN114360742A CN 114360742 A CN114360742 A CN 114360742A CN 202210027905 A CN202210027905 A CN 202210027905A CN 114360742 A CN114360742 A CN 114360742A
- Authority
- CN
- China
- Prior art keywords
- data
- drug
- database
- adverse reaction
- unit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention discloses a medicine warning data updating system and method based on data analysis, and belongs to the technical field of medical treatment. The invention comprises the following steps: the method comprises the following steps: acquiring a personal safety report, and classifying the personal safety report according to the adverse reaction type, wherein the personal safety report comprises a determined patient, a determined reporter, a determined adverse reaction event and a determined suspected medicine; step two: judging the effectiveness of the drug warning data in the classified individual case safety reports, and rejecting invalid individual case safety reports according to the judgment result; step three: recording the drug alert data and the adverse reaction types in the individual safety reports left in the step two, and sorting the data into a data set; step four: and matching and identifying the data set obtained in the step three with a database, and updating the drug alert data in the database and the adverse reaction type corresponding to the drug alert data according to the matching and identifying result.
Description
Technical Field
The invention relates to the technical field of medical treatment, in particular to a medicine warning data updating system and method based on data analysis.
Background
Pharmacusing is a scientific study and activity related to the discovery, evaluation, understanding, and prevention of adverse reactions or any other problems that may be associated with a drug, which starts with the safety of the drug user, discovers, evaluates, and prevents adverse reactions to a drug, thereby improving the safety of the patient in medication, treatment, and adjuvant therapy.
The existing medicine warning data updating system obtains medicine warning data based on individual case safety reports, and carries out classification processing on the data after the data are obtained, the classification processing in the process is complex, the situation of data omission is easy to occur, the system updating is not comprehensive, the application range of the system is further reduced, whether the obtained data has effectiveness or whether the obtained data has contingency cannot be determined when the medicine warning data are obtained, the obtained medicine warning data do not have reference value, the experience of a customer using the medicine warning data updating system is further reduced, and when old data in the updating system are updated, the situation that data updating contents are lost due to system faults exists, manual retrieval is needed, specific lost data are judged, the data processing time is prolonged, and the processing efficiency is low.
Disclosure of Invention
The present invention is directed to a system and a method for updating medication alert data based on data analysis, so as to solve the problems in the background art.
In order to solve the technical problems, the invention provides the following technical scheme: a method of data analysis-based vigilance data update for a medication comprising the steps of:
the method comprises the following steps: acquiring a personal safety report, and classifying the personal safety report according to the adverse reaction type, wherein the personal safety report comprises a determined patient, a determined reporter, a determined adverse reaction event and a determined suspected medicine;
step two: judging the effectiveness of the warning data of each medicine in the classified individual case safety reports, and eliminating invalid individual case safety reports according to the judgment result so as to improve the authority of system data;
step three: recording the drug alert data and the adverse reaction types in the individual safety reports left in the step two, and sorting the data into a data set;
step four: matching and identifying the data set obtained in the step three with a database, and updating the drug alert data in the database and the adverse reaction type corresponding to the drug alert data according to the matching and identifying result;
step five: hiding the update content of the database, identifying and judging the missing update data based on the hidden content, and finishing the comprehensive update of the database according to the identification and judgment result.
Further, the specific method for classifying the individual case safety reports according to the adverse reaction types in the step one and judging the validity of each drug warning data in the individual case safety reports after classification in the step two comprises the following steps:
step1. let adverse reaction type set X ═ poisoning, allergy, hereditary abnormal reaction, infection, mutation }, individual safety report Y ═1、Y2、…、YnRetrieving case security reports according to keywords appearing in the set, wherein n is 1, 2, 3 and 4 … and represents the number of case security reports;
step2, searching individual safety reports of various adverse reactions occurring in single drug alert data based on the retrieval result of Step1, wherein the specific method comprises the following steps:
judging whether a single drug alert data has a single safety report with various adverse reactions, wherein a specific judgment formula Q is as follows:
Q＝m+i+j+σ+κ-n；
wherein m, i, j, sigma and kappa respectively represent the number of safety reports of each case corresponding to each adverse reaction, and m, i, j, sigma,
When Q is equal to 0, the individual safety report that a plurality of adverse reactions occur to the single drug alert data does not exist, and when Q is equal to 0, the individual safety report that a plurality of adverse reactions occur to the single drug alert data exists;
step3, when Q is not equal to 0, putting each retrieval result into a set M, marking repeated events appearing in the set M, calculating the similarity between suspected medicines determined in the marked repeated events, and deleting invalid individual case safety reports of the set M;
and step4, integrating the safety reports of the individual cases after the classification processing according to the similarity calculation result in the step one (3).
Further, the specific methods for calculating the similarity between the suspected drugs determined in the marked repeated events and deleting the safety report of the invalid case in the set M in Step3 are as follows:
1) counting the suspected drugs determined in the marked repeated events, taking the determined adverse reaction events as abscissa and the determined suspected drugs in the marked repeated events as ordinate, and constructing a coordinate system;
2) calculating the similarity between the determined suspected drugs in the repeated events based on the coordinate system constructed in 1), wherein a specific similarity calculation formula S is as follows:
wherein (x)t，yt) Represents that the determined adverse reaction event in the repeated event is xtWhen the corresponding suspected drug is yt，
Indicates that the adverse reaction event determined among the repeated events is
When the corresponding suspected drug is
Respectively represent points (x)t，yt)、
The distance to the origin of the coordinates is,
respectively represent points (x)t，yt)、
The included angle between the connecting line between the Y axis and the origin of coordinates,
respectively representing the value yt、
When S is equal to 0, the suspected medicines determined in the marked repeated events are the same, indicating that the repeated events have validity, and when S is equal to 0, the suspected medicines determined in the marked repeated events are different, indicating that the repeated events do not have validity, and reclassifying the repeated events;
3) integrating all adverse reaction conditions in the repeated events with effectiveness in the step 2) into the matched suspected medicine, replacing the repeated events, deleting the marked repeated events, calculating whether the number of the safety reports of the current case is consistent with that of the safety reports of the original case, and if so, indicating that the classification is finished, wherein a specific judgment model E is as follows:
E＝m′+i′+j′+σ′+κ′+η-n；
wherein, m ', i ', j ', σ ' and κ ' respectively represent the number of individual safety reports corresponding to each adverse reaction after the duplicate event is deleted, η represents the replaced individual safety report, and when E is equal to 0, it represents that the number of individual safety reports at this time is consistent with the number of original individual safety reports, and the classification is completed.
Further, in the fourth step, the data set obtained in the third step is matched and identified with the database, and a specific method for updating the database based on the matching and identifying result is as follows:
step four (I), acquiring data sets and database data information;
step four (II), matching adverse reaction types in the data set and the data base based on the determined suspected drug type, updating the adverse reaction condition of the suspected drug when the matching identification result is inconsistent, and not updating the adverse reaction condition of the suspected drug if the matching identification result is consistent;
step four (III), the specific method for updating the adverse reaction condition of the suspected medicament in the step four (II) is as follows: if the adverse reaction of the suspected drug needs to be added, hiding the data in the database after the addition, not executing the matching identification processing in the step four (II), if the adverse reaction condition of the suspected drug needs to be modified, replacing the data in the database with the modified data, hiding the data, which is beneficial to reducing the data processing, improving the matching rate, and being convenient for quickly identifying and judging the updated missing data.
Further, the specific method for identifying and judging missing update data in the database in the fifth step is as follows:
acquiring updated hidden data in a database;
step five (II), constructing a coincidence degree calculation model, and calculating the coincidence degree of the hidden data and the data in the data set, wherein the specific coincidence degree calculation model is as follows:
representing the hidden data and the data in the data set in a coordinate form based on the coordinate system constructed in Step 3;
constructed contact ratio calculation model DlComprises the following steps:
wherein, l ═ b-a, a and b represent two adjacent adverse reactions of different types, l represents the distance between two adverse reactions, TlRepresenting the number of coordinates belonging to the data set within the distance range, f (x) representing a probability distribution function,
indicating the number of coordinates belonging to the hidden data within the distance range when DlWhen 1, it means that the hidden data and the data in the data set completely coincide, and when DlWhen the number of the hidden data is not equal to 1, indicating that the hidden data and the data in the data set are not completely overlapped, and missing update data exists at the moment;
and step five (III) judging whether the database has missing updated data or not based on the calculation result in the step five (II), and if so, updating the missing data part according to the judgment result.
A drug alert data updating system based on data analysis comprises a drug alert data validity judging module, a data integration module, a data updating module and an updating data missing judging module;
the drug warning data effectiveness judging module is used for classifying the acquired individual case security reports according to adverse reaction types, judging the effectiveness of each drug warning data in the individual case security reports according to classification processing results, eliminating invalid individual case security reports according to the judgment results, and transmitting the individual case security reports after the elimination processing to the data integration module;
the data integration module is used for receiving the personal safety report transmitted by the drug alert data effectiveness judgment module, recording the drug alert data and the adverse reaction type in the personal safety report based on the received content, sorting the data into a data set and transmitting the data set to the data updating module;
the data updating module is used for receiving the data set transmitted by the data integration module, calling the database, performing matching identification on the data set and the database, updating the drug alert data in the database and the adverse reaction type corresponding to the drug alert data according to the matching identification result, and transmitting the updated content to the updated data missing judgment module;
the update data missing judgment module is used for receiving the update content transmitted by the data update module, matching the update content with the individual case safety report obtained by the drug alert data effectiveness judgment module, and judging whether the database is updated comprehensively according to the matching result.
Further, the drug alert data validity judgment module comprises a case safety report acquisition unit, a case safety report classification processing unit, a classification result verification unit, a similarity calculation unit and a drug alert data validity judgment unit;
the personal safety report acquisition unit acquires a personal safety report drawn by medical staff, extracts adverse reaction types in the personal safety report, and transmits the acquired personal safety report and extraction information to the personal safety report classification processing unit;
the personal safety report classification processing unit receives the personal safety report and the extraction information transmitted by the personal safety report acquisition unit, preliminarily classifies the personal safety report according to the adverse reaction keywords, and transmits the preliminary classification result to the classification result verification unit;
the classification result verifying unit receives the primary classification result transmitted by the individual case safety report classification processing unit, judges whether the number of individual case safety reports obtained after classification is the same as the number of original individual case safety reports or not according to the primary classification result, transmits the primary classification result to the drug alert data validity judging unit if the individual case safety reports obtained after classification are the same as the original individual case safety reports, marks the repeated individual case safety reports if the individual case safety reports are not the same, and transmits the marked repeated events to the similarity calculating unit;
the similarity calculation unit receives the marked repeated events transmitted by the classification result verification unit, constructs a similarity calculation model, calculates the similarity between the suspected drugs determined in the marked repeated events, and transmits the calculation result to the drug alert data validity judgment unit;
the drug alert data validity judgment unit receives the preliminary classification result transmitted by the classification result verification unit and the calculation result transmitted by the similarity calculation unit, and when the received content is the preliminary classification result, directly considering that the classified individual safety report has validity, when the received content is the calculation result, if the calculation result is zero, the repeat event is considered to be effective, the adverse reaction events in the effective repeat event are all integrated into the matching suspected drug, replacing the repeated events, deleting the marked repeated events, if the calculation result is not zero, determining that the repeated events have no validity, and transmitting the repeated events to the individual case safety report classification processing unit for classification processing again, and transmitting the individual case safety reports after judgment processing to the data integration module.
Further, the data integration module receives the individual case safety report transmitted by the drug alert data validity judgment module, arranges the drug alert data and the adverse reaction types in the individual case safety report into a data set form based on the received content base, and transmits the data set to the data update module
Further, the data updating module comprises a data acquiring unit, a data matching and identifying unit and a database updating unit;
the data acquisition unit receives the data set transmitted by the data integration module, calls a drug alert data database, and transmits the data set and the database to the data matching identification unit;
the data matching and identifying unit receives the data set and the database transmitted by the data acquiring unit, matches adverse reaction types in the data set and the database based on the determined suspected drug type, and transmits a matching and identifying result to the database updating unit;
the database updating unit receives the matching identification result transmitted by the data matching identification unit, updates the adverse reaction condition of the suspected drug when the matching identification result is inconsistent, hides the data in the database after adding when the adverse reaction of the suspected drug needs to be added, replaces the data in the database with the modified data when the adverse reaction condition of the suspected drug needs to be modified, hides the modified data, does not need to update the adverse reaction condition of the suspected drug if the matching identification result is consistent, and transmits the hidden data in the database to the updated data missing judgment module.
Further, the update data loss judgment module comprises a coincidence degree calculation model construction unit and an update data loss judgment unit;
the contact ratio calculation model construction unit receives the hidden data transmitted by the database updating unit, constructs a contact ratio calculation model to calculate the contact ratio between the hidden data and the data set, and transmits the calculation result to the updating data missing judgment unit;
the update data missing judgment unit receives the calculation result transmitted by the overlap ratio calculation model construction unit, judges whether missing update data exists in the database or not based on the calculation result, and if the missing update data exists, updates the missing data part according to the judgment result.
Compared with the prior art, the invention has the following beneficial effects:
1. according to the method, the safety reports of the individual cases are classified according to the adverse reaction types, the total data of the samples before and after classification are compared, the condition that multiple adverse reactions occur in the single drug warning data in the sample data is analyzed, whether the data omission condition exists after classification is judged, the situation that the system is not comprehensive in updating is avoided, and the application range of the system is further improved.
2. The method calculates the similarity between the suspected medicines determined in the repeated events by constructing a similarity calculation model, judges whether the repeated events belong to system errors or the condition that a plurality of adverse reactions occur in single medicine alert data according to the calculation result, and deletes or replaces the repeated events based on the judgment result, so that the obtained medicine alert data has reference value, and the experience of a customer using the medicine alert data to update a system is further enhanced.
3. According to the method, the contact ratio calculation model is built, the contact ratio of the hidden data and the data in the data set is calculated, whether the missing updated data exists in the database or not is judged, manual retrieval is not needed in the process, the data processing time is shortened, and the processing efficiency is high.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
fig. 1 is a schematic workflow diagram of a data analysis-based vigilance data update system and method for pharmaceuticals according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, the present invention provides a technical solution: a method for updating vigilance data of a medication based on data analysis, the method comprising the steps of:
the method comprises the following steps: acquiring a personal safety report, and classifying the personal safety report according to the adverse reaction type, wherein the personal safety report comprises a determined patient, a determined reporter, a determined adverse reaction event and a determined suspected medicine;
step two: judging the effectiveness of the drug warning data in the classified individual case safety reports, and rejecting invalid individual case safety reports according to the judgment result;
the specific method for classifying the individual case safety report according to the adverse reaction type in the step one and judging the effectiveness of the drug warning data in the individual case safety report after classification in the step two comprises the following steps:
step1. let adverse reaction type set X ═ poisoning, allergy, hereditary abnormal reaction, infection, mutation }, individual safety report Y ═1、Y2、…、YnRetrieving case security reports according to keywords appearing in the set, wherein n is 1, 2, 3 and 4 … and represents the number of case security reports;
step2, searching individual safety reports of various adverse reactions occurring in single drug alert data based on the retrieval result of Step1, wherein the specific method comprises the following steps:
judging whether a single drug alert data has a single safety report with various adverse reactions, wherein a specific judgment formula Q is as follows:
Q＝m+i+j+σ+κ-n；
wherein m, i, j, sigma and kappa respectively represent the number of safety reports of each case corresponding to each adverse reaction, and m, i, j, sigma,
When Q is equal to 0, the individual safety report that a plurality of adverse reactions occur to the single drug alert data does not exist, and when Q is equal to 0, the individual safety report that a plurality of adverse reactions occur to the single drug alert data exists;
step3, when Q is not equal to 0, putting each retrieval result into a set M, marking repeated events appearing in the set M, calculating the similarity between suspected medicines determined in the marked repeated events, and deleting the invalid individual case safety report of the set M, wherein the specific method comprises the following steps:
1) counting the suspected drugs determined in the marked repeated events, taking the determined adverse reaction events as abscissa and the determined suspected drugs in the marked repeated events as ordinate, and constructing a coordinate system;
2) calculating the similarity between the determined suspected drugs in the repeated events based on the coordinate system constructed in 1), wherein a specific similarity calculation formula S is as follows:
wherein (x)t，yt) Represents that the determined adverse reaction event in the repeated event is xtWhen the corresponding suspected drug is yt，
Indicates that the adverse reaction event determined among the repeated events is
When the corresponding suspected drug is
Respectively represent points (x)t，yt)、
The distance to the origin of the coordinates is,
respectively represent points (x)t，yt)、
The included angle between the connecting line between the Y axis and the origin of coordinates,
respectively representing the value yt、
When S is equal to 0, the suspected medicines determined in the marked repeated events are the same, indicating that the repeated events have validity, and when S is equal to 0, the suspected medicines determined in the marked repeated events are different, indicating that the repeated events do not have validity, and reclassifying the repeated events;
3) integrating all adverse reaction conditions in the repeated events with effectiveness in the step 2) into the matched suspected medicine, replacing the repeated events, deleting the marked repeated events, calculating whether the number of the safety reports of the current case is consistent with that of the safety reports of the original case, and if so, indicating that the classification is finished, wherein a specific judgment model E is as follows:
E＝m′+i′+j′+σ′+κ′+η-n；
wherein m ', i ', j ', σ ' and κ ' respectively represent the number of individual case security reports corresponding to each adverse reaction after the duplicate event is deleted, η represents a replaced individual case security report, and when E is 0, it represents that the number of individual case security reports at this time is consistent with the number of original individual case security reports, and the classification is completed;
step4, integrating the safety reports of the individual cases after classification processing according to the similarity calculation result in the step one (3);
step three: recording the drug alert data and the adverse reaction types in the individual safety reports left in the step two, and sorting the data into a data set;
step four: matching and identifying the data set obtained in the step three with a database, and updating the drug alert data in the database and the adverse reaction type corresponding to the drug alert data according to the matching and identifying result, wherein the specific method comprises the following steps:
step four (I), acquiring data sets and database data information;
step four (II), matching adverse reaction types in the data set and the data base based on the determined suspected drug type, updating the adverse reaction condition of the suspected drug when the matching identification result is inconsistent, and not updating the adverse reaction condition of the suspected drug if the matching identification result is consistent;
step four (III), the specific method for updating the adverse reaction condition of the suspected medicament in the step four (II) is as follows: if the adverse reaction of the suspected medicine needs to be added, hiding the data in the database after the addition, not executing the matching identification processing in the step four (II), if the adverse reaction condition of the suspected medicine needs to be modified, replacing the data in the database with the modified data, hiding the data, which is beneficial to reducing the data processing, improving the matching rate, and being convenient for quickly identifying and judging the updated missing data;
step five: hiding the update content of the database, identifying and judging missing update data based on the hidden content, and finishing the overall update of the database according to the identification and judgment result, wherein the specific method comprises the following steps:
acquiring updated hidden data in a database;
step five (II), constructing a coincidence degree calculation model, and calculating the coincidence degree of the hidden data and the data in the data set, wherein the specific coincidence degree calculation model is as follows:
representing the hidden data and the data in the data set in a coordinate form based on the coordinate system constructed in Step 3;
constructed contact ratio calculation model DlComprises the following steps:
wherein, l ═ b-a, a and b represent two adjacent adverse reactions of different types, l represents the distance between two adverse reactions, TlRepresenting the number of coordinates belonging to the data set within the distance range, f (x) representing a probability distribution function,
indicating the number of coordinates belonging to the hidden data within the distance range when DlWhen 1, it means that the hidden data and the data in the data set completely coincide, and when DlWhen the number of the hidden data is not equal to 1, indicating that the hidden data and the data in the data set are not completely overlapped, and missing update data exists at the moment;
and step five (III) judging whether the database has missing updated data or not based on the calculation result in the step five (II), and if so, updating the missing data part according to the judgment result.
A kind of alert data updating system of medicine based on data analysis, including the alert data validity judgement module S1 of medicine, data integration module S2, data updating module S3 and updating the data and missing the judgement module S4;
the drug alert data validity judgment module S1 is used for classifying the acquired individual case security reports according to adverse reaction types, judging the validity of each drug alert data in the individual case security reports according to the classification processing results, eliminating invalid individual case security reports according to the judgment results, and transmitting the individual case security reports after the elimination processing to the data integration module S2; the drug alert data validity judgment module S1 includes a case security report acquisition unit S11, a case security report classification processing unit S12, a classification result verification unit S13, a similarity calculation unit S14, and a drug alert data validity judgment unit S15;
the case safety report acquiring unit S11 acquires a case safety report drawn by medical staff, extracts adverse reaction types in the case safety report, and transmits the acquired case safety report and the extracted information to the case safety report classification processing unit S12;
the personal safety report classification processing unit S12 receives the personal safety report and the extraction information transmitted by the personal safety report acquisition unit S11, preliminarily classifies the personal safety report according to the adverse reaction keyword, and transmits the preliminary classification result to the classification result verification unit S13;
the classification result verifying unit S13 receives the preliminary classification result transmitted by the case security report classification processing unit S12, determines whether the number of case security reports obtained after classification is the same as the number of original case security reports according to the preliminary classification result, transmits the preliminary classification result to the drug alert data validity determining unit S15 if the number of case security reports obtained after classification is the same as the number of original case security reports, marks recurring case security reports if the number of case security reports is not the same, and transmits the marked recurring events to the similarity calculating unit S14;
the similarity calculation unit S14 receives the labeled repeat events transmitted by the classification result verification unit S13, constructs a similarity calculation model, calculates the similarity between the suspected drugs determined in the labeled repeat events, and transmits the calculation result to the drug alert data validity judgment unit S15;
the vigilance data validity judging unit S15 receives the preliminary classification result transmitted by the classification result verifying unit S13 and the calculation result transmitted by the similarity calculating unit S14, and when the received content is the preliminary classification result, directly considering that the classified individual safety report has validity, when the received content is the calculation result, if the calculation result is zero, the repeat event is considered to be effective, the adverse reaction events in the effective repeat event are all integrated into the matching suspected drug, replacing the repeated events, deleting the marked repeated events, if the calculation result is not zero, determining that the repeated events have no validity, the repeat event is transmitted to the case security report classification processing unit S12 for classification again, and the case security report after judgment is transmitted to the data integration module S2;
the data integration module S2 is used for receiving the case safety report transmitted by the case safety data validity judgment module S1, arranging the case safety report and the drug alert data and the adverse reaction types in the case safety report into a data set form based on the received content, and transmitting the data set to the data updating module S3;
the data updating module S3 is configured to receive the data set transmitted by the data integrating module S2, invoke the database, perform matching identification on the data set and the database, update the drug alert data in the database and the adverse reaction type corresponding to the drug alert data according to the matching identification result, and transmit the updated content to the updated data missing determining module S4; the data updating module S3 includes a data obtaining unit S31, a data matching identification unit S32 and a database updating unit S33;
the data acquisition unit S31 receives the data set transmitted by the data integration module S2, invokes the medication alert data database, and transmits the data set and the database to the data matching identification unit S32;
the data matching and identifying unit S32 receives the data set and the database transmitted by the data acquiring unit S31, matches the adverse reaction types in the data set and the database based on the determined suspected drug types, and transmits the matching and identifying result to the database updating unit S33;
the database updating unit S33 receives the matching identification result transmitted by the data matching identification unit S32, when the matching identification result is inconsistent, updates the adverse reaction condition of the suspected drug, when the adverse reaction of the suspected drug needs to be added, hides the data in the database after the addition, when the adverse reaction condition of the suspected drug needs to be modified, replaces the data in the database with the modified data, hides the modified data, and if the matching identification result is consistent, does not need to update the adverse reaction condition of the suspected drug, and transmits the hidden data in the database to the updated data missing judgment module S4;
the update data missing judgment module S4 is configured to receive the update content transmitted by the data update module S3, match the update content with the individual case security report obtained by the drug alert data validity judgment module S1, and judge whether the database is updated comprehensively according to the matching result; the update data missing judgment module S4 includes a contact ratio calculation model construction unit S41 and an update data missing judgment unit S42;
the overlap ratio calculation model construction unit S41 receives the hidden data transmitted by the database updating unit S33, constructs an overlap ratio calculation model to calculate the overlap ratio between the hidden data and the data set, and transmits the calculation result to the update data loss judgment unit S42;
the update data missing determination unit S42 receives the calculation result transmitted by the overlap ratio calculation model construction unit S41, determines whether missing update data exists in the database based on the calculation result, and if so, updates the missing data portion according to the determination result.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that changes may be made in the embodiments and/or equivalents thereof without departing from the spirit and scope of the invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A method for updating drug alert data based on data analysis is characterized in that: the method comprises the following steps:
the method comprises the following steps: acquiring a personal safety report, and classifying the personal safety report according to the adverse reaction type, wherein the personal safety report comprises a determined patient, a determined reporter, a determined adverse reaction event and a determined suspected medicine;
step two: judging the effectiveness of the drug warning data in the classified individual case safety reports, and rejecting invalid individual case safety reports according to the judgment result;
step three: recording the drug alert data and the adverse reaction types in the individual safety reports left in the step two, and sorting the data into a data set;
step four: matching and identifying the data set obtained in the step three with a database, and updating the drug alert data in the database and the adverse reaction type corresponding to the drug alert data according to the matching and identifying result;
step five: hiding the update content of the database, identifying and judging the missing update data based on the hidden content, and finishing the comprehensive update of the database according to the identification and judgment result.
2. A method for updating vigilance data of a medication based on data analysis according to claim 1, wherein: the specific method for classifying the individual case safety report according to the adverse reaction type in the step one and judging the effectiveness of the drug warning data in the individual case safety report after classification in the step two comprises the following steps:
step1. let adverse reaction type set X ═ poisoning, allergy, hereditary abnormal reaction, infection, mutation }, individual safety report Y ═1、Y2、…、YnRetrieving case security reports according to keywords appearing in the set, wherein n is 1, 2, 3 and 4 … and represents the number of case security reports;
step2, searching individual safety reports of various adverse reactions occurring in single drug alert data based on the retrieval result of Step1, wherein the specific method comprises the following steps:
judging whether a single drug alert data has a single safety report with various adverse reactions, wherein a specific judgment formula Q is as follows:
Q＝m+i+j+σ+κ-n；
wherein m, i, j, sigma and kappa respectively represent the number of safety reports of each case corresponding to each adverse reaction, and m, i, j, sigma,
When Q is equal to 0, the individual safety report that a plurality of adverse reactions occur to the single drug alert data does not exist, and when Q is equal to 0, the individual safety report that a plurality of adverse reactions occur to the single drug alert data exists;
step3, when Q is not equal to 0, putting each retrieval result into a set M, marking repeated events appearing in the set M, calculating the similarity between suspected medicines determined in the marked repeated events, and deleting invalid individual case safety reports of the set M;
and step4, integrating the safety reports of the individual cases after the classification processing according to the similarity calculation result in the step one (3).
3. A method of data analysis-based vigilance data update according to claim 2, wherein: the specific methods for calculating the similarity between the suspected drugs determined in the marked repeated events and deleting the invalid individual case safety report in the set M in Step3 are as follows:
1) counting the suspected drugs determined in the marked repeated events, taking the determined adverse reaction events as abscissa and the determined suspected drugs in the marked repeated events as ordinate, and constructing a coordinate system;
2) calculating the similarity between the determined suspected drugs in the repeated events based on the coordinate system constructed in 1), wherein a specific similarity calculation formula S is as follows:
wherein (x)t，yt) Represents that the determined adverse reaction event in the repeated event is xtWhen the corresponding suspected drug is yt，
Indicates that the adverse reaction event determined among the repeated events is
When the corresponding suspected drug is
When S is equal to 0, the suspected medicines determined in the marked repeated events are the same, indicating that the repeated events have validity, and when S is equal to 0, the suspected medicines determined in the marked repeated events are different, indicating that the repeated events do not have validity, and reclassifying the repeated events;
3) integrating all adverse reaction conditions in the repeated events with effectiveness in the step 2) into the matched suspected medicine, replacing the repeated events, deleting the marked repeated events, calculating whether the number of the safety reports of the current case is consistent with that of the safety reports of the original case, and if so, indicating that the classification is finished, wherein a specific judgment model E is as follows:
E＝m′+i′+j′+σ′+κ′+η-n；
wherein, m ', i ', j ', sigma ' and kappa ' respectively represent the number of individual safety reports corresponding to each adverse reaction after the repeated events are deleted, and eta represents the replaced individual safety reports.
4. The system and method for updating vigilance data of a medication according to claim 3, wherein: in the fourth step, the data set obtained in the third step is matched and identified with the database, and the specific method for updating the database based on the matching and identifying result comprises the following steps:
step four (I), acquiring data sets and database data information;
step four (II), matching adverse reaction types in the data set and the data base based on the determined suspected drug type, updating the adverse reaction condition of the suspected drug when the matching identification result is inconsistent, and not updating the adverse reaction condition of the suspected drug if the matching identification result is consistent;
step four (III), the specific method for updating the adverse reaction condition of the suspected medicament in the step four (II) is as follows: and if the adverse reaction condition of the suspected drug needs to be modified, replacing the data in the database with modified data, and hiding the modified data.
5. A data analysis-based vigilance data update method according to claim 4, wherein: the concrete method for identifying and judging missing update data in the database in the fifth step is as follows:
acquiring updated hidden data in a database;
step five (II), constructing a coincidence degree calculation model, and calculating the coincidence degree of the hidden data and the data in the data set, wherein the specific coincidence degree calculation model is as follows:
representing the hidden data and the data in the data set in a coordinate form based on the coordinate system constructed in Step 3;
constructed contact ratio calculation model DlComprises the following steps:
wherein, l ═ b-a,a. b represents two adjacent adverse reactions of different types, l represents the distance between two adverse reactions, and TlRepresenting the number of coordinates belonging to the data set within the distance range, f (x) representing a probability distribution function,
indicating the number of coordinates belonging to the hidden data within the distance range;
and step five (III) judging whether the database has missing updated data or not based on the calculation result in the step five (II), and if so, updating the missing data part according to the judgment result.
6. A medication alert data update system based on data analysis, characterized by: the system comprises a drug alert data validity judging module (S1), a data integration module (S2), a data updating module (S3) and an update data missing judging module (S4);
the drug alert data validity judging module (S1) is used for classifying the acquired individual case security reports according to adverse reaction types, judging the validity of each drug alert data in the individual case security reports according to the classification processing results, eliminating invalid individual case security reports according to the judgment results, and transmitting the individual case security reports after the elimination processing to the data integration module (S2);
the data integration module (S2) is used for receiving the case safety report transmitted by the case safety data validity judgment module (S1), recording the case safety data and the adverse reaction type in the case safety report based on the received content, sorting the case safety report into a data set, and transmitting the data set to the data updating module (S3);
the data updating module (S3) is used for receiving the data set transmitted by the data integration module (S2), calling the database, performing matching identification on the data set and the database, updating the drug alert data in the database and the adverse reaction type corresponding to the drug alert data according to the matching identification result, and transmitting the updated content to the updated data missing judgment module (S4);
the update data missing judgment module (S4) is configured to receive the update content transmitted by the data update module (S3), match the update content with the case security report obtained by the drug alert data validity judgment module (S1), and judge whether the database is updated completely according to a matching result.
7. A data analysis-based vigilance data update system according to claim 6, wherein: the drug alert data validity judgment module (S1) comprises an individual case safety report acquisition unit (S11), an individual case safety report classification processing unit (S12), a classification result verification unit (S13), a similarity calculation unit (S14) and a drug alert data validity judgment unit (S15);
the case safety report acquiring unit (S11) acquires a case safety report drawn by medical staff, extracts adverse reaction types in the case safety report, and transmits the acquired case safety report and the extracted information to the case safety report classification processing unit (S12);
the personal safety report classification processing unit (S12) receives the personal safety report and the extraction information transmitted by the personal safety report acquisition unit (S11), preliminarily classifies the personal safety report according to the adverse reaction keyword, and transmits the preliminary classification result to the classification result verification unit (S13);
the classification result verifying unit (S13) receives the primary classification result transmitted by the individual case safety report classification processing unit (S12), judges whether the number of individual case safety reports obtained after classification is the same as the number of original individual case safety reports or not according to the primary classification result, transmits the primary classification result to the drug alert data validity judging unit (S15) if the individual case safety reports obtained after classification are the same as the original individual case safety reports, marks the repeated individual case safety reports if the individual case safety reports are not the same, and transmits the marked repeated events to the similarity calculating unit (S14);
the similarity calculation unit (S14) receives the marked recurring events transmitted by the classification result verification unit (S13), constructs a similarity calculation model, calculates the similarity between the suspected drugs determined in the marked recurring events, and transmits the calculation result to the drug alert data validity judgment unit (S15);
the medication alert data validity judgment unit (S15) receives the preliminary classification result transmitted by the classification result verification unit (S13) and the calculation result transmitted by the similarity calculation unit (S14), and when the received content is the preliminary classification result, directly considering that the classified individual safety report has validity, when the received content is the calculation result, if the calculation result is zero, the repeat event is considered to be effective, the adverse reaction events in the effective repeat event are all integrated into the matching suspected drug, replacing the repeated events, deleting the marked repeated events, if the calculation result is not zero, determining that the repeated events have no validity, the duplicate event is transmitted to the individual security report classification processing unit (S12) to be subjected to classification processing again, and transmits the individual security report after the judgment processing to the data integration module (S2).
8. A data analysis-based vigilance data update system according to claim 7, wherein: the data integration module (S2) receives the case security report transmitted by the case security data validity judgment module (S1), collates the case security report with the drug alert data and the adverse reaction type into a data set form based on the received content, and transmits the data set to the data update module (S3).
9. A data analysis-based vigilance data update system according to claim 8, wherein: the data update module (S3) includes a data acquisition unit (S31), a data matching identification unit (S32), and a database update unit (S33);
the data acquisition unit (S31) receiving the data set transmitted by the data integration module (S2), invoking a medication alert data database, and transmitting the data set and database to the data match identification unit (S32);
the data matching identification unit (S32) receives the data set and the database transmitted by the data acquisition unit (S31), matches the adverse reaction types in the data set and the database based on the determined suspected drug type, and transmits the matching identification result to the database updating unit (S33);
the database updating unit (S33) receives the matching identification result transmitted by the data matching identification unit (S32), when the matching identification result is inconsistent, the adverse reaction condition of the suspected drug is updated, when the adverse reaction of the suspected drug needs to be added, the data in the database is hidden after the addition, when the adverse reaction condition of the suspected drug needs to be modified, the data in the database is replaced by the modified data, the modified data is hidden, if the matching identification result is consistent, the adverse reaction condition of the suspected drug does not need to be updated, and the hidden data in the database is transmitted to the updating data missing judgment module (S4).
10. A data analysis-based vigilance data update system according to claim 9, wherein: the update data missing judgment module (S4) comprises a coincidence degree calculation model construction unit (S41) and an update data missing judgment unit (S42);
the coincidence degree calculation model building unit (S41) receives the hidden data transmitted by the database updating unit (S33), builds a coincidence degree calculation model to calculate the coincidence degree between the hidden data and the data set, and transmits the calculation result to the updating data missing judgment unit (S42);
the update data missing judgment unit (S42) receives the calculation result transmitted by the overlap ratio calculation model construction unit (S41), judges whether missing update data exists in the database based on the calculation result, and if so, updates the missing data part according to the judgment result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210027905.2A CN114360742A (en) | 2022-01-11 | 2022-01-11 | Medicine warning data updating system and method based on data analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210027905.2A CN114360742A (en) | 2022-01-11 | 2022-01-11 | Medicine warning data updating system and method based on data analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114360742A true CN114360742A (en) | 2022-04-15 |
Family
ID=81109571
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210027905.2A Pending CN114360742A (en) | 2022-01-11 | 2022-01-11 | Medicine warning data updating system and method based on data analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114360742A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115840915A (en) * | 2022-11-22 | 2023-03-24 | 广州城轨科技有限公司 | Automatic identification method, system, terminal and storage medium for electromechanical equipment fault |
| CN116153462A (en) * | 2023-04-20 | 2023-05-23 | 南京引光医药科技有限公司 | Drug alert system and drug alert feedback data processing method |
-
2022
- 2022-01-11 CN CN202210027905.2A patent/CN114360742A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115840915A (en) * | 2022-11-22 | 2023-03-24 | 广州城轨科技有限公司 | Automatic identification method, system, terminal and storage medium for electromechanical equipment fault |
| CN116153462A (en) * | 2023-04-20 | 2023-05-23 | 南京引光医药科技有限公司 | Drug alert system and drug alert feedback data processing method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114360742A (en) | Medicine warning data updating system and method based on data analysis | |
| US9792324B2 (en) | Method and system for uniquely identifying a person to the exclusion of all others | |
| NL2012435C2 (en) | Data processing techniques. | |
| US7970759B2 (en) | System and method for deriving a hierarchical event based database optimized for pharmaceutical analysis | |
| CN113159502B (en) | Method for assessing risk of clinical trial | |
| GB2514239A (en) | Data processing techniques | |
| CN110162975A (en) | A kind of multistep abnormal point detecting method based on neighbour's propagation clustering algorithm | |
| CN111598753A (en) | Suspect recommendation method and device, electronic equipment and storage medium | |
| CN111883253A (en) | Disease data analysis method and lung cancer risk prediction system based on medical knowledge base | |
| CN115274122A (en) | Health medical data management method, system, electronic device and storage medium | |
| CN115938608A (en) | Clinical decision early warning method and system based on prompt learning model | |
| CN111784453A (en) | Block chain-based cross-platform medicine collection price synchronization method and related device | |
| Ng et al. | Detecting non-compliant consumers in spatio-temporal health data: A case study from Medicare Australia | |
| Hill et al. | Anonymous record linkage of census and mortality records: 1981, 1986, 1991, 1996 census cohorts | |
| CN111986819B (en) | Adverse drug reaction monitoring method and device, electronic equipment and readable storage medium | |
| CN118016309A (en) | Optimal clinical treatment path recommending method and system | |
| Phillips et al. | A rule and graph-based approach for targeted identity resolution on policing data | |
| CN115662656B (en) | Evaluation method and system for side effects of medicine and electronic equipment | |
| CN110032607A (en) | A kind of auditing method based on big data | |
| CN111966838B (en) | Adverse event and medication relevance judging method and system | |
| Ferrante | Developing an offender-based tracking system: The Western Australia INOIS project | |
| Bianchi Santiago et al. | Record linkage of crashes with injuries and medical cost in Puerto Rico | |
| CN111986815A (en) | Project combination mining method based on co-occurrence relation and related equipment | |
| AU2020202131A1 (en) | Fraud detection in healthcare | |
| Arellano et al. | A probabilistic approach to the patient identification problem |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |