A data integration analysis management platform

Abstract

The application discloses a data integration analysis management platform, relates to the technical field of medical data analysis and processing, and has the technical scheme as follows: a multi-source data acquisition module acquires basic medical data and epidemic detection data; a data cleaning and fusion module acquires the basic medical data and the epidemic detection data and outputs standardized medical data and standardized epidemic data; a data standardization processing module acquires the standardized medical data and generates unified format medical data and a historical medicine consumption data set; and an intelligent decision analysis module outputs a medicine demand prediction index and a stock early warning index based on the unified format medical data and the historical medicine consumption data set. The data cleaning and fusion module can greatly reduce data noise and significantly improve data quality; and the intelligent decision analysis module can predict demand fluctuation in advance, actively prompt replenishment or clearance of accumulated goods, effectively reduce the medicine shortage rate and the expired and scrapped rate according to actual seasonal demand.

Description

Technical Field

[0001] This invention relates to the field of medical data analysis and processing technology, and more specifically, to a data integration, analysis and management platform. Background Technology

[0002] With the rapid development of precision medicine and digital clinical trials, the sources of data in the pharmaceutical field are becoming increasingly rich and complex, and medical institutions at all levels, pharmaceutical companies, and drug regulatory authorities have accumulated a massive amount of pharmaceutical data resources.

[0003] Existing pharmaceutical data typically originates from multiple different systems, exhibiting significant differences in format, standards, and semantics. This data suffers from severe data silos, inconsistent coding systems, and difficulties in cross-system data correlation, making traditional data integration techniques ineffective. Furthermore, existing clinical data generally suffers from redundancy, missing information, logical contradictions, and inconsistent value ranges, leading to numerous errors and duplicate information. In addition, current drug management relies heavily on periodic inventory checks, failing to dynamically predict demand based on historical consumption patterns, seasonal fluctuations, and epidemic trends. During peak influenza seasons, demand may surge, resulting in drug shortages.

[0004] Therefore, a new solution is needed to address this problem. Summary of the Invention

[0005] The purpose of this invention is to provide a data integration, analysis and management platform to solve the above-mentioned problems.

[0006] The above-mentioned technical objective of this invention is achieved through the following technical solution: a data integration, analysis and management platform, comprising the following steps:

[0007] By connecting to hospital information systems, laboratory information systems, drug management systems, medical insurance settlement systems, electronic medical record systems, and disease prevention and control information systems through multi-source data acquisition modules, basic medical data and epidemiological detection data can be obtained.

[0008] The data cleaning and fusion module acquires the basic medical data and epidemiological detection data and performs outlier detection and repair, missing value imputation, duplicate record deduplication, and data consistency verification, and outputs standardized medical data and standardized epidemiological data.

[0009] The data standardization processing module acquires the standardized medical data and performs field mapping, unit normalization, and encoding unification on the standardized medical data to generate unified format medical data. The data standardization processing module extracts historical consumption records of drugs from the standardized medical data to form a historical drug consumption dataset.

[0010] The intelligent decision analysis module, based on the unified format pharmaceutical data and historical drug consumption dataset, uses a multi-dimensional analysis engine to calculate and output the drug demand forecast index and inventory warning index.

[0011] The visualization and early warning module acquires the drug demand forecast index, inventory early warning index, and prescription rationality score, generates a visualization dashboard, and outputs early warning notifications under preset trigger conditions.

[0012] The present invention is further configured such that: the multi-source data acquisition module includes a data interface adapter, a real-time data stream monitoring unit, and a batch data import unit;

[0013] The data interface adapter is used to connect to the application programming interfaces of hospital information systems, laboratory information systems, drug management systems, medical insurance settlement systems, electronic medical record systems, and disease prevention and control information systems.

[0014] The real-time data stream monitoring unit uses a change data capture mechanism to collect incremental basic medical data and real-time epidemic monitoring data in real time.

[0015] The batch data import unit is used to import historical basic medical data and historical epidemic monitoring data in batches using a scheduled task.

[0016] The disease prevention and control information system provides quantitative indicators to generate an activity intensity index of epidemic pathogens, with a value range of 0 to 100.

[0017] The present invention is further configured such that: the data cleaning and fusion module includes an outlier detection unit, a missing value imputation unit, a duplicate record processing unit, and a data consistency verification unit;

[0018] The outlier detection unit uses the three sigma criterion to identify outliers in the basic medical data and epidemiological detection data that exceed three times the mean standard deviation and corrects or removes them based on domain knowledge.

[0019] The missing value imputation unit uses the mode imputation method for missing values ​​of categorical variables in the basic medical data and epidemiological detection data, and uses the K nearest neighbor imputation method for missing values ​​of continuous variables.

[0020] The duplicate record processing unit uses a similarity matching algorithm to identify duplicate records of the same entity and retains the record with the highest completeness for merging;

[0021] The data consistency verification unit performs logical consistency verification on basic medical data and epidemic detection data across systems based on a predefined business rule base. When the basic medical data and epidemic detection data do not conform to the business rules, an automatic correction or manual review process is triggered. Combined with the outlier detection unit, missing value imputation unit, and duplicate record processing unit, standardized medical data and standardized epidemic data are output.

[0022] The present invention is further configured such that: the data standardization processing module includes a field mapping unit, a unit normalization unit, a code unification unit, and a historical data extraction unit; the unified format medical data includes a unified target field, a unified unit of measurement, and an international standard code.

[0023] The field mapping unit maps different field names in standardized medical data to a unified target field based on a pre-configured source-target field mapping table;

[0024] The unit normalization unit converts drug measurement units from different sources in standardized pharmaceutical data into a unified measurement unit;

[0025] The coding unification unit maps the coding systems of various systems in standardized medical data to international standard coding.

[0026] The historical data extraction unit summarizes and statistically analyzes the daily consumption of each drug from standardized medical data according to drug name and date, and stores it in chronological order as the historical drug consumption dataset.

[0027] The present invention is further configured such that: the intelligent decision analysis module includes a drug demand forecasting unit and an inventory early warning unit;

[0028] The drug demand forecasting unit obtains the historical drug consumption dataset from the data standardization processing module, and obtains the average daily consumption of drugs based on the historical drug consumption data. It also obtains the activity intensity index of the epidemic pathogens in the standardized epidemiological data from the data cleaning and fusion module, and calculates the proportion of each drug's consumption to the total annual consumption by month based on the historical drug consumption dataset. The normalized proportion is then used as a seasonality coefficient.

[0029] The inventory early warning unit obtains the current drug inventory C1 in real time from the drug management system accessed by the multi-source data acquisition module, obtains the daily drug consumption rate from the historical drug consumption dataset, and obtains drug expiration date data from the data standardization processing module.

[0030] The present invention is further configured such that: the drug demand forecasting unit calculates and outputs the drug demand forecasting index (PDI), and the calculation formula for PDI is:

[0031] PDI=β1×(D1 / D2)+β2×S+β3×E

[0032] Wherein, D1 represents the average daily consumption of the target drug over the past 30 days; D2 represents the average daily consumption of similar drugs throughout the year, and both D1 and D2 are calculated from historical drug consumption datasets; S represents the seasonality coefficient; E represents the epidemic trend coefficient, which is obtained by dividing the activity intensity index of the epidemic pathogen by 100; β1, β2, and β3 are preset weighting coefficients, and β1+β2+β3=1;

[0033] When the PDI is greater than or equal to 1.2, the drug demand forecasting unit outputs a replenishment suggestion while outputting the PDI; when the PDI is less than or equal to 0.6, the drug demand forecasting unit outputs an inventory backlog warning while outputting the PDI; when the PDI is between 0.6 and 1.2, the drug demand forecasting unit outputs the PDI normally.

[0034] The present invention is further configured such that the calculation formula for the inventory early warning index IWI is:

[0035] IWI = γ1×(C1 / C2)+γ2×(T1 / T2)

[0036] Wherein, C2 represents the safety stock threshold, and the inventory early warning unit also dynamically adjusts the safety stock threshold based on the PDI output by the drug demand forecasting unit; T1 represents the remaining days until the expiration date of the drug, which is calculated by the inventory early warning unit based on the difference between the drug expiration date data and the current system time; T2 represents the safe shelf life days; γ1 and γ2 are preset weighting coefficients, and γ1 + γ2 = 1; when IWI is greater than or equal to 1.5, the inventory early warning unit outputs IWI and outputs a first-level early warning notification, and outputs an inventory backlog prompt; when IWI is less than or equal to 0.3, the inventory early warning unit outputs IWI and outputs a second-level early warning notification, and the drug demand forecasting unit outputs a replenishment suggestion; when IWI is between 0.3 and 1.5, the inventory early warning unit outputs a normal IWI.

[0037] The formula for calculating C2 is:

[0038] C2=T3×D1×(1+k×(PDI-1))

[0039] Wherein, T3 represents the inventory early warning unit based on the drug procurement cycle; k is a preset adjustment coefficient with a value range of 0.2-0.5. According to the preset drug demand, when PDI>1, the demand is higher than normal, and k is adjusted upward; when PDI<1, the demand is lower than normal, and k is adjusted downward.

[0040] The present invention is further configured such that: the visualization and early warning module includes a real-time data dashboard unit and an early warning push unit;

[0041] The real-time data dashboard unit dynamically displays the real-time values ​​and historical trend curves of the drug demand forecast index and inventory warning index in the form of a dashboard.

[0042] The warning push unit pushes warning notifications to preset recipients via SMS, email, or in-app messages under preset trigger conditions.

[0043] The present invention is further configured such that: the computer-readable storage medium stores a computer program, which, when executed by a computer or processor, implements any of the methods described above.

[0044] The present invention is further configured such that: the computer program product includes a computer program, which, when executed by a computer or processor, causes the computer or processor to perform the method described in any of the above-mentioned embodiments.

[0045] In summary, the present invention has the following beneficial effects:

[0046] By setting up a multi-source data acquisition module, it is possible to simultaneously access multiple core data sources such as hospital information systems, laboratory information systems, drug management systems, medical insurance settlement systems, electronic medical record systems, and disease prevention and control information systems. Through the data standardization processing module, the information collected from each system is mapped, unitized, and coded in a unified manner to generate unified format medical data, laying the foundation for subsequent intelligent analysis.

[0047] The data cleaning and fusion module detects outliers, fills in missing values, filters duplicate records, and verifies medical data. It performs comprehensive cleaning of basic medical data and epidemiological detection data, significantly reducing data noise and improving data quality.

[0048] Through the intelligent decision analysis module, combined with unified format pharmaceutical data and historical drug consumption datasets, and integrating industry average consumption data, seasonality coefficients, and epidemic trend coefficients, a multi-dimensional analysis engine is used to calculate and output drug demand forecast index and inventory warning index. It outputs replenishment or stockpiling suggestions and displays and issues warnings through visualization and warning modules. This allows for the early detection of demand fluctuations based on actual seasonal demand, proactively prompting replenishment or clearing of stockpiles, and effectively reducing drug shortage rates and expiration rates. Attached Figure Description

[0049] Figure 1 This is a flowchart illustrating a data integration, analysis, and management platform according to the present invention. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] In one possible embodiment, please refer to Figure 1 As shown, a data integration, analysis, and management platform includes the following steps:

[0052] By connecting to hospital information systems, laboratory information systems, drug management systems, medical insurance settlement systems, electronic medical record systems, and disease prevention and control information systems through multi-source data acquisition modules, basic medical data and epidemiological detection data can be obtained.

[0053] The data cleaning and fusion module acquires basic medical data and epidemiological detection data, performs outlier detection and repair, missing value imputation, duplicate record deduplication, and data consistency verification, and outputs standardized medical data and standardized epidemiological data.

[0054] The data standardization processing module acquires standardized medical data and performs field mapping, unit normalization, and coding unification on the standardized medical data to generate medical data in a unified format. The data standardization processing module extracts historical consumption records of drugs from the standardized medical data to form a historical drug consumption dataset.

[0055] The intelligent decision analysis module is based on unified format pharmaceutical data and historical drug consumption datasets, and uses a multi-dimensional analysis engine to calculate and output drug demand forecast index and inventory early warning index.

[0056] The visualization and early warning module obtains the drug demand forecast index, inventory early warning index, and prescription rationality score, generates a visualization dashboard, and outputs early warning notifications under preset trigger conditions.

[0057] For details, please refer to Figure 1 As shown, by setting up a multi-source data acquisition module, it is possible to simultaneously access multiple core data sources such as hospital information systems, laboratory information systems, drug management systems, medical insurance settlement systems, electronic medical record systems, and disease prevention and control information systems. Through the data standardization processing module, the information collected from each system is mapped, unitized, and coded in a unified manner to generate unified format medical data, laying the foundation for subsequent intelligent analysis.

[0058] The data cleaning and fusion module detects outliers, fills in missing values, filters duplicate records, and verifies medical data. It performs comprehensive cleaning of basic medical data and epidemiological detection data, significantly reducing data noise and improving data quality.

[0059] Through the intelligent decision analysis module, combined with unified format pharmaceutical data and historical drug consumption datasets, and integrating industry average consumption data, seasonality coefficients, and epidemic trend coefficients, a multi-dimensional analysis engine is used to calculate and output drug demand forecast index and inventory warning index. It outputs replenishment or stockpiling suggestions and displays and issues warnings through visualization and warning modules. This allows for the early detection of demand fluctuations based on actual seasonal demand, proactively prompting replenishment or clearing of stockpiles, and effectively reducing drug shortage rates and expiration rates.

[0060] For further details, please refer to Figure 1 As shown, the multi-source data acquisition module includes a data interface adapter, a real-time data stream monitoring unit, and a batch data import unit. Through the data interface adapter, the real-time data stream monitoring unit, and the batch data import unit, data from various systems can be imported from multiple sources.

[0061] The data interface adapter is used to connect to the application programming interface of hospital information systems, laboratory information systems, drug management systems, medical insurance settlement systems, electronic medical record systems, and disease prevention and control information systems.

[0062] The real-time data stream monitoring unit uses a change data capture mechanism to collect incremental basic medical data and real-time epidemic monitoring data in real time;

[0063] The batch data import unit is used to import historical basic medical data and historical epidemic monitoring data in batches via scheduled tasks.

[0064] The disease prevention and control information system provides quantitative indicators to generate an activity intensity index of epidemic pathogens, with a value range of 0 to 100.

[0065] For further details, please refer to Figure 1 As shown, the data cleaning and fusion module includes an outlier detection unit, a missing value imputation unit, a duplicate record processing unit, and a data consistency verification unit. The raw data of basic medical data and epidemiological detection data collected by the multi-source data acquisition module often have problems such as missing, duplicate, errors, and inconsistent formats. These dirty data will seriously affect the accuracy of subsequent analysis results. The data cleaning and fusion platform can clean these raw data to facilitate subsequent analysis and processing.

[0066] The outlier detection unit uses the three sigma criterion to identify outliers in basic medical data and epidemiological detection data that exceed three standard deviations above or below the mean, and corrects or removes them based on domain knowledge to reduce the interference of outliers.

[0067] The missing value imputation unit uses the mode imputation method for missing values ​​of categorical variables in basic medical data and epidemiological detection data, and the K nearest neighbor imputation method for missing values ​​of continuous variables. By using different imputation methods to fill in missing values, the coherence of the data is increased.

[0068] The duplicate record processing unit uses a similarity matching algorithm to identify duplicate records of the same entity and retains the record with the highest completeness for merging;

[0069] The data consistency verification unit performs logical consistency verification on basic medical data and epidemic detection data across systems based on a predefined business rule base. When the basic medical data and epidemic detection data do not conform to the business rules, it triggers an automatic correction or manual review process. Combined with the outlier detection unit, missing value imputation unit, and duplicate record processing unit, it outputs standardized medical data and standardized epidemic data.

[0070] For further details, please refer to Figure 1 As shown, the data standardization processing module includes a field mapping unit, a unit normalization unit, a coding unification unit, and a historical data extraction unit. The unified format of medical data includes unified target fields, unified units of measurement, and international standard codes. The data standardization processing module can effectively solve the pain points of the existing medical data coding system being inconsistent and cross-system data being difficult to correlate. By standardizing the medical data extracted from various systems, the medical data extracted from various systems can achieve the same coding and realize the correlation of cross-system data.

[0071] The field mapping unit maps different field names in standardized medical data to a unified target field based on a pre-configured source-target field mapping table;

[0072] Unit normalization converts drug measurement units from different sources in standardized pharmaceutical data into a unified measurement unit;

[0073] The coding standardization unit maps the coding systems of various systems in standardized medical data to international standard coding.

[0074] The historical data extraction unit summarizes and statistically analyzes the daily consumption of each drug from standardized medical data according to drug name and date, and stores it in chronological order as a historical drug consumption dataset.

[0075] For further details, please refer to Figure 1 As shown, the intelligent decision analysis module includes a drug demand forecasting unit and an inventory early warning unit;

[0076] The drug demand forecasting unit obtains historical drug consumption datasets from the data standardization processing module, and obtains the average daily consumption of drugs based on the historical drug consumption data. It also obtains the activity intensity index of epidemic pathogens from the standardized epidemiological data from the data cleaning and fusion module, and calculates the proportion of each drug's consumption to the total annual consumption by month based on the historical drug consumption dataset. This proportion is then normalized and used as a seasonality coefficient.

[0077] The drug demand forecasting unit calculates and outputs the Drug Demand Forecast Index (PDI). The formula for calculating the PDI is:

[0078] PDI=β1×(D1 / D2)+β2×S+β3×E

[0079] Wherein, D1 represents the average daily consumption of the target drug over the past 30 days; D2 represents the average daily consumption of the same type of drug throughout the year, and both D1 and D2 are calculated from historical drug consumption datasets; S represents the seasonality coefficient; E represents the epidemic trend coefficient, which is obtained by dividing the activity intensity index of the epidemic pathogen by 100; β1, β2, and β3 are preset weight coefficients, and β1+β2+β3=1;

[0080] When PDI is greater than or equal to 1.2, the drug demand forecasting unit outputs PDI and replenishment suggestions at the same time; when PDI is less than or equal to 0.6, the drug demand forecasting unit outputs PDI and inventory backlog warnings at the same time; when PDI is between 0.6 and 1.2, the drug demand forecasting unit outputs PDI normally.

[0081] Specifically, taking antiviral drugs for influenza as an example, statistics show that the total export volume of drug A in a certain month when influenza occurred was 4500 boxes, and the average daily consumption of A was D1, which equals 150. From the historical drug consumption dataset, the average daily consumption of drug A throughout the year, D2, is 80. The proportion of drug A consumption in that month to the total annual consumption in the hospital over the past three years was 18.4%, 19.5%, and 22.1%, respectively. Taking the average of 20%, and normalizing it (using the 12-month average as 1, the actual monthly proportion should be 1 / 12, or 0.0833), S... =0.2 / 0.0833=2.4, S greater than 1 indicates that the demand in this month is significantly higher than the monthly average. The epidemic pathogen activity intensity index provided by the disease prevention and control information system is 82, so E equals 0.82. Based on the influence of drug A in this month, output β1=0.3, β2=0.4, β3=0.3, then PDI=0.3×(150 / 80)+0.4×2.4+0.3×0.82=1.7685. So the current PDI is much greater than 1.2. The drug demand forecasting unit outputs replenishment suggestions.

[0082] For further details, please refer to Figure 1 As shown, the inventory early warning unit obtains the current drug inventory C1 in real time from the drug management system accessed by the multi-source data acquisition module, obtains the daily drug consumption rate from the historical drug consumption dataset, and obtains the drug expiration date data from the data standardization processing module.

[0083] The formula for calculating the Inventory Warning Index (IWI) is as follows:

[0084] IWI = γ1×(C1 / C2)+γ2×(T1 / T2)

[0085] Wherein, C2 represents the safety stock threshold, and the inventory warning unit also dynamically adjusts the safety stock threshold based on the PDI output by the drug demand forecasting unit; T1 represents the number of days remaining until the expiration date of the drug, which is calculated by the inventory warning unit based on the difference between the drug expiration date data and the current system time; T2 represents the number of days of safe shelf life; γ1 and γ2 are preset weighting coefficients, and γ1 + γ2 = 1; when IWI is greater than or equal to 1.5, the inventory warning unit outputs IWI and outputs a first-level warning notification, and outputs an inventory backlog prompt; when IWI is less than or equal to 0.3, the inventory warning unit outputs IWI and outputs a second-level warning notification, and the drug demand forecasting unit outputs a replenishment suggestion; when IWI is between 0.3 and 1.5, the inventory warning unit outputs a normal IWI;

[0086] The formula for calculating C2 is:

[0087] C2=T3×D1×(1+k×(PDI-1))

[0088] Wherein, T3 represents the inventory early warning unit based on the drug procurement cycle; k is a preset adjustment coefficient with a value range of 0.2-0.5. According to the preset drug demand, when PDI>1, the demand is higher than normal, and k is adjusted upward; when PDI<1, the demand is lower than normal, and k is adjusted downward.

[0089] Specifically, the procurement cycle T3 for drug A is 4 days. At this time, PDI is much greater than 1, and k equals 0.5. Then C2 = 4 × 150 × (1 + 0.5 × (1.7685 - 1)) = 830.55, which is rounded up to 831 boxes. The current inventory of drug A in the hospital is C1, which is 600 boxes. The remaining days T1 of drug A until its expiration date is 180, and the safe shelf life of drug A is T2, which is 210. With the preset γ1 = 0.6 and γ2 = 0.4, the current inventory warning index IWI is calculated as 0.6 × (600 / 831) + 0.4 × (180 / 210) = 0.776. Therefore, the current inventory warning index IWI is within the normal range.

[0090] For further details, please refer to Figure 1As shown, the visualization and early warning module includes a real-time data dashboard unit and an early warning push unit;

[0091] The real-time data dashboard unit dynamically displays the real-time values ​​and historical trend curves of the drug demand forecast index and inventory warning index in the form of a dashboard. This helps procurement personnel to observe the current drug inventory and usage demand, and to consider whether to purchase based on the historical trend curve. For example, the current inventory of drug A is slightly low but still within a controllable range. However, considering the strong demand, a replenishment suggestion is still output.

[0092] The early warning push unit sends early warning notifications to preset recipients via SMS, email, or in-app messages under preset trigger conditions, enabling procurement personnel to receive the messages in a timely manner and make drug purchases.

[0093] This application also provides a computer-readable storage medium, which can be any available medium that a computing device can store or a data storage device such as a data center that contains one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive). The computer-readable storage medium contains instructions that instruct the computing device to perform the aforementioned time synchronization method.

[0094] This application also provides a computer program product containing instructions. The computer program product may be a software or program product containing instructions that can run on a computing device or be stored in any available medium. When the computer program product runs on a computer device, it causes the computing device to perform the aforementioned time synchronization method.

[0095] The preferred embodiments of the present invention disclosed above are only for illustrating the present invention. These preferred embodiments do not describe all details exhaustively, nor do they limit the invention to the specific implementations described. Obviously, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention.

Claims

1. A data integration, analysis, and management platform, characterized in that, Includes the following steps: By connecting to hospital information systems, laboratory information systems, drug management systems, medical insurance settlement systems, electronic medical record systems, and disease prevention and control information systems through multi-source data acquisition modules, basic medical data and epidemiological detection data can be obtained. The data cleaning and fusion module acquires the basic medical data and epidemiological detection data and performs outlier detection and repair, missing value imputation, duplicate record deduplication, and data consistency verification, and outputs standardized medical data and standardized epidemiological data. The data standardization processing module acquires the standardized medical data and performs field mapping, unit normalization, and encoding unification on the standardized medical data to generate unified format medical data. The data standardization processing module extracts historical consumption records of drugs from the standardized medical data to form a historical drug consumption dataset. The intelligent decision analysis module, based on the unified format pharmaceutical data and historical drug consumption dataset, uses a multi-dimensional analysis engine to calculate and output the drug demand forecast index and inventory warning index. The visualization and early warning module acquires the drug demand forecast index, inventory early warning index, and prescription rationality score, generates a visualization dashboard, and outputs early warning notifications under preset trigger conditions.

2. The data integration, analysis, and management platform according to claim 1, characterized in that: The multi-source data acquisition module includes a data interface adapter, a real-time data stream monitoring unit, and a batch data import unit; The data interface adapter is used to connect to the application programming interfaces of hospital information systems, laboratory information systems, drug management systems, medical insurance settlement systems, electronic medical record systems, and disease prevention and control information systems. The real-time data stream monitoring unit uses a change data capture mechanism to collect incremental basic medical data and real-time epidemic monitoring data in real time. The batch data import unit is used to import historical basic medical data and historical epidemic monitoring data in batches using a scheduled task. The disease prevention and control information system provides quantitative indicators to generate an activity intensity index of epidemic pathogens, with a value range of 0 to 100.

3. The data integration, analysis, and management platform according to claim 1, characterized in that: The data cleaning and fusion module includes an outlier detection unit, a missing value imputation unit, a duplicate record processing unit, and a data consistency verification unit. The outlier detection unit uses the three sigma criterion to identify outliers in the basic medical data and epidemiological detection data that exceed three times the mean standard deviation and corrects or removes them based on domain knowledge. The missing value imputation unit uses the mode imputation method for missing values ​​of categorical variables in the basic medical data and epidemiological detection data, and uses the K nearest neighbor imputation method for missing values ​​of continuous variables. The duplicate record processing unit uses a similarity matching algorithm to identify duplicate records of the same entity and retains the record with the highest completeness for merging; The data consistency verification unit performs logical consistency verification on basic medical data and epidemic detection data across systems based on a predefined business rule base. When the basic medical data and epidemic detection data do not conform to the business rules, an automatic correction or manual review process is triggered. Combined with the outlier detection unit, missing value imputation unit, and duplicate record processing unit, standardized medical data and standardized epidemic data are output.

4. The data integration, analysis, and management platform according to claim 1, characterized in that: The data standardization processing module includes a field mapping unit, a unit normalization unit, a coding unification unit, and a historical data extraction unit. The unified format medical data includes unified target fields, unified units of measurement, and international standard coding. The field mapping unit maps different field names in standardized medical data to a unified target field based on a pre-configured source-target field mapping table; The unit normalization unit converts drug measurement units from different sources in standardized pharmaceutical data into a unified measurement unit; The coding unification unit maps the coding systems of various systems in standardized medical data to international standard coding. The historical data extraction unit summarizes and statistically analyzes the daily consumption of each drug from standardized medical data according to drug name and date, and stores it in chronological order as the historical drug consumption dataset.

5. A data integration, analysis, and management platform according to claim 2, characterized in that: The intelligent decision analysis module includes a drug demand forecasting unit and an inventory early warning unit; The drug demand forecasting unit obtains the historical drug consumption dataset from the data standardization processing module, and obtains the average daily consumption of drugs based on the historical drug consumption data. It also obtains the activity intensity index of the epidemic pathogens in the standardized epidemiological data from the data cleaning and fusion module, and calculates the proportion of each drug's consumption to the total annual consumption by month based on the historical drug consumption dataset. The normalized proportion is then used as a seasonality coefficient. The inventory early warning unit obtains the current drug inventory C1 in real time from the drug management system accessed by the multi-source data acquisition module, obtains the daily drug consumption rate from the historical drug consumption dataset, and obtains drug expiration date data from the data standardization processing module.

6. The data integration, analysis, and management platform according to claim 5, characterized in that: The drug demand forecasting unit calculates and outputs the drug demand forecasting index (PDI). The formula for calculating the PDI is: PDI=β1×(D1 / D2)+β2×S+β3×E Wherein, D1 represents the average daily consumption of the target drug over the past 30 days; D2 represents the average daily consumption of similar drugs throughout the year, and both D1 and D2 are calculated from historical drug consumption datasets; S represents the seasonality coefficient; E represents the epidemic trend coefficient, which is obtained by dividing the activity intensity index of the epidemic pathogen by 100; β1, β2, and β3 are preset weighting coefficients, and β1+β2+β3=1; When the PDI is greater than or equal to 1.2, the drug demand forecasting unit outputs a replenishment suggestion while outputting the PDI; when the PDI is less than or equal to 0.6, the drug demand forecasting unit outputs an inventory backlog warning while outputting the PDI; when the PDI is between 0.6 and 1.2, the drug demand forecasting unit outputs the PDI normally.

7. A data integration, analysis, and management platform according to claim 6, characterized in that: The formula for calculating the Inventory Warning Index (IWI) is as follows: IWI = γ1×(C1 / C2)+γ2×(T1 / T2) Wherein, C2 represents the safety stock threshold, and the inventory early warning unit also dynamically adjusts the safety stock threshold based on the PDI output by the drug demand forecasting unit; T1 represents the remaining days until the expiration date of the drug, which is calculated by the inventory early warning unit based on the difference between the drug expiration date data and the current system time; T2 represents the safe shelf life days; γ1 and γ2 are preset weighting coefficients, and γ1 + γ2 = 1; when IWI is greater than or equal to 1.5, the inventory early warning unit outputs IWI and outputs a first-level early warning notification, and outputs an inventory backlog prompt; when IWI is less than or equal to 0.3, the inventory early warning unit outputs IWI and outputs a second-level early warning notification, and the drug demand forecasting unit outputs a replenishment suggestion; when IWI is between 0.3 and 1.5, the inventory early warning unit outputs a normal IWI. The formula for calculating C2 is: C2=T3×D1×(1+k×(PDI-1)) Wherein, T3 represents the inventory early warning unit based on the drug procurement cycle; k is a preset adjustment coefficient with a value range of 0.2-0.

5. According to the preset drug demand, when PDI>1, the demand is higher than normal, and k is adjusted upward; when PDI<1, the demand is lower than normal, and k is adjusted downward.

8. The data integration, analysis, and management platform according to claim 1, characterized in that: The visualization and early warning module includes a real-time data dashboard unit and an early warning push unit; The real-time data dashboard unit dynamically displays the real-time values ​​and historical trend curves of the drug demand forecast index and inventory warning index in the form of a dashboard. The warning push unit pushes warning notifications to preset recipients via SMS, email, or in-app messages under preset trigger conditions.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program that, when executed by a computer or processor, implements the method described in any one of claims 1-8.

10. A computer program product, characterized in that: The computer program product includes a computer program that, when executed by a computer or processor, causes the computer or processor to perform the method as described in any one of claims 1-8.

Images