An automatic early warning system for abnormal drug sales

Abstract

This invention provides an automatic early warning system for abnormal drug sales, comprising: a data management module for periodically collecting and processing patient business data to obtain an underlying data pool; a manual control module for selecting control objects, configuring drug sales early warning rules, determining whether the control objects meet the early warning conditions, and automatically sending early warning information if the determination result is yes; a sales anomaly judgment module for extracting abnormal data from the underlying data pool, constructing a sales anomaly prediction model, and performing real-time drug sales monitoring to obtain sales anomaly trends; and a sales prediction module for extracting sales influencing factors from the underlying data pool, constructing a sales prediction model, and performing real-time drug sales prediction to obtain predicted purchase volume and predicted purchase timing. This invention can effectively identify abnormal drug sales trends, predict drug sales trends, achieve personalized automatic early warning for abnormal drug sales, monitor drug usage and consumption, and promptly detect and respond to drugs with abnormal sales.

Description

Technical Field

[0001] This invention relates to the field of drug management technology, and in particular to an automatic early warning system for abnormal drug sales. Background Technology

[0002] Typically, hospital drug inventory systems are equipped with basic early warning functions, alerting pharmacy managers to replenish stock when medications are running low, or reminding them to increase inventory during peak flu seasons. However, in the face of sudden disease outbreaks in clinical practice, such as the sudden outbreaks of SARS, H1N1, or other novel influenza species, as well as unexpected events such as environmental pollution, natural disasters, and major accidents, drug demand can surge. In these situations, traditional inventory early warning systems cannot predict and warn of sharp increases in drug consumption in a timely manner, nor can they learn and adapt to complex drug consumption patterns, leading to drug shortages. This not only wastes valuable time for doctors and patients, severely impacting the patient experience, but may also adversely affect hospital operations and management. Summary of the Invention

[0003] To overcome the shortcomings of existing technologies, the purpose of this invention is to provide an automatic early warning system for abnormal drug sales, which can effectively identify abnormal trends in drug sales, predict drug sales trends, realize multi-algorithm and personalized automatic early warning of abnormal drug sales in hospitals, and also allow manual setting of rules to accurately monitor drug use and consumption through various methods, and promptly detect and respond to drugs with abnormal sales.

[0004] To achieve the above objectives, the present invention provides the following solution: an automatic early warning system for abnormal drug sales, comprising:

[0005] The data management module is used to collect patient business data from multiple hospitals on a daily schedule, and to perform quality inspection, data normalization and data standardization operations on the patient business data to obtain the underlying data pool.

[0006] The manual control module is used to select control objects, configure drug sales warning rules according to the control objects, and then determine whether the control objects meet the warning conditions according to the drug sales warning rules. If the judgment result is yes, the warning information is automatically sent.

[0007] The sales anomaly judgment module is used to extract abnormal data from the underlying data pool, mark and separate the abnormal data to obtain anomaly judgment data, use the anomaly judgment data to perform machine learning modeling to obtain a sales anomaly prediction model, and then use the sales anomaly prediction model to monitor the real-time sales of drugs to obtain sales anomaly trends.

[0008] The sales forecasting module is used to extract the sales influencing factors from the underlying data pool, perform data cleaning, feature selection and optimization on the sales influencing factors, and then use the sales influencing factors to perform machine learning modeling to obtain a sales forecasting model. The sales forecasting model is then used to predict the real-time sales of drugs to obtain the predicted purchase quantity and the predicted purchase timing.

[0009] The data management module, the manual control module, the sales anomaly judgment module, and the sales forecast module are interconnected.

[0010] Optionally, the data management module includes:

[0011] The data acquisition unit is used to collect patient business data daily from multiple hospital inpatient management systems using view technology, stored procedure technology, or Web service technology, record the daily collection results, perform incremental updates and data synchronization based on the collection results, and determine whether there is erroneous data in the historical collection data. If the determination result is yes, the data synchronization is re-performed and a data synchronization exception alert is issued.

[0012] The data processing unit is used to automatically detect the patient business data using data quality detection tools, perform data normalization and preprocessing on the patient business data, and standardize the drug-related data in the patient business data to obtain the underlying data pool; wherein, the drug-related data includes the generic name of the drug and the drug pharmacology category;

[0013] A data storage unit is used to select a data warehouse and store the underlying data pool in the data warehouse using columnar storage.

[0014] Optionally, the manual control module includes:

[0015] The rule-making unit is used to select control objects, configure control dimensions, control cycles, warning thresholds, and warning sending targets according to the control objects, and obtain drug sales warning rules; wherein, each control dimension and control cycle is configured with different judgment warning rules.

[0016] The data comparison unit is used to determine whether the controlled object meets the warning conditions according to the drug sales warning rules. If the judgment result is yes, the relevant data and warning result of the controlled object are sent to the warning sending object.

[0017] The early warning sending unit is used to view the early warning results via hyperlinks, and the early warning results are displayed in a comparative manner using a combination of tables and charts.

[0018] Optionally, the control targets include all drugs, specific drugs, generic drug names, and pharmacological categories; the control dimensions include the entire hospital, departments, and large departments; the control period includes daily statistics, monthly statistics, non-monthly statistics, and weekly statistics; and the warning thresholds include the maximum amount of the previous N periods, the average amount of the previous N periods, and a fixed value.

[0019] Optionally, the sales anomaly detection module includes:

[0020] The system rule unit is used to build automatically generated system rules;

[0021] The abnormal data unit is used to extract abnormal data from the underlying data pool according to the system rules and the drug sales early warning rules, establish statistical reports based on the abnormal data, perform preliminary marking of the abnormal situation of the controlled object, obtain initial abnormal results, and then manually screen the initial abnormal results to obtain abnormal judgment data.

[0022] The first model building unit is used to select a first machine learning model, divide the anomaly judgment data into a first experimental group and a first reference group, train the machine learning model using machine learning algorithms and the first experimental group, and use the first reference group for performance evaluation, parameter adjustment and model optimization during the training process to obtain a sales anomaly prediction model.

[0023] Anomaly Sending Unit: Used to monitor real-time sales of medicines using the sales anomaly prediction model, obtain abnormal sales trends, generate abnormal sales information from the abnormal sales trends, and send it to the relevant staff.

[0024] Optionally, the process of constructing the statistical report includes: based on the control period, calculating the historical sales amount, sales growth rate, sales volume, average sales amount, multi-period sales amount of the controlled object, as well as the number of patients, prescriptions and related diagnoses of the controlled object in the same period, and completing the construction of the statistical report.

[0025] Optionally, the sales forecasting module includes:

[0026] The influencing factor extraction unit is used to establish a modeling database including basic information, consumption data, replenishment cycle and other factors based on the underlying data pool, obtain the sales influencing factors, and perform regression to fill in the missing data in the sales influencing factors.

[0027] The feature selection unit is used to estimate and match the propensity scores of the factors influencing sales using the pandas library in Python, as well as statsmodels and imblearn. Then, the matched indexes and data are subjected to balance verification and causal effect analysis to obtain the correlation coefficient between the factors influencing sales and the results.

[0028] The second model building unit is used to select a second machine learning model, divide the sales influencing factors into a second experimental group and a second reference group, train the machine learning model using machine learning algorithms and the second experimental group, and use the second reference group for performance evaluation, parameter adjustment and model optimization during the training process to obtain a sales prediction model.

[0029] The early warning sending unit is used to use the sales forecasting model to predict the real-time sales volume of medicines, obtain the predicted purchase quantity and the predicted purchase timing, and send the predicted purchase quantity and the predicted purchase timing to the relevant staff.

[0030] Optionally, the basic information includes name, dosage form, specifications, manufacturer, code, drug type, generic name of drug and pharmacological category; the consumption data includes the billing time, sales volume and sales trend of the controlled object; the replenishment cycle includes the warehousing time, outbound time, inventory and replenishment cycle interval of the controlled object; and the other factors include seasonal markers, holiday markers, number of patients and high-frequency diagnoses.

[0031] This invention discloses the following technical effects by providing an automatic early warning system for abnormal drug sales:

[0032] 1. Early Warning and Advance Preparation: Traditional inventory management relies heavily on inventory levels, which managers use to assess drug storage conditions. This invention, building upon inventory warning systems, adds an abnormal sales warning function. This function uses consumption rate as a benchmark indicator, keenly capturing and analyzing changes in drug consumption rates and identifying early consumption trends. When the machine learning model detects an abnormal increase in sales, it immediately alerts pharmacy staff, enabling them to take proactive measures, strengthen preparation and record-keeping, and improve the efficiency and responsiveness of medical services in actual hospital operations. This allows for rapid drug access in emergencies, avoiding delays in treatment due to waiting for pharmacy dispensing.

[0033] 2. Self-updating early warning standards, optimizing pharmacy staff work patterns: The system combines the results of manual control with data from built-in rules, while utilizing machine learning algorithms for self-iteration and optimization. This means the system can learn and identify patterns from historical data, automatically adjusting rules to adapt to constantly changing environments and needs, without manual intervention from pharmacy staff. The system shifts from passively issuing warnings based on rule parameters to proactively updating its early warning standards, demonstrating strong generalization capabilities and changing the work patterns of pharmacy staff, thus reducing their workload.

[0034] 3. Comprehensiveness of Drug Consumption and Inventory Management: 1) This invention not only allows users to manage inventory through manually set rules, but also employs two machine learning models, one for judging abnormal drug sales trends and the other for sales forecasting. This multi-dimensional early warning mechanism enables the system to cover both early-stage sudden changes and later-stage replenishment needs, improving the comprehensiveness of drug consumption and inventory management. 2) This invention also considers that in actual hospital procurement, the same drug may be purchased from multiple manufacturers with different specifications. Furthermore, in clinical applications, a single drug may have multiple similar substitutes. Therefore, the system can not only manage individual drugs but also group drugs, such as by generic name or pharmacological category. This drug group-based management approach makes abnormal sales judgment and inventory management more flexible.

[0035] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a schematic diagram of the system architecture provided in an embodiment of the present invention;

[0038] Figure 2 This is a schematic diagram of the system judgment process provided in an embodiment of the present invention. Detailed Implementation

[0039] 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.

[0040] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0041] like Figure 1-2 As shown, this invention provides an automatic early warning system for abnormal drug sales, comprising an interconnected data management module, a manual control module, a sales anomaly judgment module, and a sales prediction module. Wherein:

[0042] 1. Data Management Module

[0043] This system is used to collect patient data from multiple hospitals on a daily schedule, and performs quality checks, data normalization, and data standardization on the patient data to obtain the underlying data pool. The data management module includes:

[0044] 1.1 Data Acquisition Unit

[0045] This system utilizes view technology, stored procedure technology, or Web service technology to collect patient business data daily from multiple hospital inpatient management systems, record the daily collection results, and perform real-time incremental updates based on the collection results.

[0046] A service-oriented model is adopted, incorporating fault tolerance mechanisms to ensure data accuracy. Data synchronization with HIS (Hospital Inpatient Management System) and other systems is performed daily, recording the daily collection results to achieve data synchronization. The system also checks for erroneous data in historical collections; if an error is found, data synchronization is re-performed, and an anomaly alert is issued.

[0047] 1.2 Data Processing Unit

[0048] This system is used to automatically detect the accuracy and standardization of patient business data using data quality inspection tools, thereby improving data quality. The system performs data normalization (e.g., standardizing date formats, handling illegal data, etc.) and preprocessing on the patient business data, and standardizes drug-related data within the patient business data (e.g., standardizing drug generic names, pharmacological labeling, etc.) to obtain the underlying data pool; wherein, the drug-related data includes drug generic names and drug pharmacological categories;

[0049] 1.3 Data storage unit

[0050] This is used to select a data warehouse (such as ClickHouse) and store the underlying data pool in the data warehouse using columnar storage to improve the efficiency of large-scale data aggregation queries.

[0051] 2. Manual control module

[0052] The system is used to select control targets, configure drug sales early warning rules based on the control targets, and then determine whether the control targets meet the early warning conditions according to the drug sales early warning rules. If the determination result is yes, an early warning message is automatically sent. The manual control module includes:

[0053] 2.1 Rule-making unit

[0054] This is used to select control targets, and to configure control dimensions, control periods, warning thresholds, and warning recipients based on the control targets, thereby obtaining drug sales warning rules; wherein, each control dimension and control period is configured with different judgment and warning rules.

[0055] 1) Targets of control include all drugs, specific drugs, generic drug names, and pharmacological categories;

[0056] 2) The control dimensions of configuration rules include the entire hospital, departments, and large departments;

[0057] 3) Configure the control period for rules, including statistics by day, statistics by calendar month / non-calendar month, and statistics by week;

[0058] 4) Configure warning thresholds for the rules, including the maximum amount of the previous N periods, the average amount of the previous N periods, and a fixed value;

[0059] 5) Early warning rules can be configured separately for each control dimension and period. Taking daily statistics as an example, the rules are as follows:

[0060] Daily sales revenue increase compared to the warning threshold > YY%, daily sales revenue > XX yuan;

[0061] The increase in the average daily sales amount over the past M days compared to the warning threshold is greater than YY%, and the average daily sales amount is greater than XX yuan;

[0062] The increase in sales amount compared to the warning threshold for M consecutive days is greater than YY%, and the average daily sales amount is greater than XX yuan;

[0063] 6) Configure the warning sending object. When the warning condition is triggered, the system will automatically send a warning.

[0064] 2.2 Data Comparison Unit

[0065] This is used to determine whether the controlled object meets the warning conditions according to the drug sales warning rules. If the judgment result is yes, the relevant data and warning result of the controlled object are sent to the warning sending object.

[0066] 2.3 Early Warning Sending Unit

[0067] When the warning conditions are met, a warning message is sent to the recipient. After the warning message is sent, users can view the warning results through hyperlinks on the system platform, SMS platform, and WeChat Work platform. The warning results can be displayed in a combination of tables and charts. Clicking on the table data will automatically generate a bar chart / line chart of the results for the short term, clearly showing the comparison between the warning items and dynamic thresholds for the drug in the short term.

[0068] 3. Sales Anomaly Detection Module

[0069] This module is used to extract abnormal data from the underlying data pool, mark and separate the abnormal data to obtain abnormal judgment data, use the abnormal judgment data for machine learning modeling to obtain a sales anomaly prediction model, and then use the sales anomaly prediction model to monitor real-time drug sales and obtain sales anomaly trends. The sales anomaly judgment module includes:

[0070] 3.1 System Rule Unit

[0071] This is used to build automatically generated system rules. The system comes pre-configured with some rules to identify abnormal data, while manual control also configures rules. The abnormal data detected by these two types of rules is then merged and aggregated into the dataset needed for modeling.

[0072] 3.2 Abnormal Data Unit

[0073] This is used to extract abnormal data from the underlying data pool according to the system rules and the drug sales early warning rules, establish statistical reports based on the abnormal data, perform preliminary marking of abnormal situations of the controlled objects, obtain initial abnormal results, and then manually screen the initial abnormal results to further select valuable and meaningful anomalies and separate the data to obtain abnormal judgment data.

[0074] The process of constructing the statistical report includes: based on the control period, calculating the historical sales amount, sales growth rate, sales volume, average sales amount, multi-period sales amount of the controlled object, as well as the number of patients, prescriptions and related diagnoses of the controlled object in the same period, and completing the construction of the statistical report.

[0075] 3.3 First Model Building Unit

[0076] To select the first machine learning model, the anomaly judgment data is divided into a first experimental group and a first reference group. The machine learning model is trained using a machine learning algorithm (such as XGBoost / LightGBM / TabNet / SVM algorithm) and the first experimental group. During the training process, the first reference group is used for performance evaluation, parameter adjustment and model optimization to obtain the sales anomaly prediction model.

[0077] The performance of a model can be judged by mean square error, root mean square error, mean absolute error, F1 / F2 score, and ROC curve. The best-performing model can be selected, and the model parameters can be adjusted based on the test results to optimize prediction accuracy.

[0078] 3.4 Exception Sending Unit

[0079] This is used to monitor real-time sales of pharmaceuticals using the aforementioned sales anomaly prediction model, obtain abnormal sales trends, generate abnormal sales information from these trends, and send it to the relevant staff.

[0080] 4. Sales Forecasting Module

[0081] This module is used to extract sales-influencing factors from the underlying data pool, perform data cleaning, feature selection, and optimization on these factors, then use these factors for machine learning modeling to obtain a sales prediction model. Finally, the sales prediction model is used to predict real-time drug sales, resulting in predicted purchase volume and timing. The sales prediction module includes:

[0082] 4.1 Influencing Factor Extraction Unit

[0083] This is used to build a modeling database based on the underlying data pool, including basic information, consumption data, replenishment cycle, and other factors, to obtain factors influencing sales, and to perform regression imputation on missing data in these factors. Specifically:

[0084] Basic information: name, dosage form, specifications, manufacturer, code, drug type, generic name, pharmacological category, etc.

[0085] Consumption data: billing time, sales volume, sales trends, etc. for drugs / generic drug names / pharmacological drugs;

[0086] Replenishment cycle: Inbound time, outbound time, inventory level, and replenishment cycle interval of medicines / generic names / pharmacological drugs;

[0087] Other factors: seasonal markers, holiday markers, number of patients, high-frequency diagnoses, etc.

[0088] Data cleaning: For missing data, the pandas library in Python is used in conjunction with statsmodels and scikit-learn to perform regression imputation. After imputation, the remaining missing data is deleted.

[0089] 4.2 Feature Selection Unit

[0090] This method utilizes the pandas library in Python, along with statsmodels and imblearn, to estimate and match the propensity scores of the factors influencing sales. Then, it performs balance checks and causal effect analysis on the matched indices and data to obtain the correlation coefficient between the factors influencing sales and the results. The correlation coefficient is used to adjust the weights of each factor in the machine learning training model.

[0091] 4.3 Second Model Building Unit

[0092] To select a second machine learning model, the sales influencing factors are divided into a second experimental group and a second reference group. The machine learning model is trained using machine learning algorithms (such as XGBoost / LightGBM / TabNet / SVM algorithms) and the second experimental group. During the training process, the second reference group is used for performance evaluation, parameter adjustment and model optimization to obtain a sales prediction model.

[0093] The performance of a model can be judged by mean square error, root mean square error, and mean absolute error. The best-performing model can be selected, and the model parameters can be adjusted based on the test results to optimize prediction accuracy.

[0094] 4.4 Early Warning Sending Unit

[0095] This is used to perform real-time sales forecasting of pharmaceuticals using the sales forecasting model, obtain the predicted purchase quantity and the predicted purchase timing, and send the predicted purchase quantity and the predicted purchase timing to the relevant staff.

[0096] 5. Access Control

[0097] Different accounts can be assigned permissions, with each role having different permissions corresponding to different functions. These include: creating and managing user accounts, assigning roles and permissions; access to various functional modules of the program; permission to set drug control rules; and permission to view alert information.

[0098] Therefore, this invention provides an automatic early warning system for abnormal drug sales, which can effectively identify abnormal trends in drug sales, predict drug sales trends, and realize automatic early warning of abnormal drug sales in hospitals using multiple algorithms and personalization. In addition, the system also allows manual setting of rules to accurately monitor drug usage and consumption through various methods, and promptly detect and respond to drugs with abnormal sales.

[0099] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0100] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. An automatic early warning system for abnormal drug sales, characterized in that, include: The data management module is used to collect patient business data from multiple hospitals on a daily schedule, and to perform quality inspection, data normalization and data standardization operations on the patient business data to obtain the underlying data pool. The manual control module is used to select control objects, configure drug sales warning rules according to the control objects, and then determine whether the control objects meet the warning conditions according to the drug sales warning rules. If the judgment result is yes, the warning information is automatically sent. The sales anomaly judgment module is used to extract abnormal data from the underlying data pool, mark and separate the abnormal data to obtain anomaly judgment data, use the anomaly judgment data to perform machine learning modeling to obtain a sales anomaly prediction model, and then use the sales anomaly prediction model to monitor the real-time sales of drugs to obtain sales anomaly trends. The sales forecasting module is used to extract the sales influencing factors from the underlying data pool, perform data cleaning, feature selection and optimization on the sales influencing factors, and then use the sales influencing factors to perform machine learning modeling to obtain a sales forecasting model. The sales forecasting model is then used to predict the real-time sales of drugs to obtain the predicted purchase quantity and the predicted purchase timing. The data management module, the manual control module, the sales anomaly judgment module, and the sales forecast module are interconnected. The data management module includes: The data acquisition unit is used to collect patient business data daily from multiple hospital inpatient management systems using view technology, stored procedure technology, or Web service technology, record the daily collection results, perform incremental updates and data synchronization based on the collection results, and determine whether there is erroneous data in the historical collection data. If the determination result is yes, the data synchronization is re-performed and a data synchronization exception alert is issued. The data processing unit is used to automatically detect the patient business data using data quality detection tools, perform data normalization and preprocessing on the patient business data, and standardize the drug-related data in the patient business data to obtain the underlying data pool; wherein, the drug-related data includes the generic name of the drug and the drug pharmacology category; A data storage unit is used to select a data warehouse and store the underlying data pool in the data warehouse using columnar storage. The manual control module includes: The rule-making unit is used to select control objects, configure control dimensions, control cycles, warning thresholds, and warning sending targets according to the control objects, and obtain drug sales warning rules; wherein, each control dimension and control cycle is configured with different judgment warning rules. The data comparison unit is used to determine whether the controlled object meets the warning conditions according to the drug sales warning rules. If the judgment result is yes, the relevant data and warning result of the controlled object are sent to the warning sending object. The early warning sending unit is used to view the early warning results via hyperlinks, and the early warning results are displayed comparatively using a combination of tables and charts; The sales anomaly detection module includes: The system rule unit is used to build automatically generated system rules; The abnormal data unit is used to extract abnormal data from the underlying data pool according to the system rules and the drug sales early warning rules, establish statistical reports based on the abnormal data, perform preliminary marking of the abnormal situation of the controlled object, obtain initial abnormal results, and then manually screen the initial abnormal results to obtain abnormal judgment data. The first model building unit is used to select a first machine learning model, divide the anomaly judgment data into a first experimental group and a first reference group, train the machine learning model using machine learning algorithms and the first experimental group, and use the first reference group for performance evaluation, parameter adjustment and model optimization during the training process to obtain a sales anomaly prediction model. Anomaly Sending Unit: Used to monitor real-time sales of medicines using the sales anomaly prediction model, obtain abnormal sales trends, generate abnormal sales information from the abnormal sales trends, and send it to the relevant staff. The sales forecasting module includes: The influencing factor extraction unit is used to establish a modeling database including basic information, consumption data, replenishment cycle and other factors based on the underlying data pool, obtain the sales influencing factors, and perform regression to fill in the missing data in the sales influencing factors. The feature selection unit is used to estimate and match the propensity scores of the factors influencing sales using the pandas library in Python, as well as statsmodels and imblearn. Then, the matched indexes and data are subjected to balance verification and causal effect analysis to obtain the correlation coefficient between the factors influencing sales and the results. The second model building unit is used to select a second machine learning model, divide the sales influencing factors into a second experimental group and a second reference group, train the machine learning model using machine learning algorithms and the second experimental group, and use the second reference group for performance evaluation, parameter adjustment and model optimization during the training process to obtain a sales prediction model. The early warning sending unit is used to use the sales forecasting model to predict the real-time sales volume of medicines, obtain the predicted purchase quantity and the predicted purchase timing, and send the predicted purchase quantity and the predicted purchase timing to the relevant staff.

2. The automatic early warning system for abnormal drug sales according to claim 1, characterized in that, The controlled objects include all drugs, generic drug names, and pharmacological categories. The controlled dimensions include the entire hospital, departments, and large departments. The controlled period includes daily statistics, monthly statistics, non-monthly statistics, and weekly statistics. The warning thresholds include the maximum amount of the previous N periods, the average amount of the previous N periods, and a fixed value.

3. The automatic early warning system for abnormal drug sales according to claim 2, characterized in that, The process of constructing the statistical report includes: based on the control period, calculating the historical sales amount, sales growth rate, sales volume, average sales amount, multi-period sales amount of the controlled object, as well as the number of patients, prescriptions and related diagnoses of the controlled object in the same period, and completing the construction of the statistical report.

4. The automatic early warning system for abnormal drug sales according to claim 3, characterized in that, The basic information includes name, dosage form, specifications, manufacturer, code, drug type, generic name and pharmacological category. The consumption data includes the billing time, sales volume and sales trend of the controlled object. The replenishment cycle includes the warehousing time, outbound time, inventory and replenishment cycle interval of the controlled object. Other factors include seasonal markers, holiday markers, number of patients and high-frequency diagnoses.

Images