CN116525060B - Intelligent area monitoring method and system based on medical product supervision model - Google Patents

Intelligent area monitoring method and system based on medical product supervision model Download PDF

Info

Publication number
CN116525060B
CN116525060B CN202310744089.1A CN202310744089A CN116525060B CN 116525060 B CN116525060 B CN 116525060B CN 202310744089 A CN202310744089 A CN 202310744089A CN 116525060 B CN116525060 B CN 116525060B
Authority
CN
China
Prior art keywords
data
medical product
consumer
payment
price
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.)
Active
Application number
CN202310744089.1A
Other languages
Chinese (zh)
Other versions
CN116525060A (en
Inventor
肖锎
文庭孝
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Central South University
Original Assignee
Central South University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Central South University filed Critical Central South University
Priority to CN202310744089.1A priority Critical patent/CN116525060B/en
Publication of CN116525060A publication Critical patent/CN116525060A/en
Application granted granted Critical
Publication of CN116525060B publication Critical patent/CN116525060B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
    • G16H20/10—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to drugs or medications, e.g. for ensuring correct administration to patients
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/044—Recurrent networks, e.g. Hopfield networks
    • G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G06Q30/0206—Price or cost determination based on market factors
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Health & Medical Sciences (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Finance (AREA)
  • Strategic Management (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Accounting & Taxation (AREA)
  • Evolutionary Computation (AREA)
  • Development Economics (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Mathematical Physics (AREA)
  • Biomedical Technology (AREA)
  • Computing Systems (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Computational Linguistics (AREA)
  • Molecular Biology (AREA)
  • Software Systems (AREA)
  • Biophysics (AREA)
  • Economics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Game Theory and Decision Science (AREA)
  • Public Health (AREA)
  • Primary Health Care (AREA)
  • Medical Informatics (AREA)
  • Epidemiology (AREA)
  • Medicinal Chemistry (AREA)
  • Chemical & Material Sciences (AREA)
  • Medical Treatment And Welfare Office Work (AREA)

Abstract

The invention discloses an intelligent area monitoring method and system based on a medical product supervision model, comprising the steps of generating a training set by adopting a sliding window method; training by adopting a long-short-term memory neural network model to obtain payment data of the medical product; establishing a medical product safety supervision model, judging consumer payment data in a preset area, and obtaining a consumer medical product payment result; judging whether the payment of the consumer medical product exceeds the limit range of the medical product; if yes, collecting data of a collecting end, and feeding back price abnormality information to a matched consumer paying end; and feeding back the abnormal information of the price of the medical product containing the data of the collecting end to a medical product price supervision terminal of the preset area. According to the intelligent area monitoring method and system, the price of the medical product in the set area can be dynamically monitored by combining the big data with the neural network, so that the monitoring efficiency is improved, and further, the price out of control of the medical product is effectively reduced.

Description

Intelligent area monitoring method and system based on medical product supervision model
Technical Field
The invention relates to the technical field of medicine product price supervision, in particular to an intelligent area monitoring method, system, computer medium and computer based on a medicine product supervision model.
Background
In the medical field, the cost of medical services is always a global focus, with the continuous increase of the types and prices of medical products, in order to control medical expenses, the price of the medical products is usually limited, namely the highest selling price of the medical products is limited, and at present, in some areas, the price limiting measures of the price of the medical products are implemented, but in actual situations, the price of the medical products exceeds the price limit frequently, namely due to the reasons of asymmetric information, high supervision cost and the like, some private hospitals, dealers and medical stores avoid the price limit through illegal means or loopholes, so that the price of the medical products is out of control and the cost of the medical services is increased, and the management and the fairness of medical insurance foundation are seriously influenced;
at present, the price supervision of the existing medical products generally adopts the modes of manual inspection by supervision personnel, complaint report of consumers, random spot check by supervision personnel and the like, and specifically comprises the following steps: firstly, a large amount of manpower and material resources are required to be input for manual inspection, and comprehensive monitoring and supervision are difficult to realize; secondly, consumer complaints report the problems that consumers need to find and report the problems by themselves, and consumers can not find and report the problems in time due to incomplete understanding of the prices of the medical products for various reasons; thirdly, the supervision personnel can not cover the sales conditions of all the medical products easily in a random spot check mode; meanwhile, because the limited prices of the medical products in different periods are different, consumers often lack real and objective knowledge of the prices of the medical products.
Therefore, an intelligent area monitoring method capable of dynamically monitoring the price of the medical product in the set area by combining big data with a neural network, improving the monitoring efficiency and further effectively reducing the price runaway of the medical product is needed at present.
Disclosure of Invention
Therefore, the intelligent area monitoring method, system, computer medium and computer based on the medical product supervision model can dynamically monitor the medical product payment data of the set area, analyze the medical product payment result of the consumer, compare the medical product payment result with the matched medical product limit data, further judge whether the medical product payment of the consumer exceeds the limit range, and quickly acquire medical product price abnormality information containing the data of the collection end to provide for the relevant supervision terminal once the medical product price is found to be abnormal, thereby effectively reducing the out-of-control price of the medical product.
In order to solve the technical problems, the invention provides an intelligent area monitoring method based on a medical product supervision model,
comprising the following steps:
step S1: generating a training set by adopting a sliding window method;
step S2: the training set is continuously trained by adopting a long-and-short-term memory neural network model to obtain corrected medicine product payment data, and the method comprises the following steps: presetting payment data generated by a sliding window method to form related data of medical products, extracting preset features from the related data, training the preset features by adopting a cross-validation method, and obtaining corrected payment data of the medical products;
Step S3: establishing a medical product supervision model, judging consumer payment data in a preset area, and obtaining a consumer medical product payment result;
step S4: comparing the payment result of the consumer medical product with the medical product limit data of the preset area, and judging whether the payment of the consumer medical product exceeds the medical product limit range of the preset area;
step S5: if yes, collecting the data of the collecting end contained in the consumer payment data, and simultaneously feeding back the abnormal information of the price of the medical product to the matched consumer payment end, if not, returning to the step S3, and judging the consumer payment data of the preset area again;
step S6: and feeding back the abnormal information of the price of the medical product containing the data of the collecting end to a medical product price supervision terminal of the preset area.
By adopting the technical scheme, the data can be rapidly processed, so that abnormal price conditions of the medical products can be rapidly found, whether the payment of the medical products of consumers exceeds the limit range of the medical products in a preset area can be accurately judged, and the supervision accuracy is improved; through the supervision to the medicine goods payment data, can in time discover medicine goods price abnormal condition to feedback abnormal information to the consumer, protect consumer's interests, and through the supervision to medicine goods price, can maintain market order effectively, prevent medicine goods price out of control, and then reduce medical service cost.
As a preferred mode of the present invention, the method for generating the training set by adopting the sliding window method includes:
step S10: collecting payment data of a preset time sequence;
step S11: using sliding window method to lengthAccording to the length of +.>And (3) performing time window cutting to generate a training set:
；
wherein , is the number of samples, and each sample contains +.>A number of input features and a number of output tags,
payment data representing the last month in the time window.
By adopting the technical scheme, the data can be more comprehensively collected, so that the traceability of supervision is improved, and investigation and treatment of illegal situations are facilitated.
As a preferable mode of the invention, the method for carrying out preset processing on the payment data comprises the following steps:
step S20: and processing payment data by adopting a data cleaning method:
；
wherein , is +.>Sample number-> and />Is the data range, +.>Is the data after washing if +.>Beyond the data range->Setting it as a missing value;
step S21: carrying out data standardization processing on payment data after data cleaning:
；
wherein ,is +.f. in the payment data after cleaning >Sample No. H>Personal characteristics (I)>Is normalized payment data, +.> and />Are respectively->Minimum and maximum values of the individual features.
Step S22: and carrying out feature extraction on payment data after data normalization:
；
wherein , is the first part of the extraction>Personal characteristics (I)>Is a feature extraction function, ++>Is the characteristic of the cleaned payment data;
step S23: data aggregation is carried out on the extracted features to generate related data of medical products:
；
wherein , is the data related to the medical products after feature aggregation, < >>Is a function of the aggregation and,
is the feature data that needs to be aggregated.
As a preferred mode of the present invention, the method for extracting the preset features from the related data includes:
step S24: extracting the price of the medical product and the payment amount characteristics of the medical product of the consumer from the related data of the medical product:
；
wherein , representing the price characteristics of the medical product->Characterizing the amount paid by the consumer pharmaceutical product +.> 、A comprehensive function representing the characteristics of the price of the pharmaceutical product and the payment amount of the pharmaceutical product of the consumer, and +.>Representing data related to the pharmaceutical product.
As a preferred mode of the present invention, the method for training the preset feature is as follows:
Step S25: to extract the total characteristicsRandomly divide into->Parts, wherein each part is the same size and +.>The parts are used as training sets, and the rest 1 part is used as a test set;
step S26: for each featureTesting is carried out on the test sets respectively;
step S27: repeating step S26 until each data is used as a test set to obtainThe accuracy rate;
step S28: for each featureWill->The average of the individual accuracies was taken as its cross-validation accuracy:
；
wherein , indicate->Accuracy of secondary verification;
step S29: comparing each featureAnd selecting the feature with the highest accuracy as the final feature.
By adopting the technical scheme, the method and the device can effectively utilize data, reduce the problem of data waste, and evaluate the generalization capability of the model, so that the performance of the model is better known, the influence caused by randomness can be reduced, and the robustness of the model is improved.
As a preferable mode of the invention, the method for establishing the medical product supervision model comprises the following steps:
step S30: the long-short-term memory neural network model is evaluated on a test set, and a sensitivity score of each weight is generated:
；
wherein , Indicate->Personal weight(s)>Is a loss function;
step S31: calculating an average value of sensitivity scores of each LSTM layer of the pharmaceutical product supervision model, and generating an average score vector:
；
wherein ,is->A set of subscripts for all weights of a layer;
step S32: the average score vectors are sorted in ascending order, and the proportion of each score to all scores is calculated to generate a proportion vector:
；
wherein , is->Is>An element;
step S33: calculating the weight quantity to be reserved according to the pruning proportion:
；
wherein , is pruning proportion;
step S34: according to the proportion vector and the weight quantity, selecting the weight with the highest score as a reserved weight:
；
wherein , is->Is>A maximum value;
step S35: pruning the medical product supervision modelThe values of all weights except for 0;
step S36: repeating the steps S31-S35 for a preset number of times, generating an optimized long-short-time memory neural network model, and building the optimized long-short-time memory neural network model into a medicine product price supervision terminal of a preset area to establish a medicine product supervision model.
By adopting the technical scheme, the calculated amount and the storage amount of the long-and-short-term memory neural network model can be reduced, and the model is simplified.
As a preferred mode of the present invention, when determining consumer payment data of a preset area, the method further comprises the following steps:
step S300: judging the consumer payment data of the preset area, and collecting the collection end data contained in the consumer payment data, so as to obtain the consumer medicine product payment result and the collection end collection data in the preset time interval;
step S301: judging whether the payment result of the consumer medical product is matched with the collection data of the collection end;
step S302: if yes, returning to the step S300, judging the consumer payment data of the preset area again, and if not, further judging whether the consumer medicine product payment result is higher than the collection data of the collection end;
step S303: if yes, the medical product price monitoring terminal of the preset area is fed back with the medical product price abnormality information containing the collection end data, if not, the step S30 is returned, and the customer payment data of the preset area is judged again.
By adopting the technical scheme, when a consumer purchases the medical product, the seller can be identified to replace the type and the quantity of the medical product purchased by the consumer by other illegal means or holes, so that the medical product data input by the seller into the system is prevented from being inconsistent with the medical product purchased by the consumer, or the seller sells false and expired medical products to the consumer, and the health and the life safety of the consumer are influenced.
As a preferable mode of the invention, after judging that the payment result of the consumer medical product is higher than the collection data of the collection end, the invention further comprises the following steps:
step S304: judging whether the payment result of the consumer medical product is lower than the collection data of the collection end;
step S305: if so, acquiring the historical collection data of the collection end, collecting the cash income data in the historical collection data,
if not, returning to the step S304, and judging the payment result of the consumer medical product again;
step S306: judging whether the ratio of the cash income data in the historical collection data exceeds a preset value or not;
step S307: if yes, the medical product price monitoring terminal of the preset area is fed back with medical product price abnormality information containing the data of the collection end, if not, the step S34 is returned, and the payment result of the medical product of the consumer is judged again.
Through adopting above-mentioned technical scheme, can discern the consumer when buying the medical goods, selling the price that the side improves the medical goods through the price means of false alarm, then utilize cash to collect, the partial cash partial online payment form collection, prevent to cause economic loss to the consumer, and then influence the normal operating in medical market.
The invention also provides a computer medium, wherein the computer medium is stored with a computer program, and the computer program is executed by a processor to realize the intelligent area monitoring method based on the medical product supervision model.
The invention also provides an intelligent area monitoring system based on the medical product supervision model, which comprises:
the data processing module is used for generating training by adopting a sliding window method, and continuously training a training set by adopting a long-and-short-term memory neural network model to obtain corrected medicine product payment data, and comprises the following steps: presetting payment data generated by a sliding window method to form related data of medical products, extracting preset features from the related data, training the preset features by adopting a cross-validation method, and obtaining corrected payment data of the medical products;
the supervision module is used for embedding the optimized long-short-term memory neural network model into a medical product price supervision terminal of a preset area and establishing a medical product supervision model;
the price limiting processing module is used for judging the consumer payment data of the preset area and obtaining the payment result of the consumer medical product;
Comparing the payment result of the consumer medical product with the medical product limit data of the preset area, and judging whether the payment of the consumer medical product exceeds the medical product limit range of the preset area; if yes, collecting the data of the collecting end contained in the consumer payment data, feeding back the abnormal information of the price of the medical product to the matched consumer payment end, and if not, judging the consumer payment data of the preset area again;
the abnormal feedback module is used for feeding back abnormal information of the price of the medical product containing the data of the collection end to the medical product price supervision terminal in the preset area.
Compared with the prior art, the technical scheme of the invention has the following advantages:
1. according to the intelligent area monitoring method and system based on the medical product supervision model, the data can be collected more comprehensively by adopting a sliding window method, so that the traceability of supervision is improved, and investigation and treatment of illegal conditions are facilitated; the neural network model is memorized for a long time, so that data can be rapidly processed, further abnormal price conditions of the medical products can be rapidly found, whether the payment of the medical products of the consumers exceeds the limit range of the medical products in a preset area can be accurately judged, and the supervision accuracy is improved;
2. The abnormal condition of the price of the medical product can be timely found through the supervision of the payment data of the medical product, abnormal information is fed back to consumers, the benefits of the consumers are protected, and the market order can be effectively maintained through the supervision of the price of the medical product, the price of the medical product is prevented from being out of control, and the medical service cost is further reduced;
3. when a consumer purchases the medical product, the seller can be identified to replace the type and the quantity of the medical product purchased by the consumer by other illegal means or holes, so that the medical product data input by the seller into the system is prevented from being inconsistent with the medical product purchased by the consumer, or the seller sells false and expired medical products to the consumer, and the health and the life safety of the consumer are influenced;
4. and the selling price of the medical product can be increased by a seller through a false alarm price means when the consumer purchases the medical product, and then the cash collection and partial cash partial online payment form collection are utilized, so that the economic loss to the consumer is prevented, and the normal operation of the medical market is further influenced.
Drawings
In order that the contents of the present invention may be more clearly understood, a description will be given below of specific embodiments according to the present invention in conjunction with the accompanying drawings,
The present invention will be described in further detail.
Fig. 1 is a flow chart of the intelligent area monitoring method of the present invention.
Fig. 2 is a flow chart of a method of generating a training set of the present invention.
Fig. 3 is a flowchart of a payment data preset processing method of the present invention.
FIG. 4 is a flow chart of the preset feature training method of the present invention.
Fig. 5 is a flow chart of a method of establishing a pharmaceutical product safety supervision model according to the present invention.
FIG. 6 is a flow chart of a model data enhancement optimization method of the present invention.
FIG. 7 is a flow chart of the model ensemble learning optimization method of the present invention.
Fig. 8 is a flowchart of a second method for determining a payment result of a medical product according to the present invention.
Fig. 9 is a flowchart of a third method for determining a payment result of a medical product according to the present invention.
FIG. 10 is a schematic diagram of the connection of the intelligent zone monitoring system of the present invention.
Detailed Description
The present invention will be further described with reference to the accompanying drawings and specific examples, which are not intended to be limiting, so that those skilled in the art will better understand the invention and practice it.
In the description of the present invention, it is to be understood that the term "comprising" is intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements but may, optionally, include other steps or elements not listed or inherent to such process, method, article, or apparatus.
Example 1
Referring to fig. 1-5 and 10, the present invention provides an embodiment of an intelligent area monitoring method based on a medical product supervision model, the method includes the following steps:
step S1: generating a training set by adopting a sliding window method;
in step S1, referring to fig. 2, a method for generating a training set by specifically adopting a sliding window method is as follows:
step S10: collecting payment data of a preset time sequence;
step S11: using sliding window method to lengthAccording to the length of +.>And (3) performing time window cutting to generate a training set:
；
the payment data of the preset time sequence refers to the time sequence data which is set by an operator and contains the payment data of the specified time interval;for the length of the time series data, +.>Is the number of samples, and each sample contains +.>Input features and an output tag, +.>Payment data representing the last month in the time window;
wherein, sliding windowThe method divides the time series data containing the payment data into a plurality of windows, each window contains continuous time series data, in the payment data of the medical product, the payment data which are continuous for a plurality of months can be used as a time window, and then the front part of each time window is used The payment data of a month is used as an input characteristic, and the payment data of the next month of the time window is used as a label.
Step S2: the training set is continuously trained by adopting a long-and-short-term memory neural network model to obtain corrected medicine product payment data, and the method comprises the following steps: presetting payment data generated by a sliding window method to form related data of medical products, extracting preset features from the related data, training the preset features by adopting a cross-validation method, and obtaining corrected payment data of the medical products;
in step S2, the long-short-term memory neural network model includes, but is not limited to, constructing by using a deep learning framework such as Keras, tensorFlow, pyTorch, caffe, and the deep learning framework is specifically set by an operator according to the actual model requirement and cost; wherein the construction process is for example: importing a deep learning frame; constructing an LSTM (long and short term memory neural network) model: constructing an LSTM layer, setting the quantity of neurons, setting activation functions (including but not limited to Sigmoid, reLU, tanh activation functions, setting by operators according to actual model requirements and cost), setting the shape of input data, namely the quantity of time windows and input dimensions, constructing a full-connection layer, setting output dimensions, setting the activation functions of an output layer (including but not limited to softmax, sigmoid, reLU, tanh activation functions, setting by operators according to actual model requirements and cost), setting loss functions (referring to cross entropy loss functions), setting optimizers (referring to Adam optimizers), setting evaluation indexes, setting the input and output of training sets, setting the size of batches, setting the number of times of training, and setting the input and output of test sets;
When training is performed, performing preset processing on the generated training set, extracting designated features, training aiming at the designated features, and generating corrected data;
specifically, referring to fig. 3, the method for performing preset processing on the payment data includes:
step S20: and processing payment data by adopting a data cleaning method:
；
wherein , is +.>Sample number-> and />Is the data range, +.>Is the data after washing if +.>Beyond the data range->Setting it as a missing value;
step S21: carrying out data standardization processing on payment data after data cleaning:
；
wherein , is +.f. in the payment data after cleaning>Sample No. H>Personal characteristics (I)>Is normalized payment data, +.> and />Are respectively->Minimum and maximum values of the individual features.
Step S22: and carrying out feature extraction on payment data after data normalization:
；
wherein , is the first part of the extraction>Personal characteristics (I)>Is a feature extraction function, ++>Is the characteristic of the cleaned payment data;
step S23: data aggregation is carried out on the extracted features to generate related data of medical products:
；
wherein , is the data related to the medical products after feature aggregation, < > >Is a function of the aggregation and,
is the feature data that needs to be aggregated.
Specifically, the method for extracting the preset features from the related data comprises the following steps:
step S24: extracting the price of the medical product and the payment amount characteristics of the medical product of the consumer from the related data of the medical product:
；
wherein , representing the price characteristics of the medical product->Characterizing the amount paid by the consumer pharmaceutical product +.>、A comprehensive function representing the characteristics of the price of the pharmaceutical product and the payment amount of the pharmaceutical product of the consumer, and +.>Representing a pharmaceutical product
Related data; other characteristics (such as medicine payment mode characteristics and medicine characteristics, which are set by operators according to actual supervision requirements and cost) in the medicine related data can be extractedRepresenting the extracted total features;
specifically, referring to fig. 5, the preset features are trained by adopting a cross-validation method, and a specific training process is as follows:
step S25: to extract the total characteristicsRandomly divide into->Parts, wherein each part is the same size and +.>The parts are used as training sets, and the rest 1 part is used as a test set;
step S26: for each featureTesting is carried out on the test sets respectively;
step S27: repeating step S26 until each data is used as a test set to obtain The accuracy rate;
step S28: for each featureWill->The average of the individual accuracies was taken as its cross-validation accuracy:
；
wherein ,indicate->Accuracy of secondary verification；
Step S29: comparing each featureAnd selecting the feature with the highest accuracy as the final feature.
The specific process can be expressed as:
；
wherein , method for representing sliding window->For original payment data->For the length of the time series data, +.>For the preset treatment, < >>For payment data generated by sliding window method, < >>For feature extraction, cryptophan Diels>Is the related data of the medical products after the preset treatment,for cross-validation->For the extracted feature data, < > for>To divide training setTest set(s)>To train LSTM model, < >>To consolidate the multiple cross-validated predictions, +.>For the training set and the test set obtained after the cross-validation, < + >>Is +.>Secondary verification (Tech)>To divide the training set into training set and verification set, < >>For the modified data obtained by LSTM model training of the training set,/for the training set>To->And merging the prediction results obtained by the secondary cross validation to obtain a final prediction result.
Step S3: establishing a medical product supervision model, judging consumer payment data in a preset area, and obtaining a consumer medical product payment result;
In step S3, referring to fig. 6, the method for establishing a supervision model of a pharmaceutical product is as follows:
step S30: the long-short-term memory neural network model is evaluated on a test set, and a sensitivity score of each weight is generated:
；
wherein , indicate->Personal weight(s)>Is a loss function;
step S31: calculating an average value of sensitivity scores of each LSTM layer of the pharmaceutical product safety supervision model, and generating an average score vector:
；
wherein , is->A set of subscripts for all weights of a layer;
step S32: the average score vectors are sorted in ascending order, and the proportion of each score to all scores is calculated to generate a proportion vector:
；
wherein , is->Is>An element;
step S33: calculating the weight quantity to be reserved according to the pruning proportion:
；
wherein , is pruning proportion;
step S34: according to the proportion vector and the weight quantity, selecting the weight with the highest score as a reserved weight:
；
wherein , is->Is>A maximum value;
step S35: pruning the medical product supervision modelThe values of all weights except for 0;
step S36: repeating the steps S31-S35 for a preset number of times to generate an optimized long-short-time memory neural network model, and combining the long-short-time memory neural network model with a medical product price monitoring terminal in a preset area to form a medical product monitoring model;
The neural network model is memorized for a long time after model pruning optimization, so that the calculated amount and the memory amount can be reduced, and the model is simplified; the consumer medicine product payment result includes, but is not limited to, medicine product type, quantity, unit price, total price, payment mode (such as online medical insurance payment, online payment of a payment platform, online payment of online bank, etc.), money receiving end data, medicine product data corresponding to a money receiving end (medicine product type, quantity, unit price, total price), wherein the money receiving end refers to a seller terminal; the preset area is set by the operator according to the actual supervision requirement and cost, including but not limited to province, city, district, village and street, and in this embodiment, the area range of the city is referred to.
Step S4: comparing the payment result of the consumer medical product with the medical product limit data of the preset area, and judging whether the payment of the consumer medical product exceeds the medical product limit range of the preset area;
in step S4, when comparing, the price of the corresponding price for the consumer to purchase the medical product is compared with the price limit of the medical product in the preset area, and whether the price of the consumer exceeds the price limit to purchase the medical product is determined.
Step S5: if yes, collecting the data of the collecting end contained in the consumer payment data, and simultaneously feeding back the abnormal information of the price of the medical product to the matched consumer payment end, if not, returning to the step S3, and judging the consumer payment data of the preset area again;
in step S5, when the consumer purchases the medical product beyond the limit, the abnormal information of the price of the medical product is fed back to the payment end corresponding to the consumer so as to remind the consumer of purchasing the medical product beyond the limit, and the information is reserved so as to facilitate the maintenance of the right; the payment end refers to terminal equipment of a consumer.
Step S6: feeding back abnormal information of the price of the medical product containing the data of the collecting end to a medical product price supervision terminal of a preset area;
in step S6, the pharmaceutical product price monitoring terminal is designated by an operator as a department terminal having pharmaceutical product monitoring authority; the abnormal price information of the medical products includes, but is not limited to, the type, quantity, unit price, total price, payment mode (such as online payment of medical insurance, online payment of a payment platform, online payment of online bank, etc.), data of the receiving end and data of the medical products corresponding to the receiving end (type, quantity, unit price, total price of the medical products).
When the medical supervision model is constructed and information is fed back to the medical product price supervision terminal, relevant laws and regulations and privacy protection regulations are required to be complied with, and the collected and processed data are ensured not to reveal the privacy information of consumers.
Example two
Referring to fig. 6 and 7, the second embodiment is basically the same as the first embodiment except that: the method for optimizing the long-short-term memory neural network model further comprises the following steps:
step S37: and (3) performing regular optimization on the long-short-term memory neural network model:
；
wherein , for the length of the time series data, +.>Is a binary mask vector, each element +.>Equal probability value is 0 or 1, wherein +.>The representation will->The%>Individual element discard->Representing a reservation;
the mask vectorThe acquisition process of (1) is as follows: />Wherein->The proportion of elements to be discarded is indicated as a super parameter, preferably 0.2 or 0.5 in this embodiment.
Specifically, referring to fig. 6, the method for optimizing the long-short-term memory neural network model further includes:
step S38: data enhancement operation set for setting long-short-time memory neural network modelWherein the data enhancement operation set +. >Including but not limited to rotation, translation, scaling, and noise addition operations;
step S39: for each original training sample of long-short-term memory neural network model training setRandom slave data enhancement operation set +.>Is selected from a group of operations of: /> ； wherein />Paying data for original medical products;
step S40: for each operationOriginal training sample->Applied to operations->Generating a new sample->；
Step S41: all samples to be generatedAdding the enhanced training set, and further performing model training on the enhanced training set;
wherein each training sample is enhanced during the optimization processSecond time, the final enhanced training set size is therefore +.>For the situation that the safety supervision data of the medical products is less, the training set can be expanded by the data enhancement method.
Specifically, referring to fig. 7, the method for optimizing the long-short-term memory neural network model further includes:
step S42: dividing the long-short-term memory neural network model training set intoThe non-overlapping portions:
；
step S43: for each ofTraining->The LSTM model:
。
step S44: for each test sampleUse +.>The individual model gets- >The following results: />
。
Step S45: will beThe individual results are concatenated to form a matrix:
。
step S46: for all test samplesMatrix +.>Spliced together to form a large matrix:
；
step S47: will be large matrixInput into the Meta Model, output the final result:
；
the Meta Model comprises, but is not limited to, linear regression, logistic regression, decision trees, random forests and gradient lifting tree (Gradient Boosting Tree) models, and is specifically set by operators according to actual optimization requirements and cost; the accuracy and the robustness of the long-and-short-term memory neural network model are improved by adopting the integrated learning mode.
Example III
Referring to fig. 8 and 10, the third embodiment is substantially the same as the first embodiment except that: when judging the consumer payment data of the preset area, the method further comprises the following steps:
step S300: judging the consumer payment data of the preset area, and collecting the collection end data contained in the consumer payment data, so as to obtain the consumer medicine product payment result and the collection end collection data in the preset time interval;
in step S300, the preset time interval in this embodiment is set by the operator according to the actual supervision time interval requirement.
Step S301: judging whether the payment result of the consumer medical product is matched with the collection data of the collection end;
step S302: if yes, returning to the step S300, judging the consumer payment data of the preset area again, and if not, further judging whether the consumer medicine product payment result is higher than the collection data of the collection end;
in step S302, if it is determined that the payment result of the consumer pharmaceutical product is higher than the collection data of the collection end, it is indicated that the collection end, i.e. the pharmaceutical product data entered by the seller into the system may not be identical to the pharmaceutical product purchased by the consumer, for example: inputting false and expired medical product data into a system, and then enabling consumers to pay the difference price independently by using other payment modes; the price of the medical product is virtually reported, so that a consumer pays more fees, and then the other consumers can automatically intercept the input system with the same limit price of the medical product in the fees paid by the consumer; the packaging specification and the quantity of the medical products are adjusted, namely, the unit price is lower than the limit standard, but the total price is higher than the limit standard by reducing the quantity of each medical product or increasing the packaging specification; or, the variety and specification of the medical products are regulated, namely, in the sales of the price-limiting medical products, the seller can be replaced by the high-end specification or high-price variety of the same type of medical products, so that the selling price is improved; the price of the medical products is changed, namely, other fees or additional values, such as service fees, delivery fees, detection fees, gifts and the like, are added on the selling price of the medical products.
Step S303: if yes, the medical product price monitoring terminal of the preset area is fed back with the medical product price abnormal information containing the collection end data, if not, the step S300 is returned, and the customer payment data of the preset area is judged again.
By adopting the technical scheme, when a consumer purchases the medical product, the seller can be identified to replace the type and the quantity of the medical product purchased by the consumer by other illegal means or holes, so that the medical product data input by the seller into the system is prevented from being inconsistent with the medical product purchased by the consumer, or the seller sells false and expired medical products to the consumer, and the health and the life safety of the consumer are influenced.
Example IV
Referring to fig. 9 to 10, the fourth embodiment is substantially the same as the first embodiment except that: after judging that the payment result of the consumer medical product is higher than the collection data of the collection end, the method further comprises the following steps:
step S304: judging whether the payment result of the consumer medical product is lower than the collection data of the collection end;
step S305: if yes, acquiring historical collection data of a collection end, acquiring cash income data in the historical collection data, and if not, returning to the step S304 to judge the payment result of the consumer medical product again;
In step S305, when it is determined that the payment result of the consumer' S medical product is lower than the collection data of the collection end, it is indicated that the seller can make the consumer pay by cash payment, part of cash and part of online payment.
Step S306: judging whether the ratio of the cash income data in the historical collection data exceeds a preset value or not;
in step S306, the specific preset value is set by the operator according to the actual supervision requirement.
Step S307: if yes, the medical product price monitoring terminal of the preset area is fed back with medical product price abnormality information containing the data of the collection end, if not, the step S304 is returned, and the payment result of the medical product of the consumer is judged again.
If the consumer pays by cash payment, part of cash and part of online payment, the source and the flow direction of the medical product are difficult to track, and the quality safety problem of the medical product is easy to cause, namely, unlike the online payment, the cash transaction has clear transaction records, the source and the flow direction of the purchased medical product cannot be tracked, and the problems of counterfeit and expired medical product and the like possibly exist, so that the health of the consumer is endangered; and, be unfavorable for the maintenance of consumer's equity, namely cash transaction is difficult to produce effective proof of purchase, and the consumer is difficult to provide evidence when the right is held, leads to consumer's equity to be difficult to obtain effective guarantee.
By adopting the technical scheme, when a consumer purchases the medical product, the seller can improve the selling price of the medical product through the false alarm price means, and then cash collection and partial cash partial online payment form collection are utilized, so that economic loss to the consumer is prevented, and the normal operation of the medical market is further influenced.
Example five
The invention also provides a computer medium, wherein the computer medium is stored with a computer program, and the computer program is executed by a processor to realize the intelligent area monitoring method based on the medical product supervision model.
The invention also provides a computer, comprising the computer medium.
Example six
Referring to fig. 1-10, the present invention further provides an embodiment of an intelligent area monitoring system based on a medical product supervision model, including:
the data processing module is used for generating training by adopting a sliding window method, and continuously training a training set by adopting a long-and-short-term memory neural network model to obtain corrected medicine product payment data, and comprises the following steps: presetting payment data generated by a sliding window method to form related data of medical products, extracting preset features from the related data, training the preset features by adopting a cross-validation method, and obtaining corrected payment data of the medical products;
The supervision module is used for embedding the optimized long-short-term memory neural network model into a medical product price supervision terminal of a preset area and establishing a medical product supervision model;
the price limiting processing module is used for judging the consumer payment data of the preset area and obtaining the payment result of the consumer medical product;
comparing the payment result of the consumer medical product with the medical product limit data of the preset area, and judging whether the payment of the consumer medical product exceeds the medical product limit range of the preset area; if yes, collecting the data of the collecting end contained in the consumer payment data, feeding back the abnormal information of the price of the medical product to the matched consumer payment end, and if not, judging the consumer payment data of the preset area again;
the abnormal feedback module is used for feeding back abnormal information of the price of the medical product containing the data of the collection end to the medical product price supervision terminal in the preset area.
It is apparent that the above examples are given by way of illustration only and are not limiting of the embodiments. Other variations and modifications of the present invention will be apparent to those of ordinary skill in the art in light of the foregoing description. It is not necessary here nor is it exhaustive of all embodiments. And obvious variations or modifications thereof are contemplated as falling within the scope of the present invention.

Claims (9)

1. An intelligent area monitoring method based on a medical product supervision model is characterized by comprising the following steps of:
step S1: generating a training set by adopting a sliding window method;
step S2: the training set is continuously trained by adopting a long-and-short-term memory neural network model to obtain corrected medicine product payment data, and the method comprises the following steps: presetting payment data generated by a sliding window method to form medical product related data, extracting preset features from the medical product related data, training the preset features by adopting a cross verification method, and obtaining corrected medical product payment data;
step S3: establishing a medical product safety supervision model, judging consumer payment data of a preset area, and obtaining a consumer medical product payment result;
step S4: comparing the consumer medical product payment result with the medical product limit data of the preset area, and judging whether the consumer medical product payment result exceeds the medical product limit range of the preset area;
step S5: if yes, collecting end data contained in the consumer payment data, and simultaneously feeding back abnormal information of the price of the medical product to the matched consumer payment end, and if not, returning to the step S3;
Step S6: feeding back abnormal information of the price of the medical product containing the data of the collecting end to a medical product price supervision terminal of a preset area;
the method for establishing the medicine product safety supervision model comprises the following steps:
step S30: the long-short-term memory neural network model is evaluated on a test set, and a sensitivity score of each weight is generated:
；
wherein ,indicate->Personal weight(s)>Is a loss function;
step S31: calculating the average value of the sensitivity scores of each LSTM layer of the medical product safety supervision model to generate an average score vector;
step S32: the average score vectors are sequenced in ascending order, and the proportion of each sensitivity score to all sensitivity scores is calculated to generate a proportion vector;
step S33: calculating the weight quantity to be reserved according to the pruning proportion;
step S34: selecting the weight with the highest score as a reserved weight according to the proportion vector and the weight quantity;
step S35: performing model pruning of the medical product safety supervision model, and setting the value of all weights except the reserved weight to be 0;
step S36: repeating the steps S31-S35 for a preset number of times, generating an optimized long-short-time memory neural network model, and building the optimized long-short-time memory neural network model into a medicine product price supervision terminal of a preset area to establish a medicine product safety supervision model.
2. The intelligent area monitoring method based on the medical product supervision model according to claim 1, wherein the method for generating the training set by adopting the sliding window method is as follows:
step S10: collecting payment data of a preset time sequence;
step S11: and cutting the payment data with the length of T according to the time window with the length of T by adopting a sliding window method, and generating a training set.
3. The intelligent area monitoring method based on the medical product supervision model according to claim 1, wherein the method for performing preset processing on the payment data is as follows:
step S20: and processing payment data by adopting a data cleaning method:
；
wherein ,is the +.>Sample number-> and />Is the data range, +.>Is the payment data after cleaning, ifBeyond the data range->Setting it as a missing value;
step S21: carrying out data standardization processing on payment data after data cleaning:
；
wherein ,is +.f. in the payment data after cleaning>Sample No. H>Personal characteristics (I)>Is the payment data after normalization, and />Are respectively->Sample No. H>Minimum and maximum values of the individual features;
Step S22: extracting characteristics of payment data after data normalization;
step S23: and carrying out data aggregation on the extracted features to generate related data of the medical products.
4. The intelligent area monitoring method based on a medical product supervision model according to claim 3, wherein the method for extracting preset features from the medical product related data is as follows:
step S24: extracting the price characteristics of the medical products and the payment amount characteristics of the consumer medical products in the related data of the medical products:
;
wherein ,representing the price characteristics of the medical product->Characterizing the amount paid by the consumer pharmaceutical product +.>、/>Representing the integrated function of extracting the price characteristics of the pharmaceutical product and the payment amount characteristics of the consumer pharmaceutical product, ++>Representing data related to the pharmaceutical product.
5. The intelligent area monitoring method based on the medical product supervision model according to claim 4, wherein the method for training the preset features is as follows:
step S25: to extract the total characteristicsRandomly divide into->Parts, wherein each part is the same size and +.>The parts are used as training sets, and the rest 1 part is used as a test set;
step S26: for each feature, testing on a test set;
Step S27: repeating step S26 until each feature is used as a test set to obtainThe accuracy rate;
step S28: for each feature, willThe average of the individual accuracies was taken as its cross-validation accuracy:
；
wherein ,indicate->Accuracy of secondary verification;
step S29: and comparing the cross verification accuracy of each feature, and selecting the feature with the highest accuracy as the final feature.
6. The intelligent area monitoring method based on the medical product supervision model according to claim 1, wherein when judging consumer payment data of a preset area, the method further comprises the following steps:
step S300: judging the consumer payment data of the preset area, and collecting the collection end data contained in the consumer payment data, so as to obtain the consumer medicine product payment result and the collection end collection data in the preset time interval;
step S301: judging whether the payment result of the consumer medical product is matched with the collection data of the collection end;
step S302: if yes, returning to the step S300, judging the consumer payment data of the preset area again, and if not, further judging whether the consumer medicine product payment result is higher than the collection data of the collection end;
Step S303: if yes, the medical product price monitoring terminal of the preset area is fed back with the medical product price abnormal information containing the collection end data, if not, the step S300 is returned, and the customer payment data of the preset area is judged again.
7. The intelligent area monitoring method based on a pharmaceutical product supervision model according to claim 6, further comprising the steps of:
step S304: judging whether the payment result of the consumer medical product is lower than the collection data of the collection end;
step S305: if yes, acquiring historical collection data of a collection end, acquiring cash income data in the historical collection data, and if not, returning to the step S304 to judge the payment result of the consumer medical product again;
step S306: judging whether the ratio of the cash income data in the historical collection data exceeds a preset value or not;
step S307: if yes, the medical product price monitoring terminal of the preset area is fed back with medical product price abnormality information containing the data of the collection end, if not, the step S304 is returned, and the payment result of the medical product of the consumer is judged again.
8. A computer medium having stored thereon a computer program for execution by a processor to implement a method of intelligent region monitoring based on a pharmaceutical product supervision model according to any one of claims 1-7.
9. An intelligent area monitoring system based on a medical product supervision model, which is characterized by comprising:
the data processing module is used for generating training by adopting a sliding window method, and continuously training a training set by adopting a long-and-short-term memory neural network model to obtain corrected medicine product payment data, and comprises the following steps: presetting payment data generated by a sliding window method to form medical product related data, extracting preset features from the medical product related data, training the preset features by adopting a cross verification method, and obtaining corrected medical product payment data;
the supervision module is used for embedding the optimized long-short-term memory neural network model into a medical product price supervision terminal of a preset area and establishing a medical product safety supervision model;
the price limiting processing module is used for judging the consumer payment data of the preset area and obtaining the payment result of the consumer medical product; comparing the consumer medical product payment result with the medical product limit data of the preset area, and judging whether the consumer medical product payment result exceeds the medical product limit range of the preset area; if yes, collecting end data contained in the consumer payment data, and feeding back abnormal information of the price of the medical product to the matched consumer payment end;
The abnormal feedback module is used for feeding back abnormal information of the price of the medical product containing the data of the collection end to the medical product price supervision terminal in the preset area;
wherein, the supervision module is further configured to:
step S30: the long-short-term memory neural network model is evaluated on a test set, and a sensitivity score of each weight is generated:
；
wherein ,indicate->Personal weight(s)>Is a loss function;
step S31: calculating the average value of the sensitivity scores of each LSTM layer of the medical product safety supervision model to generate an average score vector;
step S32: the average score vectors are sequenced in ascending order, and the proportion of each sensitivity score to all sensitivity scores is calculated to generate a proportion vector;
step S33: calculating the weight quantity to be reserved according to the pruning proportion;
step S34: selecting the weight with the highest score as a reserved weight according to the proportion vector and the weight quantity;
step S35: performing model pruning of a medical product safety supervision model, and setting the value of all weights except for the model pruning to be 0;
step S36: repeating the steps S31-S35 for a preset number of times, generating an optimized long-short-time memory neural network model, and building the optimized long-short-time memory neural network model into a medicine product price supervision terminal of a preset area to establish a medicine product safety supervision model.
CN202310744089.1A 2023-06-25 2023-06-25 Intelligent area monitoring method and system based on medical product supervision model Active CN116525060B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310744089.1A CN116525060B (en) 2023-06-25 2023-06-25 Intelligent area monitoring method and system based on medical product supervision model

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310744089.1A CN116525060B (en) 2023-06-25 2023-06-25 Intelligent area monitoring method and system based on medical product supervision model

Publications (2)

Publication Number Publication Date
CN116525060A CN116525060A (en) 2023-08-01
CN116525060B true CN116525060B (en) 2023-10-03

Family

ID=87406594

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310744089.1A Active CN116525060B (en) 2023-06-25 2023-06-25 Intelligent area monitoring method and system based on medical product supervision model

Country Status (1)

Country Link
CN (1) CN116525060B (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109598630A (en) * 2018-12-04 2019-04-09 平安医疗健康管理股份有限公司 The monitoring and managing method and relevant device of fixed point medical institution based on living things feature recognition
WO2020119175A1 (en) * 2018-12-13 2020-06-18 平安医疗健康管理股份有限公司 Method for monitoring medical expense abnormalities, monitoring server and storage medium
CN111784453A (en) * 2020-06-29 2020-10-16 平安国际智慧城市科技股份有限公司 Block chain-based cross-platform medicine collection price synchronization method and related device
CN114037541A (en) * 2021-11-05 2022-02-11 湖南创研科技股份有限公司 Fixed-point medicine institution supervision method based on biological feature recognition and related equipment
CN115130966A (en) * 2022-07-12 2022-09-30 广东省药品交易中心有限公司 Medicine and medical consumable purchasing platform and purchasing method thereof
CN115762705A (en) * 2022-11-18 2023-03-07 厦门易联众易方科技有限公司 Dual-channel drug service management platform, and establishment method, device and equipment thereof
CN116168410A (en) * 2023-04-21 2023-05-26 江苏羲辕健康科技有限公司 Medicine box information identification method and system based on neural network

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140358578A1 (en) * 2013-05-31 2014-12-04 American Pharmacotherapy, Llc System and method for comparing pharmaceutical prices and medication utilization
US11023979B2 (en) * 2014-03-24 2021-06-01 Ventegra, Inc. Reference price index for pharmaceutical products
US11861637B2 (en) * 2016-09-29 2024-01-02 Clear Demand, Inc. System of demand modeling and price calculation based on interpolated market price elasticity functions

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109598630A (en) * 2018-12-04 2019-04-09 平安医疗健康管理股份有限公司 The monitoring and managing method and relevant device of fixed point medical institution based on living things feature recognition
WO2020114221A1 (en) * 2018-12-04 2020-06-11 平安医疗健康管理股份有限公司 Biometric recognition-based designated medical institution supervision method and related device
WO2020119175A1 (en) * 2018-12-13 2020-06-18 平安医疗健康管理股份有限公司 Method for monitoring medical expense abnormalities, monitoring server and storage medium
CN111784453A (en) * 2020-06-29 2020-10-16 平安国际智慧城市科技股份有限公司 Block chain-based cross-platform medicine collection price synchronization method and related device
CN114037541A (en) * 2021-11-05 2022-02-11 湖南创研科技股份有限公司 Fixed-point medicine institution supervision method based on biological feature recognition and related equipment
CN115130966A (en) * 2022-07-12 2022-09-30 广东省药品交易中心有限公司 Medicine and medical consumable purchasing platform and purchasing method thereof
CN115762705A (en) * 2022-11-18 2023-03-07 厦门易联众易方科技有限公司 Dual-channel drug service management platform, and establishment method, device and equipment thereof
CN116168410A (en) * 2023-04-21 2023-05-26 江苏羲辕健康科技有限公司 Medicine box information identification method and system based on neural network

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
Evaluating Price and Availability of Essential Medicines in China: A Mixed Cross-Sectional and Longitudinal Study;Caijun Yang等;Frontiers in Pharmacology;第11卷;第1-11页 *
威海定点民营医疗机构价格管理情况调研;毕兴全等;中国医疗保险(第05期);第32-35页 *
毕兴全 ; 王晓华 ; .威海定点民营医疗机构价格管理情况调研.中国医疗保险.2020,(05),第32-35页. *
网络健康信息综合聚合模式构建研究;廖花林等;情报理论与实践;第43卷(第09期);第137-143页 *

Also Published As

Publication number Publication date
CN116525060A (en) 2023-08-01

Similar Documents

Publication Publication Date Title
US11734692B2 (en) Data breach detection
Cumming et al. Disentangling crowdfunding from fraudfunding
US10997599B2 (en) Method for detecting merchant data breaches with a computer network server
US20190130407A1 (en) Real-time cross-channel fraud protection
CN103294592B (en) User instrument is utilized to automatically analyze the method and system of the defect in its service offering alternately
CN110689438A (en) Enterprise financial risk scoring method and device, computer equipment and storage medium
CN110246030A (en) In many ways risk management method, terminal, device and storage medium after the loan to link
US20180130071A1 (en) Identifying and labeling fraudulent store return activities
CN106415576A (en) System for the measurement and automated accumulation of diverging cyber risks, and corresponding method thereof
Perols et al. Information fusion in continuous assurance
Kuenkaikaew Predictive Audit Analytics: Evolving to a new era
Gupta et al. Audit selection strategy for improving tax compliance: application of data mining techniques
CN111583033A (en) Association analysis method and device based on relation between listed company and stockholder
CN113609193A (en) Method and device for training prediction model for predicting customer transaction behavior
CN113052705A (en) Supply chain financial service process management method and device
Dawe Conducting national money laundering or financing of terrorism risk assessment
Beltzung et al. Real-time detection of fake-shops through machine learning
Brand Corporate whistleblowing, smart regulation and regtech: The coming of the whistlebot?
FR3048840A1 (en)
CN113434575B (en) Data attribution processing method, device and storage medium based on data warehouse
Dai et al. Audit analytics: A field study of credit card after-sale service problem detection at a major bank
CN116525060B (en) Intelligent area monitoring method and system based on medical product supervision model
Ferreira et al. Fraud in organic foods
KR102499182B1 (en) Loan regular auditing system using artificia intellicence
CN115564591A (en) Financing product determination method and related equipment

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
GR01 Patent grant
GR01 Patent grant