A method and system for managing inventory information of medical supplies

Abstract

This invention relates to the field of data processing, specifically to a method and system for managing inventory information of medical supplies. It extracts various habitual features from complete historical material requisition data. Simultaneously, it verifies current material requisition records. If incomplete records exist, such as missing information or illegible handwriting, the method matches the habitual features with known information in the requisition record. If a match is found, the incomplete record is completed using the target habitual feature and its confidence level. This application utilizes historical data analysis to obtain requisition habitual features and uses these features to automatically repair missing registration data, effectively reducing the verification workload for management personnel.

Description

Technical Field

[0001] This invention relates to the field of data processing, specifically to a method and system for managing inventory information of medical supplies. Background Technology

[0002] Inventory management of medical supplies is a complex but crucial process that directly impacts the quality and efficiency of healthcare services. Effective inventory management ensures a safe and efficient supply of medical supplies while minimizing waste. Most existing inventory management systems rely on information technology, recording information such as the receipt, issuance, and inventory checks of medical supplies into the management system. This facilitates usage registration, inventory alerts, and other management functions.

[0003] Currently, warehouse material requisition and outbound information registration generally relies on manual entry. However, with the transformation of warehouses towards informatization, this information will be transferred to computer management systems. Manually entered records, lacking mandatory management mechanisms, are prone to incompleteness and illegible handwriting. This results in incomplete information when the data from the register is entered into the information management system. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a method and system for managing inventory information of medical supplies, so as to solve the problems in the background art.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] The present invention provides a method for managing inventory information of medical supplies, comprising the following steps:

[0007] Obtain material requisition records for the current management cycle and historical material requisition data, wherein the material requisition records are paper materials and the historical material requisition data are electronic data;

[0008] Extract various material requisition habit features and their confidence levels from the historical material requisition data;

[0009] The material requisition record is subjected to image and text recognition to obtain preliminary material requisition data; and the preliminary material requisition data is verified to obtain incomplete records.

[0010] Based on the known data of the incomplete record, the incomplete record is matched with the various material requisition habit features. When the incomplete record matches any target material requisition habit feature, the incomplete record is completed based on the target material requisition habit feature and the confidence level of the target material requisition habit feature.

[0011] In one embodiment of this application, the historical material requisition data includes multiple material requisition records, each of which includes values ​​for multiple parameters. The extraction of multiple material requisition habit features and confidence levels for these habits from the historical material requisition data includes:

[0012] Remove useless parameters from historical material requisition data to obtain the target parameters;

[0013] Combine any two parameters to obtain the sample data template. ,in, Indicates the first target parameter. Indicates the second objective parameter;

[0014] Based on the sample data template Extract multiple sample data from the historical material requisition data. , Indicates the first The value of the first target parameter, Indicates the first The value of the second objective parameter;

[0015] For multiple sample data Clustering was performed to obtain multiple clusters. ,in, The cluster number;

[0016] Based on the multiple clusters Verify the characteristics of material requisition habits and the confidence level of material requisition habits.

[0017] In one embodiment of this application, based on the plurality of clusters Verify the characteristics of material requisition habits and the confidence level of these habits, including:

[0018] For the multiple clusters Labeling is performed to obtain the multiple clusters. Category label of the first target parameter Category label of the second target parameter ;

[0019] Statistical clusters Number of samples within a cluster Conforms to category labels Number of sample data and conforming to category labels Number of sample data ;

[0020] Calculate separately Number of samples within a cluster With category label Number of sample data ratio Number of samples within a cluster With category label Number of sample data ratio ;

[0021] In ratio When the proportion exceeds a preset threshold, the category label is determined. exist The characteristics of material requisition habits; in the ratio When the proportion exceeds a preset threshold, the category label is determined. exist The material requisition habits are characterized; and the ratio is used as the confidence level of the material requisition habits.

[0022] In one embodiment of this application, based on the sample data template Extract multiple sample data from the historical material requisition data. ,include:

[0023] When the sample data template When none of the parameters are medical supply names, the data is based on the sample data template. Sample data is constructed using the values ​​of parameters belonging to the same material requisition record. ;

[0024] When the sample data template When the stored parameter is the name of a medical supply, the name of the medical supply is vectorized and processed based on the sample data template. Sample data is constructed using the values ​​of parameters belonging to the same material requisition record. .

[0025] In one embodiment of this application, multiple sample data are... Clustering was performed to obtain multiple clusters. ,include:

[0026] When both the first and second target parameters are numeric or vector types, construct a two-dimensional parameter. and two-dimensional parameters Density clustering was performed to obtain multiple clusters. ;

[0027] When one of the first target parameter and the second target parameter is a character type, and the other is a numeric or vector type; the target parameter based on the character type is used for multiple sample data. The data is divided into multiple data units; within each data unit, multiple sample data are processed based on either numeric or vector type. Clustering was performed to obtain multiple clusters. .

[0028] In one embodiment of this application, for the plurality of clusters Labeling is performed to obtain the multiple clusters. Category label of the first target parameter Category label of the second target parameter ,include:

[0029] When either the first or second target parameter is a numeric type, calculate the average value of the target parameters with numeric data types within the cluster. and standard deviation and based on the average value and the standard deviation Build reference range labels ,in, For range adjustment parameters;

[0030] When the first target parameter or the second target parameter is a vector type, label the target parameter whose numerical type is a vector.

[0031] When the first target parameter or the second target parameter is of type character, the character itself is used as the label for the target parameter whose data type is character.

[0032] In one embodiment of this application, matching the incomplete record with the various material requisition habit characteristics based on the known data of the incomplete record includes:

[0033] The known data of the incomplete records are matched with the labels of the various material requisition habit features, and the material requisition habit features whose labels match the known data are taken as the target material requisition habit features.

[0034] In one embodiment of this application, the incomplete record is completed based on the target material requisition habit characteristics and the confidence level of the target material requisition habit characteristics, including:

[0035] Based on the target material requisition habit characteristics and known data, candidate labels for incomplete records are determined. and candidate tags confidence level , ,in, For the first Confidence level corresponding to the target material requisition habit characteristics of known data;

[0036] Filter out target labels with a confidence level greater than the confidence threshold;

[0037] The incomplete records are completed based on the target labels.

[0038] In one embodiment of this application, it further includes:

[0039] The candidate labels with the highest confidence levels Send to the corresponding recipient for confirmation.

[0040] This application provides a medical supplies inventory information management system, including:

[0041] The acquisition module is used to acquire the material requisition records and historical material requisition data for the current management cycle, wherein the material requisition records are paper materials and the historical material requisition data are electronic data;

[0042] The feature extraction module is used to extract various material requisition habit features and confidence levels of various material requisition habit features from the historical material requisition data;

[0043] The verification module is used to perform image and text recognition on the material requisition record to obtain preliminary material requisition data; and to verify the preliminary material requisition data to obtain incomplete records.

[0044] The data completion module is used to match the incomplete record with the various material requisition habit features based on the known data of the incomplete record, and when the incomplete record matches any target material requisition habit feature, to complete the incomplete record based on the target material requisition habit feature and the confidence level of the target material requisition habit feature.

[0045] The beneficial effects of this invention are as follows: This invention provides a method and system for managing inventory information of medical supplies, which extracts various habitual features from complete historical material requisition data. Simultaneously, it verifies current material requisition records. If incomplete records exist, such as missing information or illegible handwriting, the method matches the known information in the material requisition record with the habitual features. If a match is found, the incomplete record is completed using the target habitual features and their confidence level. This application utilizes historical data analysis to obtain habitual features and uses these features to automatically repair missing registration data, thereby effectively reducing the verification workload for management personnel. Attached Figure Description

[0046] The present invention will be further described below with reference to the accompanying drawings and embodiments:

[0047] Figure 1 This is a flowchart illustrating a method for managing inventory information of medical supplies in one embodiment of this application;

[0048] Figure 2 This is a flowchart illustrating the extraction of material requisition habit features in one embodiment of this application;

[0049] Figure 3 This is a structural diagram of a medical supplies inventory information management system shown in one embodiment of this application. Detailed Implementation

[0050] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0051] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the layers related to the present invention and are not drawn according to the actual number, shape and size of the layers in the actual implementation. In the actual implementation, the form, number and proportion of each layer can be arbitrarily changed, and the layer layout may also be more complex.

[0052] Numerous details are explored in the following description to provide a more thorough explanation of embodiments of the invention; however, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details.

[0053] Figure 1 This is a flowchart illustrating a method for managing inventory information of medical supplies in one embodiment of this application, as shown below. Figure 1 As shown, a method for managing inventory information of medical supplies according to this embodiment may include the following steps:

[0054] S110, Obtain the material requisition records and historical material requisition data for the current management cycle, wherein the material requisition records are paper materials and the historical material requisition data are electronic data;

[0055] Historical material requisition records refer to complete material requisition records that have been verified, while material requisition records for the current management cycle refer to all paper registration records within the current entry cycle that require material requisition information to be extracted through image recognition. Material requisition records are stored in the form of images.

[0056] S120, extract various material requisition habit features and confidence levels of various material requisition habit features from the historical material requisition data;

[0057] The following table is a material requisition record table in one embodiment of this application:

[0058]

[0059] The material requisition habit characteristics in this application refer to the high probability of simultaneous occurrence of any two material requisition parameters within their respective value ranges. For example, there is a high probability of requisitioning routine medical consumables such as masks, cotton swabs, and iodine solution between 9:00 and 10:00 AM. In other words, there exists a material requisition habit characteristic: {Time: 9:00-10:00; Type of material requisition: Routine medical consumables}. This application is not limited to the above habit characteristics, but also includes the requisitioner and requisition time, the requisitioner and requisition type, and the requisition type and quantity, etc.

[0060] Figure 2 This is a flowchart illustrating the extraction of material requisition habit features in one embodiment of this application, such as... Figure 2 As shown, specifically, this application employs the following process to extract multiple material requisition habit features and their confidence levels, including:

[0061] S121, Remove useless parameters from historical material requisition data to obtain target parameters;

[0062] For example, remove information such as serial number, department, specification / unit, and approver. This information does not reflect common patterns or can be directly derived from existing information (material requisitioner - unit; material name - unit / specification).

[0063] S122, combine any two parameters to obtain the sample data template. ,in, Indicates the first target parameter. Indicates the second objective parameter;

[0064] A sample data template refers to a template composed of two header parameters, such as material requisition time / material name, material requisition time / quantity, material name / quantity, etc.

[0065] S123, based on the sample data template Extract multiple sample data from the historical material requisition data. , Indicates the first The value of the first target parameter, Indicates the first The value of the second objective parameter;

[0066] After constructing the sample data template Next, the sample data template Compare the data with the table, and then extract the corresponding data samples.

[0067] However, clustering is required in subsequent processing. Therefore, the material names need to be vectorized. The specific process is as follows:

[0068] (1) When the sample data template When none of the parameters are medical supply names, the data is based on the sample data template. Sample data is constructed using the values ​​of parameters belonging to the same material requisition record. ;

[0069] If sample data template If any parameter is not a name of a medical supply (i.e., its data type is character or numeric), then its original information will be retained. Examples include the recipient, quantity, and time.

[0070] (2) When the sample data template When the stored parameter is the name of a medical supply, the name of the medical supply is vectorized and processed based on the sample data template. Sample data is constructed using the values ​​of parameters belonging to the same material requisition record. .

[0071] If the parameter is a name of a medical supply, it is vectorized. This application utilizes a pre-trained Word2vec model to vectorize the names of medical supplies, thereby ensuring a high cosine similarity between word vectors corresponding to medical supply names with similar functions and principles. The training process of the Word2vec model is as follows:

[0072] (1) Collect medical supply name data: Collect the names of medical supplies from open knowledge bases, official databases or professional websites in the medical field. Ensure that the data covers a wide range of medical supplies, especially those with similar functional principles.

[0073] (2) Data preprocessing: The collected medical supply names are segmented into words to remove stop words and noise. The segmented results are then converted into a format suitable for training the Word2vec model.

[0074] (3) Select the Word2vec model and set the model parameters: Determine the dimension of the vector (e.g., 100-dimensional, 200-dimensional, etc.), which will affect the expressive power and computational complexity of the vector. Set the window size (i.e., consider the number of context words). The larger the window size, the richer the context information that the model can capture. Determine the number of iterations and the training algorithm to ensure that the model can fully learn the relationships between words.

[0075] (4) Train the Word2vec model using the preprocessed medical supply name data. During the training process, the model will continuously adjust its parameters to learn the similarity between medical supply names.

[0076] After training, the model can map each medical supply name to a high-dimensional vector space. By querying the model, the vector representation of each medical supply name can be obtained.

[0077] S124, for multiple sample data Clustering was performed to obtain multiple clusters. ,in, The cluster number;

[0078] After extracting the sample data, this application clusters multiple sample data sets, thereby grouping sample data sets with similar values ​​for the first and second target parameters together. The specific clustering method of this application is as follows:

[0079] (1) When both the first target parameter and the second target parameter are numeric or vector types, construct a two-dimensional parameter. and two-dimensional parameters Density clustering was performed to obtain multiple clusters. ;

[0080] If both the first and second target parameters are numerical or vector types, the DBSCAN clustering algorithm can be used directly for clustering.

[0081] If one of the first and second objective parameters is a numeric type and the other is a vector type, the first and second objective parameters are first combined to generate a multidimensional vector, and then clustered. After clustering, the numeric objective parameter can be separated out.

[0082] For example, the word vectors transformed earlier are 100-dimensional. The values ​​of the two target parameters are combined into a 101-dimensional vector before clustering. After clustering, the 101st dimension parameter of all data samples within a cluster is separated to complete the reconstruction.

[0083] (2) When one of the first target parameter and the second target parameter is a character type and the other is a numeric type or a vector type; the target parameter based on the character type is used for multiple sample data. The data is divided into multiple data units; within each data unit, multiple sample data are processed based on either numeric or vector type. Clustering was performed to obtain multiple clusters. .

[0084] If either the first target parameter or the second target parameter is a character type, then the target parameter is first divided according to the character type, and then the other target parameter is clustered.

[0085] For example, if the first objective parameter is the person who received the materials, and the second objective parameter is the word vector of the material name, then we first divide the data into multiple units based on the person who received the materials. Then, we cluster the data samples within each unit according to the word vectors. Since the number of authorized recipients of materials is valid, dividing the data by the person who received the materials will not result in over-segmentation.

[0086] Clustering is used because some parameters have too many possible values, making it difficult to obtain representative probability values ​​during subsequent probability calculations. For example, the possible values ​​for material name, time, and quantity may be hundreds. Therefore, clustering groups data samples with similar values ​​together as data samples representing certain patterns. The probability values ​​calculated based on samples within each cluster can reflect these patterns.

[0087] S125, for the multiple clusters Labeling is performed to obtain the multiple clusters. Category label of the first target parameter Category label of the second target parameter ;

[0088] Since the data samples were clustered in the preceding text, this application labels the clusters to summarize the clustering results. Specifically, this includes:

[0089] (1) When the first target parameter or the second target parameter is a numeric type, calculate the average value of the target parameters with numeric data type within the cluster. and standard deviation and based on the average value and the standard deviation Build reference range labels ,in, For range adjustment parameters;

[0090] If the target parameter type within the cluster is numerical, then range labels can be constructed using the mean and standard deviation of the target parameter within the cluster. For example, the time range is 9:00-10:00, and the quantity range is 200-300;

[0091] Specifically, under normal circumstances The value is 3, meaning the range label satisfies the three-standard-deviation principle.

[0092] (2) When the first target parameter or the second target parameter is a vector type, label the type of the target parameter whose numerical type is a vector.

[0093] If the target parameter within a cluster is a vector type, i.e., the name of the material, then since we clustered material names with similar functions and principles as described earlier, they can be labeled according to their functions and principles. Examples include disinfection supplies, antibiotics, and surgical instruments.

[0094] (3) When the first target parameter or the second target parameter is of type character, the character itself is used as the label of the target parameter of type character.

[0095] If it is a character type, then use the application itself as the tag, using the name of the person receiving the material.

[0096] S126, Statistical Cluster Number of samples within a cluster Conforms to category labels Number of sample data and conforming to category labels Number of sample data ;

[0097] S127, calculate respectively Number of samples within a cluster With category label Number of sample data ratio Number of samples within a cluster With category label Number of sample data ratio ;

[0098] The ratio represents the probability that, when the value of one target parameter matches the label, the value of another target parameter also matches the corresponding label. For example, when the person requisitioning materials is XX, there is an X% probability that the material requisition type will also be surgical instruments. Therefore, this application can find various candidate material requisition habit characteristics and their probabilities by calculating the ratio.

[0099] S128, in the ratio When the proportion exceeds a preset threshold, the category label is determined. exist The characteristics of material requisition habits; in the ratio When the proportion exceeds a preset threshold, the category label is determined. exist The material requisition habits are characterized; and the ratio is used as the confidence level of the material requisition habits.

[0100] If the probability is greater than a preset threshold, such as 10%, then it can be determined that this type of material requisition habit exists in past material requisition records.

[0101] Furthermore, regarding confidence levels, even for the same material collection habit characteristic, the confidence level differs across different category labels. For example, for material collector XX, there is a 30% probability that they will collect medical supplies between 9:00 AM and 10:00 AM. However, during this same time period, many people collect medical supplies, and the probability that XX is among them is only 10%. Therefore, it is necessary to calculate the existence of material collection habit characteristics for different labels separately. The confidence level is crucial for the subsequent matching process.

[0102] S130, perform image and text recognition on the material requisition record to obtain preliminary material requisition data; and verify the preliminary material requisition data to obtain incomplete records;

[0103] This application uses a scanner or multifunction printer to scan paper forms into digital image files in PDF or common image formats such as JPEG and PNG. Then, an OCR service is used to convert the PDF or image files into text. This text is not 100% accurate, so manual verification is necessary to correct common errors, resulting in preliminary material requisition data.

[0104] During the manual verification process, errors that cannot be corrected can be marked or deleted. By utilizing missing information, illegible handwriting, or smears, incomplete records can be obtained.

[0105] S140, based on the known data of the incomplete record, the incomplete record is matched with the multiple material requisition habit features, and when the incomplete record matches any target material requisition habit feature, the incomplete record is completed based on the target material requisition habit feature and the confidence level of the target material requisition habit feature.

[0106] In this application, multiple material requisition habit features are matched based on known data from incomplete records, and any material requisition habit feature that matches the known data with any label is taken as the target material requisition habit feature.

[0107] For example, the incomplete record is as follows: It is evident that the information of the person who received the materials is missing;

[0108]

[0109] At this point, multiple material requisition habit features are matched with the known data "requisition time: 9:30", "material name: mask", and "quantity: 200" to obtain multiple target material requisition habit features that match the above known data.

[0110] For example, the characteristics of target material requisition habits include:

[0111] {Material requisition time: 9:00-10:00; Requisitioner: A}; Confidence level based on material requisition time: 15%;

[0112] {Material requisition time: 9:10-10:00; Requisitioner: B}; Confidence level based on material requisition time: 18%;

[0113] {Material Name: Daily Consumables; Recipient: A}; Confidence level based on material name: 23%;

[0114] {Material Name: Daily Consumables; Recipient: B}; Confidence level based on material name: 11%;

[0115] {Data volume: 150-300; Material requisitioner: A}; Confidence level based on material name: 15%;

[0116] {Data volume: 100-250; Material requisitioner: B}; Confidence level from the perspective of material name: 28%;

[0117] Then, based on the confidence level of the aforementioned target material requisition habit characteristics, the total confidence level of multiple candidate results can be calculated, that is, candidate labels for incomplete records can be determined based on the target material requisition habit characteristics and known data. and candidate tags confidence level , ,in, For the first Confidence level corresponding to the target material requisition habit characteristics of known data;

[0118] Taking the aforementioned target material requisition habits as an example;

[0119] The confidence level for the missing parameter A in the incomplete record is: 15% + 23% + 15% = 53%;

[0120] The confidence level for the missing parameter B in the incomplete record is: 18% + 11% + 28% = 57%;

[0121] Then, target labels with a confidence level greater than the confidence threshold are selected; in this embodiment, the confidence threshold is set to 50%, meaning that both A and B can be used as target labels.

[0122] Finally, the incomplete records are completed based on the target labels.

[0123] Both A and B are automatically filled in to fill in the missing parts, and the confidence level is marked;

[0124] For example, the completed material requisition record would be:

[0125]

[0126] Since the true result cannot be directly inferred in this application, but only the most likely result is found based on existing clues, this application selects the multiple candidate labels with the highest confidence. Send the confirmation to the corresponding object. The target object is the material recipient in the record. If the missing part is the material recipient, send the confirmation to multiple inferred candidate material recipients.

[0127] If a confirmation message is received from the candidate who requested the materials, the actual record can be corrected and added to the table.

[0128] The above process can greatly improve the information repair efficiency of information managers, help them narrow down the scope of incomplete information confirmation to a smaller range, and reduce the communication costs associated with confirmation and supplementation work.

[0129] This invention discloses a method for managing medical supply inventory information. It extracts various habitual features from complete historical requisition data. Simultaneously, it verifies current requisition records. If incomplete records exist, such as missing information or illegible handwriting, the method matches the requisition habitual features with known information in the requisition record. If a match is found, the incomplete record is completed using the target requisition habitual feature and its confidence level. This application utilizes historical data analysis to obtain requisition habitual features and uses these features to automatically repair missing registration data, effectively reducing the verification workload for management personnel.

[0130] like Figure 3 As shown, this application provides a medical supplies inventory information management system, including:

[0131] The acquisition module is used to acquire the material requisition records and historical material requisition data for the current management cycle, wherein the material requisition records are paper materials and the historical material requisition data are electronic data;

[0132] The feature extraction module is used to extract various material requisition habit features and confidence levels of various material requisition habit features from the historical material requisition data;

[0133] The verification module is used to perform image and text recognition on the material requisition record to obtain preliminary material requisition data; and to verify the preliminary material requisition data to obtain incomplete records.

[0134] The data completion module is used to match the incomplete record with the various material requisition habit features based on the known data of the incomplete record, and when the incomplete record matches any target material requisition habit feature, to complete the incomplete record based on the target material requisition habit feature and the confidence level of the target material requisition habit feature.

[0135] This invention discloses a medical supplies inventory information management system that extracts various habitual features from complete historical material requisition data. Simultaneously, it verifies current material requisition records. If incomplete records exist, such as missing information or illegible handwriting, the system matches the known information in the requisition record with the requisition habitual features. If a match is found, the incomplete record is completed using the target requisition habitual features and their confidence level. This application utilizes historical data analysis to obtain requisition habitual features and uses these features to automatically repair missing registration data, thereby effectively reducing the verification workload for management personnel.

[0136] This embodiment also provides an electronic terminal, including: a processor and a memory;

[0137] The memory is used to store computer programs, and the processor is used to execute the computer programs stored in the memory so that the terminal performs any of the methods in this embodiment.

[0138] As will be understood by those skilled in the art, the computer-readable storage medium described in this embodiment allows for the implementation of all or part of the steps in the above method embodiments by computer program-related hardware. The aforementioned computer program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0139] The electronic terminal provided in this embodiment includes a processor, a memory, a transceiver, and a communication interface. The memory and the communication interface are connected to the processor and the transceiver and complete communication between them. The memory is used to store computer programs, the communication interface is used to perform communication, and the processor and the transceiver are used to run the computer programs, so that the electronic terminal performs the steps of the above method.

[0140] In this embodiment, the memory may include random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device.

[0141] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0142] In the above embodiments, although the invention has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. The embodiments of the invention are intended to cover all such substitutions, modifications, and variations falling within the broad scope of the appended claims.

[0143] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A method for managing inventory information of medical supplies, characterized in that, Including the following steps: Obtain material requisition records for the current management cycle and historical material requisition data, wherein the material requisition records are paper materials and the historical material requisition data are electronic data; The historical material requisition data is used to extract multiple material requisition habit features and their confidence levels. The material requisition habit features are the value ranges of any two material requisition parameters that have a probability of simultaneous occurrence, where the probability of simultaneous occurrence is greater than a preset probability threshold. The material requisition parameters include requisition time, material name, quantity, requisition location, and the department to which the material belongs. The historical material requisition data includes multiple requisition records, each containing values ​​for multiple parameters. Extracting the multiple material requisition habit features and their confidence levels from the historical material requisition data includes: removing useless parameters from the historical material requisition data to obtain target parameters; and combining any two parameters to obtain a sample data template. ,in, Indicates the first target parameter. Indicates the second target parameter; based on the sample data template Extract multiple sample data from the historical material requisition data. , Indicates the first The value of the first target parameter, Indicates the first The value of the second objective parameter; for multiple sample data Clustering was performed to obtain multiple clusters. ,in, The cluster number; based on the multiple clusters Verify the characteristics of material requisition habits and the confidence level of material requisition habits; based on the multiple clusters Verifying material requisition habit characteristics and the confidence level of material requisition habits includes: for the multiple clusters Labeling is performed to obtain the multiple clusters. Category label of the first target parameter Category label of the second target parameter ; Statistical cluster Number of samples within a cluster Conforms to category labels Number of sample data and conforming to category labels Number of sample data ; calculate separately Number of samples within a cluster With category label Number of sample data ratio Number of samples within a cluster With category label Number of sample data ratio ; in ratio When the proportion exceeds a preset threshold, the category label is determined. exist The characteristics of material requisition habits; in the ratio When the proportion exceeds a preset threshold, the category label is determined. exist The characteristics of material requisition habits; and the ratio is used as the confidence level of the characteristics of material requisition habits; The material requisition record is subjected to image and text recognition to obtain preliminary material requisition data; and the preliminary material requisition data is verified to obtain incomplete records. Based on the known data of the incomplete record, the incomplete record is matched with the various material requisition habit features. When the incomplete record matches any target material requisition habit feature, the incomplete record is completed based on the target material requisition habit feature and its confidence level. Matching the incomplete record with the various material requisition habit features based on the known data of the incomplete record includes: matching the known data of the incomplete record with the labels of the various material requisition habit features, and using the material requisition habit feature whose label matches the known data as the target material requisition habit feature. Completing the incomplete record based on the target material requisition habit feature and its confidence level includes: determining candidate labels for the incomplete record based on the target material requisition habit feature and the known data. and candidate tags confidence level , ,in, For the first The confidence level of the target material requisition habit features of known data is used to filter out target labels with a confidence level greater than the confidence threshold; the incomplete records are then completed based on the target labels.

2. The method for managing inventory information of medical supplies according to claim 1, characterized in that, Based on the sample data template Extract multiple sample data from the historical material requisition data. ,include: When the sample data template When none of the parameters are medical supply names, the data is based on the sample data template. Sample data is constructed using the values ​​of parameters belonging to the same material requisition record. ; When the sample data template When the stored parameter is the name of a medical supply, the name of the medical supply is vectorized and processed based on the sample data template. Sample data is constructed using the values ​​of parameters belonging to the same material requisition record. .

3. The method for managing inventory information of medical supplies according to claim 1, characterized in that, For multiple sample data Clustering was performed to obtain multiple clusters. ,include: When both the first and second target parameters are numeric or vector types, construct a two-dimensional parameter. and two-dimensional parameters Density clustering was performed to obtain multiple clusters. ; When one of the first target parameter and the second target parameter is a character type, and the other is a numeric or vector type; the target parameter based on the character type is used for multiple sample data. The data is divided into multiple data units; within each data unit, multiple sample data are processed based on either numeric or vector type. Clustering was performed to obtain multiple clusters. .

4. The method for managing inventory information of medical supplies according to claim 1, characterized in that, For the multiple clusters Labeling is performed to obtain the multiple clusters. Category label of the first target parameter Category label of the second target parameter ,include: When either the first or second target parameter is a numeric type, calculate the average value of the target parameters with numeric data types within the cluster. and standard deviation and based on the average value and the standard deviation Build reference range labels ,in, For range adjustment parameters; When the first target parameter or the second target parameter is a vector type, label the target parameter whose numerical type is a vector. When the first target parameter or the second target parameter is of type character, the character itself is used as the label for the target parameter whose data type is character.

5. The method for managing inventory information of medical supplies according to claim 1, characterized in that, Also includes: The candidate labels with the highest confidence levels Send to the corresponding recipient for confirmation.

6. A medical supplies inventory information management system, characterized in that, include: The acquisition module is used to acquire the material requisition records and historical material requisition data for the current management cycle, wherein the material requisition records are paper materials and the historical material requisition data are electronic data; The feature extraction module is used to extract multiple material requisition habit features and their confidence levels from the historical material requisition data. The material requisition habit features are the value ranges of any two material requisition parameters that have a probability of simultaneous occurrence, where the probability of simultaneous occurrence is greater than a preset probability threshold. The material requisition parameters include requisition time, material name, quantity, requisition details, and the department to which the material belongs. The historical material requisition data includes multiple material requisition records, each containing values ​​for multiple parameters. Extracting multiple material requisition habit features and their confidence levels from the historical material requisition data includes: removing useless parameters from the historical material requisition data to obtain target parameters; and combining any two parameters to obtain a sample data template. ,in, Indicates the first target parameter. Indicates the second target parameter; based on the sample data template Extract multiple sample data from the historical material requisition data. , Indicates the first The value of the first target parameter, Indicates the first The value of the second objective parameter; for multiple sample data Clustering was performed to obtain multiple clusters. ,in, The cluster number; based on the multiple clusters Verify the characteristics of material requisition habits and the confidence level of material requisition habits; based on the multiple clusters Verifying material requisition habit characteristics and the confidence level of material requisition habits includes: for the multiple clusters Labeling is performed to obtain the multiple clusters. Category label of the first target parameter Category label of the second target parameter ; Statistical cluster Number of samples within a cluster Conforms to category labels Number of sample data and conforming to category labels Number of sample data ; calculate separately Number of samples within a cluster With category label Number of sample data ratio Number of samples within a cluster With category label Number of sample data ratio ; in ratio When the proportion exceeds a preset threshold, the category label is determined. exist The characteristics of material requisition habits; in the ratio When the proportion exceeds a preset threshold, the category label is determined. exist The characteristics of material requisition habits; and the ratio is used as the confidence level of the characteristics of material requisition habits; The verification module is used to perform image and text recognition on the material requisition record to obtain preliminary material requisition data; and to verify the preliminary material requisition data to obtain incomplete records. The data completion module is used to match the incomplete record with the various material requisition habit features based on the known data of the incomplete record, and to complete the incomplete record based on the target material requisition habit feature and its confidence level when the incomplete record matches any target material requisition habit feature; matching the incomplete record with the various material requisition habit features based on the known data of the incomplete record includes: matching the known data of the incomplete record with the labels of the various material requisition habit features, and taking the material requisition habit feature whose label matches the known data as the target material requisition habit feature; completing the incomplete record based on the target material requisition habit feature and its confidence level includes: determining candidate labels for the incomplete record based on the target material requisition habit feature and the known data. and candidate tags confidence level , ,in, For the first The confidence level of the target material requisition habit features of known data is used to filter out target labels with a confidence level greater than the confidence threshold; the incomplete records are then completed based on the target labels.

Images