Disease name coding method, device, computer equipment and storage medium

By deduplication of disease names in electronic medical records, and using precise matching and fuzzy matching models to code according to standard disease classification tables, the problem of low accuracy of disease names in the existing technology is solved, and the matching is improved code accuracy.

CN111695336BActive Publication Date: 2024-11-26PING AN TECH (SHENZHEN) CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202010340354.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-04-26
Publication Date
2024-11-26
Estimated Expiration
2040-04-26

AI Technical Summary

Technical Problem

The existing ICD-10 dedicated encoding query database system cannot effectively handle irregular expressions when coding disease names, resulting in low code accuracy.

Method used

The disease name list after deduplication is used, and multi-dimensional and multi-modal coding is performed through the exact matching model and the fuzzy matching model, and a disease name coding list is generated using the standard disease classification table. The exact matching model consists of several precise matching sub-models arranged in an orderly manner. The fuzzy matching model consists of several fuzzy matching sub-models, including word frequency matching, N-Gram, edit distance and cosine calculation sub-models.

Benefits of technology

The accuracy of disease name coding is improved, the calculation amount is reduced through deduplication processing, and the code is coordinated according to the standard disease classification table in both precise matching and fuzzy matching to ensure multi-dimensional and multi-modal matching.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN111695336B_ABST
    Figure CN111695336B_ABST
Patent Text Reader

Abstract

The embodiment of the present application belongs to the field of artificial intelligence, and relates to a disease name matching method, device, computer equipment and storage medium, the method comprising: obtaining a disease name list from an electronic medical record; deduplicating repeated disease names in the disease name list to obtain a deduplicated disease name list; inputting the deduplicated disease name list into an exact matching model, matching according to a standard disease classification table, obtaining a first matching result and candidate matching disease names; inputting the obtained candidate matching disease names into a fuzzy matching model, matching according to the standard disease classification table, obtaining a second matching result; generating a disease name matching list based on the first matching result and the second matching result. The present application performs multi-dimensional and multi-modal matching of disease names, which improves the accuracy of disease name matching, and the disease name list can also be stored in a blockchain to improve the privacy and security of data.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a disease name coding method, device, computer equipment and storage medium. Background Art

[0002] With the development of Internet technology, the use of electronic medical records has become more popular. Electronic medical records refer to individual health information recorded in digital information during medical activities. Electronic medical records are created, collected, managed and consulted by medical staff, and contain long-term, multi-dimensional and rich individual health information. ICD-10 (International Classification of Diseases, version 10) is an international unified disease classification method developed by the World Health Organization. It classifies diseases and represents them in a coded manner, making the disease categories an orderly combination. ICD-10 includes nearly 26,000 types of diseases, covering various diseases in all departments. The existing application research of electronic medical records in the medical field is usually based on ICD-10 coding, so it is particularly important to match the disease names in electronic medical records, that is, to correspond the disease names in electronic medical records with ICD-10 coding.

[0003] The existing ICD-10 dedicated code query database system provides a query function from Chinese disease names to ICD-10 codes, but it can only query medical record texts containing standard disease names. Due to the uneven cultural levels of medical staff in various medical institutions and their different habits of describing diseases, there are a large number of non-standard expressions in electronic medical records, such as colloquial expressions, abbreviations of disease names, and different expressions of the same disease. Therefore, the existing ICD-10 dedicated code query database system has a large amount of data that cannot be processed when matching disease names, and the accuracy of disease name matching is low. Summary of the invention

[0004] The purpose of the embodiments of the present application is to propose a disease name coding method, device, computer equipment and storage medium to solve the problem of low accuracy of disease name coding.

[0005] In order to solve the above technical problems, the present application embodiment provides a disease name coding method, which adopts the following technical solution:

[0006] Obtain a list of disease names from the electronic medical record;

[0007] Deduplication processing is performed on repeated disease names in the disease name list to obtain a deduplication disease name list;

[0008] Input the deduplicated disease name list into the exact matching model, perform coding according to the standard disease classification table, and obtain a first coding result and candidate coding disease names;

[0009] Input the obtained candidate matching disease name into the fuzzy matching model, match the code according to the standard disease classification table, and obtain a second matching result;

[0010] Generate a disease name code pairing list according to the first code pairing result and the second code pairing result.

[0011] Furthermore, the precise matching model is composed of a plurality of ordered precise matching sub-models; the step of inputting the deduplicated disease name list into the precise matching model, matching the disease names according to the standard disease classification table, and obtaining the first matching result and the candidate matching disease names specifically includes:

[0012] Input each disease name in the deduplicated disease name list into the exact matching sub-model according to the arrangement order of the exact matching sub-model in the exact matching model;

[0013] Through the current exact matching sub-model, the standard disease classification table is searched for a standard disease name that matches the input disease name;

[0014] When a matching standard disease name is found, the found standard disease name and the disease code corresponding to the standard disease name are used as the first pair of code results of the disease name;

[0015] When the current accurate matching sub-model fails to find a matching standard disease name, the disease name is input into the next accurate matching sub-model to continue matching;

[0016] If the disease name is not matched by each exact matching sub-model, the disease name is marked as a candidate matching disease name.

[0017] Furthermore, the step of inputting each disease name in the deduplicated disease name list into the exact matching sub-model according to the arrangement order of the exact matching sub-model in the exact matching model specifically includes:

[0018] Each disease name in the deduplicated disease name list is input into the precise matching sub-model according to the arrangement order of the four precise matching sub-models in the precise matching model; the four precise matching sub-models include a complete matching sub-model, a stop word removal sub-model, a primary and secondary separation sub-model and a synonym recognition sub-model.

[0019] Furthermore, the fuzzy matching model is composed of a plurality of fuzzy matching sub-models; the step of inputting the obtained candidate matching disease name into the fuzzy matching model, matching according to the standard disease classification table, and obtaining the second matching result specifically includes:

[0020] Input the obtained candidate matched code disease names into each fuzzy matching sub-model in the fuzzy matching model;

[0021] Based on the fuzzy matching sub-models, calculating the similarity between the candidate matched disease name and each standard disease name in the standard disease classification table;

[0022] A second code pairing result is generated according to the similarities calculated by the fuzzy matching sub-models.

[0023] Furthermore, the step of inputting the obtained candidate matching disease names into each fuzzy matching sub-model in the fuzzy matching model specifically includes:

[0024] The obtained candidate paired disease names are input into four fuzzy matching sub-models in the fuzzy matching model, and the four fuzzy matching sub-models include a word frequency matching sub-model, an N-Gram sub-model, an edit distance sub-model and a cosine calculation sub-model.

[0025] Furthermore, when the fuzzy matching sub-model is an edit distance sub-model, the step of calculating the similarity between the candidate matched code disease name and each standard disease name in the standard disease classification table specifically includes:

[0026] Calculating the text edit distance between the candidate matched code disease name and each standard disease name in the standard disease classification table;

[0027] Each text edit distance is normalized, and each text edit distance after normalization is used as the similarity between the candidate paired disease name and the standard disease name.

[0028] Furthermore, the step of generating a second pairing result according to the similarities calculated by the fuzzy matching sub-models specifically includes:

[0029] For each candidate code-matching disease name, the standard disease name and disease code corresponding to the maximum similarity are selected from the similarities calculated by the fuzzy matching sub-models to perform Hard Voting fusion to obtain the second code-matching result;

[0030] or,

[0031] SoftVoting fusion is performed according to the similarities calculated by the fuzzy matching models to obtain a second pair of code results.

[0032] In order to solve the above technical problems, the embodiment of the present application further provides a disease name coding device, including:

[0033] List acquisition module, used to obtain a list of disease names from electronic medical records;

[0034] A list deduplication module is used to perform deduplication processing on repeated disease names in the disease name list to obtain a deduplication disease name list;

[0035] An accurate matching module, used for inputting the deduplicated disease name list into an accurate matching model, matching the disease names according to a standard disease classification table, and obtaining a first matching result and candidate matching disease names;

[0036] A fuzzy matching module, used for inputting the obtained candidate matching disease name into the fuzzy matching model, matching the disease name according to the standard disease classification table, and obtaining a second matching result;

[0037] A list generation module is used to generate a disease name code matching list according to the first code matching result and the second code matching result.

[0038] In order to solve the above technical problems, an embodiment of the present application also provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of the above-mentioned disease name coding method are implemented.

[0039] In order to solve the above technical problems, an embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned disease name coding method are implemented.

[0040] Compared with the prior art, the embodiments of the present application mainly have the following beneficial effects: first, the disease name list is deduplicated to reduce the amount of calculation; the deduplicated disease name list is input into the exact matching model for exact matching to obtain a first coding result, and the disease names that fail to achieve exact matching are input into the fuzzy matching model as candidate coding disease names for fuzzy matching to obtain a second coding result. During the two coding matches, the coding is performed according to the standard disease classification table; finally, a disease name coding list is generated based on the first coding result and the second coding result, and the disease names are matched in multiple dimensions and multiple modes through exact matching and fuzzy matching, thereby improving the accuracy of disease name matching. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0042] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0043] Figure 2 is a flow chart of an embodiment of a disease name coding method according to the present application;

[0044] Figure 3 yes Figure 2 A flowchart of a specific implementation of step S203;

[0045] Figure 4 yes Figure 2 A flowchart of a specific implementation of step S204;

[0046] Figure 5 It is a structural schematic diagram of an embodiment of a disease name coding device according to the present application;

[0047] Figure 6 It is a structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION

[0048] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0049] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0050] The blockchain referred to in this invention is a new application mode of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0051] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0052] like Figure 1 As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0053] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0054] Terminal devices 101, 102, 103 can be various electronic devices with display screens and supporting web browsing, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III), MP4 (Moving Picture Experts Group Audio Layer IV) players, laptop computers, desktop computers, etc.

[0055] The server 105 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal devices 101 , 102 , and 103 .

[0056] It should be noted that the disease name matching method provided in the embodiment of the present application is generally executed by a server, and accordingly, the disease name matching device is generally set in the server.

[0057] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to the implementation requirements.

[0058] Continue to refer Figure 2 , shows a flow chart of an embodiment of a disease name matching method according to the present application. The disease name matching method comprises the following steps:

[0059] Step 201, obtaining a list of disease names from the electronic medical record.

[0060] In this embodiment, the disease name coding method is executed on the electronic device (eg Figure 1 The server shown in the figure) can communicate with the terminal or the server through a wired connection or a wireless connection. It should be noted that the above wireless connection method may include but is not limited to 3G / 4G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods currently known or to be developed in the future.

[0061] The disease name list may be a list of disease names recorded in electronic medical records.

[0062] Specifically, the information recorded in the electronic medical record is structured. For example, the electronic medical record contains disease name information, symptom record information, and diagnosis and treatment information. The server reads a large number of disease names from the structured electronic medical record to obtain a disease name list. In the disease name list, the electronic medical record identifier is stored corresponding to the disease name. The electronic medical record identifier is the identifier of the electronic medical record, and the electronic medical record identifier can be a string of letters, numbers, special symbols, etc.

[0063] In one embodiment, the electronic medical records read by the server may come from various terminals or from a preset database.

[0064] In one embodiment, the server can set a scheduled task to perform disease name matching on a regular basis, such as matching once a month or a quarter. The scheduled task can be set to activate the matching task at a specific time of each month or a quarter. The server can use Cron (scheduled task) in Linux to trigger information synchronization instructions, and Cron can execute specific tasks at the agreed time.

[0065] Step 202, deduplication processing is performed on the repeated disease names in the disease name list to obtain a deduplication disease name list.

[0066] Specifically, there may be a large number of identical disease names in the disease name list. For example, during the peak season of influenza, many patients with influenza will go to the hospital for treatment. At this time, the electronic medical records obtained by the hospital will have many disease names such as "influenza". If deduplication is not performed, the server will need to perform a large number of repeated calculations, which will increase the amount of calculation and reduce the coding efficiency.

[0067] The server first identifies repeated disease names in the disease name list, and then performs deduplication processing to obtain a deduplicated disease name list.

[0068] Among a large number of repeated disease names, only one can be retained, and the rest of the repeated disease names can be deleted. The electronic medical record identifier corresponding to the deleted disease name and the electronic medical record identifier corresponding to the retained disease name can be associated and stored, so as to finally restore the coding results of all disease names; among them, the electronic medical record identifier corresponding to the retained disease name can use the initial electronic medical record identifier, or the electronic medical record identifier corresponding to the retained disease name can be reset.

[0069] Step 203, input the deduplicated disease name list into the exact matching model, perform coding according to the standard disease classification table, and obtain the first coding result and candidate coding disease names.

[0070] Specifically, the exact matching model performs an exact match on the disease name from the text level, and the exact matching model codes the disease name according to the standard disease classification table, that is, the disease name is exactly matched with the standard disease name in the standard disease classification table. The standard disease classification table stores the standard disease name and the disease code corresponding to the standard disease name, and the standard disease classification table can be the 10th version of the International Classification of Diseases (ICD): ICD-10.

[0071] When a disease name can be accurately matched with a standard disease name, the standard disease name and the disease code corresponding to the standard disease name are used as the first coding result. Disease names that cannot be matched by the exact matching model are used as candidate coding disease names and undergo the second round of coding.

[0072] Step 204, input the obtained candidate matching disease name into the fuzzy matching model, match it according to the standard disease classification table, and obtain a second matching result.

[0073] Specifically, the fuzzy matching model performs fuzzy matching on the candidate disease names through similarity calculation, and uses the standard disease name that can achieve fuzzy matching with the candidate disease name and the disease code corresponding to the standard disease name as the second coding result.

[0074] In one embodiment, the server can use different types of fuzzy matching methods to calculate the similarity between the candidate matching disease name and the standard disease name in the standard disease classification table, and combine the similarities calculated by different types of fuzzy matching methods to determine the standard disease name that achieves fuzzy matching with the candidate matching disease name, and use the standard disease name and the disease code corresponding to the standard disease name as the second matching result.

[0075] Step 205, generating a disease name code matching list according to the first code matching result and the second code matching result.

[0076] Specifically, the server merges the first code pairing result and the second code pairing result into a list, in which the electronic medical record identifier, the disease name, the standard disease name matching the disease name, and the disease code corresponding to the standard disease name are stored correspondingly. For the disease name deleted in the deduplication process, the server uses the first code pairing result or the second code pairing result of the disease name associated with the deleted disease name as the code pairing result of the deleted disease name, thereby obtaining a complete disease name code pairing list.

[0077] In one embodiment, the server may also upload the generated disease name pairing list to the blockchain to improve the privacy and security of the disease name pairing list.

[0078] In this embodiment, the disease name list is first deduplicated to reduce the amount of calculation; the deduplicated disease name list is input into the exact matching model for exact matching to obtain a first coding result, and the disease names that fail to achieve exact matching are input into the fuzzy matching model as candidate coding disease names for fuzzy matching to obtain a second coding result. Both codings are performed according to the standard disease classification table; finally, a disease name coding list is generated based on the first coding result and the second coding result, and the disease names are matched in multiple dimensions and multiple modes through exact matching and fuzzy matching, thereby improving the accuracy of disease name matching.

[0079] Furthermore, the exact matching model is composed of several ordered exact matching sub-models, such as Figure 3 As shown, the above step 203 may include:

[0080] Step 2031, input each disease name in the disease name list after deduplication into the exact matching sub-model according to the arrangement order of the exact matching sub-model in the exact matching model.

[0081] Specifically, the exact matching model can be composed of several different, ordered exact matching sub-models. The exact matching sub-model can first perform simple text-level preprocessing on the input disease name, and then perform exact matching. The preprocessing can be processing the characters or phrases in the disease name, such as correcting typos, removing repeated characters or phrases, converting synonyms, removing meaningless characters, etc. Different exact matching sub-models can perform different text-level preprocessing on the disease name; it is understandable that there can also be an exact matching sub-model that does not preprocess the disease name.

[0082] The server inputs the disease names in the deduplicated disease name list into the exact match sub-model first according to the arrangement order of the exact match sub-model in the exact match model.

[0083] Step 2032, using the current exact matching sub-model, query the standard disease classification table for a standard disease name that matches the input disease name.

[0084] Specifically, the current exact matching sub-model preprocesses the input disease name according to the preprocessing procedure, obtains the standard disease classification table after preprocessing, compares the disease name with each standard disease name in the standard disease classification table one by one, and queries for the standard disease name that can be matched.

[0085] Step 2033, when a matching standard disease name is found, the found standard disease name and the disease code corresponding to the standard disease name are used as the first pair of code results of the disease name.

[0086] Specifically, when a standard disease name that matches a disease name is queried, the matched standard disease name and the disease code corresponding to the standard disease name are used as the first pair of code results for the disease name.

[0087] After the exact matching sub-model completes the matching of a disease name, it starts to process the next input disease name. When a disease name can be matched by a certain exact matching sub-model, the processing of the disease name ends and the disease name is no longer matched by the remaining exact matching sub-models.

[0088] Step 2034: When the current exact matching sub-model does not find a matching standard disease name, the disease name is input into the next exact matching sub-model to continue matching.

[0089] Specifically, if the current exact matching sub-model fails to find a standard disease name that matches the disease name in the standard disease classification table, the disease name is input into the next exact matching sub-model according to the arrangement order of the exact matching sub-models to continue matching.

[0090] Step 2035: If the disease name is not matched by each exact matching sub-model, the disease name is marked as a candidate matching disease name.

[0091] Specifically, when the exact matching sub-model cannot match the disease name, the disease name is input into the next exact matching sub-model for matching. When each exact matching sub-model cannot match the disease name, the disease name is marked as a candidate matching disease name.

[0092] In this embodiment, each disease name in the deduplicated disease name list is input into the precise matching sub-model in the arrangement order of the precise matching sub-model for matching. If a match is possible, the first code pairing result is generated. If a match is not possible, the next precise matching sub-model is input to continue matching. The precise matching sub-models are different, which ensures that the disease names can be accurately matched from multiple dimensions, thereby improving the accuracy of disease name coding.

[0093] In one embodiment, the above step 203 may specifically include: inputting each disease name in the disease name list after deduplication into the precise matching sub-model according to the arrangement order of the four precise matching sub-models in the precise matching model. The four precise matching sub-models include a complete matching sub-model, a stop word removal sub-model, a primary and secondary separation sub-model, and a synonym recognition sub-model.

[0094] Specifically, there are four exact matching sub-models in the exact matching model, and the four exact matching sub-models are arranged in the order of complete matching sub-model, stop word removal sub-model, primary and secondary separation sub-model and synonym recognition sub-model. The server first inputs each disease name in the disease name list after deduplication into the exact matching sub-model.

[0095] Complete match sub-model: used to completely match the disease name, compare the input disease name with the standard disease name in the standard disease classification table in sequence, and if the disease name is exactly the same as a certain standard disease name, it is determined that the disease name completely matches the standard disease name. The complete match sub-model uses the matched standard disease name and the disease code corresponding to the standard disease name as the first pair of code results. Disease names that cannot be matched by the complete match sub-model are input into the stop word removal sub-model.

[0096] Stop word removal sub-model: Pre-process the disease names to remove stop words before matching. First, remove meaningless punctuation marks in the disease names (such as "?@%¥#,; / ", etc.); then access the pre-built stop word library for medical diseases, which records numbers, directional words, and some specific terms; call the stop word library for medical diseases to remove stop words in the disease name (for example, if the disease name is "left metatarsal fracture", remove "left"); match the disease name after removing stop words with the standard disease name in the standard disease classification table in turn, and the matched standard disease name and the disease code corresponding to the standard disease name are used as the first pair of code results. Unmatched disease names are input into the primary and secondary separation sub-model.

[0097] Primary-secondary separation sub-model: The disease name is pre-processed for primary-secondary disease separation before matching. The disease name may be composed of multiple disease names connected together. The primary-secondary separation sub-model extracts the primary disease name and the secondary disease name. (For example, the disease name is "1. Diabetes 2. Hypertension", extract the primary disease name "diabetes" and the secondary disease name "hypertension"). Match the primary disease name and the secondary disease name with the standard disease names in the standard disease classification table in turn to obtain the first coding result. Among them, the primary disease name can be the disease name recognized first, and the secondary disease name can be the disease name recognized later. If the disease name is composed of multiple disease names connected together, multiple coding results will be obtained, the primary disease name corresponds to the primary coding result, and the secondary disease name corresponds to the secondary coding result. Disease names that are not matched by the primary-secondary separation sub-model are input into the synonym recognition sub-model.

[0098] Synonymous recognition sub-model: Pre-process the disease names for synonymous conversion before matching. The synonymous recognition sub-model accesses the pre-built synonymous disease word library, which records different representations of the same body part, different representations of the same symptom, and different representations of the same disease. The synonymous disease word library is called to replace the synonyms in the disease name, for example, "malignant tumor" is uniformly replaced with "cancer", "hyperthyroidism" is replaced with "hyperthyroidism", etc. Then, the disease names after synonymous replacement are matched with the standard disease names in the standard disease classification table in sequence to obtain the first pair of code results.

[0099] The disease names that failed to match the four exact matching sub-models are marked as candidate paired disease names.

[0100] It can be understood that the above four exact matching sub-models can also be arranged in any order.

[0101] In this embodiment, the disease name is input into the precise matching sub-model according to the arrangement order of the four precise matching sub-models in the precise matching model. The four precise matching sub-models are respectively a complete matching sub-model, a stop word removal sub-model, a primary and secondary separation sub-model and a synonym recognition sub-model. Different methods can be used to match the disease name based on the four precise matching sub-models, thereby improving the accuracy of matching the disease name.

[0102] Furthermore, the fuzzy matching model is composed of several fuzzy matching sub-models, such as Figure 4 As shown, the above step 204 may specifically include:

[0103] Step 2041, input the obtained candidate matching disease names into each fuzzy matching sub-model in the fuzzy matching model.

[0104] Specifically, the fuzzy matching model can be composed of several fuzzy matching sub-models, and the candidate code-paired disease name is input into each fuzzy matching sub-model in the fuzzy matching model. Each fuzzy matching sub-model can match the candidate code-paired disease name through different fuzzy matching methods.

[0105] In one embodiment, the step of inputting the obtained candidate code-paired disease names into each fuzzy matching sub-model in the fuzzy matching model specifically includes: inputting the obtained candidate code-paired disease names into four fuzzy matching sub-models in the fuzzy matching model, and the four fuzzy matching sub-models include a word frequency matching sub-model, an N-Gram sub-model, an edit distance sub-model and a cosine calculation sub-model.

[0106] Specifically, the fuzzy matching model consists of four fuzzy matching sub-models, including a word frequency matching sub-model, an N-Gram sub-model, an edit distance sub-model, and a cosine calculation sub-model. Each candidate code-matching disease name will be input into the four fuzzy matching sub-models for different fuzzy matching.

[0107] The word frequency matching sub-model parses the candidate code-matching disease names and the standard disease names in the standard disease classification table into a set of single characters (for example, "diabetes" is parsed into {"sugar", "urine", "disease"}). The Jaccard coefficient (Jaccard index, also known as the Jaccard similarity coefficient, is used to compare the similarities and differences between finite sample sets. The larger the Jaccard coefficient value, the higher the sample similarity.) is used as the similarity between the character set of the candidate code-matching disease name and the character set of each standard disease name, and a control parameter is added during the calculation process for smoothing operations.

[0108] For example, the candidate matching disease name is "diabetes", and the word frequency matching sub-model calculates the Jaccard coefficient between "diabetes" and 26,000 standard disease names in ICD-10 one by one. The calculation formula is as follows:

[0109]

[0110] Among them, A is the character set of candidate disease names, and B is the character set of standard disease names; Jaccard(A,B) is the similarity between A and B; lenA represents the length of set A, that is, the number of characters in set A; lenB represents the length of set B, that is, the number of characters in set B; len(A∩B) represents the number of identical characters in set A and set B, α and β are control parameters, which are set manually, for example, α can be set to 1 and β can be set to 0.5.

[0111] Then the Jaccard of the candidate code disease name "diabetes" and the standard disease name "diabetic foot" is calculated as:

[0112] N-Gram sub-model: N-Gram (also known as N-gram model) is often used in natural language processing. The N-Gram of a text represents the phrases obtained by segmenting the text into length N, where the value of N is generally 2 or 3. The N-Gram sub-model parses the candidate paired disease names and the standard disease names into sets of phrases. For example, "diabetes" is parsed into {"$sugar", "diabetes", "diabetes", "disease$"}, where $ is a filler character. The similarity between the candidate paired disease name phrase set and the standard disease name phrase set is then calculated using the following formula:

[0113]

[0114] Among them, M is the phrase set of candidate disease names, and N is the phrase set of standard disease names; Jaccard(M,N) is the similarity between M and N; lenM represents the length of set M, that is, the number of phrases in set M; lenN represents the length of set N, that is, the number of phrases in set N; len(M∩N) represents the number of identical phrases in set M and set N, and δ and ε are control parameters, which are set manually.

[0115] Edit distance sub-model: used to calculate the Levenshtein distance between the candidate paired disease names and each standard disease name. The smaller the distance, the higher the similarity.

[0116] Levenshtein distance (also known as text edit distance) refers to the minimum number of operations required to transform one string into another. Transformation operations include insertion, deletion, and replacement.

[0117] For example: to transform "eeba" into "abac":

[0118] eeba (delete the first e) → eba

[0119] eba (replace the remaining e with a) → aba

[0120] aba (insert c at the end) → abac

[0121] Then the Levenshtein distance between "eeba" and "abac" is 3.

[0122] Cosine calculation sub-model: The cosine calculation sub-model needs to be trained first. First, crawl medical-related data from the Internet to build a medical corpus (for example, crawl data from Wikipedia, Baidu Encyclopedia, and Medical Encyclopedia to build a medical corpus), and use the crawled data to train the Word2Vec model, which is a model for generating word vectors. The cosine calculation sub-model first segments the candidate code disease name, and then uses the trained Word2Vec model to convert the segmented part into word vectors, and calculates the vector average of each word vector as the disease name vector. For example, the disease name "upper respiratory tract infection" is segmented into "upper", "respiratory tract" and "infection". The disease name vector of "upper respiratory tract infection" can be represented by the average value of the word vectors of the three segmented words "upper", "respiratory tract" and "infection". Similarly, calculate the disease name vector of the standard disease name. Use the PCA model to reduce the dimension of the disease name vector, and translate the disease name vector to the area centered on the origin to increase the difference between the vectors. The cosine similarity between the disease name vector of the candidate paired disease name after PCA correction and the disease name vector of each standard disease name is calculated. Among them, PCA (principal components analysis, also known as principal component analysis technology) is mainly used for data dimensionality reduction.

[0123] In this embodiment, the candidate coded disease names are input into four fuzzy matching sub-models in the fuzzy matching model. The four fuzzy matching sub-models are word frequency matching sub-model, N-Gram sub-model, edit distance sub-model and cosine calculation sub-model. Each fuzzy matching sub-model matches the candidate coded disease names, thereby ensuring the accuracy of the coding of the candidate coded disease names.

[0124] It can be understood that the standard disease classification table is stored in the server's database, and the storage address of the standard disease classification table is pre-stored in each precise matching sub-model and each fuzzy matching sub-model; when matching, each precise matching sub-model and each fuzzy matching sub-model obtains the standard disease classification table according to the storage address and performs matching based on the standard disease classification table.

[0125] Step 2042, based on each fuzzy matching sub-model, calculate the similarity between the candidate paired disease name and each standard disease name in the standard disease classification table.

[0126] Specifically, for each fuzzy matching sub-model, the similarity between the input candidate paired disease name and each standard disease name in the standard disease classification table is calculated.

[0127] In one embodiment, when the fuzzy matching sub-model is an edit distance sub-model, the steps for calculating the similarity between the candidate coded disease name and each standard disease name in the standard disease classification table specifically include: calculating the text edit distance between the candidate coded disease name and each standard disease name in the standard disease classification table; normalizing each text edit distance, and using each normalized text edit distance as the similarity between the candidate coded disease name and each standard disease name.

[0128] Specifically, for each candidate matching disease name, the edit distance sub-model calculates the text edit distance between the candidate matching disease name and each standard disease name. The text edit distance is an integer, and the smaller the text edit distance, the higher the similarity. In order to perform subsequent operations with the similarity calculated by other fuzzy matching sub-models, the text edit distance needs to be normalized, the value of the text edit distance needs to be compressed to the interval [0,1], and the normalized text edit distance is used as the similarity between the candidate matching disease name and each standard disease name.

[0129] The edit distance sub-model can normalize the text edit distance through linear normalization, standardized normalization and nonlinear normalization.

[0130] In this embodiment, the edit distance sub-model calculates the text edit distance between the candidate matched disease name and each standard disease name, and uses the normalized text edit distance as the similarity between the candidate matched disease name and the standard disease name to ensure that the second matched disease result can be generated by combining the similarities calculated by the other fuzzy matching sub-models.

[0131] Step 2043: Generate a second pairing result according to the similarity calculated by each fuzzy matching sub-model.

[0132] Specifically, from the similarities calculated by the fuzzy matching sub-models, the server can select the standard disease name and its disease code corresponding to the highest similarity as the sub-code matching result of the fuzzy matching sub-model. The sub-code matching result with the most occurrences among the sub-code matching results of each fuzzy matching sub-model is used as the second code matching result.

[0133] In one embodiment, each fuzzy matching sub-model is preset with a corresponding weight. After obtaining each sub-coding result, the weight of each sub-coding result is calculated according to the weight of each fuzzy matching sub-model, and the sub-coding result with the highest weight ratio is selected as the second coding result. For example, assuming that there are 4 fuzzy matching sub-models, the sub-coding results of two fuzzy matching sub-models are both X, and the sub-coding results of two fuzzy matching sub-models are both Y; the weights of the two fuzzy matching sub-models with sub-coding results of X are both 0.2, and the weights of the two fuzzy matching sub-models with sub-coding results of Y are both 0.3, then the weight ratio of Y (0.6) is greater than the weight ratio of X (0.4), and Y is used as the second coding result.

[0134] In one embodiment, the step of generating a second code pairing result based on the similarities calculated by each fuzzy matching sub-model specifically includes: for each candidate code pairing disease name, from the similarities calculated by each fuzzy matching sub-model, screening the standard disease name and disease code corresponding to the maximum similarity for HardVoting fusion to obtain the second code pairing result; or, performing SoftVoting fusion based on the similarities calculated by each fuzzy matching model to obtain the second code pairing result.

[0135] Among them, HardVoting fusion is to select the standard disease name and its disease code corresponding to the highest similarity from the similarities calculated by each fuzzy matching sub-model, and determine the second code pairing result according to the rule of minority obeys majority; SoftVoting fusion is to average the similarities between the candidate code pairing disease names output by each fuzzy matching sub-model and each standard disease name, and select the standard disease name and its disease code with the highest average similarity as the second code pairing result.

[0136] When using HardVoting fusion, for each candidate matching disease name, first take the standard disease name with the highest similarity calculated by each fuzzy matching sub-model and its corresponding disease code to obtain several groups of sub-matching results, and then take the standard disease name and its corresponding disease code that appear the most times in the several groups of sub-matching results as the second matching result.

[0137] For example, in the calculation results of the word frequency matching sub-model, the candidate matching disease name has the highest similarity with "peripheral neuropathy", which is 90%; the N-Gram sub-model is "peripheral neuropathy", with a similarity of 80%; the edit distance sub-model is "peripheral neuropathy", with a similarity of 100%; the cosine calculation sub-model is "peripheral neuritis", with a similarity of 85%. In the four groups of sub-matching results, "peripheral neuropathy" appears 3 times, "peripheral neuritis" appears 1 time, and the number of occurrences of "peripheral neuropathy" is greater than the number of occurrences of "peripheral neuritis", so "peripheral neuropathy" and its corresponding disease code are used as the second matching result.

[0138] When using SoftVoting fusion, for each candidate matching disease name, obtain the similarity between each fuzzy matching sub-model and all standard disease names. When there are four fuzzy matching sub-models, a total of 4*26000 similarities are obtained. Then, combined with the results of each fuzzy matching sub-model, the weighted average of the similarity between each standard disease name and the candidate matching disease name is calculated, and the standard disease name and its disease code with the highest average similarity are used as the second matching result.

[0139] For example (each fuzzy matching sub-model only lists two standard disease names as an example):

[0140] Word frequency matching submodel: peripheral neuropathy - similarity 99%; peripheral neuritis - similarity 1%;

[0141] N-Gram submodels: peripheral neuropathy - 49% similarity; peripheral neuritis - 51% similarity;

[0142] Edit distance submodels: peripheral neuropathy - similarity 40%; peripheral neuritis - similarity 60%;

[0143] Cosine calculation submodels: peripheral neuropathy - similarity 90%; peripheral neuritis - similarity 10%;

[0144] When the weights of the fuzzy matching sub-models are the same, then:

[0145] The weighted average of the similarity of “peripheral neuropathy” is: (99%+49%+40%+90%)÷4=69.5%;

[0146] The weighted average of the similarity of “peripheral neuritis”: (1%+51%+60%+10%) / 2=30.5%;

[0147] If the weighted average of the similarities of "peripheral neuropathy" is greater than the weighted average of the similarities of "peripheral neuritis", "peripheral neuropathy" and its corresponding disease code will be used as the second pair of code results.

[0148] In this embodiment, HardVoting fusion or SoftVoting fusion is performed based on the similarity calculated by each fuzzy matching sub-model, and the results of each fuzzy matching sub-model are taken into consideration to generate the second pair of code results, thereby improving the accuracy of generating the second pair of code results.

[0149] In this embodiment, the candidate matching disease name is input into each fuzzy matching sub-model in the fuzzy matching model. Each fuzzy matching sub-model uses a different method to calculate the similarity between the candidate matching disease name and each standard disease name. The similarities calculated by each fuzzy matching sub-model are then combined to generate a second matching result, thereby improving the matching accuracy of the candidate matching disease name.

[0150] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0151] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0152] Further references Figure 4 , as a response to the above Figure 2 The present application provides an embodiment of a disease name coding device, and the device embodiment is similar to Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0153] like Figure 4 As shown, the disease name coding device 300 described in this embodiment includes: a list acquisition module 301, a list deduplication module 302, an exact matching module 303, a fuzzy matching module 304 and a list generation module 305, wherein:

[0154] The list acquisition module 301 is used to acquire a list of disease names from the electronic medical record.

[0155] The list deduplication module 302 is used to perform deduplication processing on repeated disease names in the disease name list to obtain a deduplication disease name list.

[0156] The precise matching module 303 is used to input the deduplicated disease name list into the precise matching model, perform coding according to the standard disease classification table, and obtain the first coding result and candidate coding disease names.

[0157] The fuzzy matching module 304 is used to input the obtained candidate matching disease name into the fuzzy matching model, perform matching according to the standard disease classification table, and obtain a second matching result.

[0158] The list generation module 305 is used to generate a disease name code matching list according to the first code matching result and the second code matching result.

[0159] In this embodiment, the disease name list is first deduplicated to reduce the amount of calculation; the deduplicated disease name list is input into the exact matching model for exact matching to obtain a first coding result, and the disease names that fail to achieve exact matching are input into the fuzzy matching model as candidate coding disease names for fuzzy matching to obtain a second coding result. Both codings are performed according to the standard disease classification table; finally, a disease name coding list is generated based on the first coding result and the second coding result, and the disease names are matched in multiple dimensions and multiple modes through exact matching and fuzzy matching, thereby improving the accuracy of disease name matching.

[0160] In some optional implementations of this embodiment, the above-mentioned precise matching module 303 includes: a name input submodule, a name query submodule, a first generation submodule and a name marking submodule, wherein:

[0161] The name input submodule is used to input each disease name in the deduplicated disease name list into the exact matching submodel according to the arrangement order of the exact matching submodel in the exact matching model.

[0162] The name query submodule is used to query the standard disease name that matches the input disease name in the standard disease classification table through the current exact matching submodel.

[0163] The first generating submodule is used to, when a matching standard disease name is found, use the found standard disease name and the disease code corresponding to the standard disease name as the first pair of code results of the disease name.

[0164] The name input submodule is also used to input the disease name into the next accurate matching submodel to continue matching when the current accurate matching submodel fails to find a matching standard disease name.

[0165] The name marking submodule is used to mark the disease name as a candidate matching disease name if the disease name is not matched by each exact matching submodel.

[0166] In this embodiment, each disease name in the deduplicated disease name list is input into the precise matching sub-model in the arrangement order of the precise matching sub-model for matching. If a match is possible, the first code pairing result is generated. If a match is not possible, the next precise matching sub-model is input to continue matching. The precise matching sub-models are different, which ensures that the disease names can be accurately matched from multiple dimensions, thereby improving the accuracy of disease name coding.

[0167] In some optional implementations of the present embodiment, the above-mentioned name input submodule is also used to: input each disease name in the deduplicated disease name list into the precise matching submodel according to the arrangement order of the four precise matching submodels in the precise matching submodel; the four precise matching submodels include a complete matching submodel, a stop word removal submodel, a primary and secondary separation submodel, and a synonym recognition submodel.

[0168] In this embodiment, the disease name is input into the precise matching sub-model according to the arrangement order of the four precise matching sub-models in the precise matching model. The four precise matching sub-models are respectively a complete matching sub-model, a stop word removal sub-model, a primary and secondary separation sub-model and a synonym recognition sub-model. Different methods can be used to match the disease name based on the four precise matching sub-models, thereby improving the accuracy of matching the disease name.

[0169] In some optional implementations of this embodiment, the fuzzy matching module 304 includes: an input submodule, a calculation submodule and a second generation submodule, wherein:

[0170] The input submodule is used to input the obtained candidate matching disease names into each fuzzy matching submodel in the fuzzy matching model.

[0171] The calculation submodule is used to calculate the similarity between the candidate matching disease names and the standard disease names in the standard disease classification table based on each fuzzy matching submodel.

[0172] The second generating submodule is used to generate a second code pairing result according to the similarity calculated by each fuzzy matching submodel.

[0173] In this embodiment, the candidate matching disease name is input into each fuzzy matching sub-model in the fuzzy matching model. Each fuzzy matching sub-model uses a different method to calculate the similarity between the candidate matching disease name and each standard disease name. The similarities calculated by each fuzzy matching sub-model are then combined to generate a second matching result, thereby improving the matching accuracy of the candidate matching disease name.

[0174] In some optional implementations of this embodiment, the above-mentioned input submodule is also used to: input the obtained candidate code disease names into four fuzzy matching sub-models in the fuzzy matching model, and the four fuzzy matching sub-models include a word frequency matching sub-model, an N-Gram sub-model, an edit distance sub-model and a cosine calculation sub-model.

[0175] In this embodiment, the candidate coded disease names are input into four fuzzy matching sub-models in the fuzzy matching model. The four fuzzy matching sub-models are word frequency matching sub-model, N-Gram sub-model, edit distance sub-model and cosine calculation sub-model. Each fuzzy matching sub-model matches the candidate coded disease names, thereby ensuring the accuracy of the coding of the candidate coded disease names.

[0176] In some optional implementations of this embodiment, when the fuzzy matching sub-model is an edit distance sub-model, the calculation sub-module includes: a distance calculation unit and a distance normalization unit, wherein:

[0177] A distance calculation unit, used for calculating the text editing distance between the candidate matched disease name and each standard disease name in the standard disease classification table;

[0178] The distance normalization unit is used to normalize each text editing distance, and use each text editing distance after normalization as the similarity between the candidate paired disease name and the standard disease name.

[0179] In this embodiment, the edit distance sub-model calculates the text edit distance between the candidate matched disease name and each standard disease name, and uses the normalized text edit distance as the similarity between the candidate matched disease name and the standard disease name to ensure that the second matched disease result can be generated by combining the similarities calculated by the other fuzzy matching sub-models.

[0180] In some optional implementations of this embodiment, the second generation submodule includes: a HardVoting unit or a SoftVoting unit, wherein:

[0181] The HardVoting unit is used to select the standard disease name and disease code corresponding to the maximum similarity from the similarities calculated by each fuzzy matching sub-model for each candidate code-pairing disease name, and perform HardVoting fusion to obtain the second code-pairing result.

[0182] The SoftVoting unit is used to perform SoftVoting fusion according to the similarities calculated by each fuzzy matching model to obtain a second pair of code results.

[0183] In this embodiment, HardVoting fusion or SoftVoting fusion is performed based on the similarity calculated by each fuzzy matching sub-model, and the results of each fuzzy matching sub-model are taken into consideration to generate the second pair of code results, thereby improving the accuracy of generating the second pair of code results.

[0184] To solve the above technical problems, the present application also provides a computer device. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0185] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected through a system bus. It should be noted that the figure only shows a computer device 4 with components 41-43, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

[0186] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0187] The memory 41 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 41 can be an internal storage unit of the computer device 4, such as a hard disk or memory of the computer device 4. In other embodiments, the memory 41 can also be an external storage device of the computer device 4, such as a plug-in hard disk equipped on the computer device 4, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. Of course, the memory 41 can also include both the internal storage unit of the computer device 4 and its external storage device. In this embodiment, the memory 41 is generally used to store the operating system and various application software installed on the computer device 4, such as the program code of the disease name matching method, etc. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or are to be output.

[0188] The processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments. The processor 42 is generally used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to run the program code stored in the memory 41 or process data, such as running the program code of the disease name matching method.

[0189] The network interface 43 may include a wireless network interface or a wired network interface. The network interface 43 is generally used to establish a communication connection between the computer device 4 and other electronic devices.

[0190] The computer device provided in this embodiment can execute the steps of the above-mentioned disease name matching method. The steps of the disease name matching method here can be the steps of the disease name matching method in each of the above-mentioned embodiments.

[0191] In this embodiment, the disease name list is first deduplicated to reduce the amount of calculation; the deduplicated disease name list is input into the exact matching model for exact matching to obtain a first coding result, and the disease names that fail to achieve exact matching are input into the fuzzy matching model as candidate coding disease names for fuzzy matching to obtain a second coding result. Both codings are performed according to the standard disease classification table; finally, a disease name coding list is generated based on the first coding result and the second coding result, and the disease names are matched in multiple dimensions and multiple modes through exact matching and fuzzy matching, thereby improving the accuracy of disease name matching.

[0192] The present application also provides another embodiment, namely, providing a computer-readable storage medium, which stores a disease name coding program, and the disease name coding program can be executed by at least one processor to enable the at least one processor to perform the steps of the disease name coding method as described above.

[0193] In this embodiment, the disease name list is first deduplicated to reduce the amount of calculation; the deduplicated disease name list is input into the exact matching model for exact matching to obtain a first coding result, and the disease names that fail to achieve exact matching are input into the fuzzy matching model as candidate coding disease names for fuzzy matching to obtain a second coding result. Both codings are performed according to the standard disease classification table; finally, a disease name coding list is generated based on the first coding result and the second coding result, and the disease names are matched in multiple dimensions and multiple modes through exact matching and fuzzy matching, thereby improving the accuracy of disease name matching.

[0194] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0195] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A disease name coding method, It is characterized in that The steps include: Obtain a list of disease names from electronic medical records; Deduplication processing is performed on repeated disease names in the disease name list to obtain a deduplication disease name list; Input the deduplicated disease name list into the exact matching model, perform coding according to the standard disease classification table, and obtain a first coding result and candidate coding disease names; Input the obtained candidate matching disease name into the fuzzy matching model, match the code according to the standard disease classification table, and obtain a second matching result; Generate a disease name code matching list according to the first code matching result and the second code matching result; The precise matching model is composed of a number of ordered precise matching sub-models; the step of inputting the deduplicated disease name list into the precise matching model, matching the disease names according to the standard disease classification table, and obtaining the first matching result and candidate matching disease names specifically includes: Input each disease name in the deduplicated disease name list into the exact matching sub-model according to the arrangement order of the exact matching sub-model in the exact matching model; Through the current exact matching sub-model, the standard disease classification table is searched for a standard disease name that matches the input disease name; When a matching standard disease name is found, the found standard disease name and the disease code corresponding to the standard disease name are used as the first pair of code results of the disease name; When the current accurate matching sub-model fails to find a matching standard disease name, the disease name is input into the next accurate matching sub-model to continue matching; If the disease name is not matched by each exact matching sub-model, the disease name is marked as a candidate matching disease name; The step of inputting each disease name in the deduplicated disease name list into the exact matching sub-model according to the arrangement order of the exact matching sub-model in the exact matching model specifically includes: Each disease name in the deduplicated disease name list is input into the precise matching sub-model according to the arrangement order of the four precise matching sub-models in the precise matching model; the four precise matching sub-models include a complete matching sub-model, a stop word removal sub-model, a primary and secondary separation sub-model and a synonym recognition sub-model.

2. The disease name coding method according to claim 1, It is characterized in that The fuzzy matching model is composed of a plurality of fuzzy matching sub-models; the step of inputting the obtained candidate matching disease name into the fuzzy matching model, matching according to the standard disease classification table, and obtaining the second matching result specifically includes: Input the obtained candidate matched code disease names into each fuzzy matching sub-model in the fuzzy matching model; Based on the fuzzy matching sub-models, calculating the similarity between the candidate matched disease name and each standard disease name in the standard disease classification table; A second code pairing result is generated according to the similarities calculated by the fuzzy matching sub-models.

3. The method according to claim 2, It is characterized in that The step of inputting the obtained candidate matching disease names into each fuzzy matching sub-model in the fuzzy matching model specifically includes: The obtained candidate paired disease names are input into four fuzzy matching sub-models in the fuzzy matching model, and the four fuzzy matching sub-models include a word frequency matching sub-model, an N-Gram sub-model, an edit distance sub-model and a cosine calculation sub-model.

4. The disease name coding method according to claim 3, It is characterized in that When the fuzzy matching sub-model is an edit distance sub-model, the steps of calculating the similarity between the candidate paired disease name and each standard disease name in the standard disease classification table specifically include: Calculating the text edit distance between the candidate matched code disease name and each standard disease name in the standard disease classification table; Each text edit distance is normalized, and each text edit distance after normalization is used as the similarity between the candidate paired disease name and the standard disease name.

5. The disease name coding method according to claim 2, It is characterized in that The step of generating a second pairing result according to the similarities calculated by the fuzzy matching sub-models specifically includes: For each candidate code-matching disease name, the standard disease name and disease code corresponding to the maximum similarity are selected from the similarities calculated by the fuzzy matching sub-models to perform Hard Voting fusion to obtain the second code-matching result; or, SoftVoting fusion is performed according to the similarities calculated by the fuzzy matching models to obtain a second pair of code results.

6. A disease name coding device, It is characterized in that include: List acquisition module, used to obtain a list of disease names from electronic medical records; A list deduplication module is used to perform deduplication processing on repeated disease names in the disease name list to obtain a deduplication disease name list; An accurate matching module, used for inputting the deduplicated disease name list into an accurate matching model, matching the disease names according to a standard disease classification table, and obtaining a first matching result and candidate matching disease names; A fuzzy matching module, used for inputting the obtained candidate matching disease name into the fuzzy matching model, matching the disease name according to the standard disease classification table, and obtaining a second matching result; A list generating module, used for generating a disease name code matching list according to the first code matching result and the second code matching result; The precise matching model is composed of a number of ordered precise matching sub-models; the deduplicated disease name list is input into the precise matching model, and the code is matched according to the standard disease classification table to obtain the first matching result and the candidate matching disease names including: Input each disease name in the deduplicated disease name list into the exact matching sub-model according to the arrangement order of the exact matching sub-model in the exact matching model; Through the current exact matching sub-model, the standard disease classification table is searched for a standard disease name that matches the input disease name; When a matching standard disease name is found, the found standard disease name and the disease code corresponding to the standard disease name are used as the first pair of code results of the disease name; When the current accurate matching sub-model fails to find a matching standard disease name, the disease name is input into the next accurate matching sub-model to continue matching; If the disease name is not matched by each exact matching sub-model, the disease name is marked as a candidate matching disease name; The step of inputting each disease name in the deduplicated disease name list into the exact matching sub-model according to the arrangement order of the exact matching sub-model in the exact matching model comprises: Each disease name in the deduplicated disease name list is input into the precise matching sub-model according to the arrangement order of the four precise matching sub-models in the precise matching model; the four precise matching sub-models include a complete matching sub-model, a stop word removal sub-model, a primary and secondary separation sub-model and a synonym recognition sub-model.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the disease name coding method as described in any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the disease name coding method according to any one of claims 1 to 5.

Citation Information

Patent Citations

  • Knowledge-based electronic medical record quality control method

    CN106682397A

  • Method for disambiguating entities in medical disease diagnosis record

    CN106951684A

  • Automatic disease coding method and system

    CN107705839A