Standardized processing method and device for medical text data

By performing word segmentation processing and calculation of the similarity of medical knowledge graph inversion index on medical knowledge data, the problem of inconsistent medical data specifications between different hospitals is solved, the standardized processing of medical text data is realized, and the efficiency of academic communication and telemedicine is improved.

CN114004228BActive Publication Date: 2024-11-08泰康保险集团股份有限公司
View PDF 9 Cites 0 Cited by

Patent Information

Application Number
CN202111262573.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-10-28
Publication Date
2024-11-08
Estimated Expiration
2041-10-28

AI Technical Summary

Technical Problem

The degree of informatization and medical data recording standards between different hospitals are inconsistent, which makes it difficult for medical text data to be interconnected, affecting academic exchanges and telemedicine data transmission.

Method used

By receiving the original medical text data after desensitization, word segmentation is performed, and using the pre-constructed medical knowledge graph and inverted index method, the similarity is calculated to determine the candidate medical standard word collection and standardization is performed.

Benefits of technology

It improves the efficiency of communication between hospitals, ensures the standardization of medical data, and supports more effective academic exchanges and telemedicine.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114004228B_ABST
    Figure CN114004228B_ABST
Patent Text Reader

Abstract

The present invention discloses a method and device for standardizing medical text data, wherein the method comprises: receiving desensitized original medical text data; performing word segmentation processing on the received original medical text data to obtain an original medical text word segmentation set corresponding to the original medical text data; based on a pre-constructed medical knowledge graph, using an inverted index method to determine one or more candidate medical standard word sets corresponding to the original medical text word segmentation set; calculating the similarity between the original medical text word segmentation set and each candidate medical standard word set, and determining the candidate medical standard word set with the greatest similarity as the target medical standard word set of the original medical text data; and performing standardization processing on the original medical text data according to the target medical standard word set. The present invention utilizes standardized medical text data for academic exchanges between hospitals and inter-hospital telemedicine, which can greatly improve the efficiency of academic exchanges and telemedicine.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and device for standardizing processing of medical text data. Background Art

[0002] This section is intended to provide a background or context to the embodiments of the invention recited in the claims. No admission is made that the description herein is prior art by inclusion in this section.

[0003] Due to the different development levels of information technology in various hospitals, the degree of informatization among different hospitals varies greatly, and the standards followed for recording medical data are also uneven, which has caused great difficulties for academic exchanges and telemedicine data transmission between hospitals. Therefore, how to provide a method for standardizing medical text data and quickly realize the standardized processing of medical text data is a technical problem that needs to be solved urgently. Summary of the invention

[0004] The embodiment of the present invention provides a standardized processing method for medical text data, which is used to solve the technical problem that there are many non-standard medical words in the original medical text data of the existing hospital system, and it is difficult to communicate with each other. The method includes: receiving the desensitized original medical text data; performing word segmentation processing on the collected original medical text data to obtain an original medical text word segmentation set corresponding to the original medical text data, wherein the original medical text word segmentation set includes: one or more original medical text word segmentations; based on a pre-constructed medical knowledge graph, using an inverted index method to determine one or more candidate medical standard word sets corresponding to the original medical text word segmentation set, wherein the medical knowledge graph includes: multiple medical standards and one or more candidate medical standard words corresponding to each medical standard; calculating the similarity between the original medical text word segmentation set and each candidate medical standard word set, and determining the candidate medical standard word set with the largest similarity as the target medical standard word set of the original medical text data; and performing standardized processing on the original medical text data according to the target medical standard word set.

[0005] The embodiment of the present invention also provides a standardized processing device for medical text data, which is used to solve the technical problem that there are many non-standard medical words in the original medical text data of the existing hospital system, which is difficult to communicate. The device includes: a medical text data acquisition module, which is used to receive the original medical text data after desensitization; a text segmentation module, which is used to perform segmentation processing on the collected original medical text data to obtain an original medical text segmentation set corresponding to the original medical text data, wherein the original medical text segmentation set includes: one or more original medical text segmentations; a knowledge graph query module, which is used to determine one or more candidate medical standard word sets corresponding to the original medical text segmentation set based on a pre-constructed medical knowledge graph using an inverted index method, wherein the medical knowledge graph includes: multiple medical standards and one or more candidate medical standard words corresponding to each medical standard; a medical standard word set determination module, which is used to calculate the similarity between the original medical text segmentation set and each candidate medical standard word set, and determine the candidate medical standard word set with the largest similarity as the target medical standard word set of the original medical text data; a medical standardization processing module, which is used to perform standardized processing on the original medical text data according to the target medical standard word set.

[0006] An embodiment of the present invention also provides a computer device to solve the technical problem that the original medical text data in the existing hospital system contains many non-standard medical terms and is difficult to communicate. The computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the above-mentioned standardized processing method for medical text data is implemented.

[0007] An embodiment of the present invention also provides a computer-readable storage medium to solve the technical problem that the original medical text data in the existing hospital system contains many non-standard medical terms and is difficult to communicate. The computer-readable storage medium stores a computer program for executing the above-mentioned standardized processing method of medical text data.

[0008] The standardized processing method, device, computer equipment and computer-readable storage medium for medical text data provided in the embodiments of the present invention collect original medical text data, perform word segmentation processing on the original medical text data to obtain an original medical text word segmentation set corresponding to the original medical text data, and then determine one or more candidate medical standard word sets corresponding to the original medical text word segmentation set based on a pre-constructed medical knowledge graph using an inverted index method, calculate the similarity between the original medical text word segmentation set and each candidate medical standard word set, and determine the candidate medical standard word set with the greatest similarity as the target medical standard word set of the original medical text data, and finally perform standardized processing on the original medical text data according to the target medical standard word set.

[0009] Compared with the technical solution in the prior art that directly uses original medical text data for communication, the embodiment of the present invention can greatly improve efficiency by standardizing the original medical text data and then using the standardized medical text data for academic exchanges and telemedicine. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0011] Figure 1 A flow chart of a method for standardizing medical text data provided in an embodiment of the present invention;

[0012] Figure 2 A flowchart of determining a candidate medical standard word set based on a medical text semantic classification model provided in an embodiment of the present invention;

[0013] Figure 3 A flowchart of performing machine learning on a medical text semantic classification model provided in an embodiment of the present invention;

[0014] Figure 4 A flowchart of determining a candidate medical standard word set based on a medical knowledge graph provided in an embodiment of the present invention;

[0015] Figure 5 A flow chart of a method for constructing a medical knowledge graph provided in an embodiment of the present invention;

[0016] Figure 6 A flowchart of performing telemedicine using standardized medical text data provided in an embodiment of the present invention;

[0017] Figure 7 A flowchart of storing business data based on blockchain provided in an embodiment of the present invention;

[0018] Figure 8 A schematic diagram of a standardized processing device for medical text data provided in an embodiment of the present invention;

[0019] Fig. 9 A schematic diagram of an optional standardized processing device for medical text data provided in an embodiment of the present invention;

[0020] Fig.10 A schematic diagram of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0022] An embodiment of the present invention provides a standardized processing method for medical text data, which can be applied to but not limited to academic exchanges and telemedicine.

[0023] Figure 1 FIG. 1 is a flow chart of a method for standardizing medical text data provided in an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0024] S101, receiving the desensitized original medical text data.

[0025] The original medical text data in S101 above can be the medical data entered by medical staff. For some non-text medical data, text data can be obtained after certain processing, for example, text data obtained by optical character recognition of paper documents such as test reports. These desensitized original medical text data are data that can be collected, stored, and used for subsequent medical treatment or insurance underwriting, claims, product recommendations, etc. with the authorization of the user.

[0026] S102, performing word segmentation processing on the collected original medical text data to obtain an original medical text word segmentation set corresponding to the original medical text data, wherein the original medical text word segmentation set includes: one or more original medical text word segmentations.

[0027] After the original medical text data is collected, the original medical text data is segmented to obtain one or more corresponding original medical text segmentations, forming an original medical text segmentation set corresponding to the original medical text data.

[0028] S103, based on a pre-constructed medical knowledge graph, an inverted index method is used to determine one or more candidate medical standard word sets corresponding to the original medical text word set, wherein the medical knowledge graph includes: multiple medical standards and one or more candidate medical standard words corresponding to each medical standard.

[0029] It should be noted that the medical knowledge graph in the embodiment of the present invention is a pre-constructed one that contains multiple medical standards and one or more candidate medical standard words corresponding to each medical standard. Each original medical text segmentation in the original medical text segmentation set is matched with each candidate medical standard word in the medical knowledge graph. The medical standard corresponding to the successfully matched candidate medical standard word can be determined as the medical standard corresponding to the original medical text data, and then the original medical text data is standardized based on the corresponding medical standards.

[0030] Since the candidate medical standard words corresponding to different medical standards may be the same, there may be more than one medical standard obtained through word segmentation matching, and there is also more than one corresponding candidate medical standard word set.

[0031] S104, calculating the similarity between the original medical text word set and each candidate medical standard word set, and determining the candidate medical standard word set with the greatest similarity as the target medical standard word set of the original medical text data.

[0032] Since there may be more than one candidate medical standard word set corresponding to the original medical text word set, the candidate medical standard word set with the greatest similarity can be determined as the target medical standard word set of the original medical text data based on a similarity algorithm.

[0033] S105, standardizing the original medical text data according to the target medical standard word set.

[0034] After determining the target medical standard word set of the original medical text data, the original medical text data is standardized using the medical standard words in the target medical standard word set to obtain standardized medical text data.

[0035] When the standardized processing method for medical text data provided in the embodiment of the present invention is applied to a telemedicine system, it is necessary to compare the medical personnel information in one hospital system with the medical personnel information in another hospital system. No matter which hospital's medical personnel information is sent to the other party for comparison, there is a risk of data leakage. In the embodiment of the present invention, the medical personnel information is encrypted based on an irreversible encryption algorithm based on salted hashing. Each hospital system uses the same encryption algorithm to encrypt the sensitive information of the same medical personnel (for example, ID number). The encrypted ciphertext is consistent. Therefore, the comparison can be based on the encrypted user sensitive information ciphertext. In this way, the same medical personnel in different hospitals can be compared and the leakage of sensitive information can be avoided.

[0036] It should be noted that for medical text segmentation words that are not in the medical knowledge graph, it may be impossible to find the corresponding medical standard words based on the knowledge graph. Therefore, in one embodiment, after the collected original medical text data is segmented to obtain the original medical text segmentation set corresponding to the original medical text data, such as Figure 2 As shown, the standardized processing method for medical text data provided in the embodiment of the present invention can also determine the candidate medical standard word set corresponding to the original medical text word set through the following steps:

[0037] S201, obtaining a pre-trained medical text semantic classification model, wherein the medical text semantic classification model is used to predict a candidate medical standard word set corresponding to an original medical text word set;

[0038] S202, input each original medical text segmentation contained in the original medical text segmentation set into a pre-trained medical text semantic classification model, and output one or more candidate medical standard word sets corresponding to the original medical text segmentation set.

[0039] In one embodiment, Figure 3 As shown, the standardized processing method for medical text data provided in the embodiment of the present invention can also perform machine learning on the medical text semantic classification model through the following machine learning process:

[0040] S301, obtaining multiple medical text segmentations with marked medical classifications and standard words as sample data;

[0041] S302, using sample data, performing machine learning on the conditional random field model to train a medical text semantic classification model that can predict the original medical text word set corresponding to the candidate medical standard word set.

[0042] In one embodiment, Figure 4 As shown, the standardized processing method for medical text data provided in the embodiment of the present invention can be implemented by the following steps when determining one or more candidate medical standard word sets corresponding to the original medical text word set based on the pre-built medical knowledge graph and using the inverted index method:

[0043] S401, matching each original medical text segmentation contained in the original medical text segmentation set with the text segmentation of each medical standard in the medical knowledge graph, to obtain a plurality of medical standards containing one or more original medical text segmentations in the original medical text segmentation set;

[0044] S402, determining the intersection of the original medical text word set and the word set of each medical standard;

[0045] S403: Determine the medical standards in the segmentation intersection that contain a number of segmentations exceeding a preset threshold as a candidate medical standard word set corresponding to the original medical text segmentation set.

[0046] Each medical standard is generally stored in the database as follows: ID->standard code, standard name, remarks, ...; the standard name of each medical standard is segmented, and then the following inverted index is made: segmentation X->ID of the mth standard of standard A, ID of the nth standard of standard B, ... This means that the name of the mth standard of standard A, the name of the nth standard of standard B, and so on contain the above segmentation X after segmentation. In this way, we can know which standard the segmentations contained in a certain medical text data can correspond to after segmentation processing. The segmentations contained in this medical health data and the segmentations contained in the standards that can be found to be related are then calculated for intersection. The standards with the largest intersection are the candidate medical standard word sets for the next step.

[0047] Furthermore, before determining one or more candidate medical standard word sets corresponding to the original medical text word set by using the inverted index method based on the pre-built medical knowledge graph, such as Figure 5 As shown, the standardized processing method for medical text data provided in the embodiment of the present invention can also construct a medical knowledge graph through the following steps:

[0048] S501, obtaining multiple medical standards and multiple medical standard words contained in each medical standard;

[0049] S502, constructing a medical knowledge graph according to multiple medical standards and multiple medical standard words contained in each medical standard.

[0050] Since the medical knowledge graph stores the detailed word segmentation of each medical standard, this can solve the problem that the word library on the market has poor word segmentation effect on medical and health data. For example, the word "mesothelial tissue" could not be accurately segmented before using the medical knowledge graph. Now this word segmentation is no longer a problem. Spaces, punctuation marks, and other symbols that have no medical significance are stored in the knowledge graph as stop words with weight attributes. These stop words are interference information for mapping standard fields, so they need to be removed before word segmentation.

[0051] It should be noted that the medical knowledge graph constructed in the embodiment of the present invention is automatically updateable. Based on the medical knowledge graph, all received medical health data are segmented and stop words are removed. Then, Elasticsearch is used to perform inverted indexing on mature medical standards to obtain a medical knowledge graph containing medical standard words of multiple medical standards, so that the original medical text data can be medically classified through word segmentation comparison.

[0052] For example, "D18.0" is a standard code for a sub-category in ICD-10, and its corresponding name is "hemangioma, any part". Its parent category is "D18", and its corresponding name is "hemangioma and lymphangioma, any part", and the two nodes "hemangioma" and "any part" are their participles. Assuming that a piece of original medical text data is a certain hemangioma, first look for such a hemangioma with the same name in the ICD-10 standard. If not, look for similar ones in the details under the standard "D18.0 Hemangioma, any part". If no similar ones are found, the corresponding standard is "D18.0 Hemangioma, any part" of ICD-10. If similar ones are found (such as "D18.003 Facial hemangioma"), the corresponding standard is "D18.003 Facial hemangioma" of ICD-10.

[0053] In one embodiment, after the original medical text data is standardized according to the target medical standard word set, as shown in FIG. Figure 6 As shown, the standardized processing method for medical text data provided in the embodiment of the present invention may also include the following steps:

[0054] S601, obtaining standardized medical text data through original medical text data;

[0055] S602, constructing health record data based on standardized medical text data;

[0056] S603, assisting telemedicine based on the health record data of the same patient in different hospitals.

[0057] In order to ensure that medical data is not tampered with, the medical data can be stored on the blockchain network. However, since the larger the amount of data stored on the blockchain network, the lower the data storage efficiency will be, in one embodiment, Figure 7 As shown, the standardized processing method for medical text data provided in the embodiment of the present invention may also include the following steps to store business data:

[0058] S701, receiving the desensitized original medical text data;

[0059] S702, storing the hash value of the original medical text data in the blockchain network, and storing the original medical text data in the relational database.

[0060] As can be seen from the above, the standardized processing method for medical text data provided in the embodiment of the present invention, after performing word segmentation processing on the original medical text data, based on the pre-constructed medical knowledge graph and combined with the inverted index method, determines one or more candidate standard data corresponding to the original medical text data; for medical data for which the mapping standard cannot be determined through the knowledge graph, after semantic classification of the original medical text data using a pre-trained semantic classification model, predicts one or more candidate standard data corresponding to the original medical text data; finally, combined with the word segmentation similarity algorithm, obtains the standard data to which the original medical text data is ultimately mapped from the candidate standard data, and finally, assists in telemedicine, academic exchanges and other services based on the standard data to which the original medical text data is ultimately mapped.

[0061] Based on the same inventive concept, the present invention also provides a standardized processing device for medical text data, as described in the following embodiments. Since the principle of the device to solve the problem is similar to the standardized processing method for medical text data, the implementation of the device can refer to the implementation of the standardized processing method for medical text data, and the repeated parts will not be repeated.

[0062] Figure 8 FIG. 1 is a schematic diagram of a standardized processing device for medical text data provided in an embodiment of the present invention. Figure 8 As shown, the device includes: a medical text data acquisition module 801, a text segmentation module 802, a knowledge graph query module 803, a medical standard word set determination module 804 and a medical standardization processing module 805.

[0063] Among them, the medical text data collection module 801 is used to receive the original medical text data after desensitization; the text segmentation module 802 is used to perform segmentation processing on the received original medical text data to obtain the original medical text segmentation set corresponding to the original medical text data, wherein the original medical text segmentation set includes: one or more original medical text segmentations; the knowledge graph query module 803 is used to determine one or more candidate medical standard word sets corresponding to the original medical text segmentation set based on the pre-constructed medical knowledge graph using an inverted index method, wherein the medical knowledge graph includes: multiple medical standards and one or more candidate medical standard words corresponding to each medical standard; the medical standard word set determination module 804 is used to calculate the similarity between the original medical text segmentation set and each candidate medical standard word set, and determine the candidate medical standard word set with the largest similarity as the target medical standard word set of the original medical text data; the medical standardization processing module 805 is used to perform standardization processing on the original medical text data according to the target medical standard word set.

[0064] It should be noted that the above-mentioned medical text data collection module 801, text segmentation module 802, knowledge graph query module 803, medical standard word set determination module 804 and medical standardization processing module 805 correspond to S101 to S105 in the method embodiment, and the examples and application scenarios implemented by the above-mentioned modules and corresponding steps are the same, but are not limited to the contents disclosed in the above-mentioned method embodiment. It should be noted that the above-mentioned modules as part of the device can be executed in a computer system such as a set of computer executable instructions.

[0065] In one embodiment, Fig. 9 As shown, the standardized processing device for medical text data provided in the embodiment of the present invention may also include: a semantic classification prediction module 806, which is used to obtain a pre-trained medical text semantic classification model, and input each original medical text segmentation contained in the original medical text segmentation set into the pre-trained medical text semantic classification model, and output one or more candidate medical standard word sets corresponding to the original medical text segmentation set, wherein the medical text semantic classification model is used to predict the candidate medical standard word set corresponding to the original medical text segmentation set.

[0066] It should be noted that the semantic classification prediction module 806 corresponds to S201 to S202 in the method embodiment, and the examples and application scenarios implemented by the modules and corresponding steps are the same, but are not limited to the contents disclosed in the method embodiment. It should be noted that the modules as part of the device can be executed in a computer system such as a set of computer executable instructions.

[0067] In one embodiment, Fig. 9 As shown, the standardized processing device for medical text data provided in the embodiment of the present invention may also include: a machine learning module 807, which is used to obtain multiple medical text segmentations that have been labeled with medical classifications and standard words as sample data; and use the sample data to perform machine learning on the conditional random field model to train a medical text semantic classification model that can predict the original medical text segmentation set corresponding to the candidate medical standard word set.

[0068] It should be noted that the machine learning module 807 corresponds to S301 to S302 in the method embodiment, and the examples and application scenarios implemented by the modules and corresponding steps are the same, but are not limited to the contents disclosed in the method embodiment. It should be noted that the modules as part of the device can be executed in a computer system such as a set of computer executable instructions.

[0069] In one embodiment, Fig. 9As shown, in the standardized processing device for medical text data provided in the embodiment of the present invention, the knowledge graph query module 803 specifically includes: a segmentation matching unit 8031, which is used to match each original medical text segmentation contained in the original medical text segmentation set with the text segmentation of each medical standard in the medical knowledge graph to obtain multiple medical standards containing one or more original medical text segmentations in the original medical text segmentation set; a segmentation intersection determination unit 8032, which is used to determine the segmentation intersection of the original medical text segmentation set and each medical standard; a candidate medical standard word set determination unit 8033, which is used to determine the medical standard whose number of segments in the segmentation intersection exceeds a preset threshold as the candidate medical standard word set corresponding to the original medical text segmentation set.

[0070] It should be noted that the above-mentioned word segmentation matching unit 8031, word segmentation intersection determination unit 8032 and candidate medical standard word set determination unit 8033 correspond to S401 to S403 in the method embodiment, and the above-mentioned modules and corresponding steps implement the same examples and application scenarios, but are not limited to the contents disclosed in the above-mentioned method embodiment. It should be noted that the above-mentioned modules as part of the device can be executed in a computer system such as a set of computer executable instructions.

[0071] In one embodiment, Fig. 9 As shown, the standardized processing device for medical text data provided in the embodiment of the present invention may also include: a knowledge graph construction module 808, which is used to obtain multiple medical standards and multiple medical standard words contained in each medical standard; and construct a medical knowledge graph based on the multiple medical standards and the multiple medical standard words contained in each medical standard.

[0072] It should be noted that the above-mentioned knowledge graph construction module 808 corresponds to S501-S502 in the method embodiment, and the examples and application scenarios implemented by the above-mentioned modules and corresponding steps are the same, but are not limited to the contents disclosed in the above-mentioned method embodiment. It should be noted that the above-mentioned module as part of the device can be executed in a computer system such as a set of computer executable instructions.

[0073] In one embodiment, Fig. 9 As shown, the standardized processing device for medical text data provided in the embodiment of the present invention may also include: a business processing module 809, which is used to obtain standardized medical text data through original medical text data; construct health record data based on standardized medical text data; and assist telemedicine based on the health record data of the same medical person in different hospitals.

[0074] It should be noted that the business handling module 809 corresponds to S601 to S603 in the method embodiment, and the examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the contents disclosed in the above method embodiment. It should be noted that the above modules, as part of the device, can be executed in a computer system such as a set of computer executable instructions.

[0075] In one embodiment, Fig. 9 As shown, the standardized processing device for medical text data provided in the embodiment of the present invention may also include: a blockchain data storage module 810, which is used to store the hash value of the original medical text data in the blockchain network, and store the original medical text data in a relational database.

[0076] It should be noted that the above-mentioned blockchain data storage module 810 corresponds to S701-S702 in the method embodiment, and the examples and application scenarios implemented by the above-mentioned modules and corresponding steps are the same, but are not limited to the contents disclosed in the above-mentioned method embodiment. It should be noted that the above-mentioned module as part of the device can be executed in a computer system such as a set of computer executable instructions.

[0077] Based on the same inventive concept, the embodiment of the present invention also provides a computer device to solve the technical problem that there are many non-standard medical words in the original medical text data of the existing hospital system, which makes it difficult to communicate with each other. Fig.10 A schematic diagram of a computer device provided in an embodiment of the present invention is shown in FIG. Fig.10 As shown, the computer device 100 includes a memory 1001, a processor 1002, and a computer program stored in the memory 1001 and executable on the processor 1002. When the processor 1002 executes the computer program, the above-mentioned standardized processing method for medical text data is implemented.

[0078] Based on the same inventive concept, an embodiment of the present invention also provides a computer-readable storage medium to solve the technical problem that the original medical text data in the existing hospital system contains many non-standard medical terms and is difficult to communicate with each other. The computer-readable storage medium stores a computer program for executing the above-mentioned standardized processing method of medical text data.

[0079] In summary, the standardized processing method, apparatus, computer device and computer-readable storage medium for medical text data provided in the embodiments of the present invention, after receiving the original medical text data, perform word segmentation processing on the original medical text data to obtain the original medical text word segmentation set corresponding to the original medical text data, and then based on the pre-constructed medical knowledge graph, use the inverted index method to determine one or more candidate medical standard word sets corresponding to the original medical text word segmentation set, calculate the similarity between the original medical text word segmentation set and each candidate medical standard word set, and determine the candidate medical standard word set with the largest similarity as the target medical standard word set of the original medical text data, and finally perform standardized processing on the original medical text data according to the target medical standard word set.

[0080] Compared with the technical solutions in the prior art that directly use various original medical documents and texts for remote medical care and academic exchanges, the embodiments of the present invention can greatly improve efficiency by standardizing the original medical text data and then using the standardized medical text data for remote medical care and academic exchanges.

[0081] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0082] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0083] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0084] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0085] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A standardized processing method for medical text data, It is characterized in that include: Receive the desensitized original medical text data; Performing word segmentation processing on the collected original medical text data to obtain an original medical text word segmentation set corresponding to the original medical text data, wherein the original medical text word segmentation set includes: one or more original medical text word segmentations; Based on a pre-constructed medical knowledge graph, an inverted index method is used to determine one or more candidate medical standard word sets corresponding to the original medical text word set, wherein the medical knowledge graph includes: multiple medical standards and one or more candidate medical standard words corresponding to each medical standard; Calculating the similarity between the original medical text word set and each candidate medical standard word set, and determining the candidate medical standard word set with the greatest similarity as the target medical standard word set of the original medical text data; According to the target medical standard word set, the original medical text data is standardized; For medical text segmentation words that are not in the medical knowledge graph, it is impossible to find the corresponding medical standard words based on the knowledge graph. After the collected original medical text data is segmented to obtain the original medical text segmentation set corresponding to the original medical text data, the method further includes: Acquire a pre-trained medical text semantic classification model, wherein the medical text semantic classification model is used to predict a candidate medical standard word set corresponding to the original medical text word set; Input each original medical text segmentation contained in the original medical text segmentation set into a pre-trained medical text semantic classification model, and output one or more candidate medical standard word sets corresponding to the original medical text segmentation set; Based on the pre-constructed medical knowledge graph, an inverted index method is used to determine one or more candidate medical standard word sets corresponding to the original medical text word set, including: Matching each original medical text segmentation contained in the original medical text segmentation set with the text segmentation of each medical standard in the medical knowledge graph to obtain a plurality of medical standards containing one or more original medical text segmentations in the original medical text segmentation set; Determine the intersection of the original medical text word segmentation set and the word segmentation of each medical standard; Determine the medical standards whose segmentation number in the segmentation intersection exceeds a preset threshold as the candidate medical standard word set corresponding to the original medical text segmentation set; After the original medical text data is standardized according to the target medical standard word set, the method further includes: Obtain standardized medical text data; Construct health record data based on standardized medical text data; Telemedicine is assisted by the health record data of the same patient in different hospitals.

2. The method according to claim 1, It is characterized in that Get pre-trained medical text semantic classification models, including: Obtain multiple medical text segmentations with annotated medical classifications and standard words as sample data; The sample data is used to perform machine learning on the conditional random field model to train a medical text semantic classification model that can predict the candidate medical standard word set corresponding to the original medical text word set.

3. The method according to claim 1, It is characterized in that Before determining one or more candidate medical standard word sets corresponding to the original medical text word set by using an inverted index method based on a pre-constructed medical knowledge graph, the method further includes: Acquire multiple medical standards and multiple medical standard words contained in each medical standard; A medical knowledge graph is constructed based on the multiple medical standards and the multiple medical standard words contained in each medical standard.

4. The method according to claim 1, Features: The hash value of the original medical text data is stored in the blockchain network, and the original medical text data is stored in the relational database.

5. A standardized processing device for medical text data, It is characterized in that include: A medical text data collection module is used to receive the original medical text data after desensitization; A text segmentation module is used to perform segmentation processing on the collected original medical text data to obtain an original medical text segmentation set corresponding to the original medical text data, wherein the original medical text segmentation set includes: one or more original medical text segmentations; A knowledge graph query module is used to determine one or more candidate medical standard word sets corresponding to the original medical text word set based on a pre-built medical knowledge graph by using an inverted index method, wherein the medical knowledge graph includes: multiple medical standards and one or more candidate medical standard words corresponding to each medical standard; A medical standard word set determination module, used to calculate the similarity between the original medical text word set and each candidate medical standard word set, and determine the candidate medical standard word set with the greatest similarity as the target medical standard word set of the original medical text data; A medical standardization processing module, used for performing standardization processing on the original medical text data according to the target medical standard word set; For medical text segmentation that is not in the medical knowledge graph, it is impossible to find the corresponding medical standard words based on the knowledge graph. The text segmentation module is also used for: Acquire a pre-trained medical text semantic classification model, wherein the medical text semantic classification model is used to predict a candidate medical standard word set corresponding to the original medical text word set; Input each original medical text segmentation contained in the original medical text segmentation set into a pre-trained medical text semantic classification model, and output one or more candidate medical standard word sets corresponding to the original medical text segmentation set; The knowledge graph query module is specifically used for: Matching each original medical text segmentation contained in the original medical text segmentation set with the text segmentation of each medical standard in the medical knowledge graph to obtain a plurality of medical standards containing one or more original medical text segmentations in the original medical text segmentation set; Determine the intersection of the original medical text word segmentation set and the word segmentation of each medical standard; Determine the medical standards whose segmentation number in the segmentation intersection exceeds a preset threshold as the candidate medical standard word set corresponding to the original medical text segmentation set; Medical standardization processing module is also used for: Obtain standardized medical text data; Construct health record data based on standardized medical text data; Telemedicine is assisted by the health record data of the same patient in different hospitals.

6. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the standardized processing method of medical text data described in any one of claims 1 to 4 is implemented.

7. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores a computer program for executing the method for standardizing medical text data according to any one of claims 1 to 4.

Citation Information

Patent Citations

  • Medical text semantic recognition method, medical text semantic recognition device, electronic equipment and readable storage medium

    CN111128394A

  • Medical term standardization method and device, computer equipment and storage medium

    CN111581976A

  • Medical information search method and device, electronic device and storage medium

    CN112199494A

  • Medical bill code matching method and device

    CN113139498A

  • Automatic questioning and answering processing method and automatic questioning and answering system

    US20190243900A1