WO2020052162A1 - 疾病数据映射方法、装置、计算机设备和存储介质 - Google Patents

疾病数据映射方法、装置、计算机设备和存储介质 Download PDF

Info

Publication number
WO2020052162A1
WO2020052162A1 PCT/CN2018/123365 CN2018123365W WO2020052162A1 WO 2020052162 A1 WO2020052162 A1 WO 2020052162A1 CN 2018123365 W CN2018123365 W CN 2018123365W WO 2020052162 A1 WO2020052162 A1 WO 2020052162A1
Authority
WO
WIPO (PCT)
Prior art keywords
matching
node
word
words
disease data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2018/123365
Other languages
English (en)
French (fr)
Inventor
胡帆
胡雪莹
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An Technology Shenzhen Co Ltd
Original Assignee
Ping An Technology Shenzhen Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ping An Technology Shenzhen Co Ltd filed Critical Ping An Technology Shenzhen Co Ltd
Publication of WO2020052162A1 publication Critical patent/WO2020052162A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H70/00—ICT specially adapted for the handling or processing of medical references
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/243—Classification techniques relating to the number of classes
    • G06F18/24323—Tree-organised classifiers
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/237—Lexical tools
    • G06F40/247—Thesauruses; Synonyms
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/279—Recognition of textual entities
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
    • G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Definitions

  • the present application relates to a method, an apparatus, a computer device, and a storage medium for mapping disease data.
  • a method, an apparatus, a computer device, and a storage medium for mapping disease data are provided.
  • a disease data mapping method includes:
  • a matching path where the current node is located is determined, and target disease data is obtained according to the matching path.
  • a disease data mapping device includes:
  • a segmentation result acquisition module configured to obtain disease data to be mapped, and perform segmentation on the disease data to be mapped to obtain a segmentation result
  • a first matching module configured to match a word in the segmentation result with a root node of each matching tree in a matching forest, and select a target root node as a current node according to the root node matching result;
  • a second matching module configured to match a word in the segmentation result with the next-level child node when the current node has a corresponding next-level child node
  • the current node acquisition module is configured to select a target child node as a current node according to the child node matching result, and enter a corresponding next-level child node when the current node has the word in the segmentation result and the next node. Steps for matching sub-nodes;
  • a target disease data acquisition module is configured to determine a matching path where the current node is located when the current node does not have a corresponding next-level child node, and obtain target disease data according to the matching path.
  • a computer device includes a memory and one or more processors.
  • the memory stores computer-readable instructions.
  • the steps of the disease data mapping method provided in any embodiment of the present application are implemented.
  • One or more non-volatile computer-readable storage media storing computer-readable instructions, and when the computer-readable instructions are executed by one or more processors, the one or more processors implement one of the embodiments of the present application Steps of the provided disease data mapping method.
  • FIG. 1 is an application scenario diagram of a disease data mapping method according to one or more embodiments.
  • FIG. 2 is a schematic flowchart of a disease data mapping method according to one or more embodiments.
  • FIG. 3 is a schematic diagram of a matching tree according to one or more embodiments.
  • FIG. 4 is a schematic flowchart of a disease data mapping method in another embodiment.
  • FIG. 5 is a block diagram of a disease data mapping apparatus according to one or more embodiments.
  • FIG. 6 is a block diagram of a computer device according to one or more embodiments.
  • the disease data mapping method provided in this application can be applied to the application environment shown in FIG. 1.
  • the terminal 102 communicates with the server 104 through a network.
  • the server 104 obtains the disease data to be mapped from the terminal, segmentes the disease data, and obtains the segmentation result; matches the words in the segmentation result with the root node of each matching tree in the matching tree set, and selects the target root according to the root node matching result.
  • Node as the current node; match the words in the word segmentation result with the child nodes corresponding to the current node; select the target child node as the current node according to the child node matching results, and enter the child node corresponding to the words in the word segmentation result with the current node Step of matching; when the current node does not have a corresponding child node, determine the matching path where the current node is located, obtain the target disease data according to the matching path, and finally, the server 104 may return the target disease data to the terminal 102.
  • the terminal 102 may be, but is not limited to, various personal computers, notebook computers, smart phones, and tablet computers.
  • the server 104 may be implemented by an independent server or a server cluster composed of multiple servers.
  • a disease data mapping method is provided.
  • the method is applied to the server in FIG. 1 as an example, and includes the following steps:
  • Step S202 Obtain disease data to be mapped, perform word segmentation on the disease data to be mapped, and obtain a word segmentation result.
  • the disease data to be mapped refers to the disease names that need to be mapped to standard disease data in a predetermined base table library (such as ICD-10).
  • the word segmentation result refers to the word sequence obtained by the word segmentation. For example, if the word segmentation is performed on “open cerebellar hemorrhage”, the word segmentation result can be “open / cerebellar / bleeding”.
  • the terminal sends a mapping request to the server and carries the disease data to be mapped in the mapping request.
  • the server parses the mapping request to obtain the disease data to be mapped.
  • the terminal uploads the disease data to be mapped to the server, and the server queries the disease data to be mapped from the database according to a preset period.
  • the disease data to be mapped is obtained by the server, the disease data is segmented to obtain a segmentation result.
  • step S204 the words in the segmentation result are matched with the root node of each matching tree in the matching forest, and the target root node is selected as the current node according to the root node matching result.
  • the matching tree is generated according to the standard disease data in a predetermined base table library. Each node on the matching tree is composed of words obtained by segmenting the standard disease data.
  • the matching tree is used to match words in the segmentation result corresponding to the disease data to be mapped to determine a matching path corresponding to the disease data to be mapped.
  • the root node matching result refers to a matching result obtained by matching a word in the segmentation result with a word corresponding to the root node.
  • the matching result of the root node includes a successful match with the root node and an unsuccessful match with the root node.
  • matching the words in the segmentation result with the root node of each matching tree in the matching tree set may be comparing each word in the segmentation result with a word corresponding to each root node. When the two words are the same, the match is successful; otherwise, the match is unsuccessful.
  • matching the words in the segmentation result with the root node of each matching tree in the matching tree set may be calculating the matching degree of each word in the segmentation result and the word corresponding to each root node. , When the maximum value of the matching degree is greater than a preset threshold, the word with the highest matching degree is a successful matching word; otherwise, the matching is unsuccessful.
  • the root node is the target root node, and the target root node is used as the current node.
  • step S206 it is determined whether a corresponding next-level child node exists in the current node.
  • step S208 if there is a match, the words in the segmentation result are matched with the next-level child node.
  • next-level node refers to a node that has a parent-child relationship with the current node.
  • the words in the segmentation result are continuously matched with the next-level child nodes corresponding to the current node.
  • matching the words in the segmentation result with the next-level child node corresponding to the current node may be comparing the words in the segmentation result with the words corresponding to each child node respectively. When they are the same, the match is successful; otherwise, the match is unsuccessful.
  • matching the words in the segmentation result with the next-level child node corresponding to the current node may be to calculate the matching degree between the word in the segmentation result and the word corresponding to each child node.
  • the maximum value is greater than a preset threshold, the word with the highest degree of matching is the word that has been successfully matched; otherwise, the match is unsuccessful.
  • Step S210 Select the target child node as the current node according to the child node matching result.
  • the child node matching result refers to a matching result obtained by matching a word in a segmentation result with a child node corresponding to a current node, and the child node matching result includes a successful match with a child node and an unsuccessful match with a child node.
  • the child node is the target child node, and the server uses the target child node as the current node.
  • step S206 the server repeatedly executes step S206.
  • step S212 if it does not exist, a matching path where the current node is located is determined, and target disease data is obtained according to the matching path.
  • the current node when there is no lower-level child node of the current node, the current node is the last node of the matching tree.
  • the matching path refers to a path composed of all nodes that match the words in the segmentation result corresponding to the disease data to be mapped from the first matching node (the root node) to the last matching node.
  • a matching path may be uniquely determined according to the current node. As shown in FIG. 3, if the root node of a disease data match is A, and the last current node (the last successful match node) is a leaf node H, the matching path determined according to H is A-C-F-H.
  • each matching path on the matching tree has a mapping relationship with a standard disease data in the base table library. After the matching path is determined, the corresponding standard disease data can be obtained according to the matching path.
  • the standard The disease data is the target disease data corresponding to the disease data to be mapped.
  • the diseased data to be mapped is segmented to obtain the segmented result, and then the words in the segmented result are matched with the matching tree in the matching tree set, and finally Determine the matching path, and then obtain the standard disease data corresponding to the matching path, and use the standard disease data as the target disease data corresponding to the disease data to be mapped.
  • the server when the server obtains the disease data to be mapped, it automatically matches with the matching tree. To obtain standardized target disease data, compared with the manual mapping in traditional technology, the efficiency and accuracy are significantly improved.
  • the method before obtaining the disease data to be mapped in step S202, the method further includes: a step of generating a matching forest.
  • This step specifically includes: segmenting each disease data in a predetermined base table library, and according to each disease data. The corresponding word segmentation result is used to obtain the word set, and the word type to which each word in the word set belongs; multiple words with the first preset type of word type are extracted from the word set as the root node of the matching tree; generated based on the root node Multiple matching trees to get matching forest.
  • the base table library refers to standard disease coding libraries, such as local coding libraries, ICD-10, and so on.
  • a base table library is determined in advance, and the disease data in the disease database is used as the standard data expected to be mapped.
  • Word segmentation is performed for each disease data in the base table database, and the words obtained after word segmentation for all disease data form a word set. Obtain the type of words to which each word belongs in the word set.
  • the word types include: disease, abnormal tissue, abnormal body, abnormal symptoms, anatomical part, disease type, disease nature, degree, and location.
  • the first preset type is at least one of a disease, an abnormal tissue, an abnormal body, and an abnormal symptom. Words of the first preset type are, for example, bleeding, fracture, deformity, and walking kidney.
  • all words in the word set with a preset type are extracted, and each word is used as a root node, and then a matching tree is established according to each root node to obtain a matching forest.
  • generating multiple matching trees according to a root node to obtain a matching forest includes:
  • step S402 a word having a word type of a second preset type is extracted from the word set to obtain a first subset.
  • the second preset type is one or more types other than the first preset type among types corresponding to all words in the word set.
  • the second preset type is an anatomical part, and words of the second preset type are, for example, cerebellum and ear.
  • step S404 co-occurrence words corresponding to the root node are extracted from the first sub-set to obtain a current leaf node set corresponding to the matching tree.
  • Co-occurrence refers to two words that appear together in the same disease data. For example, the three words corresponding to "open cerebellar hemorrhage" are open, cerebellum, and hemorrhage are co-occurring words. After co-occurrence words are extracted, the co-occurrence words are taken as the next-level child nodes of their corresponding root nodes. At this time, these child nodes are the current leaf nodes of the matching tree.
  • extracting the co-occurrence words corresponding to the root node from the first subset includes: obtaining codes of all disease data corresponding to each root node, obtaining a first code set corresponding to each root node; obtaining a first The coding of all disease data corresponding to each word in a sub-set, to obtain the second coding set corresponding to each word in the first sub-set; when the first coding set corresponding to the root node and the second coding set corresponding to the word have an intersection , Determine the word as a co-occurrence word corresponding to the root node.
  • Step S406 extracting a word of a third preset type from a word set to obtain a second sub-set.
  • the third preset type is one or more types other than the first preset type and the second preset type among types corresponding to all words in the word set.
  • the second preset type is at least one of a disease type, a nature and a degree of the disease, and words of a third preset type are, for example, open, comminuted, acute, and chronic.
  • Step S408 the co-occurrence words of the current leaf node set are extracted from the second sub-set to update the current leaf node set corresponding to the matching tree.
  • the co-occurrence word is used as the next-level node of the leaf node. At this time, It is equivalent to the matching tree continuing to grow at the leaf node, and the co-occurrence word of the leaf node is taken as the new current leaf node.
  • step S410 a word with a word type of a fourth preset type is extracted from the word set to obtain a third sub-set.
  • the fourth preset type is one or more types other than the first preset type, the second preset type, and the third preset type among types corresponding to all words in the word set.
  • the fourth preset type is orientation, and words of the fourth preset type are: left side, right side.
  • step S412 the co-occurrence word of the updated current leaf node set is extracted from the third sub-set, and the target leaf node set corresponding to the matching tree is obtained according to the updated co-occurrence word of the current leaf node set, and a matching tree is generated.
  • the target leaf node refers to the leaf node when the matching tree is finally generated.
  • the matching tree shown in FIG. 3 when the matching tree grows to include nodes A, B, and C, B and C are leaf nodes; when the matching tree grows to include nodes A, B, C, D, and E , B, D, E are leaf nodes; when the matching tree is finally generated, including nodes A, B, C, D, E, F, G, at this time, the leaf nodes are B, D, F, G. It can be seen that As the matching tree grows, the set of leaf nodes also changes.
  • the method further includes: determining disease data corresponding to a matching path where each target leaf node on the matching tree is located, and establishing a mapping relationship between the matching path and its corresponding disease data.
  • the matching path where each target leaf node is located can uniquely identify a disease data, and the words corresponding to each node on the matching path can be matched with each disease data in the base table library.
  • a disease data When the words corresponding to each node on the matching path are completely contained, the disease data is the disease data corresponding to the matching path.
  • a disease path is: dislocation-hip-half-right, and the corresponding disease data is: right hip subluxation.
  • mapping relationship is established between the matching path and its corresponding disease data, and all mapping relationships can be saved as a mapping table. After the matching path is determined, the corresponding disease data can be queried according to the mapping table.
  • the words in the segmentation result are matched with the root node of each matching tree in the matching forest, and the target root node is selected as the current node according to the root node matching result, including when the word in the segmentation result and any one
  • the synonyms corresponding to each root node are searched from a pre-established thesaurus; the words in the segmentation result are matched with the synonyms, and the target root node is selected as the current node according to the synonym matching result.
  • Synonym matching results refer to the matching results obtained by matching words in the segmentation result with synonyms.
  • a thesaurus is established in advance, and the thesaurus includes synonyms of words corresponding to the base table library.
  • the matching tree can be further expanded through synonyms to improve the matching accuracy, and finally improve the accuracy of the disease data mapping.
  • matching the words in the segmentation result with the root node of each matching tree in the matching forest, and selecting the target root node as the current node according to the root node matching result including when there are multiple matching root nodes
  • the mutual exclusion weight corresponding to each root node is obtained from a pre-established mutex lexicon, and the root node with the larger weight is used as the target root node.
  • the two words are mutually exclusive and mutually exclusive.
  • a mutually exclusive dictionary can be established in advance, and mutually exclusive weights can be set for each pair of mutually exclusive words.
  • the server can determine whether there are mutex words in multiple matching root nodes by searching from the mutex dictionary. When there are mutex words, get the mutex weight corresponding to each mutex word and increase the mutex weight.
  • the root node corresponding to the word is the target root node. For example, in a soft tissue injury half fracture, if the mutual exclusion of the fracture is greater than the injury, the root node corresponding to the fracture is taken as the target root node.
  • a disease with B disease and A disease causing B disease, etc.
  • multiple root nodes that have been successfully matched are used as target root nodes.
  • a root node continues to determine the matching path, and finally obtains target disease data corresponding to each target root node, and then combines the target disease data to obtain standard disease data corresponding to the disease data to be mapped.
  • the segmentation result obtained after segmentation of certain disease data to be mapped is: A / B / C / D / E, where C corresponds to a successfully matched root node, and E corresponds to a successfully matched root node, then A,
  • the matching tree corresponding to B and C is continuously matched to obtain the target disease data A1, and the matching tree corresponding to D and E is further matched to obtain the target disease data A2, and the final mapping result is A1A2.
  • the matching efficiency can be improved, and ultimately the mapping efficiency of the disease data is improved.
  • steps in FIG. 2 and FIG. 4 are sequentially displayed according to the directions of the arrows, these steps are not necessarily performed sequentially in the order indicated by the arrows. Unless explicitly stated in this document, the execution of these steps is not strictly limited, and these steps can be performed in other orders. Moreover, at least a part of the steps in FIG. 2 and FIG. 4 may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily performed at the same time, but may be performed at different times. These sub-steps or The execution order of the phases is not necessarily performed sequentially, but may be performed in turn or alternately with other steps or at least a part of the sub-steps or phases of other steps.
  • a disease data mapping device 500 which includes a word segmentation result acquisition module 502, a first matching module 504, a second matching module 506, a current node acquisition module 508, and a target disease.
  • Data acquisition module 510 where:
  • the segmentation result acquisition module 502 is configured to acquire disease data to be mapped, perform segmentation on the disease data to be mapped, and obtain a segmentation result;
  • a first matching module 504 configured to match the words in the segmentation result with the root node of each matching tree in the matching forest, and select the target root node as the current node according to the root node matching result;
  • a second matching module 506, configured to match a word in the segmentation result with the next-level child node when the current node has a corresponding next-level child node;
  • the current node acquisition module 508 is configured to select a target child node as the current node according to the child node matching result, and enter a matching word in the word segmentation result with the child node at the next level when the current node has a corresponding child node at the next level.
  • the target disease data acquisition module 510 is configured to determine a matching path where the current node is located when the current node does not have a corresponding next-level child node, and obtain target disease data according to the matching path.
  • the above device further includes: a matching forest generation module, configured to perform word segmentation on each disease data in a predetermined base table library, obtain a word set according to the word segmentation result corresponding to each disease data, and obtain a word set The type of each word that each word belongs to; multiple words with the first preset type of word type are extracted from the word set as the root nodes of the matching tree; multiple matching trees are generated according to the root node to obtain the matching forest.
  • a matching forest generation module configured to perform word segmentation on each disease data in a predetermined base table library, obtain a word set according to the word segmentation result corresponding to each disease data, and obtain a word set The type of each word that each word belongs to; multiple words with the first preset type of word type are extracted from the word set as the root nodes of the matching tree; multiple matching trees are generated according to the root node to obtain the matching forest.
  • the matching forest generation module is further configured to extract a word of a second preset type from a word set to obtain a first sub-set; and extract a co-occurrence word corresponding to a root node from the first sub-set, Get the current leaf node set corresponding to the matching tree; extract words with a third preset type from the word set to get the second sub-set; extract co-occurrence words of the current leaf node set from the second sub-set to update The current leaf node set corresponding to the matching tree; extracting words of the fourth preset type from the word set to obtain a third subset; and extracting the updated co-occurring words of the current leaf node set from the third subset, According to the updated co-occurrence words of the current leaf node set, the target leaf node set corresponding to the matching tree is obtained, and a matching tree is generated.
  • the above apparatus further includes a mapping relationship establishing module, configured to determine disease data corresponding to a matching path where each target leaf node on the matching tree is located, and establish a mapping relationship between the matching path and its corresponding disease data.
  • a mapping relationship establishing module configured to determine disease data corresponding to a matching path where each target leaf node on the matching tree is located, and establish a mapping relationship between the matching path and its corresponding disease data.
  • the matching forest generation module is further configured to obtain codes of all disease data corresponding to each root node, to obtain a first code set corresponding to each root node; to obtain all the codes corresponding to each word in the first subset. Coding the disease data to obtain the second coding set corresponding to each word in the first subset; when the first coding set corresponding to the root node and the second coding set corresponding to the word intersect, it is determined that the word is a common corresponding to the root node. Present word.
  • the first matching module 504 is configured to search a synonym corresponding to each root node from a pre-established thesaurus when a word in the word segmentation result does not match any root node; Words and synonyms are matched, and the target root node is selected as the current node according to the result of the synonyms match.
  • the first matching module 504 is configured to obtain each root from a pre-established mutex dictionary when there are multiple matching root nodes and there are mutually exclusive relationships between the multiple matching root nodes.
  • the mutual exclusion weight corresponding to the node, the root node with the larger weight is taken as the target root node; when there are multiple matching root nodes and there is no mutual exclusion relationship among multiple matching root nodes, multiple matching is successful As the target root node.
  • Each module in the above-mentioned disease data mapping device may be implemented in whole or in part by software, hardware, and a combination thereof.
  • the above-mentioned modules may be embedded in the hardware form or independent of the processor in the computer device, or may be stored in the memory of the computer device in the form of software, so that the processor calls and performs the operations corresponding to the above modules.
  • a computer device is provided.
  • the computer device may be a server, and its internal structure diagram may be as shown in FIG. 6.
  • the computer device includes a processor, a memory, a network interface, and a database connected through a system bus.
  • the processor of the computer device is used to provide computing and control capabilities.
  • the memory of the computer device includes a non-volatile computer-readable storage medium and an internal memory.
  • the non-volatile computer-readable storage medium stores an operating system, computer-readable instructions, and a database.
  • the internal memory provides an environment for operating systems and computer-readable instructions in a non-volatile computer-readable storage medium.
  • the computer equipment database is used to store various disease data.
  • the network interface of the computer device is used to communicate with an external terminal through a network connection.
  • the computer-readable instructions are executed by a processor to implement a disease data mapping method.
  • FIG. 6 is only a block diagram of a part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer equipment to which the scheme of the present application is applied.
  • the specific computer equipment may be Include more or fewer parts than shown in the figure, or combine certain parts, or have a different arrangement of parts.
  • a computer device includes a memory and one or more processors.
  • the memory stores computer-readable instructions.
  • the steps of the disease data mapping method provided in any embodiment of the present application are implemented.
  • One or more non-volatile computer-readable storage media storing computer-readable instructions, and when the computer-readable instructions are executed by one or more processors, the one or more processors implement one of the embodiments of the present application Steps of the provided disease data mapping method.
  • Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory.
  • Volatile memory can include random access memory (RAM) or external cache memory.
  • RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous chain Synchlink DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
  • SRAM static RAM
  • DRAM dynamic RAM
  • SDRAM synchronous DRAM
  • DDRSDRAM dual data rate SDRAM
  • ESDRAM enhanced SDRAM
  • SLDRAM synchronous chain Synchlink DRAM
  • Rambus direct RAM
  • DRAM direct memory bus dynamic RAM
  • RDRAM memory bus dynamic RAM

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Epidemiology (AREA)
  • Computational Linguistics (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Medical Informatics (AREA)
  • Primary Health Care (AREA)
  • Public Health (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种疾病数据映射方法，包括：获取待映射的疾病数据，对待映射的疾病数据进行分词，得到分词结果；将分词结果中的词语与匹配森林中每一个匹配树的根节点进行匹配，根据根节点匹配结果选取目标根节点作为当前节点；当当前节点存在对应的下一级子节点时，将分词结果中的词语与下一级子节点进行匹配；根据子节点匹配结果选取目标子节点作为当前节点，并进入当当前节点存在对应的下一级子节点时，将分词结果中的词语与下一级子节点进行匹配的步骤；当当前节点不存在对应的下一级子节点时，确定当前节点所在的匹配路径，根据匹配路径获取目标疾病数据。

Description

疾病数据映射方法、装置、计算机设备和存储介质
相关申请的交叉引用
本申请要求于2018年09月10日提交中国专利局，申请号为2018110519111，申请名称为“疾病数据映射方法、装置、计算机设备和存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及一种疾病数据映射方法、装置、计算机设备和存储介质。
背景技术
由于医生习惯或区域性特点导致医用术语不规范，疾病诊断名称差异性大，海量医疗数据信息无法互联互通，形成数据孤岛，无法进行有价值的医疗大数据分析研究。为了解决这一问题，需要将医生的疾病诊断数据映射为标准化的疾病名称。
传统技术中，医院中通常由专门的人负责对这些疾病名称数据进行人工映射，然而，发明人意识到，这种方式不仅效率低下，而且准确性并不高。
发明内容
根据本申请公开的各种实施例，提供一种疾病数据映射方法、装置、计算机设备和存储介质。
一种疾病数据映射方法包括：
获取待映射的疾病数据，对所述待映射的疾病数据进行分词，得到分词结果；
将所述分词结果中的词语与匹配森林中每一个匹配树的根节点进行匹配，根据根节点匹配结果选取目标根节点作为当前节点；
当所述当前节点存在对应的下一级子节点时，将所述分词结果中的词语与所述下一级子节点进行匹配；
根据子节点匹配结果选取目标子节点作为当前节点，并进入当所述当前节点存在对应的下一级子节点时，将所述分词结果中的词语与所述下一级子节点进行匹配的步骤；及
当所述当前节点不存在对应的下一级子节点时，确定所述当前节点所在的匹配路径，根据所述匹配路径获取目标疾病数据。
一种疾病数据映射装置包括：
分词结果获取模块，用于获取待映射的疾病数据，对所述待映射的疾病数据进行分词，得到分词结果；
第一匹配模块，用于将所述分词结果中的词语与匹配森林中每一个匹配树的根节点进行匹配，根据根节点匹配结果选取目标根节点作为当前节点；
第二匹配模块，用于当所述当前节点存在对应的下一级子节点时，将所述分词结果中的词语与所述下一级子节点进行匹配；
当前节点获取模块，用于根据子节点匹配结果选取目标子节点作为当前节点，并进入当所述当前节点存在对应的下一级子节点时，将所述分词结果中的词语与所述下一级子节点进行匹配的步骤；及
目标疾病数据获取模块，用于当所述当前节点不存在对应的下一级子节点时，确定所述当前节点所在的匹配路径，根据所述匹配路径获取目标疾病数据。
一种计算机设备，包括存储器和一个或多个处理器，存储器中存储有计算机可读指令，计算机可读指令被处理器执行时实现本申请任意一个实施例中提供的疾病数据映射方法的步骤。
一个或多个存储有计算机可读指令的非易失性计算机可读存储介质，计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器实现本申请任意一个实施例中提供的疾病数据映射方法的步骤。
本申请的一个或多个实施例的细节在下面的附图和描述中提出。本申请的其它特征和优点将从说明书、附图以及权利要求书变得明显。
附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其它的附图。
图1为根据一个或多个实施例中疾病数据映射方法的应用场景图。
图2为根据一个或多个实施例中疾病数据映射方法的流程示意图。
图3为根据一个或多个实施例中匹配树的示意图。
图4为另一个实施例中疾病数据映射方法的流程示意图。
图5为根据一个或多个实施例中疾病数据映射装置的框图。
图6为根据一个或多个实施例中计算机设备的框图。
具体实施方式
为了使本申请的技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
本申请提供的疾病数据映射方法，可以应用于如图1所示的应用环境中。终端102通过网络与服务器104行通信。服务器104从终端获取待映射的疾病数据，对疾病数据进行分词，得到分词结果；将分词结果中的词语与匹配树集合中每一个匹配树的根节点进行匹配，根据根节点匹配结果选取目标根节点作为当前节点；将分词结果中的词语与当前节点对应的子节点进行匹配；根据子节点匹配结果选取目标子节点作为当前节点，并进入将分词结果中的词语与当前节点对应的子节点进行匹配的步骤；当当前节点不存在对应的子节点时，确定当前节点所在的匹配路径，根据匹配路径获取目标疾病数据，最后，服务器104可将目标疾病数据返回至终端102。
终端102可以但不限于是各种个人计算机、笔记本电脑、智能手机、平板电脑，服务器104可以用独立的服务器或者是多个服务器组成的服务器集群来实现。
在一些实施例中，如图2所示，提供了一种疾病数据映射方法，以该方法应用于图1 中的服务器为例进行说明，包括以下步骤：
步骤S202，获取待映射的疾病数据，对待映射的疾病数据进行分词，得到分词结果。
待映射的疾病数据指的是需要被映射至预先确定的基表库(如ICD-10)中的标准疾病数据的疾病名称。分词结果指的是分词得到的词序列，如对“开放性小脑出血”进行分词，得到的分词结果可以为“开放性/小脑/出血”
在一些实施例中，终端向服务器发送映射请求，并在映射请求中携带待映射的疾病数据，服务器器接收到映射请求后，解析该映射请求，获取到待映射的疾病数据。在另一些实施例中，终端将待映射的疾病数据上传至服务器，服务器按照预设的周期从数据库中查询待映射的疾病数据。
进一步，服务器获取到待映射的疾病数据后，对疾病数据进行分词，得到分词结果。
步骤S204，将分词结果中的词语与匹配森林中每一个匹配树的根节点进行匹配，根据根节点匹配结果选取目标根节点作为当前节点。
匹配树根据预先确定的基表库中的标准疾病数据生成，匹配树上的每一个节点由标准疾病数据分词后得到的词语组成。匹配树用于和待映射的疾病数据对应的分词结果中的词语进行匹配，以确定待映射的疾病数据对应的匹配路径。根节点匹配结果指的是将分词结果中的词语与根节点对应的词语进行匹配得到的匹配结果，根节点匹配结果包括与根节点匹配成功以及与根节点匹配不成功。
在一些实施例中，将分词结果中的词语与匹配树集合中每一个匹配树的根节点进行匹配可以是，将分词结果中的每一个词语分别与每一个根节点对应的词语进行比对，当两个词相同时，则为匹配成功，否则，则匹配不成功。
在另一些实施例中，将分词结果中的词语与匹配树集合中每一个匹配树的根节点进行匹配可以是，分别计算分词结果中的每一个词语与每一个根节点对应的词语的匹配度，当匹配度最大值大于预设阈值时，则匹配度最大的词语为匹配成功的词语，否则，则匹配不成功。
进一步，当匹配成功的根节点只有一个时，该根节点即为目标根节点，将该目标根节点作为当前节点。
步骤S206，判断当前节点是否存在对应的下一级子节点。
步骤S208，若存在，则将分词结果中的词语与下一级子节点进行匹配。
具体地，下一级节点指的是与当前节点存在父子关系的节点。在本实施例中，当获取到当前节点后，将分词结果中的词语继续与当前节点对应的下一级子节点进行匹配。
在一些实施例中，将分词结果中的词语与当前节点对应的下一级子节点进行匹配可以是，将分词结果中的词语分别与每一个子节点对应的词语进行比对，当两个词相同时，则为匹配成功，否则，则匹配不成功。
在另一些实施例中，将分词结果中的词语与当前节点对应的下一级子节点进行匹配可以是，分别计算分词结果中的词语与每一个子节点对应的词语的匹配度，当匹配度最大值大于预设阈值时，则匹配度最大的词语为匹配成功的词语，否则，则匹配不成功。
可以理解，在匹配时，可以只将分词结果中未曾进行匹配过的词与下一级子节点进行匹配，从而提高匹配效率。
步骤S210，根据子节点匹配结果选取目标子节点作为当前节点。
具体地，子节点匹配结果指的是将分词结果中的词语与当前节点对应的下一级子节点进行匹配得到的匹配结果，子节点匹配结果包括与子节点匹配成功以及与子节点匹配不成功。当匹配成功的子节点只有一个时，该子节点即为目标子节点，服务器将该目标子节点作为当前节点。
进一步，服务器重复执行步骤S206。
步骤S212，若不存在，则确定当前节点所在的匹配路径，根据匹配路径获取目标疾病数据。
具体地，当当前节点不存在下一级子节点时，当前节点即为匹配树的末级节点。匹配路径指的是从第一个匹配成功的节点(根节点)到最后一个匹配成功的节点，中间的所有与待映射的疾病数据对应的分词结果中的词语相匹配的节点所组成的路径。在一些实施例中，当当前节点不存在下一级子节点即当前节点为叶子节点时，根据当前节点可唯一确定一条匹配路径。如图3所示，若某个疾病数据匹配成功的根节点为A，最后一个当前节点(最后一个匹配成功的节点)为叶子节点H，则根据H确定的匹配路径为A-C-F-H。
在本实施例中，匹配树上的每一条匹配路径都与基表库中的一个标准疾病数据存在映射关系，当匹配路径确定好后，可根据匹配路径获取到对应的标准疾病数据，该标准疾 病数据即为待映射的疾病数据对应的目标疾病数据。
上述疾病数据映射方法中，服务器在获取到待映射的疾病数据后，对待映射的疾病数据进行分词，得到分词结果，然后通过将分词结果中词语与匹配树集合中的匹配树进行匹配，并最终确定匹配路径，然后获取匹配路径对应的标准疾病数据，将该标准疾病数据作为待映射的疾病数据对应的目标疾病数据，本申请中，服务器在获取到待映射的疾病数据自动与匹配树进行匹配来得到标准化的目标疾病数据，相较于传统技术中的人工映射，效率和准确率都得到显著提升。
在一些实施例中，步骤S202获取待映射的疾病数据之前，还包括：生成匹配森林的步骤，该步骤具体包括：对预先确定的基表库中每一个疾病数据进行分词，根据每一个疾病数据对应的分词结果得到词语集合，并获取词语集合中每一个词语所属的词语类型；从词语集合中提取词语类型为第一预设类型的多个词语分别作为匹配树的根节点；根据根节点生成多个匹配树，得到匹配森林。
基表库指的是标准疾病编码库，例如地方编码库、ICD-10等等。在本实施例中，预先确定一个基表库，将疾病库中的疾病数据作为期望映射得打的标准数据。对基表库中的每一个疾病数据进行分词，所有疾病数据进行分词后得到的词语组成一个词语集合。获取词语集合中，每一个词语所属的词语类型，词语类型包括：疾病、异常组织、异常机体、异常症状、解剖部位、疾病类型、疾病性质、程度、方位。
第一预设类型为疾病、异常组织、异常机体、异常症状中的至少一种，第一预设类型的词语例如：出血、骨折、畸形、游走肾。
在本实施例中，将词语集合中所有词语类型为预设类型的词语都提取出来，每一个词语作为一个根节点，然后根据每一个根节点分别建立一棵匹配树，得到匹配森林。
在一些实施例中，如图4所示，根据根节点生成多个匹配树，得到匹配森林，包括：
步骤S402，从词语集合中提取词语类型为第二预设类型的词语，得到第一子集合。
具体地，第二预设类型为词语集合中所有词语对应的类型中除第一预设类型之外的一种或多种类型。在一些实施例中，第二预设类型为解剖部位，第二预设类型的词语例如：小脑、耳。
步骤S404，从第一子集合中提取根节点对应的共现词，得到匹配树对应的当前叶子 节点集合。
共现词指的是共同出现在同一个疾病数据中的两个词。如“开放性小脑出血”对应的三个词开放性、小脑、出血互为共现词。提取到共现词后，将共现词作为其对应的根节点的下一级子节点，此时，这些子节点为匹配树的当前叶子节点。
在一些实施例中，从第一子集合中提取根节点对应的共现词，包括：获取每一个根节点对应的所有疾病数据的编码，得到每一个根节点对应的第一编码集合；获取第一子集合中每一个词语对应的所有疾病数据的编码，得到第一子集合中每一个词语对应的第二编码集合；当根节点对应的第一编码集合与词语对应的第二编码集合存在交集时，确定词语为根节点对应的共现词。
步骤S406，从词语集合中提取词语类型为第三预设类型的词语，得到第二子集合。
第三预设类型为词语集合中所有词语对应的类型中除第一预设类型、第二预设类型之外的一种或多种类型。在一些实施例中，第二预设类型为疾病类型、疾病性质、程度中的至少一种，第三预设类型的词语例如，开放性、粉碎性、急性、慢性。
步骤S408，从第二子集合中提取当前叶子节点集合的共现词，以更新匹配树对应的当前叶子节点集合。
具体地，对于匹配树上的每一个当前叶子节点，判断其在第二子集合中是否存在共现词，若存在，则将该共现词作为该叶子节点的下一级节点，此时，相当于匹配树在该叶子节点处继续生长，该叶子节点的共现词作为新的当前叶子节点。
可以理解，对于不存在共现词的当前叶子节点，仍然作为当前叶子节点。
步骤S410，从词语集合中提取词语类型为第四预设类型的词语，得到第三子集合。
第四预设类型为词语集合中所有词语对应的类型中除第一预设类型、第二预设类型、第三预设类型之外的一种或多种类型。在一些实施例中，第四预设类型为方位，第四预设类型的词语如：左侧、右侧。
步骤S412，从第三子集合中提取更新后的当前叶子节点集合的共现词，根据更新后的当前叶子节点集合的共现词得到匹配树对应的目标叶子节点集合，生成匹配树。，目标叶子节点指的是匹配树最终生成时候的叶子节点。
举例说明，如图3所示的匹配树中，当匹配树生长到包括节点A、B、C时，B和C 为叶子节点；当匹配树生长到包括节点A、B、C、D、E，B、D、E为叶子节点；当匹配树最终生成时，包括节点A、B、C、D、E、F、G，此时，叶子节点为B、D、F、G，可见，随着匹配树的不断生长，叶子节点集合也跟着不断变化。
在一些实施例中，生成匹配树之后，还包括：确定匹配树上每一个目标叶子节点所在的匹配路径对应的疾病数据，将匹配路径与其对应的疾病数据建立映射关系。
具体地，每一个目标叶子节点所在的匹配路径可以唯一确定一个疾病数据，可将匹配路径上各个节点对应的词语与基表库中的各个疾病数据进行匹配，匹配成功时，即某个疾病数据完全包含匹配路径上各个节点对应的词语时，该疾病数据即为匹配路径对应的疾病数据。例如，某个疾病路径为：脱位-髋-半-右侧，其对应的疾病数据为：右侧髋关节半脱位。
进一步，将匹配路径与其对应的疾病数据建立映射关系，可将所有的映射关系保存为一个映射表。当匹配路径确定后，可根据该映射表来查询其对应的疾病数据。
在一些实施例中，将分词结果中的词语与匹配森林中每一个匹配树的根节点进行匹配，根据根节点匹配结果选取目标根节点作为当前节点，包括：当分词结果中的词语与任意一个根节点都不匹配时，从预先建立的同义词库中查找每一个根节点对应的同义词；将分词结果中的词语与同义词进行匹配，根据同义词匹配结果选取目标根节点作为当前节点。
同义词匹配结果指的是将分词结果中的词语与同义词进行匹配得到的匹配结果。在本实施例中，预先建立起一个同义词库，同义词库中包括基表库对应的词语的同义词。
上述实施例中，通过同义词可以对匹配树进行进一步扩充，提高匹配准确度，进而最终提高疾病数据映射的准确性。
在一些实施例中，将分词结果中的词语与匹配森林中每一个匹配树的根节点进行匹配，根据根节点匹配结果选取目标根节点作为当前节点，包括：当存在多个匹配成功的根节点时且多个匹配成功的根节点存在互斥关系时，从预先建立的互斥词库中获取每一个根节点对应的互斥权重，将权重较大的根节点作为目标根节点；当存在多个匹配成功的根节点时且多个匹配成功的根节点不存在互斥关系时，判断分词结果中是否存在预设的连接词，若是，则将多个匹配成功的根节点作为目标根节点。
具体地，当两个词同时出现时，其中一个词的语义可以忽略时，这两个词存在互斥关系，互为互斥词。如，软组织损伤半骨折中，损伤与骨折为互斥词。可预先建立一个互斥词典，并对每一对互斥词分别设定互斥权重。服务器可通过从互斥词典中进行查找，判断多个匹配成功的根节点中是否存在互斥词，当存在互斥词，获取每一个互斥词对应的互斥权重，将互斥权重较大的词语对应的根节点作为目标根节点。如，软组织损伤半骨折中，若骨折的互斥权重大于损伤，则将骨折对应的根节点作为目标根节点。
进一步，对于某些疾病数据中，有可能包含两种疾病，如A疾病伴B疾病、A疾病导致B疾病等，此时，将多个匹配成功的根节点都作为目标根节点，分别根据每一个根节点继续确定匹配路径，并最终获得每一个目标根节点对应的目标疾病数据，然后将各个目标疾病数据组合得到待映射的疾病数据对应的标准疾病数据。举例说明，某个待映射的疾病数据分词后得到的分词结果为：A/B/C/D/E，C对应一个匹配成功的根节点，E对应一个匹配成功的根节点，则将A、B与C对应的匹配树进行继续匹配以得到目标疾病数据A1，将D与E对应的匹配树进行继续匹配以得到目标疾病数据A2，最后得到的映射结果为A1A2。
上述实施例中，通过判断出互斥词，可以提高匹配效率，从而最终提高疾病数据的映射效率。
应该理解的是，虽然图2及图4中的各个步骤按照箭头的指示依次显示，但是这些步骤并不是必然按照箭头指示的顺序依次执行。除非本文中有明确的说明，这些步骤的执行并没有严格的顺序限制，这些步骤可以以其它的顺序执行。而且，图2及图4中的至少一部分步骤可以包括多个子步骤或者多个阶段，这些子步骤或者阶段并不必然是在同一时刻执行完成，而是可以在不同的时刻执行，这些子步骤或者阶段的执行顺序也不必然是依次进行，而是可以与其它步骤或者其它步骤的子步骤或者阶段的至少一部分轮流或者交替地执行。
在一些实施例中，如图5所示，提供了一种疾病数据映射装置500，包括：分词结果获取模块502、第一匹配模块504、第二匹配模块506、当前节点获取模块508和目标疾病数据获取模块510，其中：
分词结果获取模块502，用于获取待映射的疾病数据，对待映射的疾病数据进行分词， 得到分词结果；
第一匹配模块504，用于将分词结果中的词语与匹配森林中每一个匹配树的根节点进行匹配，根据根节点匹配结果选取目标根节点作为当前节点；
第二匹配模块506，用于当当前节点存在对应的下一级子节点时，将分词结果中的词语与下一级子节点进行匹配；
当前节点获取模块508，用于根据子节点匹配结果选取目标子节点作为当前节点，并进入当当前节点存在对应的下一级子节点时，将分词结果中的词语与下一级子节点进行匹配的步骤；
目标疾病数据获取模块510，用于当当前节点不存在对应的下一级子节点时，确定当前节点所在的匹配路径，根据匹配路径获取目标疾病数据。
在一些实施例中，上述装置还包括：匹配森林生成模块，用于对预先确定的基表库中每一个疾病数据进行分词，根据每一个疾病数据对应的分词结果得到词语集合，并获取词语集合中每一个词语所属的词语类型；从词语集合中提取词语类型为第一预设类型的多个词语分别作为匹配树的根节点；根据根节点生成多个匹配树，得到匹配森林。
在一些实施例中，匹配森林生成模块还用于从词语集合中提取词语类型为第二预设类型的词语，得到第一子集合；从第一子集合中提取根节点对应的共现词，得到匹配树对应的当前叶子节点集合；从词语集合中提取词语类型为第三预设类型的词语，得到第二子集合；从第二子集合中提取当前叶子节点集合的共现词，以更新匹配树对应的当前叶子节点集合；从词语集合中提取词语类型为第四预设类型的词语，得到第三子集合；从第三子集合中提取更新后的当前叶子节点集合的共现词，根据更新后的当前叶子节点集合的共现词得到匹配树对应的目标叶子节点集合，生成匹配树。
在一些实施例中，上述装置还包括映射关系建立模块，用于确定匹配树上每一个目标叶子节点所在的匹配路径对应的疾病数据，将匹配路径与其对应的疾病数据建立映射关系。
在一些实施例中，匹配森林生成模块还用于获取每一个根节点对应的所有疾病数据的编码，得到每一个根节点对应的第一编码集合；获取第一子集合中每一个词语对应的所有疾病数据的编码，得到第一子集合中每一个词语对应的第二编码集合；当根节点对应的第 一编码集合与词语对应的第二编码集合存在交集时，确定词语为根节点对应的共现词。
在一些实施例中，第一匹配模块504用于当分词结果中的词语与任意一个根节点都不匹配时，从预先建立的同义词库中查找每一个根节点对应的同义词；将分词结果中的词语与同义词进行匹配，根据同义词匹配结果选取目标根节点作为当前节点。
在一些实施例中，第一匹配模块504用于当存在多个匹配成功的根节点时且多个匹配成功的根节点存在互斥关系时，从预先建立的互斥词库中获取每一个根节点对应的互斥权重，将权重较大的根节点作为目标根节点；当存在多个匹配成功的根节点时且多个匹配成功的根节点不存在互斥关系时，则将多个匹配成功的根节点作为目标根节点。
关于疾病数据映射装置的具体限定可以参见上文中对于疾病数据映射方法的限定，在此不再赘述。上述疾病数据映射装置中的各个模块可全部或部分通过软件、硬件及其组合来实现。上述各模块可以硬件形式内嵌于或独立于计算机设备中的处理器中，也可以以软件形式存储于计算机设备中的存储器中，以便于处理器调用执行以上各个模块对应的操作。
在一些实施例中，提供了一种计算机设备，该计算机设备可以是服务器，其内部结构图可以如图6所示。该计算机设备包括通过系统总线连接的处理器、存储器、网络接口和数据库。该计算机设备的处理器用于提供计算和控制能力。该计算机设备的存储器包括非易失性计算机可读存储介质、内存储器。该非易失性计算机可读存储介质存储有操作系统、计算机可读指令和数据库。该内存储器为非易失性计算机可读存储介质中的操作系统和计算机可读指令的运行提供环境。该计算机设备的数据库用于存储各种疾病数据。该计算机设备的网络接口用于与外部的终端通过网络连接通信。该计算机可读指令被处理器执行时以实现一种疾病数据映射方法。
本领域技术人员可以理解，图6中示出的结构，仅仅是与本申请方案相关的部分结构的框图，并不构成对本申请方案所应用于其上的计算机设备的限定，具体的计算机设备可以包括比图中所示更多或更少的部件，或者组合某些部件，或者具有不同的部件布置。
一种计算机设备，包括存储器和一个或多个处理器，存储器中存储有计算机可读指令，计算机可读指令被处理器执行时实现本申请任意一个实施例中提供的疾病数据映射方法 的步骤。
一个或多个存储有计算机可读指令的非易失性计算机可读存储介质，计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器实现本申请任意一个实施例中提供的疾病数据映射方法的步骤。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机可读指令来指令相关的硬件来完成，所述的计算机可读指令可存储于一非易失性计算机可读取存储介质中，该计算机可读指令在执行时，可包括如上述各方法的实施例的流程。本申请所提供的各实施例中所使用的对存储器、存储、数据库或其它介质的任何引用，均可包括非易失性和/或易失性存储器。非易失性存储器可包括只读存储器(ROM)、可编程ROM(PROM)、电可编程ROM(EPROM)、电可擦除可编程ROM(EEPROM)或闪存。易失性存储器可包括随机存取存储器(RAM)或者外部高速缓冲存储器。作为说明而非局限，RAM以多种形式可得，诸如静态RAM(SRAM)、动态RAM(DRAM)、同步DRAM(SDRAM)、双数据率SDRAM(DDRSDRAM)、增强型SDRAM(ESDRAM)、同步链路(Synchlink)DRAM(SLDRAM)、存储器总线(Rambus)直接RAM(RDRAM)、直接存储器总线动态RAM(DRDRAM)、以及存储器总线动态RAM(RDRAM)等。
以上实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本申请的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本申请构思的前提下，还可以做出若干变形和改进，这些都属于本申请的保护范围。因此，本申请专利的保护范围应以所附权利要求为准.。。

Claims (20)

  1. 一种疾病数据映射方法，包括：
    获取待映射的疾病数据，对所述待映射的疾病数据进行分词，得到分词结果；
    将所述分词结果中的词语与匹配森林中每一个匹配树的根节点进行匹配，根据根节点匹配结果选取目标根节点作为当前节点；
    当所述当前节点存在对应的下一级子节点时，将所述分词结果中的词语与所述下一级子节点进行匹配；
    根据子节点匹配结果选取目标子节点作为当前节点，并进入当所述当前节点存在对应的下一级子节点时，将所述分词结果中的词语与所述下一级子节点进行匹配的步骤；及
    当所述当前节点不存在对应的下一级子节点时，确定所述当前节点所在的匹配路径，根据所述匹配路径获取目标疾病数据。
  2. 根据权利要求1所述的方法，其特征在于，在所述将所述分词结果中的词语与匹配森林中每一个匹配树的根节点进行匹配之前，所述方法还包括：
    对预先确定的基表库中每一个疾病数据进行分词，根据每一个所述疾病数据对应的分词结果得到词语集合，并获取所述词语集合中每一个词语所属的词语类型；
    从所述词语集合中提取词语类型为第一预设类型的多个词语分别作为匹配树的根节点；及
    根据所述根节点生成多个匹配树，得到匹配森林。
  3. 根据权利要求2所述的方法，其特征在于，所述根据所述根节点生成多个匹配树，得到匹配森林，包括：
    从所述词语集合中提取词语类型为第二预设类型的词语，得到第一子集合；
    从所述第一子集合中提取所述根节点对应的共现词，得到所述匹配树对应的当前叶子节点集合；
    从所述词语集合中提取词语类型为第三预设类型的词语，得到第二子集合；
    从所述第二子集合中提取当前叶子节点集合的共现词，以更新所述匹配树对应的当前叶子节点集合；
    从所述词语集合中提取词语类型为第四预设类型的词语，得到第三子集合；及
    从所述第三子集合中提取更新后的当前叶子节点集合的共现词，根据更新后的当前叶子节点集合的共现词得到所述匹配树对应的目标叶子节点集合，生成匹配树。
  4. 根据权利要求3所述的方法，其特征在于，在所述根据更新后的当前叶子节点集合的共现词得到所述匹配树对应的目标叶子节点集合，生成匹配树之后，所述方法还包括：
    确定所述匹配树上每一个所述目标叶子节点所在的匹配路径对应的疾病数据，及将所述匹配路径与其对应的疾病数据建立映射关系。
  5. 根据权利要求3所述的方法，其特征在于，所述从所述第一子集合中提取所述根节点对应的共现词，包括：
    获取每一个所述根节点对应的所有疾病数据的编码，得到每一个所述根节点对应的第一编码集合；
    获取所述第一子集合中每一个词语对应的所有疾病数据的编码，得到所述第一子集合中每一个词语对应的第二编码集合；及
    当所述根节点对应的第一编码集合与所述词语对应的第二编码集合存在交集时，确定所述词语为所述根节点对应的共现词。
  6. 根据权利要求1至5任意一项所述的方法，其特征在于，将所述分词结果中的词语与匹配森林中每一个匹配树的根节点进行匹配，根据根节点匹配结果选取目标根节点作为当前节点，包括：
    当所述分词结果中的词语与任意一个所述根节点都不匹配时，从预先建立的同义词库中查找每一个所述根节点对应的同义词；及
    将所述分词结果中的词语与所述同义词进行匹配，根据同义词匹配结果选取目标根节点作为当前节点。
  7. 根据权利要求1至5任意一项所述的方法，其特征在于，将所述分词结果中的词语与匹配森林中每一个匹配树的根节点进行匹配，根据根节点匹配结果选取目标根节点作为当前节点，包括：
    当存在多个匹配成功的根节点时且所述多个匹配成功的根节点存在互斥关系时，从预先建立的互斥词库中获取每一个根节点对应的互斥权重，将权重较大的根节点作为目标根节点；及
    当存在多个匹配成功的根节点时且所述多个匹配成功的根节点不存在互斥关系时，则将所述多个匹配成功的根节点作为目标根节点。
  8. 一种疾病数据映射装置，包括：
    分词结果获取模块，用于获取待映射的疾病数据，对所述待映射的疾病数据进行分词，得到分词结果；
    第一匹配模块，用于将所述分词结果中的词语与匹配森林中每一个匹配树的根节点进行匹配，根据根节点匹配结果选取目标根节点作为当前节点；
    第二匹配模块，用于当所述当前节点存在对应的下一级子节点时，将所述分词结果中的词语与所述下一级子节点进行匹配；
    当前节点获取模块，用于根据子节点匹配结果选取目标子节点作为当前节点，并进入当所述当前节点存在对应的下一级子节点时，将所述分词结果中的词语与所述下一级子节点进行匹配的步骤；及
    目标疾病数据获取模块，用于当所述当前节点不存在对应的下一级子节点时，确定所述当前节点所在的匹配路径，根据所述匹配路径获取目标疾病数据。
  9. 根据权利要求8所述的装置，其特征在于，还包括：匹配森林生成模块，用于对预先确定的基表库中每一个疾病数据进行分词，根据每一个疾病数据对应的分词结果得到词语集合，并获取词语集合中每一个词语所属的词语类型；从词语集合中提取词语类型为第一预设类型的多个词语分别作为匹配树的根节点；根据根节点生成多个匹配树，得到匹配森林。
  10. 一种计算机设备，包括存储器及一个或多个处理器，所述存储器中储存有计算机可读指令，所述计算机可读指令被所述一个或多个处理器执行时，使得所述一个或多个处理器执行以下步骤：
    获取待映射的疾病数据，对所述待映射的疾病数据进行分词，得到分词结果；
    将所述分词结果中的词语与匹配森林中每一个匹配树的根节点进行匹配，根据根节点匹配结果选取目标根节点作为当前节点；
    当所述当前节点存在对应的下一级子节点时，将所述分词结果中的词语与所述下一级 子节点进行匹配；
    根据子节点匹配结果选取目标子节点作为当前节点，并进入当所述当前节点存在对应的下一级子节点时，将所述分词结果中的词语与所述下一级子节点进行匹配的步骤；及
    当所述当前节点不存在对应的下一级子节点时，确定所述当前节点所在的匹配路径，根据所述匹配路径获取目标疾病数据。
  11. 根据权利要求10所述的计算机设备，其特征在于，所述处理器执行所述计算机可读指令时还执行以下步骤：
    对预先确定的基表库中每一个疾病数据进行分词，根据每一个所述疾病数据对应的分词结果得到词语集合，并获取所述词语集合中每一个词语所属的词语类型；
    从所述词语集合中提取词语类型为第一预设类型的多个词语分别作为匹配树的根节点；及
    根据所述根节点生成多个匹配树，得到匹配森林。
  12. 根据权利要求11所述的计算机设备，其特征在于，所述处理器执行所述计算机可读指令时还执行以下步骤：
    从所述词语集合中提取词语类型为第二预设类型的词语，得到第一子集合；
    从所述第一子集合中提取所述根节点对应的共现词，得到所述匹配树对应的当前叶子节点集合；
    从所述词语集合中提取词语类型为第三预设类型的词语，得到第二子集合；
    从所述第二子集合中提取当前叶子节点集合的共现词，以更新所述匹配树对应的当前叶子节点集合；
    从所述词语集合中提取词语类型为第四预设类型的词语，得到第三子集合；及
    从所述第三子集合中提取更新后的当前叶子节点集合的共现词，根据更新后的当前叶子节点集合的共现词得到所述匹配树对应的目标叶子节点集合，生成匹配树。
  13. 根据权利要求12所述的计算机设备，其特征在于，所述处理器执行所述计算机可读指令时还执行以下步骤：
    确定所述匹配树上每一个所述目标叶子节点所在的匹配路径对应的疾病数据，及将所述匹配路径与其对应的疾病数据建立映射关系。
  14. 根据权利要求12所述的计算机设备，其特征在于，所述处理器执行所述计算机可读指令时还执行以下步骤：
    获取每一个所述根节点对应的所有疾病数据的编码，得到每一个所述根节点对应的第一编码集合；
    获取所述第一子集合中每一个词语对应的所有疾病数据的编码，得到所述第一子集合中每一个词语对应的第二编码集合；及
    当所述根节点对应的第一编码集合与所述词语对应的第二编码集合存在交集时，确定所述词语为所述根节点对应的共现词。
  15. 根据权利要求10至14任意一项所述的计算机设备，其特征在于，所述处理器执行所述计算机可读指令时还执行以下步骤：
    当所述分词结果中的词语与任意一个所述根节点都不匹配时，从预先建立的同义词库中查找每一个所述根节点对应的同义词；及
    将所述分词结果中的词语与所述同义词进行匹配，根据同义词匹配结果选取目标根节点作为当前节点。
  16. 一个或多个存储有计算机可读指令的非易失性计算机可读存储介质，所述计算机可读指令被一个或多个处理器执行时，使得所述一个或多个处理器执行以下步骤：
    获取待映射的疾病数据，对所述待映射的疾病数据进行分词，得到分词结果；
    将所述分词结果中的词语与匹配森林中每一个匹配树的根节点进行匹配，根据根节点匹配结果选取目标根节点作为当前节点；
    当所述当前节点存在对应的下一级子节点时，将所述分词结果中的词语与所述下一级子节点进行匹配；
    根据子节点匹配结果选取目标子节点作为当前节点，并进入当所述当前节点存在对应的下一级子节点时，将所述分词结果中的词语与所述下一级子节点进行匹配的步骤；及
    当所述当前节点不存在对应的下一级子节点时，确定所述当前节点所在的匹配路径，
    根据所述匹配路径获取目标疾病数据。
  17. 根据权利要求16所述的存储介质，其特征在于，所述计算机可读指令被所述处 理器执行时还执行以下步骤：
    对预先确定的基表库中每一个疾病数据进行分词，根据每一个所述疾病数据对应的分词结果得到词语集合，并获取所述词语集合中每一个词语所属的词语类型；
    从所述词语集合中提取词语类型为第一预设类型的多个词语分别作为匹配树的根节点；及
    根据所述根节点生成多个匹配树，得到匹配森林。
  18. 根据权利要求17所述的存储介质，其特征在于，所述计算机可读指令被所述处理器执行时还执行以下步骤：
    从所述词语集合中提取词语类型为第二预设类型的词语，得到第一子集合；
    从所述第一子集合中提取所述根节点对应的共现词，得到所述匹配树对应的当前叶子节点集合；
    从所述词语集合中提取词语类型为第三预设类型的词语，得到第二子集合；
    从所述第二子集合中提取当前叶子节点集合的共现词，以更新所述匹配树对应的当前叶子节点集合；
    从所述词语集合中提取词语类型为第四预设类型的词语，得到第三子集合；及
    从所述第三子集合中提取更新后的当前叶子节点集合的共现词，根据更新后的当前叶子节点集合的共现词得到所述匹配树对应的目标叶子节点集合，生成匹配树。
  19. 根据权利要求18所述的存储介质，其特征在于，所述计算机可读指令被所述处理器执行时还执行以下步骤：
    确定所述匹配树上每一个所述目标叶子节点所在的匹配路径对应的疾病数据，及将所述匹配路径与其对应的疾病数据建立映射关系。
  20. 根据权利要求18所述的存储介质，其特征在于，所述计算机可读指令被所述处理器执行时还执行以下步骤：
    获取每一个所述根节点对应的所有疾病数据的编码，得到每一个所述根节点对应的第一编码集合；
    获取所述第一子集合中每一个词语对应的所有疾病数据的编码，得到所述第一子集合中每一个词语对应的第二编码集合；及
    当所述根节点对应的第一编码集合与所述词语对应的第二编码集合存在交集时，确定所述词语为所述根节点对应的共现词。
PCT/CN2018/123365 2018-09-10 2018-12-25 疾病数据映射方法、装置、计算机设备和存储介质 Ceased WO2020052162A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811051911.1A CN109448860B (zh) 2018-09-10 2018-09-10 疾病数据映射方法、装置、计算机设备和存储介质
CN201811051911.1 2018-09-10

Publications (1)

Publication Number Publication Date
WO2020052162A1 true WO2020052162A1 (zh) 2020-03-19

Family

ID=65531037

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/123365 Ceased WO2020052162A1 (zh) 2018-09-10 2018-12-25 疾病数据映射方法、装置、计算机设备和存储介质

Country Status (2)

Country Link
CN (1) CN109448860B (zh)
WO (1) WO2020052162A1 (zh)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111696635A (zh) * 2020-05-13 2020-09-22 平安科技（深圳）有限公司 疾病名称标准化方法及装置
CN111985246A (zh) * 2020-08-27 2020-11-24 武汉东湖大数据交易中心股份有限公司 一种基于主要症状与伴随症状词的疾病认知系统
CN112035623A (zh) * 2020-09-11 2020-12-04 杭州海康威视数字技术股份有限公司 智能问答方法、装置、电子设备及存储介质
CN114064942A (zh) * 2021-11-11 2022-02-18 深圳市酷开软件技术有限公司 一种作品内容池的确定方法、装置、设备及存储介质
CN114860887A (zh) * 2022-05-26 2022-08-05 康键信息技术（深圳）有限公司 基于智能联想的疾病内容推送方法、装置、设备及介质
CN116049588A (zh) * 2022-12-30 2023-05-02 格兰菲智能科技有限公司 页面加载方法、装置、计算机设备和存储介质
CN120564942A (zh) * 2025-08-01 2025-08-29 中国人民解放军联勤保障部队第九六八医院 一种咽喉术后恢复信息的智能分析管理方法、设备及介质

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110309132B (zh) * 2019-05-08 2022-02-22 广东中建普联科技股份有限公司 一种工程概算表的定额标准化方法
CN115280303B (zh) * 2020-05-28 2026-02-27 深圳市欢太科技有限公司 文本处理方法、电子设备和计算机可读存储介质
CN112308857B (zh) * 2020-12-25 2021-05-11 数坤（北京）网络科技有限公司 一种血管中心线的确定方法、装置及可读存储介质
CN113436730A (zh) * 2021-07-14 2021-09-24 山大地纬软件股份有限公司 一种医院疾病诊断分类自动编码的方法及系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102841919A (zh) * 2012-06-30 2012-12-26 北京神州泰岳软件股份有限公司 一种会话文本中表情解析的方法和系统
CN107577826A (zh) * 2017-10-25 2018-01-12 山东众阳软件有限公司 基于原始诊断数据的疾病分类编码方法及系统
CN107705839A (zh) * 2017-10-25 2018-02-16 山东众阳软件有限公司 疾病自动编码方法及系统
CN107731269A (zh) * 2017-10-25 2018-02-23 山东众阳软件有限公司 基于原始诊断数据和病历文件数据的疾病编码方法及系统

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105095665B (zh) * 2015-08-13 2018-07-06 易保互联医疗信息科技（北京）有限公司 一种中文疾病诊断信息的自然语言处理方法及系统
CN106372439A (zh) * 2016-09-21 2017-02-01 北京大学 基于病例库的疾病症状及其权重知识的获取和处理方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102841919A (zh) * 2012-06-30 2012-12-26 北京神州泰岳软件股份有限公司 一种会话文本中表情解析的方法和系统
CN107577826A (zh) * 2017-10-25 2018-01-12 山东众阳软件有限公司 基于原始诊断数据的疾病分类编码方法及系统
CN107705839A (zh) * 2017-10-25 2018-02-16 山东众阳软件有限公司 疾病自动编码方法及系统
CN107731269A (zh) * 2017-10-25 2018-02-23 山东众阳软件有限公司 基于原始诊断数据和病历文件数据的疾病编码方法及系统

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111696635A (zh) * 2020-05-13 2020-09-22 平安科技（深圳）有限公司 疾病名称标准化方法及装置
CN111985246A (zh) * 2020-08-27 2020-11-24 武汉东湖大数据交易中心股份有限公司 一种基于主要症状与伴随症状词的疾病认知系统
CN111985246B (zh) * 2020-08-27 2023-08-15 武汉东湖大数据交易中心股份有限公司 一种基于主要症状与伴随症状词的疾病认知系统
CN112035623A (zh) * 2020-09-11 2020-12-04 杭州海康威视数字技术股份有限公司 智能问答方法、装置、电子设备及存储介质
CN114064942A (zh) * 2021-11-11 2022-02-18 深圳市酷开软件技术有限公司 一种作品内容池的确定方法、装置、设备及存储介质
CN114860887A (zh) * 2022-05-26 2022-08-05 康键信息技术（深圳）有限公司 基于智能联想的疾病内容推送方法、装置、设备及介质
CN116049588A (zh) * 2022-12-30 2023-05-02 格兰菲智能科技有限公司 页面加载方法、装置、计算机设备和存储介质
CN120564942A (zh) * 2025-08-01 2025-08-29 中国人民解放军联勤保障部队第九六八医院 一种咽喉术后恢复信息的智能分析管理方法、设备及介质

Also Published As

Publication number Publication date
CN109448860B (zh) 2024-09-17
CN109448860A (zh) 2019-03-08

Similar Documents

Publication Publication Date Title
WO2020052162A1 (zh) 疾病数据映射方法、装置、计算机设备和存储介质
CN109215796B (zh) 搜索方法、装置、计算机设备和存储介质
CN111459978B (zh) 查询方法、装置、计算机设备和存储介质
CN111341456B (zh) 糖尿病足知识图谱生成方法、装置及可读存储介质
CN112151141A (zh) 医疗数据处理方法、装置、计算机设备和存储介质
WO2020034808A1 (zh) 决策数据获取方法、装置、计算机设备和存储介质
CN109215771B (zh) 医疗映射关系库建立方法、装置、计算机设备和存储介质
WO2021114632A1 (zh) 疾病名称标准化方法、装置、设备及存储介质
WO2021151325A1 (zh) 基于医疗知识图谱的分诊模型训练方法、装置及设备
CN108874773B (zh) 关键词新增方法、装置、计算机设备和存储介质
WO2021151358A1 (zh) 基于解释模型的分诊信息推荐方法、装置、设备及介质
CN109213775B (zh) 搜索方法、装置、计算机设备和存储介质
CN111666370B (zh) 面向多源异构航天数据的语义索引方法和装置
CN106055909A (zh) 基于大数据知识库的治疗方案推荐系统
WO2020007027A1 (zh) 线上问答方法、装置、计算机设备和存储介质
CN109192258B (zh) 医疗数据转化方法、装置、计算机设备和存储介质
CN104620241A (zh) 多语言文档聚类
CN108595437B (zh) 文本查询纠错方法、装置、计算机设备和存储介质
CN113204942A (zh) 病案编码方法、装置、终端设备及可读存储介质
CN111429991B (zh) 药品预测方法、装置、计算机设备及存储介质
CN109065129A (zh) 科室推荐方法、装置、计算机设备和存储介质
WO2018228151A1 (zh) 分诊方法、装置和设备以及计算机可读存储介质
CN114416847A (zh) 一种数据转换的方法、装置、服务器及存储介质
CN110808095B (zh) 诊断结果识别、模型训练的方法、计算机设备及存储介质
CN116631614A (zh) 治疗方案生成方法、装置、电子设备及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18933208

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18933208

Country of ref document: EP

Kind code of ref document: A1