WO2021032220A2 - 一种基于证素的中医辅助诊断系统 - Google Patents

一种基于证素的中医辅助诊断系统 Download PDF

Info

Publication number
WO2021032220A2
WO2021032220A2 PCT/CN2020/121967 CN2020121967W WO2021032220A2 WO 2021032220 A2 WO2021032220 A2 WO 2021032220A2 CN 2020121967 W CN2020121967 W CN 2020121967W WO 2021032220 A2 WO2021032220 A2 WO 2021032220A2
Authority
WO
WIPO (PCT)
Prior art keywords
syndrome
tcm
patient
disease
symptoms
Prior art date
Application number
PCT/CN2020/121967
Other languages
English (en)
French (fr)
Other versions
WO2021032220A3 (zh
Inventor
孙钊
吴军
樊昭磊
李涛
郑彬茹
冯德杰
Original Assignee
山东众阳健康科技集团有限公司
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 山东众阳健康科技集团有限公司 filed Critical 山东众阳健康科技集团有限公司
Publication of WO2021032220A2 publication Critical patent/WO2021032220A2/zh
Publication of WO2021032220A3 publication Critical patent/WO2021032220A3/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/334—Query execution
    • G06F16/3344—Query execution using natural language analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
    • G06F16/367—Ontology
    • 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
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems

Definitions

  • the invention belongs to the technical field of medical information data processing, and in particular relates to a syndrome-based Chinese medicine auxiliary diagnosis system.
  • TCM doctors are still extremely scarce, especially in China’s vast rural areas, where medical resources are scarce, and the number of doctors is seriously insufficient compared to the number of people to be served.
  • village doctors are generally low in medical skills, and they know more about TCM. Very little. Therefore, the development of a TCM clinical auxiliary diagnosis system is of great significance for solving the shortage of basic TCM doctors or assisting doctors in carrying out TCM-based medical behaviors.
  • Some systems use syndrome elements to differentiate TCM syndromes, but rely too much on mathematical methods and fail to incorporate the clinical experience of TCM doctors;
  • prescriptions prescribed by these systems are not integrated with the patient's diagnosis of Western medicine, and the prescriptions lack specificity.
  • the present invention provides a TCM auxiliary diagnosis system based on syndrome elements, which can identify the patient's oral symptom descriptions, according to the patient's description, based on syndrome elements and clinical practice experience of TCM syndrome differentiation , Prescribe a prescription based on the patient’s Western medical diagnosis.
  • one or more embodiments of the present invention provide the following technical solutions:
  • a TCM auxiliary diagnosis system based on syndrome elements includes:
  • the database is used to store the TCM knowledge map, which includes symptoms to syndrome elements, syndrome elements to TCM syndrome types, syndrome elements to syndrome types, TCM syndrome types and Western medicine diagnosis to prescription names, prescription names, and basic patient information The mapping relationship between the ingredients and dosage of Chinese medicines in prescriptions;
  • Patient information acquisition module used to acquire basic information, symptoms and western medical diagnosis provided by patients
  • the TCM syndrome type inference module according to the knowledge map of TCM, infer the TCM syndrome type of the patient based on the patient's symptoms;
  • the module of formulating hidden medicines recommends Chinese medicine prescriptions based on the knowledge map of Chinese medicine, the basic information of the patient, the type of Chinese medicine and the diagnosis of western medicine.
  • One or more embodiments provide a syndrome-based Chinese medicine auxiliary diagnosis system, including:
  • Client terminal used to obtain basic information, symptoms and western medical diagnosis provided by the patient, and send it to the server;
  • the server stores the TCM knowledge graph, which includes symptoms to syndrome elements, syndrome elements to TCM syndrome types, syndrome elements to syndrome types, TCM syndrome types and western medicine diagnosis to prescription name, prescription name and basic patient information to prescription
  • the mapping relationship between Chinese medicine components and dosage after receiving the information provided by the patient from the client, perform the following operations:
  • TCM knowledge map infer the TCM syndrome type of the patient based on the patient's symptoms; According to the patient's basic information, TCM syndrome type and Western medicine diagnosis, recommend TCM prescriptions.
  • the present invention provides a TCM auxiliary diagnosis system based on syndrome elements.
  • the system introduces a TCM knowledge map.
  • the knowledge map records symptoms to syndrome elements, syndrome elements to TCM syndrome types, and syndrome elements to syndrome types.
  • TCM syndrome type and western medicine diagnosis to the prescription name, prescription name and basic patient information (sex, age, etc.) to the mapping relationship of prescription Chinese medicine components and dosage, so that the system can be combined with the clinical experience of Chinese medicine and the theory of Chinese medicine based on syndrome elements
  • the basic information of the patient and the diagnosis of western medicine to give comprehensive prescription recommendations.
  • Figure 1 is a schematic diagram of a framework of a syndrome-based TCM auxiliary diagnosis system in one or more embodiments of the present invention
  • FIG. 2 is a schematic diagram of part of the TCM knowledge map structure
  • Figure 3 is a schematic diagram of the map structure of the "symptom ⁇ syndrome element" part.
  • Disease-based syndrome elements The smallest unit describing body pathological changes in TCM theory. It is divided into two types: disease-based syndrome elements and disease-location syndrome elements. Disease-based syndrome elements refer to the pathological changes that occur in the body (for example, yin deficiency and yang deficiency). Refers to the site of disease (for example, kidney);
  • TCM syndrome type In TCM, it is equivalent to the diagnosis of Western medicine. It is composed of one or several disease syndrome elements combined (or not combined) with one (or several) disease location syndrome elements.
  • the syndrome type "Kidney Yin and Yang" Deficiency syndrome is composed of disease syndrome elements “Yin deficiency”, “Yang deficiency” and disease location syndrome element "Kidney”, while syndrome type "Yin deficiency syndrome” consists of disease syndrome element "Yin deficiency”.
  • Each diseased syndrome element corresponds to a single syndrome type that only contains the diseased syndrome element.
  • the single syndrome type corresponding to "qi deficiency" is "qi deficiency syndrome";
  • Typical symptoms of syndrome elements Each symptom corresponds to one or several syndrome elements. If a symptom corresponds to only one syndrome element (sickness or disease location), the symptom is called the typical symptom of the syndrome element.
  • This embodiment discloses a TCM auxiliary diagnosis system based on syndrome elements, including:
  • the database is used to store the knowledge graph of Chinese medicine.
  • the nodes of the knowledge graph are divided into multiple categories, including: symptom, syndrome element, syndrome element category, TCM syndrome type + Western medicine diagnosis, prescription name + patient information, prescription and traditional Chinese medicine composition and dosage; edges are used to represent these categories
  • Correspondence between symptoms including: symptom ⁇ syndrome element, syndrome element ⁇ TCM syndrome type, syndrome element ⁇ syndrome element type, TCM syndrome type + western medicine diagnosis ⁇ prescription name (Chinese medicine syndrome type combined with western medicine diagnosis), prescription name + patient information (For example, gender, age) ⁇ the ingredients and dosage of Chinese medicine in the prescription (adjust the ingredients and dosage of the prescription according to the basic information of the patient).
  • the schematic diagram of the map structure is shown in Figure 2, and Figure 3 is the schematic diagram of the map structure of "symptom ⁇ syndrome element".
  • the knowledge graph is stored in the graph database neo4j in the form of a directed graph.
  • the patient information acquisition module is used to acquire and store basic information, symptoms, and western medical diagnosis provided by the patient; wherein, the basic information includes but is not limited to gender, age, etc. Those skilled in the art can understand that the patient can provide patient information through text input or voice input or other existing input methods.
  • the patient information processing module standardizes the acquired patient information, converts the symptom sentences into the symptoms stored in the TCM knowledge graph, and obtains the patient's symptom set. Specifically include the following steps:
  • the TCM syndrome type inference module infers the patient's TCM syndrome type based on the patient's symptoms. Specifically include the following steps:
  • the patient’s symptom set includes 6 symptoms (that is, 6 symptoms are identified from the patient information): Z1, Z2, Z3, Z4, Z5, Z6.
  • each symptom corresponds to The evidence is as follows:
  • e1-e7 are disease elements
  • b1-b7 are disease elements
  • the inference process of the diseased syndrome element includes:
  • step 2 If it exists, check whether the disease syndrome elements corresponding to the typical symptoms can cover all symptoms. If all symptoms are covered, use the syndrome elements corresponding to the typical symptoms as the output result of the patient's disease syndrome elements. If all symptoms cannot be covered, go to step 2 ;
  • the disease syndrome elements e1 and e7 corresponding to Z1 and Z6 cannot cover the symptoms Z2 and Z3, so proceed to step 2.
  • step 3 If there is more than one disease syndrome that appears most frequently, go to step 3;
  • step 3 is entered.
  • the disease syndrome element and the disease syndrome element corresponding to the typical symptoms (if any) are used as the output result of the patient's disease syndrome element.
  • the order of appearance of the disease syndrome element e2 is higher than that of e6, so the output results of the disease syndrome element of the patient are e1, e2, and e7.
  • the disease location syndrome element is performed on the premise that the disease syndrome element is inferred, and the inference process includes:
  • step 2 If it exists, check whether the syndrome element corresponding to the typical symptom can cover all symptoms. If it covers all symptoms, use the syndrome element corresponding to the typical symptom as the output result of the patient's syndrome element. If it cannot cover all symptoms, go to step 2 ;
  • the disease location syndrome elements b1 and b7 corresponding to Z1 and Z6 cannot cover the symptoms Z2 and Z3, so proceed to step 2.
  • the disease location syndrome elements other than the disease location syndrome elements corresponding to the typical symptoms determine the disease location syndrome element that occurs the most frequently. If there is only one disease location syndrome element that occurs the most frequently, then the disease location syndrome element and The disease location syndrome element corresponding to the typical symptom (if any) is the output result of the patient's disease location syndrome element;
  • step 3 If there is more than one disease location syndrome element that appears most frequently, go to step 3;
  • step 3 is entered.
  • the disease location syndrome element with the highest appearance order can be found, the disease location syndrome element and the disease location syndrome element corresponding to the typical symptoms (if any) are used as the output result of the patient's disease location syndrome element.
  • the order of appearance of the disease location syndrome element b2 is higher than that of b6, so the output result of the disease location syndrome element of the patient is b1, b2, b7.
  • step (3) If you can't find any disease location syndrome element according to steps 1-3, go to step (3), the patient's disease location syndrome element result is an empty set.
  • the output result of the disease location syndrome element of the patient is b1, b2, and b7.
  • step (3) Infer the candidate TCM syndrome types based on the diseased syndrome elements inferred in step (2).
  • the disease syndrome element corresponding to the syndrome type is completely equal to [Note: Not included] disease syndrome element set E. If you can find such a syndrome type (maybe more than one), output these syndromes Type; if unable to find, go to step 2;
  • step (2) Combining the disease location syndrome elements inferred in step (2), infer the most suitable TCM syndrome type from the candidate TCM syndrome types obtained in step (3); specifically including:
  • step (2) Assume that the evidence deduced in step (2) is as follows:
  • step (3) For the candidate TCM syndrome types inferred in step (3), select the corresponding disease location syndrome element set to be the TCM syndrome type combination of the set B subset. For example, assume that in step (3), the CCP inferred five sets of syndrome combinations, namely:
  • the disease location set B contains b1, b2, and b3, there are three groups of syndrome types that meet the conditions, G1, G3, and G5.
  • the prescribing and latent medicine module through the patient's symptoms, TCM syndromes, and Western medicine diagnosis, according to the TCM knowledge map, formulating latent drugs for the patient.
  • the specific implementation steps are as follows:
  • the TCM syndrome type inferred by the TCM syndrome type inference module combined with the patient's basic information (for example, gender, age) and Western medicine diagnosis obtained by the patient information acquisition module, the name of the TCM prescription, drug composition, and dosage are given in the TCM knowledge map .
  • the purpose of this embodiment is to provide a syndrome element-based TCM auxiliary diagnosis system, which includes:
  • Client terminal used to obtain basic information, symptoms and western medical diagnosis provided by the patient, and send it to the server;
  • the server stores the TCM knowledge graph, which includes symptoms to syndrome elements, syndrome elements to TCM syndrome types, syndrome elements to syndrome types, TCM syndrome types and western medicine diagnosis to prescription name, prescription name and basic patient information to prescription
  • the mapping relationship between Chinese medicine components and dosage after receiving the information provided by the patient from the client, perform the following operations:
  • TCM knowledge map infer the TCM syndrome type of the patient based on the patient's symptoms; According to the patient's basic information, TCM syndrome type and Western medicine diagnosis, recommend TCM prescriptions.
  • the present invention provides a TCM auxiliary diagnosis system based on syndrome elements.
  • the system introduces a TCM knowledge map.
  • the knowledge map records symptoms to syndrome elements, syndrome elements to TCM syndrome types, and syndrome elements to syndrome types.
  • TCM syndrome type and western medicine diagnosis to the prescription name, prescription name and basic patient information (sex, age, etc.) to the mapping relationship of prescription Chinese medicine components and dosage, so that the system can be combined with the clinical experience of Chinese medicine and the theory of Chinese medicine based on syndrome elements
  • the basic information of the patient and the diagnosis of western medicine to give comprehensive prescription recommendations.
  • the system can also identify the patient’s colloquial symptom descriptions of symptoms. After receiving the patient’s descriptions, the natural language processing method is used to process them, and the symptoms in the colloquial description correspond to standardized expressions in the TCM knowledge graph, so that The system has strong ability to recognize the expression of patients, and the recommended prescriptions are more targeted.
  • modules or steps of the present invention can be implemented by a general computer device. Alternatively, they can be implemented by program code executable by the computing device, so that they can be stored in a storage device. The device is executed by a computing device, or they are separately fabricated into individual integrated circuit modules, or multiple modules or steps in them are fabricated into a single integrated circuit module for implementation.
  • the present invention is not limited to any specific combination of hardware and software.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Public Health (AREA)
  • Biomedical Technology (AREA)
  • General Physics & Mathematics (AREA)
  • Medical Informatics (AREA)
  • Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Animal Behavior & Ethology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Primary Health Care (AREA)
  • Pathology (AREA)
  • Epidemiology (AREA)
  • Artificial Intelligence (AREA)
  • Medical Treatment And Welfare Office Work (AREA)

Abstract

本发明公开了一种基于证素的中医辅助诊断系统，包括：数据库，用于存储中医知识图谱，所述中医知识图谱包括症状到证素、证素到中医证型、证素到证素类型、中医证型和西医诊断到药方名称、药方名称和患者基本信息到药方中药成分及用量的映射关系；患者信息获取模块，用于获取患者提供的基本信息、症状及西医诊断；中医症型推断模块，根据中医知识图谱，基于患者的症状推断患者的中医证型；组方潜药模块，根据中医知识图谱，根据患者的基本信息、中医证型和西医诊断，推荐中医药方。本发明通过引入中医知识图谱，能够在进行中医药方推荐时综合考虑症状、患者基本信息和西医诊断等因素，给出更为合理的推荐结果。

Description

一种基于证素的中医辅助诊断系统 技术领域
本发明属于医疗信息数据处理技术领域，尤其涉及一种基于证素的中医辅助诊断系统。
背景技术
本部分的陈述仅仅是提供了与本公开相关的背景技术信息，不必然构成在先技术。
中医在养生保健、治未病服务、防治传染病及特色康复服务方面有重大优势，尤其在常见病、多发病和慢性病防治中能发挥独特的作用。然而在基层，中医医生还极为紧缺，尤其在中国广袤的农村地区，医疗资源匮乏，医生数量相比要服务的人群数量严重不足，另外村医的医术也普遍较低，对于中医知识更是知之甚少。因此，开发中医临床辅助诊断系统对于解决基层中医医生数量不足或辅助医生开展基于中医药的医疗行为具有重大的意义。
据发明人了解，现阶段的各种中医临床辅助诊断系统并没有在中医临床实践中发挥很大作用，主要存在以下问题：
大多数系统只通过症状进行推断，没有与中医理论深入结合；
有些系统通过证素进行中医辨证分型，但是过于依赖数学方法，没有结合中医医生的临床经验；
这些系统可以识别中医术语描述的症状，却无法识别口语化的症状描述，例如，系统可以明白“食少纳呆”，却不能理解“我不想吃饭，懒得说话”；
并且，这些系统开出的药方均没有与患者的西医诊断相结合，药方缺乏针对性。
发明内容
为克服上述现有技术的不足，本发明提供了一种基于证素的中医辅助诊断系统，它能够识别患者口语化症状描述，根据患者的描述，基于证素和中医临床的实践经验辨证分型，结合患者的西医诊断开出药方。
为实现上述目的，本发明的一个或多个实施例提供了如下技术方案：
一种基于证素的中医辅助诊断系统，包括：
数据库，用于存储中医知识图谱，所述中医知识图谱包括症状到证素、证素到中医证型、证素到证素类型、中医证型和西医诊断到药方名称、药方名称和患者基本信息到药方中药成分及用量的映射关系；
患者信息获取模块，用于获取患者提供的基本信息、症状及西医诊断；
中医症型推断模块，根据中医知识图谱，基于患者的症状推断患者的中医证型；
组方潜药模块，根据中医知识图谱，根据患者的基本信息、中医证型和西医诊断，推荐中医药方。
一个或多个实施例提供了一种基于证素的中医辅助诊断系统，包括：
客户端，用于获取患者提供的基本信息、症状及西医诊断，并发送至服务器；
服务器，存储中医知识图谱，所述中医知识图谱包括症状到证素、证素到中医证型、证素到证素类型、中医证型和西医诊断到药方名称、药方名称和患者基本信息到药方中药成分及用量的映射关系；当接收到客户端发送的患者提供的信息后，执行以下操作：
根据中医知识图谱，基于患者的症状推断患者的中医证型；根据患者的基本信息、中医证型和西医诊断，推荐中医药方。
以上一个或多个技术方案存在以下有益效果：
本发明提供了一种基于证素的中医辅助诊断系统，所述系统中引入了中医知识图谱，所述知识图谱中记录了症状到证素、证素到中医证型、证素到证素类型、中医证型和西医诊断到药方名称、药方名称和患者基本信 息(性别、年龄等)到药方中药成分及用量的映射关系，使得该系统能够与中医临床经验和基于证素的中医理论相结合，在患者提供了症状的基础上，将其与患者基本信息、西医诊断相结合，综合给出药方建议。
附图说明
构成本发明的一部分的说明书附图用来提供对本发明的进一步理解，本发明的示意性实施例及其说明用于解释本发明，并不构成对本发明的不当限定。
图1为本发明一个或多个实施例中一种基于证素的中医辅助诊断系统框架示意图；
图2为部分中医知识图谱结构示意图；
图3为“症状→证素”部分的图谱结构示意图。
具体实施方式
应该指出，以下详细说明都是示例性的，旨在对本发明提供进一步的说明。除非另有指明，本文使用的所有技术和科学术语具有与本发明所属技术领域的普通技术人员通常理解的相同含义。
需要注意的是，这里所使用的术语仅是为了描述具体实施方式，而非意图限制根据本发明的示例性实施方式。如在这里所使用的，除非上下文另外明确指出，否则单数形式也意图包括复数形式，此外，还应当理解的是，当在本说明书中使用术语“包含”和/或“包括”时，其指明存在特征、步骤、操作、器件、组件和/或它们的组合。
在不冲突的情况下，本发明中的实施例及实施例中的特征可以相互组合。
术语解释：
证素：中医理论中描述身体病变的最小单位，分为病性证素和病位证素两种，病性证素指身体出现的病变(例如，阴虚、阳虚)，病位证素指出现病变的部位(例如，肾)；
中医证型：在中医中相当于西医的诊断，由一种或几种病性证素结合(或不结合)一种(或几种)病位证素组成，例如，证型“肾阴阳两虚证”由病性证素“阴虚”、“阳虚”和病位证素“肾”组成，而证型“阴虚证”由病性证素“阴虚”组成。每个病性证素对应一个只包含该病性证素的单一证型，例如，“气虚”对应的单一证型为“气虚证”；
证素的典型症状：每种症状会对应一个或几个证素，若某种症状只对应一个(病性或病位)证素，则该症状叫做此证素的典型症状。
实施例一
本实施例公开了一种基于证素的中医辅助诊断系统，包括：
数据库，用于存储中医知识图谱。
所述知识图谱的节点被划分为多个类别，包括：症状、证素、证素类别、中医证型+西医诊断、药方名称+患者信息、药方中药的成分及用量；边用来表示这些类别之间的对应关系，包括：症状→证素，证素→中医症型，证素→证素类型，中医证型+西医诊断→药方名称(中医证型结合西医诊断)，药方名称+患者信息(例如，性别、年龄)→药方中药的成分及用量(根据患者的基本信息调整药方的成分和用量)。图谱的结构示意图如图2所示，图3为“症状→证素”的图谱结构示意图。
所述知识图谱以有向图的形式存储在图数据库neo4j中。
患者信息获取模块，用于获取患者提供的基本信息、症状及西医诊断并进行存储；其中，所述基本信息包括但不限于性别、年龄等。本领域技术人员可以理解，患者可通过文本输入或语音输入或其他现有输入方法提供患者的信息。
患者信息处理模块，对获取的患者信息进行标准化处理，将其中的症状语句转化成中医知识图谱中所存储的症状，得到该患者的症状集合。具体包括以下步骤：
(1)中医知识图谱中涉及的所有症状的集合记为症状词典，患者输入 的症状语句记为元语句，对元语句进行去标点处理，得到合语句；
(2)用分词工具(例如，结巴分词)对合语句进行分词，分词结果被称为初始分词集；
(3)对于初始分词集中的每个词，均在症状词典中进行查询，若能查询到，则将该词纳入该患者的症状集合；将不能找到，将该词纳入“待处理分词集”；
(4)对“待处理分词集”中的每一个词i，找到它在元语句中所在的断句，将该断句转成拼音(只有拼音字母，不带音调)，利用搜索匹配工具(例如，Lucene结合BM25)找到该拼音在症状词典所包含所有症状的拼音集合中得分最高的症状，将该症状作为i的识别结果，纳入该患者的症状集合。该患者症状集合中的症状按照患者输入信息中的出现顺序排列。
中医症型推断模块，根据中医知识图谱，基于患者的症状推断患者的中医证型。具体包括以下步骤：
(1)根据中医知识图谱，基于该患者的症状集合中的症状推断出证素；
本实施例中，假设该患者的症状集合中包括6个症状(即从患者信息中识别出6个症状)：Z1、Z2、Z3、Z4、Z5、Z6，依据中医知识图谱，各症状对应的证素如下：
Z1对应的证素：e1，b1；
Z2对应的证素：e2，e8，b2，b8；
Z3对应的证素：e2，e3，e4，b2，b3，b4；
Z4对应的证素：e1，e6，b1，b6；
Z5对应的证素：e6，e7，b6，b7；
Z6对应的证素：e7，b7；
其中e1-e7为病性证素，b1-b7为病位证素。
(2)根据所述步骤(1)确定的证素，进行该患者病性证素和病位证素的推断。其中，所述病性证素的推断过程包括：
①检查所有症状中是否存在某些病性证素的典型症状(即该症状只对应一个证素，如本例中的Z1、Z6)，若不存在，进入步骤②；
若存在，检查典型症状对应的病性证素是否能涵盖所有症状，若涵盖所有症状，将典型症状对应的证素作为该患者病性证素的输出结果，若不能涵盖所有症状，进入步骤②；
本实施例中，Z1、Z6对应的病性证素e1，e7不能涵盖症状Z2，Z3，故而进入步骤②。
②在典型症状对应的病性证素之外的病性证素中，确定出现次数最多的病性证素，若出现次数最多的病性证素仅有一个，则将该病性证素以及典型症状对应的病性证素(若有)作为该患者病性证素的输出结果；
若出现次数最多的病性证素不止一个，进入步骤③；
本实施例中，病性证素e2、e6出现次数都是最多的，均为2次，故而进入步骤③。
③在出现次数最多的多个病性证素中，根据症状的出现顺序，选择出现顺序最靠前的病性证素，若出现在同一个症状对应的证素中，则出现顺序相同，无法找到出现顺序最靠前的病性证素，则只将典型症状对应的病性证素(若有)作为该患者病性证素的输出结果；
若能够找到出现顺序最靠前的一个病性证素，将该病性证素以及典型症状对应的病性证素(若有)作为该患者病性证素的输出结果。
本实施例中，病性证素e2出现的顺序比e6靠前，因此该患者病性证素的输出结果为e1、e2、e7。
④若按照步骤①-③找不到任何病性证素，则输出“输入的症状不能确定证型，请输入更多症状或重新输入症状”以提示患者；待接收患者通过患者信息获取模块重新输入的症状后，重新执行病性证素推断。
所述病位证素在推断出病性证素的前提下进行，推断过程包括：
①检查所有症状中是否存在某些病位证素的典型症状(即该症状只对 应一个病位证素，如本例中的Z1、Z6)，若不存在，进入步骤②；
若存在，检查典型症状对应的病位证素是否能涵盖所有症状，若涵盖所有症状，将典型症状对应的证素作为该患者病位证素的输出结果，若不能涵盖所有症状，进入步骤②；
本实施例中，Z1、Z6对应的病位证素b1，b7不能涵盖症状Z2，Z3，故而进入步骤②。
②在典型症状对应的病位证素之外的病位证素中，确定出现次数最多的病位证素，若出现次数最多的病性证素仅有一个，则将该病位证素以及典型症状对应的病位证素(若有)作为该患者病位证素的输出结果；
若出现次数最多的病位证素不止一个，进入步骤③；
本实施例中，病位证素b2、b6出现次数都是最多的，均为2次，故而进入步骤③。
③在出现次数最多的多个病位证素中，根据症状的出现顺序，选择出现顺序最靠前的病位证素，若出现在同一个症状对应的证素中，则出现顺序相同，无法找到出现顺序最靠前的病位证素，则只将典型症状对应的病位证素(若有)作为该患者病位证素的输出结果；
若能够找到出现顺序最靠前的一个病位证素，将该病位证素以及典型症状对应的病位证素(若有)作为该患者病位证素的输出结果。
本实施例中，病位证素b2出现的顺序比b6靠前，因此该患者病位证素的输出结果为b1、b2、b7。
④若按照步骤①-③找不到任何病位证素，则进入步骤(3)，该患者的病位证素结果为空集。
本实施例中，该患者病位证素的输出结果为b1、b2、b7。
(3)根据步骤(2)中推断出的病性证素推断候选中医证型。
假设推断出的证素如下：
病性证素集合E：e1、e2、e3、e4；
病位证素集合B：b1、b2、b3；
推断原则：选择数量最少的证型覆盖所有步骤(301)中所推断出的病性证素
①在知识图谱中查找中医证型，该证型对应的病性证素完全等于【注意：不是包含】病性证素集合E，若能找到这样的证型(可能不止一个)，输出这些证型；若无法找到，进入步骤②；
②在知识图谱中查找两个中医证型，满足以下条件：它们对应的病性证素的并集完全等于【注意：不是包含】病性证素集合E，若能找到这样的证型组合(可能不止一个组合)，输出这些证型组合；若步骤②中无法找到符合条件的中医证型组合，进入步骤③；
③输出每个病性证素对应的单一中医证型。
(4)结合步骤(2)中推断出的病位证素，从步骤(3)中获得的候选中医证型中推断出最合适的中医证型；具体包括：
①对步骤(3)中获得的候选中医证型，根据中医知识图谱，找出它们对应的病位证素。
假设步骤(2)推断出的证素如下：
病性证素集合E：e1、e2、e3、e4；
病位证素集合B：b1、b2、b3；
②对于步骤(3)中推断出的候选中医证型，选择所对应的病位证素集合是集合B子集的中医证型组合。例如，假设步骤(3)中共推断出五组证型组合，分别是：
证型组合G1：b1、b2、b3；
证型组合G2：b1、b2、b4；
证型组合G3：b2、b3；
证型组合G4:b1、b4；
证型组合G5:空集；
由于病位集合B包含b1、b2、b3，所以符合条件的证型组合有三组，G1、G3、G5。
③找出候选证型组合G1、G3、G5中对应的病位证素个数最多的证型组合，若个数最多的证型组合只有一个，输出该证型组合，若找到的证型组合不止一组，可任选一组输出。
组方潜药模块，通过患者的症状、中医证型、西医诊断，依据中医知识图谱，给患者组方潜药。具体实施步骤如下：
根据中医症型推断模块推断出的中医证型，结合患者信息获取模块获取的患者基本信息(例如，性别、年龄)、西医诊断，依据中医知识图谱中给出中医药方的名称和药品成分、用量。
实施例二
本实施例的目的是提供一种基于证素的中医辅助诊断系统，所述系统包括：
客户端，用于获取患者提供的基本信息、症状及西医诊断，并发送至服务器；
服务器，存储中医知识图谱，所述中医知识图谱包括症状到证素、证素到中医证型、证素到证素类型、中医证型和西医诊断到药方名称、药方名称和患者基本信息到药方中药成分及用量的映射关系；当接收到客户端发送的患者提供的信息后，执行以下操作：
将患者提供的症状转化为中医知识图谱中所存储的症状，得到该患者的症状集合；
根据中医知识图谱，基于患者的症状推断患者的中医证型；根据患者的基本信息、中医证型和西医诊断，推荐中医药方。
以上实施例二中涉及的具体实现方法与实施例一相对应，具体可参见实施例一的相关说明部分。
以上一个或多个实施例具有以下技术效果：
本发明提供了一种基于证素的中医辅助诊断系统，所述系统中引入了中医知识图谱，所述知识图谱中记录了症状到证素、证素到中医证型、证素到证素类型、中医证型和西医诊断到药方名称、药方名称和患者基本信息(性别、年龄等)到药方中药成分及用量的映射关系，使得该系统能够与中医临床经验和基于证素的中医理论相结合，在患者提供了症状的基础上，将其与患者基本信息、西医诊断相结合，综合给出药方建议。
所述系统还能够识别患者对症状的口语化症状描述，在接收到患者的描述后，通过自然语言处理方法对其处理，将口语化描述中的症状对应到中医知识图谱中的标准化表达，使得系统对于患者的表达识别能力强，推荐药方的针对性更强。
本领域技术人员应该明白，上述本发明的各模块或各步骤可以用通用的计算机装置来实现，可选地，它们可以用计算装置可执行的程序代码来实现，从而，可以将它们存储在存储装置中由计算装置来执行，或者将它们分别制作成各个集成电路模块，或者将它们中的多个模块或步骤制作成单个集成电路模块来实现。本发明不限制于任何特定的硬件和软件的结合。
以上所述仅为本发明的优选实施例而已，并不用于限制本发明，对于本领域的技术人员来说，本发明可以有各种更改和变化。凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。
上述虽然结合附图对本发明的具体实施方式进行了描述，但并非对本发明保护范围的限制，所属领域技术人员应该明白，在本发明的技术方案的基础上，本领域技术人员不需要付出创造性劳动即可做出的各种修改或变形仍在本发明的保护范围以内。

Claims (10)

  1. 一种基于证素的中医辅助诊断系统，其特征在于，包括：
    数据库，用于存储中医知识图谱，所述中医知识图谱包括症状到证素、证素到中医证型、证素到证素类型、中医证型和西医诊断到药方名称、药方名称和患者基本信息到药方中药成分及用量的映射关系；
    患者信息获取模块，用于获取患者提供的基本信息、症状及西医诊断；
    中医症型推断模块，根据中医知识图谱，基于患者的症状推断患者的中医证型；
    组方潜药模块，根据中医知识图谱，根据患者的基本信息、中医证型和西医诊断，推荐中医药方。
  2. 如权利要求1所述的一种基于证素的中医辅助诊断系统，其特征在于，所述中医知识图谱中的节点用于表示多个类别：症状、证素、中医证型、证素类型、中医证型和西医诊断、药方名称和患者信息、药方成分及用量；边用来表示这些类别之间的对应关系。
  3. 如权利要求1所述的一种基于证素的中医辅助诊断系统，其特征在于，还包括患者信息处理模块，用于将患者提供的症状转化为中医知识图谱中所存储的症状，得到该患者的症状集合。
  4. 如权利要求3所述的一种基于证素的中医辅助诊断系统，其特征在于，将患者提供的症状转化为中医知识图谱中所存储的症状具体包括：
    提取中医知识图谱中涉及的所有症状，构成症状词典；
    将患者提供的症状语句依次进行去标点、分词处理，得到初始分词集；
    对于初始分词集中的每个词，均在症状词典中进行查询，若能查询到，则将该词写入该患者的症状集合；将不能找到，将该词写入待处理分词集；
    对于待处理分词集中的每一个词，确定其在患者提供的症状语句中所在的断句，将该断句转成拼音，基于拼音相似性在症状词典中查找症状，得到该词的识别结果，写入该患者的症状集合；该患者的症状集合中的症状按照患者提供的症状语句中出现的顺序排列。
  5. 如权利要求1所述的一种基于证素的中医辅助诊断系统，其特征在于，推断患者的中医证型包括：
    根据中医知识图谱，基于该患者提供的症状获取每个症状对应的证素；
    根据证素对症状的覆盖程度、出现次数和出现顺序，从中推断该患者的病性证素和病位证素；
    根据推断出的病性证素推断候选中医证型；
    结合推断出的病位证素，从候选中医证型中推断出最合适的中医证型。
  6. 如权利要求5所述的一种基于证素的中医辅助诊断系统，其特征在于，推断该患者的病性证素包括：
    ①查询是否存在只对应一个病性证素的典型症状，若不存在，进入步骤②；若存在，检查这些病性证素能否涵盖所有症状，若能涵盖，将这些病性证素写入该患者的病性证素集合，若不能涵盖，进入步骤②；
    ②在剩余病性证素中，确定出现次数最多的病性证素，若出现次数最多的病性证素仅有一个，将该病性证素写入该患者的病性证素集合；若出现次数最多的病性证素不止一个，进入步骤③；
    ③查看这些病性证素的出现是否存在先后顺序，若存在，将出现顺序靠前的病性证素写入该患者的病性证素集合；
    ④若按照步骤①-③找不到任何病性证素，则输出补充症状描述的提 示。
  7. 如权利要求6所述的一种基于证素的中医辅助诊断系统，其特征在于，在推断出该患者的病性证素后，进一步执行该患者病位证素的推断，包括：
    ①查询是否存在只对应一个病位证素的典型症状，若不存在，进入步骤②；若存在，检查这些病位证素能否涵盖所有症状，若能涵盖，将这些病位证素写入该患者的病位证素集合，若不能涵盖，进入步骤②；
    ②在剩余病位证素中，确定出现次数最多的病位证素，若出现次数最多的病位证素仅有一个，将该病位证素写入该患者的病位证素集合；若出现次数最多的病位证素不止一个，进入步骤③；
    ③查看这些病位证素的出现是否存在先后顺序，若存在，将出现顺序靠前的病位证素写入该患者的病位证素集合；
    ④若按照步骤①-③找不到任何病位证素，则该患者的病位证素集合为空集。
  8. 如权利要求5所述的一种基于证素的中医辅助诊断系统，其特征在于，推断候选中医证型包括：
    ①在知识图谱中查找中医证型，若能够查找到一个或多个中医证型对应的病性证素完全等于病性证素集合，该一个或多个中医证型即为候选中医证型；若无法找到，进入步骤②；
    ②进一步查找两个中医证型，所述两个中医证型对应的病性证素的并集完全等于病性证素集合，若能够查找到，所述两个中医证型即为候选中医证型；若无法找到，进入步骤③；
    ③查找每个病性证素对应的单一中医证型作为候选中医证型。
  9. 如权利要求8所述的一种基于证素的中医辅助诊断系统，其特征在于，推断最合适的中医证型包括：
    对于候选中医证型中的每个中医症型，根据中医知识图谱查找其对应的病位证素；选择所对应的病位证素集合是病位证素集合子集的中医证型组合；
    查找候选证型组合中对应的病位证素个数最多的证型组合，即为最合适的中医证型。
  10. 一种基于证素的中医辅助诊断系统，其特征在于，包括：
    客户端，用于获取患者提供的基本信息、症状及西医诊断，并发送至服务器；
    服务器，存储中医知识图谱，所述中医知识图谱包括症状到证素、证素到中医证型、证素到证素类型、中医证型和西医诊断到药方名称、药方名称和患者基本信息到药方中药成分及用量的映射关系；当接收到客户端发送的患者提供的信息后，执行以下操作：
    根据中医知识图谱，基于患者的症状推断患者的中医证型；根据患者的基本信息、中医证型和西医诊断，推荐中医药方。
PCT/CN2020/121967 2019-08-20 2020-10-19 一种基于证素的中医辅助诊断系统 WO2021032220A2 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910768831.6A CN110459321B (zh) 2019-08-20 2019-08-20 一种基于证素的中医辅助诊断系统
CN201910768831.6 2019-08-20

Publications (2)

Publication Number Publication Date
WO2021032220A2 true WO2021032220A2 (zh) 2021-02-25
WO2021032220A3 WO2021032220A3 (zh) 2021-04-15

Family

ID=68487882

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/121967 WO2021032220A2 (zh) 2019-08-20 2020-10-19 一种基于证素的中医辅助诊断系统

Country Status (2)

Country Link
CN (1) CN110459321B (zh)
WO (1) WO2021032220A2 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113241173A (zh) * 2021-05-12 2021-08-10 华中科技大学 一种用于慢性阻塞性肺疾病的中医辅助诊疗方法和系统
CN116631612A (zh) * 2023-06-09 2023-08-22 广东工业大学 一种基于多图融合的图卷积草药推荐方法及计算机
CN116821779A (zh) * 2023-08-31 2023-09-29 中南大学湘雅医院 一种针对胃肠健康的大数据识别方法

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110459321B (zh) * 2019-08-20 2020-10-23 山东众阳健康科技集团有限公司 一种基于证素的中医辅助诊断系统
CN110931124A (zh) * 2019-12-09 2020-03-27 中国中医科学院中医药信息研究所 诊断系统
CN112086206A (zh) * 2020-09-09 2020-12-15 北京小白世纪网络科技有限公司 一种利用可编辑知识图谱进行推演的处方查找方法及装置
CN112216383B (zh) * 2020-10-26 2023-02-21 山东众阳健康科技集团有限公司 基于证素和深度学习的中医智能问诊舌诊综合系统
CN112466436B (zh) * 2020-11-25 2024-02-23 北京小白世纪网络科技有限公司 基于循环神经网络的智能中医开方模型训练方法及装置
CN112700838B (zh) * 2020-12-30 2024-03-01 平安科技（深圳）有限公司 基于大数据的用药方案推荐方法、装置及相关设备
CN113140310B (zh) * 2021-05-06 2024-02-27 王耘 一种中医药智能诊疗系统及创建方法
CN114512228A (zh) * 2022-02-08 2022-05-17 吾征智能技术(北京)有限公司 一种中医疾病辅助诊断系统、设备及存储介质
CN115423052B (zh) * 2022-11-07 2023-03-24 南京大经中医药信息技术有限公司 基于多图注意力的中医证型分类方法

Family Cites Families (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN2718703Y (zh) * 2003-12-16 2005-08-17 杨富元 一种中医证治处方速查盘
CN1804850A (zh) * 2005-11-24 2006-07-19 朱文锋 中医证素辨证诊疗系统
CN102222153A (zh) * 2010-01-27 2011-10-19 洪文学 中医机器问诊数量化辩证诊断方法
CN105825064A (zh) * 2016-03-22 2016-08-03 施弘 日常性中医疗法智能咨询系统
CN107066814A (zh) * 2017-03-09 2017-08-18 南京邮电大学 一种基于望闻问切协同的中医智能辅助诊断系统
CN107145744B (zh) * 2017-05-08 2018-03-02 合肥工业大学 医学知识图谱的构建方法、装置及辅助诊断方法
CN107280648A (zh) * 2017-07-31 2017-10-24 郦永平 中医诊断系统
CN108986911A (zh) * 2018-07-06 2018-12-11 成都中医药大学 一种中医辩证论治数据处理方法
CN109448838A (zh) * 2018-09-28 2019-03-08 小伍健康科技（上海）有限责任公司 一种基于深度神经网络的症状诊断方法及设备
CN109271530A (zh) * 2018-10-17 2019-01-25 长沙瀚云信息科技有限公司 一种疾病知识图谱构建方法和平台系统、设备、存储介质
CN109766445B (zh) * 2018-12-13 2024-03-26 平安科技（深圳）有限公司 一种知识图谱构建方法及数据处理装置
CN109670054B (zh) * 2018-12-26 2020-11-10 医渡云（北京）技术有限公司 知识图谱构建方法、装置、存储介质及电子设备
CN109830299B (zh) * 2019-02-14 2021-07-27 南京大经中医药信息技术有限公司 一种基于人体模型的中医病机辨证方法、装置及存储介质
CN110085307B (zh) * 2019-04-04 2023-02-03 华东理工大学 一种基于多源知识图谱融合的智能导诊方法和系统
CN110085325B (zh) * 2019-04-30 2021-06-01 王小岗 关于中医经验数据的知识图谱的构建方法及装置
CN110459321B (zh) * 2019-08-20 2020-10-23 山东众阳健康科技集团有限公司 一种基于证素的中医辅助诊断系统

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113241173A (zh) * 2021-05-12 2021-08-10 华中科技大学 一种用于慢性阻塞性肺疾病的中医辅助诊疗方法和系统
CN116631612A (zh) * 2023-06-09 2023-08-22 广东工业大学 一种基于多图融合的图卷积草药推荐方法及计算机
CN116631612B (zh) * 2023-06-09 2024-03-19 广东工业大学 一种基于多图融合的图卷积草药推荐方法及计算机
CN116821779A (zh) * 2023-08-31 2023-09-29 中南大学湘雅医院 一种针对胃肠健康的大数据识别方法
CN116821779B (zh) * 2023-08-31 2023-11-14 中南大学湘雅医院 一种针对胃肠健康的大数据识别方法

Also Published As

Publication number Publication date
CN110459321A (zh) 2019-11-15
CN110459321B (zh) 2020-10-23
WO2021032220A3 (zh) 2021-04-15

Similar Documents

Publication Publication Date Title
WO2021032220A2 (zh) 一种基于证素的中医辅助诊断系统
US20190035506A1 (en) Intelligent auxiliary diagnosis method, system and machine-readable medium thereof
Okon et al. Natural language processing of Reddit data to evaluate dermatology patient experiences and therapeutics
US20180108443A1 (en) Apparatus and method for analyzing natural language medical text and generating a medical knowledge graph representing the natural language medical text
Masarie Jr et al. An interlingua for electronic interchange of medical information: using frames to map between clinical vocabularies
CN109299239A (zh) 一种基于es的电子病历检索方法
CN112687397B (zh) 罕见病知识库的处理方法及装置、可读存储介质
CN107578818B (zh) 一种基于深度学习的辅助开方方法及装置
CN110838368A (zh) 一种基于中医临床知识图谱的机器人主动问诊方法
CN107578797A (zh) 一种基于云计算的健康诊疗方案推荐的方法及其系统
Wang et al. Information needs mining of COVID-19 in Chinese online health communities
CN116013534A (zh) 一种基于医学指南和数据的临床辅助决策方法及系统
Travers et al. Using nurses’ natural language entries to build a concept-oriented terminology for patients’ chief complaints in the emergency department
Xiu et al. Construction of a digestive system tumor knowledge graph based on chinese electronic medical records: development and usability study
Ju et al. Doctor recommendation model based on ontology characteristics and disease text mining perspective
WO2022229964A1 (en) Method of generating a diseases database, usage of the diseases database, and system therefor
CN113064960A (zh) 一种精确搜索与患者病情相似病例的方法
Li et al. Personalized intelligent syndrome differentiation guided by TCM consultation philosophy
Yu et al. Dataset and enhanced model for eligibility criteria-to-sql semantic parsing
Nguyen et al. Mimic-iv-icd: A new benchmark for extreme multilabel classification
Galitsky Multi-case-based reasoning by syntactic-semantic alignment and discourse analysis
Yin et al. Study on construction and application of knowledge graph of TCM diagnosis and treatment of viral hepatitis B
CN115171880A (zh) 一种融合症状权重的疾病推荐系统
Varshney et al. Cdialog: A multi-turn covid-19 conversation dataset for entity-aware dialog generation
Li et al. Medical text entity recognition based on CRF and joint entity

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: 20854105

Country of ref document: EP

Kind code of ref document: A2

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20854105

Country of ref document: EP

Kind code of ref document: A2

122 Ep: pct application non-entry in european phase

Ref document number: 20854105

Country of ref document: EP

Kind code of ref document: A2

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 22/09/2022)

122 Ep: pct application non-entry in european phase

Ref document number: 20854105

Country of ref document: EP

Kind code of ref document: A2