JP2008108199A - Data base system, program, and information retrieval method in data base system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a data base system capable of easily providing knowledge regarding medical care corresponding to symptoms. <P>SOLUTION: In a storage part 101, a related phrase and disease condition combination information 111 in which a plurality of disease condition related phrases related to disease conditions are combined with each disease condition phrases included in a plurality of disease condition phrases indicating the disease conditions respectively is stored. In response to information input by a user, symptom information concerning a sentence in which the symptoms are described is accepted by a server control part 102. Next, two or more combinations of the disease condition related phrases having predetermined relation from the symptom information are recognized by predetermined language analysis processing by a natural sentence analysis part 122. Then, one or more disease condition phrases corresponding to the two or more combinations of the disease related phrases recognized by the natural sentence analysis part 122 are detected from the related phrase and disease condition combination information 111 by a disease condition detection part 124. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a database system.

  Social backgrounds surrounding medical care, such as an increase in medical expenses due to aging, changes in disease structure, calculation of medical expenses based on Diagnosis Procedure Combination (DPC) method, reduction of hospital stays, and introduction of long-term care insurance As a result, the need for home nursing / care is increasing.

  However, there are problems such as lack of medical knowledge of home caregivers and their families, lack of knowledge provision system for nursing / nursing at home.

  Accordingly, various techniques relating to medical database systems have been proposed (for example, Patent Documents 1 to 3). For example, Patent Document 1 proposes a technique for decomposing a natural sentence such as a finding sentence into words and phrases to be used as components of a database, and Patent Document 2 proposes a technique for inferring a medical condition from a patient inquiry result. Patent Document 3 proposes a technique for detecting a disease name by searching a database in which a relationship between a disease name and a symptom is described using words related to the symptom.

JP 2000-298696 A JP-A-6-176007 JP 2000-123098 A

  However, Patent Document 1 proposes a technique for constructing a database, but does not particularly mention the contents that easily provide information desired by a user using a search or the like. In the technique according to Patent Document 2, it is necessary to input numerical values and inspection results as keywords for a large number of predetermined items, and the degree of freedom of expressions and phrases that can be input as keywords is low, and fluctuations in expressions and phrases It was not compatible with. Furthermore, in the technique according to Patent Document 3, it is necessary to select a fixed term as a keyword, and it is difficult for an ordinary person who does not have basic knowledge about medical care to select a keyword.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a database system that can easily provide medical knowledge corresponding to a symptom.

  In order to solve the above-mentioned problem, the invention of claim 1 is a database system, wherein a plurality of pathological terms related to a pathological condition are included for each pathological phrase included in a plurality of pathological phrases each indicating a pathological condition. From the symptom information, a storage means for storing word combination information combined with each other, a symptom reception means for receiving symptom information related to a sentence describing symptoms in response to information input by a user, and a predetermined language analysis process Recognizing means for recognizing a combination of two or more medical condition related phrases having a predetermined relationship; and a medical condition detecting means for detecting one or more medical condition phrases corresponding to the combination of the two or more medical condition related phrases from the phrase combination information. It is characterized by having.

  The invention of claim 2 is the database system according to claim 1, wherein the acquisition means for acquiring information on a plurality of disease state related phrases by machine learning using text data provided from an external device, and the recognition means The combination of the two or more medical condition related phrases is recognized using the information on the plurality of medical condition related phrases acquired by the acquiring means.

  A third aspect of the present invention is the database system according to the first or second aspect, wherein the predetermined relationship includes a relationship between a subject and a predicate.

  A fourth aspect of the present invention is the database system according to any one of the first to third aspects, wherein the storage means is different from the predetermined representative phrase for the predetermined representative phrase and has the same meaning. At least one of information associated with one or more synonyms having the above and information associated with one or more similar words having a meaning different from the predetermined representative word and similar to the predetermined representative word The thesaurus information is stored, and the disease state detection means uses the thesaurus information to replace at least one disease state related phrase of the two or more disease state related phrases with a corresponding representative word, One or more medical condition words combined with the combination of two or more medical condition-related words after the replacement are detected in the phrase combination information.

  The invention according to claim 5 is the database system according to any one of claims 1 to 4, wherein the phrase combination information is one or more for each disease name included in a plurality of disease names. A disease name detection means for detecting a disease name corresponding to one or more disease state words detected by the disease state detection means from the disease state combination information, the disease state combination information including the disease state combination information And an output means for outputting the disease name detected by the disease name detection means in a manner that can be perceived by the user.

  The invention according to claim 6 is the database system according to claim 5, wherein the disease-pathology combination information includes information obtained by combining two or more disease-phrases for one disease name, and the disease name detection means The one disease name is detected when one or more disease state words detected by the disease state detection means match at least a part of the two or more disease state words. .

  The invention of claim 7 is the database system according to claim 6, wherein the disease name detection means detects one or more disease state words detected by the disease state detection means among the two or more disease condition words. The at least one disease name is detected when it matches at least a predetermined proportion or more of a disease state phrase.

  The invention according to claim 8 is the database system according to any one of claims 5 to 7, wherein the phrase combination information includes a combination of one or more countermeasures for each disease name. Including combination information, wherein the database system further comprises coping method detection means for detecting one or more coping methods corresponding to the disease names detected by the disease name detection means from the disease coping information. However, the coping method detected by the coping method detection means is output in a manner that can be perceived by the user.

  The invention according to claim 9 is the database system according to any one of claims 1 to 4, wherein the output is one in which a user can perceive one or more medical condition words detected by the medical condition detection means. And an information updating means for adding, in response to an input operation by the user, information obtained by combining the one or more medical condition words output by the output means, a coping method, and result information to the word combination information. Is further provided.

  The invention of claim 10 is the database system according to any one of claims 5 to 8, wherein the disease coping information includes at least one personal information and at least one information for each disease name. The database system further includes personal information accepting means for accepting one or more personal information in response to an input operation by a user, and the coping method detecting means includes the disease One or more coping methods combined with both the disease name detected by the disease name detecting unit and the one or more personal information received by the personal information receiving unit are detected from the coping information. To do.

  The invention of claim 11 is a program for causing the database system to function as the database system according to any one of claims 1 to 10 by being executed by a computer included in the database system.

  The invention of claim 12 is an information search method in a database system, wherein (a) a plurality of disease state related phrases related to a disease state are respectively provided for each disease state phrase included in a plurality of disease state words each indicating a disease state. The step of storing the combined word / phrase combination information in the storage unit, (b) in response to information input by the user, receiving the symptom information related to the sentence describing the symptom, (c) by a predetermined language analysis process, Recognizing a combination of two or more medical condition-related words having a predetermined relationship from the symptom information; and (d) detecting one or more medical condition words corresponding to the combination of the two or more medical condition-related words from the word combination information And a step of performing.

  In this specification, the word “phrase” is used to mean a phrase including a word, a symbol, a code, and a set of a plurality of words.

  According to the invention described in any one of the first to twelfth aspects, the medical condition corresponding to the symptom can be easily provided because the corresponding medical condition is detected simply by inputting the symptom by sentence.

  According to the invention described in claim 2, by enhancing information indicating a word related to a medical condition by machine learning, the accuracy of recognizing the word related to the medical condition from a sentence indicating a symptom can be improved, and the result In particular, the detection accuracy of a medical condition can be improved.

  According to the invention described in claim 3, by detecting a combination of a subject and a predicate corresponding to a phrase related to a medical condition from a sentence showing a symptom, and detecting the medical condition from the combination of the words, the detection accuracy of the medical condition is further improved. Can be increased.

  According to the invention described in claim 4, by detecting a medical condition after replacing a synonym phrase or a synonym phrase recognized from a sentence showing a symptom with a corresponding representative phrase, Since the medical condition can be detected while coping with the difference, as a result, it is possible to reduce the amount of data to be prepared in the database and the storage capacity of the storage medium.

  According to the invention described in any one of claims 5 to 7, the user can know the name of the disease corresponding to the symptom simply by inputting the symptom using a sentence.

  In particular, according to any of the inventions according to claims 6 and 7, even if the user does not input enough symptoms, the user can know the name of the disease corresponding to the symptoms.

  According to the invention described in any one of claims 8 to 10, the user can know a disease name corresponding to the symptom and a coping method only by inputting the symptom with a sentence.

  According to the invention described in claim 9, the information stored in the phrase combination information is evolved by adding, to the phrase combination information, information that is a combination of new countermeasures and results for the detected medical condition. Can do.

  According to the tenth aspect of the present invention, the user can know a coping method tailored to the patient such as age and sex.

  According to the eleventh aspect of the present invention, the same effects as those of the first to tenth aspects of the invention can be obtained.

  According to the invention of the twelfth aspect, the same effect as that of the invention of the first aspect can be obtained.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

<Outline of medical information system>
FIG. 1 is a diagram illustrating a schematic configuration of a medical information system 1 according to an embodiment of the present invention.

  As shown in FIG. 1, the medical information system 1 includes a server 100 and terminals 10 to 40, and the server 100 and the terminals 10 to 40 exchange data with each other via a network line NT such as the Internet. It is connected so that it can send and receive. Further, a terminal (management terminal) 190 for managing the medical information system 1 is connected to the server 100 so that data can be transmitted and received.

  The server 100 includes a database 110 (hereinafter referred to as “medical information DB”) that stores various types of information for supporting nursing and nursing care. This server 100 provides the terminals 10-40 with information for supporting nursing and nursing care in response to requests from the terminals 10-40. That is, the medical information system 1 functions as a database system that provides information stored in the medical information DB 110 in response to a user request.

<Functional configuration of medical information system>
FIG. 2 is a block diagram showing a functional configuration of the medical information system 1. Since the functions of the terminals 10 to 40 are the same, the terminals 10 to 40 are collectively shown in FIG. 2 for the purpose of preventing the diagram from being complicated.

○ Terminals 10 to 40:
The terminals 10 to 40 are, for example, personal computers (personal computers) used in general homes. The terminals 10 to 40 may be portable terminals (PDAs) or the like. The terminals 10 to 40 mainly include a storage unit 11, a terminal control unit 12, a display unit 13, an operation unit 14, and an interface (I / F) unit 15.

  The storage unit 11 is configured by a hard disk or the like, for example, and stores a program PG1 for realizing the functions of the medical information system 1.

  The terminal control unit 12 includes a CPU and a memory, and implements various functions and operations based on various signals from the operation unit 14 and the program PG1 stored in the storage unit 11. By the terminal control unit 12 reading and executing the program PG1 stored in the storage unit 11, various functions and operations of the medical information system 1 are realized. Various data appropriately generated in the process of information processing in the terminal control unit 12 are temporarily stored in the memory.

  The display unit 13 is configured by an output device such as a CRT or a liquid crystal panel, and visibly outputs various images in response to signals from the terminal control unit 12.

  The operation unit 14 includes, for example, a keyboard, a mouse, and the like, and responds to various operations performed by the user of the terminals 10 to 40 (such as the patient himself or a caregiver) according to various operations corresponding to the operation of the terminal control unit 12. Send a signal.

  The I / F unit 15 is an interface for controlling transmission / reception of data to / from the server 100 or the like via the network line NT.

○ Management terminal 190:
The management terminal 190 is configured by, for example, a personal computer. The management terminal 190 mainly includes a storage unit 191, a PC control unit 192, a display unit 193, an operation unit 194, and an interface (I / F) unit 195.

  The storage unit 191 is configured by, for example, a hard disk or the like, and stores a program PG2 for realizing the functions of the medical information system 1.

  The PC control unit 192 includes a CPU and a memory, and realizes various functions and operations based on various signals from the operation unit 194, the program PG2 stored in the storage unit 191, and the like. Various functions of the medical information system 1 such as a function for controlling display of various images on the display unit 193 by reading and executing the program PG2 stored in the storage unit 191 by the PC control unit 192. And the operation is realized. Note that various data appropriately generated in the process of information processing in the PC control unit 192 is temporarily stored in the memory.

  The display unit 193 is configured by, for example, a CRT or a liquid crystal panel, and outputs various images in response to a signal from the PC control unit 192.

  The operation unit 194 includes, for example, a keyboard and a mouse. When the operation unit 194 is appropriately operated by a user of the management terminal 190 (that is, an administrator), various signals corresponding to the operation are sent to the PC control unit 192. Sent.

  The I / F unit 195 is an interface for controlling data transmission / reception with the server 100.

○ Server 100:
The server 100 mainly includes a storage unit 101, a server control unit 102, and an interface (I / F) unit 105.

  The storage unit 101 is configured by a hard disk or the like, for example, and stores a program PG3 for realizing the function of the server 100. The storage unit 101 also includes data (hereinafter also referred to as “analysis data”) 101a for enabling language analysis, which will be described later, information related to a so-called thesaurus (hereinafter also referred to as “thesaurus information”) 101b, and medical information. DB 110 is stored. The storage unit 101 also stores various data for realizing the functions of the medical information system 1.

  The medical information DB 110 stores related word / phrase pathological condition combination information 111 and pathological condition / disease countermeasure combination information 112.

  The analysis data 101a includes data necessary for so-called morpheme analysis (hereinafter referred to as “morpheme analysis data”) and data necessary for dependency analysis (hereinafter referred to as “dependency analysis data”). The analysis data 101a includes data (hereinafter referred to as “keyword data”) in which various keywords such as words (terms and phrases) related to a medical condition are collected.

  For example, morphological analysis data and dependency analysis data are input in advance from the management terminal 190 and stored in the storage unit 101. The keyword data is stored in the storage unit 101 by performing machine learning using teacher data input from the management terminal 190 to the server 100. The keyword data may be partly or wholly input from the management terminal 190 and stored in a simple phrase list format.

  The thesaurus information 101b is information based on medical knowledge, and a set (synonym phrase group) of phrases (synonym phrases) composed of a plurality of synonyms and phrases having the same meaning (synonym phrases) and the synonym phrases. Consists of information (hereinafter also referred to as “synonym representative word related information”) associated with a representative word (representative word) of the group, a plurality of synonyms and phrases having similar meanings (synonymous phrases) Includes a large number of pieces of information (hereinafter also referred to as “synonym representative word related information”) in which a set (synonymous phrase group) of words (synonymous phrase) and a representative word (representative phrase) in the synonym group are associated with each other It is out.

  That is, the thesaurus information 101b includes one or more data elements indicating a predetermined representative phrase (hereinafter referred to as “representative phrase element” or “representative phrase” for short) having the same meaning as the predetermined representative phrase. Information associated with one or more data elements (hereinafter also referred to as “synonym phrase elements” or abbreviated as “synonym phrases” for short) and each data element (representative phrase element) indicating a predetermined representative phrase Among information associated with one or more data elements (hereinafter also referred to as “similar phrase elements” or “similar phrases” for short) indicating one or more similar phrases having a similar meaning to the predetermined representative phrase At least one of them.

  In the thesaurus information 101b, for example, the representative phrase “head” is associated with the synonym phrase group “head” “head” “head” “temporal head”. Note that the information included in the thesaurus information 101b is not limited to information related to phrases such as nouns, and may include any phrases such as phrases used as other parts of speech such as adjectives, phrases used as predicates, and the like. For example, synonymous phrases of predicates include not only completely different phrases but also phrases with different utilization forms. Further, when an internal term is used in a specific hospital, information relating a phrase generally used as a representative phrase for the internal term to the internal term may be added to the thesaurus information 101b. .

  The thesaurus information 101b may be stored in the storage unit 101 through machine learning using teacher data input from the management terminal 190 to the server 100, or a part or all of the thesaurus information 101b may be input from the management terminal 190. And may be stored.

  In the related word / phrase condition combination information 111, a plurality of words / phrases related to a medical condition (hereinafter referred to as “word / phrase related to medical condition”) are also referred to as “disease related words / phrases” with respect to the word / phrase indicating the medical condition (hereinafter also referred to as “pathological condition / phrase”). ) Is a large number of accumulated data. In other words, the related word / phrase combination information 111 corresponds to each disease / phrase element included in a plurality of data elements (hereinafter also referred to as “disease word / phrase element” or abbreviated as “disease word / phrase”) respectively. The information is a combination of a plurality of data elements (hereinafter also referred to as “disease related words / phrases” or abbreviated as “disease related words / phrases”).

  FIG. 3 is a diagram showing a specific example of the data content included in the related phrase disease state combination information 111. In FIG. 3, two pathology-related phrases “head” and “pain” are associated with the pathology phrase “headache”, and two pathology-related phrases “fever” “high or raised” with respect to the pathology phrase “fever”. Are associated, and the two disease state related phrases “blood pressure” and “high or increased” are associated with the disease state phrase “blood pressure increase”. “High or raised” indicates that “high” or “raised” is exclusively selected.

  The pathological condition coping information 112 stores a large number of data obtained by combining at least one pathological phrase and an appropriate coping method with respect to a phrase indicating a disease name (hereinafter also referred to as “disease name phrase”). In other words, the pathological condition / disease handling combination information 112 corresponds to each disease name element included in a plurality of data elements (hereinafter also referred to as “disease name element” or “disease name” for short) respectively indicating a plurality of disease name phrases. 1 or more data elements (hereinafter referred to as “coping”), each of which includes information combining one or more medical condition phrase elements (hereinafter also referred to as “disease medical condition combination information”) and indicating a coping method for each disease name element. It also includes information (hereinafter also referred to as “disease handling combination information”) in combination with “method element” and also referred to as “coping method” for short.

  FIG. 4 is a diagram showing a specific example of the data contents included in the pathological condition / disease handling combination information 112. FIG. 4 shows a state in which one disease name phrase “influenza” and one coping method “administer XX” are associated with three disease state phrases “fever, malaise, chills”.

  The server control unit 102 includes a CPU and a memory, and is a part that controls operations related to the medical information system 1 based on various signals from the management terminal 190 and the terminals 10 to 40, the program PG3, and the like. Here, for example, the server control unit 102 reads and executes the program PG3 stored in the storage unit 101 based on a signal from the management terminal 190, thereby realizing various functions and operations related to the medical information system 1. The Note that various data appropriately generated in the process of information processing in the server control unit 102 is temporarily stored in the memory.

  FIG. 2 physically shows a functional configuration realized by the server control unit 102 reading and executing the program PG3 by the CPU. The server control unit 102 includes a machine learning unit 121, a natural sentence analysis unit 122, a phrase replacement unit 123, a disease state detection unit 124, a disease detection unit 125, a countermeasure detection unit 126, and a data update unit 127 as functional configurations.

  The machine learning unit 121 executes machine learning using teacher data given from an external device, thereby acquiring data elements (disease state-related phrase elements) related to a plurality of disease state-related words and phrases in the storage unit 101. The keyword data of the analysis data 101a is constructed. Here, machine learning will be described.

  For example, a learning corpus or the like is given as teacher data. The learning corpus includes text data corresponding to a large amount of sentences related to a medical condition. In this text data, information indicating that each disease state related phrase is a disease state related phrase is tagged. For example, in a sentence “blood pressure is rising”, information indicating that the phrase “blood pressure” “rising” is a disease state related phrase is tagged.

  Then, the machine learning unit 121 extracts disease state related words from the learning corpus, creates data in a list format in which the disease state related words are listed (hereinafter also referred to as “disease state related word list data”), and stores the data in the storage unit 101. It is stored as keyword data in the analysis data 101a. Examples of the pathology-related words include “site”, “number of times”, “frequency” in which symptoms occur, “state” related to a certain symptom, “parameter” corresponding to an index indicating the degree of a certain symptom, and the like. .

  However, if the above machine learning method is used, only the disease state related words / phrases previously tagged with the information indicating the disease state related words / phrases in the teacher data can be identified, but the following machine learning method is used. Thus, a word / phrase that has not been tagged in advance can also be identified as a disease state related word / phrase.

  For example, the machine learning function of the machine learning unit 121 uses the morpheme analysis data included in the analysis data 101a to decompose the text data constituting the learning corpus into morphemes by morpheme analysis, and for each morpheme, the morpheme itself The morpheme part-of-speech, the morpheme inflection form, the information of morphemes before and after (for example, two before and after), etc. are used to learn the pattern in which the morpheme related to the medical condition related phrase appears. Then, the machine learning unit 121 can recognize a word / phrase that has not been tagged in advance as a medical condition-related word / phrase according to the pattern.

More specifically, for example, when the machine learning unit 121 frequently produces sentences such as “** is rising” in the learning corpus (a variety of words are entered in the “**” portion). When a medical condition-related phrase frequently appears in the “**” part, a pattern in which a medical condition-related phrase enters the “**” part can be learned. If such a pattern is used, the machine learning unit 121 uses, for example, the phrase “blood pressure / is rising” from the phrase “blood pressure / high” from the context of the word, and treats “blood pressure” as a medical condition. It can be extracted as a related phrase. The machine learning can be realized by using a so-called SVM (Support Vector Machine). As a method of extracting disease state-related phrases from a learning corpus using SVM, for example, a program “Yamcha” (http://chasen.org/ ^〜 taku / software / cabocha / Yamcha /) developed at Nara Institute of Science and Technology For example).

  The natural sentence analysis unit 122 performs predetermined language analysis processing including so-called dependency analysis on information (hereinafter also referred to as “symptom information”) of a sentence (natural sentence) in which a symptom is described. Recognize combinations of two or more disease-related phrases having the relationship The symptom information is input by the user appropriately operating the operation unit 14 of the terminals 10 to 40 and is received by the natural sentence analysis unit 122. Further, the “predetermined language analysis process” mentioned here includes recognition and dependency analysis of disease state related phrases using keyword data included in the analysis data 101a, and the “predetermined relationship” includes, for example, This includes the relationship between the subject and predicate, and the relationship between modification and modification.

  FIG. 5 is a diagram for explaining processing in the natural sentence analysis unit 122. In FIG. 5, each medical condition-related phrase is underlined, and arrows indicating the relationship between the subject and the predicate and the relationship between the modifier and the modified word are shown.

  For example, for the sentence “head hurts”, a combination of two medical condition-related words “head hurts” and “pain” having a relationship between the subject and the predicate is recognized. In addition, for the sentence “Highly fever”, there are three medical conditions related to the relationship between the subject and the predicate, and further the relationship between the modifier and the modified word (here, the relationship between the adjective corresponding to the predicate and the adverb indicating the degree). Combinations of the phrases “hot”, “pretty” and “high” are recognized.

  Also, for the sentence “blood pressure is higher than yesterday”, there are four pathology-related phrases “blood pressure”, “yesterday”, “comparison” that have a relationship between the subject and predicate and a modifier and a modified word. "" Up "combination is recognized. Furthermore, for the sentence “Many coughs due to bronchitis”, the four pathology-related phrases “bronchitis” “for” “cough” have the relationship between the subject and predicate and the relationship between the modifier and the modified word. “Many” combinations are recognized.

As a method for realizing the dependency analysis, for example, the program “Cabocha” (see http://chasen.org/ ^〜 taku / software / cabocha /) developed by Nara Institute of Science and Technology is used. A method is mentioned.

  The phrase replacement unit 123 uses the thesaurus information 101b to replace the disease state related words such as synonyms and synonyms with the representative words for the disease state related words recognized by the natural sentence analysis unit 122. By replacing many synonyms and synonyms that exist here with representative phrases, the related phrase pathology combination information 111 only needs to store combinations of phrases for the representative phrases, so the related phrase pathology combination information 111 The amount of information can be reduced, and it is possible to effectively use the load related to generation of the related phrase / phrase combination information 111 and the storage capacity of the storage unit 101.

  Specifically, the phrase replacement unit 123 performs a process of replacing the phrases “head”, “head”, and “head” with the representative phrase “head”, for example. In addition, for example, a phrase having various utilization forms for “rising” is replaced with the representative phrase “rising”. More specifically, for a medical condition related phrase “rising up” corresponding to a predicate, if the related phrase medical condition combination information 111 is registered as a representative phrase “rising up” in the form of an end form without an auxiliary verb or the like, a natural sentence When the analysis unit 122 recognizes the pathology related phrases “raised” and “raised” with the auxiliary verb, the representative phrase “raised” is replaced. This replacement is realized by a form in which the morpheme “up” is recognized by the morphological analysis and is replaced with the representative phrase “up” corresponding to the word “up”.

  The disease state detection unit 124 detects one or more disease state words / phrases corresponding to a combination of two or more disease state related words / phrases recognized by the natural sentence analysis unit 122 from the related word / phrase state combination information 111. Specifically, in the related word / phrase condition combination information 111, a disease state / phrase combined with a combination of two or more disease state related words recognized by the natural sentence analysis unit 122 and appropriately replaced by the word / phrase replacement unit 123 is detected. . When the data content as shown in FIG. 3 is included in the related phrase / phrase / symptom combination information 111, for example, the combination of the pathology-related phrase “head” and “pain” corresponds to the combination of the pathology-related phrase. The medical condition phrase “headache” is detected.

  The disease detection unit 125 detects a disease name corresponding to one or more disease state words detected by the disease state detection unit 124 from the disease state / disease handling combination information 112. When the data contents as shown in FIG. 4 are included in the pathological condition / disease coping information 112, for example, for the three pathological phrases “fever”, “fatigue”, “chill”, the combination of the pathological terms The disease name “influenza” corresponding to is detected.

  The coping detection unit 126 detects one or more coping methods corresponding to the disease name detected by the disease detection unit 125 from the pathological condition coping countermeasure information 112. When the data content as shown in FIG. 4 is included in the disease state / disease countermeasure combination information 112, for example, for the disease name “influenza”, a countermeasure method “administer XX” corresponding to the disease name. Is detected.

  The one or more medical condition phrases recognized by the medical condition detection unit 124, the disease name detected by the disease detection unit 125, and the one or more countermeasures detected by the treatment detection unit 126 are transferred to the terminal control unit 12 of the terminal 10. Then, it is visually output from the display unit 13 as appropriate under the control of the terminal control unit 12.

  On the other hand, when one or more disease state phrases are recognized by the disease state detection unit 124 but no disease name is detected by the disease detection unit 125, the disease state phrase detected by the disease state detection unit 124 is the terminal control of the terminal 10. The data is transferred to the unit 12 and output visually on the display unit 13 under the control of the terminal control unit 12. This visual output is, for example, a screen (hereinafter referred to as “provisional information”) that allows a user to input (information provision) information indicating a coping method corresponding to a medical condition phrase and information (hereinafter also referred to as “result information”) indicating a result of the coping method. It is also referred to as an “information provision screen”. A specific example of the information providing screen will be described later.

  In response to an input operation of the operation unit 14 by the user, the data update unit 127 combines the new countermeasure method and the result information with respect to the disease state phrase and adds them to the disease state disease countermeasure combination information 112. For example, when the information providing screen is visibly output on the display unit 13, the user operates the operation unit 14 in various ways, so that a new condition corresponding to the medical condition phrase visually output on the information providing screen is displayed. When the coping method and the result information are input, information obtained by combining these pathology phrases, the new coping method, and the result information is added to the pathological condition coping information 112.

  The I / F unit 105 is an interface for controlling transmission / reception of data via the network line NT to / from the terminals 10 to 40 and the like, and controlling transmission / reception of data to / from the management terminal 190.

  In the medical information system 1, the functions of the terminal control unit 12, the server control unit 102, and the PC control unit 192 cooperate to send and receive various data to each other, thereby providing medical information (hereinafter “ A process related to “medical support function” (hereinafter also referred to as “medical support process”) is realized.

<Operation of medical information system>
FIG. 6 is a flowchart showing an operation flow of medical support processing in the medical information system 1. 7 to 9 are diagrams exemplifying screens visually presented on the terminals 10 to 40 in this operation. Hereinafter, an example in which the user uses the medical information system 1 in the terminal 10 among the terminals 10 to 40 will be described.

  The operation flow shown in FIG. 6 is started when the function of the medical information system 1 using the terminal 10 is activated by the user operating the operation unit 14 in the terminal 10 in various ways, and proceeds to step S1.

  In step S <b> 1, a screen for inputting symptoms (hereinafter also referred to as “symptom input screen”) is displayed on the display unit 13 under the control of the terminal control unit 12.

  FIG. 7 is a diagram illustrating a symptom input screen G1. The symptom input screen G1 is provided with a column (hereinafter also referred to as “symptom entry column”) W1 in which a symptom can be written in a natural sentence in the approximate center of the screen, and when the user operates the keyboard included in the operation unit 14 in various ways. Information (symptom information) indicating symptoms composed of natural sentences is entered. Examples of the symptom information include information composed of natural sentences indicating symptoms such as “head ache”. When the symptom information is entered in the symptom entry column W1, when the search button SB1 is pressed by aligning the mouse pointer MP with the search button SB1 at the lower right of the screen and clicking the left button of the mouse, for example, the symptom entry column W1 is displayed. The entered symptom information can be input to instruct execution of search processing based on the symptom information. If the clear button CB1 provided at the lower right of the symptom input screen G1 is pressed with the mouse pointer MP, the symptom information entered in the symptom entry column W1 can be erased.

  In step S <b> 2, it is determined whether or not symptom information is input by the terminal control unit 12. Here, when the search button SB1 is pressed on the symptom input screen G1, it is determined that symptom information has been input, and the process proceeds to step S3. At this time, the symptom information is transmitted from the terminal control unit 12 to the server control unit 102. Here, the process of step S2 is repeated until the search button SB1 is pressed on the symptom input screen G1.

  In step S <b> 3, the symptom information transmitted from the terminal control unit 12 is received by the server control unit 102.

  In step S4, the natural sentence analysis unit 122 performs a predetermined language analysis process on the symptom information received in step S3. Here, a predetermined language analysis process is performed on the symptom information, thereby recognizing a combination of two or more disease state related phrases having a predetermined relationship (for example, a relationship between a subject and a predicate).

  In step S5, the phrase replacement unit 123 appropriately replaces two or more medical condition-related phrases recognized in step S4 with representative phrases while referring to the thesaurus information 101b. Specifically, for example, when all of the two or more medical condition-related phrases recognized in step S4 are representative phrases, the replacement with the representative phrase is not performed at all, and the two or more medical condition-related phrases recognized in step S4 are performed. If all of the phrases are not representative phrases, each disease state related phrase is replaced with a representative phrase. Further, for example, if some of the two or more medical condition-related phrases recognized in step S4 are not representative phrases, the partial medical condition-related phrases are replaced with the representative phrases. More specifically, for example, the medical condition related phrase “head” is replaced with the representative phrase “head”.

  In step S6, one or more medical condition phrases corresponding to the combination of two or more medical condition related phrases recognized in step S4 and appropriately replaced with representative words in step S5 are detected from the related phrase medical condition combination information 111. For example, one medical condition phrase “headache” corresponding to a combination of two medical condition related phrases “head” and “pain” is detected.

  In step S <b> 7, the disease name corresponding to the one or more disease state words detected in step S <b> 6 is detected from the disease state disease countermeasure combination information 112 by the disease detection unit 125. For example, a disease name “influenza” corresponding to three medical condition phrases “fever”, “malaise”, and “chill” is detected. Here, in the case where there is no disease name corresponding to one or more disease state words detected in step S6 in the disease state disease coping combination information 112, the disease name is not detected, and step S8 is performed. Proceed to

  In step S8, when a disease name is detected in step S7, the countermeasure detection unit 126 detects one or more countermeasures corresponding to the disease name detected in step S7 from the pathological condition / disease countermeasure combination information 112. . For example, the coping method “administer XX” corresponding to the disease name “influenza” is detected. On the other hand, when the disease name is not detected in step S7, the countermeasure detection unit 126 detects one or more countermeasures corresponding to the disease state phrase detected in step S6 from the disease state countermeasure combination information 112. Although the latter case will be described later, a coping method is detected when there is no coping method corresponding to the one or more medical condition words detected in step S6 in the coping information for coping with disease state disease 112. Instead, the process proceeds to step S9.

  In step S9, the server control unit 102 determines whether a coping method is detected in step S8. If a coping method is detected in step S8, the process proceeds to step S10, and if not detected, the process proceeds to step S14.

  In step S10, under the control of the server control unit 102 and the terminal control unit 12, the one or more medical condition words detected in step S6, the disease name detected in step S7, and the one or more coping methods detected in step S8 Is transferred from the server control unit 102 to the terminal control unit 12.

  In step S11, under the control of the terminal control unit 12, various information (one or more medical condition phrases, disease names, and one or more coping methods) transferred in step S10 are visibly output (hereinafter referred to as “search”). Is also displayed on the display unit 13.

  FIG. 8 is a diagram illustrating a search result screen G2. In the search result screen G2, a list TB1 that displays search results corresponding to the symptom information input by the user is arranged in the approximate center of the screen. The list TB1 is a visual output of one or more disease states / phrases detected in step S6, a disease name detected in step S7, and one or more countermeasures detected in step S8. The combination of the pathological phrases “fever”, “malaise”, “chill”, disease name “influenza” and coping method “administer XX” is visually output. By visually recognizing the search result screen G2, the user can grasp a medical condition, a disease name, and a coping method estimated from a patient's symptom.

  Therefore, the user can know the disease name corresponding to the symptom and the coping method corresponding to the symptom and the disease name simply by inputting the symptom in a natural sentence.

  In step S12, the terminal control unit 12 determines whether there is a next search. Here, when the next search button RB1 provided in the lower right part of the search result screen G2 is pressed with the mouse pointer MP, it returns to step S1 assuming that there is a next search, and the symptom input screen G1 is displayed on the display unit 13. Is displayed. On the other hand, if the next search button RB1 is not pressed with the mouse pointer MP, the process proceeds to step S13.

  In step S13, the terminal control unit 12 determines whether there is an end instruction. Here, when the end button FB1 provided at the lower right part of the search result screen G2 is pressed with the mouse pointer MP, the operation flow is ended as if the end instruction has been given. On the other hand, if the end button FB1 is not pressed with the mouse pointer MP, the process returns to step S12. That is, if neither the next search button RB1 or the end button FB1 provided on the search result screen G2 is pressed, the processes in steps S12 and S13 are repeated.

  In step S <b> 14, one or more medical conditions detected in step S <b> 6 are transferred from the server control unit 102 to the terminal control unit 12 under the control of the server control unit 102 and the terminal control unit 12.

  In step S15, under the control of the terminal control unit 12, one or more medical condition words transferred in step S14 are visibly output, and an information provision screen for receiving input of a new coping method and result information is displayed on the display unit 13. Is displayed.

  FIG. 9 is a diagram illustrating an information providing screen G3. In the information provision screen G3, a sentence LT indicating that a disease name and an appropriate coping method could not be detected is arranged at the top of the screen, and a search result display corresponding to the symptom information input by the user is displayed in the approximate center of the screen. A list TB2 having a column in which new information can be entered is arranged.

  In the list TB2, one or more pathological phrases detected in step S6 are visibly output, but the disease names and countermeasures that have not been detected are output in a blank EP state. Further, in the list table TB2, a column (entry column) NE1 for entering a coping method newly performed by the user and a column (entry column) NE2 for entering information (result information) indicating the result of executing the coping method are provided. Is provided.

  FIG. 9 shows an example in which the coping method “rest” is entered in the entry field NE1, and the result information “◯” is entered in the entry field NE2. When the clear button CB2 provided at the lower right portion of the information provision screen G3 is pressed with the mouse pointer MP, the contents entered in the entry fields NE1 and NE2 are erased.

  In step S16, it is determined whether to return to the symptom input screen G1 under the control of the terminal control unit 12. Here, when the return button RB2 provided at the lower right portion of the information provision screen G3 is pressed with the mouse pointer MP, the process returns to step S1, and the symptom input screen G1 is displayed on the display unit 13. On the other hand, if the return button RB2 is not pressed with the mouse pointer MP, the process proceeds to step S17.

  In step S17, it is determined whether new combination information has been input under the control of the terminal control unit 12. Here, in the information provision screen G3, when information is entered in the entry fields NE1 and NE2 of the list table TB2, when the input button EB1 at the lower right of the screen is pressed, new combination information is input. Proceed to step S18. At this time, the combination of the disease state phrase displayed in the list TB2 and the new coping method and the result information is input as “new combination information”, and the new combination is sent from the terminal control unit 12 to the server control unit 102. Information is sent. On the other hand, if the input button EB1 at the lower right of the screen is not pressed, the process returns to step S16. That is, if neither the return button RB2 or the input button EB1 provided on the information provision screen G3 is pressed, the processes in steps S16 and S17 are repeated.

  In step S <b> 18, new combination information transmitted from the terminal control unit 12 is received by the server control unit 102.

  In step S19, the data update unit 127 adds the new combination information received in step S18 to the symptom-disease countermeasure combination information 112, and the operation flow ends.

  FIG. 10 is a diagram illustrating data contents included in the symptom-disease handling combination information 112 after new combination information is added to the symptom-disease handling combination information 112. In FIG. 10, the coping method “rest” and the result information “O” are combined with two pathological phrases “fever” and “headache”, and the coping method “**” for the pathological phrases “vomiting” and “flank pain”. ** (various drug names are included in **) "and result information" x "are described in combination.

  In this way, new combination information obtained by combining a new coping method and result information with respect to the detected medical condition phrase is added to the symptom-disease coping information 112 and stored in the medical information DB 110. Evolution of the information that has been made.

  Then, if data as shown in FIG. 10 is stored in the symptom-disease coping information 112, a disease name is not detected in step S7, but a coping method is detected in step S8. Note that it is preferable that the handling method combined with the result information “×” is not detected in step S8 from the viewpoint of visually presenting an effective handling method to the user.

  As described above, in the medical information system 1 according to the embodiment of the present invention, a medical condition phrase corresponding to a symptom is detected only by the user inputting the patient's symptom in a natural sentence on the terminals 10 to 40. For this reason, the knowledge regarding the medical treatment corresponding to symptoms, such as a medical condition, can be easily provided with respect to a user.

  In addition, the machine learning unit 121 can enrich the disease-related words and phrases constituting the keyword data included in the analysis data 101a by machine learning using teacher data. For this reason, the recognition accuracy of a disease state-related phrase can be increased from a natural sentence indicating a symptom, and as a result, the detection accuracy of a disease state is increased.

  The natural sentence analysis unit 122 recognizes a subject / predicate combination that is a disease state related phrase from a natural sentence indicating a symptom, and the disease state detection unit 124 detects a disease state phrase from the combination of the words. For this reason, compared to the case of detecting a medical condition by keyword search using a database in which one term is associated with a medical condition, a process for detecting a medical condition phrase further combined with a combination of words is performed. Therefore, the detection accuracy of the medical condition is further increased. Even if compared with a configuration in which a disease state is detected by keyword search using a database in which a disease state is associated with a large number of phrases and sentences indicating symptoms, the amount of data for constructing the database becomes enormous. The trouble can be avoided. That is, it is possible to reduce the work required for constructing the database, effectively use the storage capacity of the storage medium by reducing the amount of data, and reduce the storage capacity.

  The synonym phrase or synonym phrase recognized from the natural sentence indicating the symptom by the natural sentence analysis unit 122 is replaced with the corresponding representative phrase by the phrase replacement unit 123, and then the disease state detection unit 124 detects the disease state phrase. Therefore, it is possible to detect a disease state word / phrase while coping with fluctuations in expressions used for each user or group and differences in words / phrases. If a database that covers all fluctuations in expressions and differences in phrases is constructed, it is possible to avoid problems such as an enormous amount of data.

<Modification>
As mentioned above, although embodiment of this invention was described, this invention is not limited to the thing of the content demonstrated above.

  For example, in the above-described embodiment, various information (such as a disease state phrase, a disease name, and a coping method) is visibly output on the display unit 13. However, the present invention is not limited to this, and for example, various information is output by voice. Anything that outputs in a manner that can be perceived by the user may be used.

  In the above embodiment, the search processing is performed using only the combination of words recognized from the symptom information as a search condition. However, the present invention is not limited to this, and the handling method includes, for example, age, gender, medical history, etc. Since it differs depending on various personal information, it may be added to the search conditions as appropriate.

  Hereinafter, an example in which age information is added to a search condition will be described.

  FIG. 11 is a diagram illustrating a symptom input screen G1A according to a modification. The symptom input screen G1A is obtained by adding a pull-down list DL for inputting age information to the symptom input screen G1 shown in FIG. The user can select the patient's age in the list DL using the mouse pointer MP by variously operating the operation unit 14, and can input information on the age by pressing the search button SB1. Although only one type of personal information is input here, the personal information may be one or more types. In such an aspect, the server control unit 102 responds to an input operation to the operation unit 14 by the user. A data element indicating one or more personal information (hereinafter also referred to as “personal information element” or abbreviated as “personal information”) is accepted.

  FIG. 12 is a diagram illustrating a specific example of data contents included in the symptom-disease handling combination information 112 according to the modification. In FIG. 12, for example, three pathological phrases “fever”, “fatigue”, and “chill” are associated with age-related personal information “under 10 years old”, one disease name “influenza”, and one countermeasure “rest”. Combined with three pathological phrases “fever”, “malaise”, “chill”, personal information “over 10 years old”, one disease name “influenza” and one coping method “administer XX” The state is shown. In such an embodiment, the symptom-disease handling combination information 112 includes information in which one or more personal information and one or more handling methods are combined for each disease name.

  In addition, in the state combined with the personal information “under 10 years old” mentioned here, there are combinations of all 10 types of personal information, pathological phrases, disease names, and countermeasures combined with each of 0-9 years old. Is combined with personal information “10 years old or older”, for example, 111 types of personal information combined with each of 10 to 120 years old, medical condition phrases, disease names and coping It is synonymous with the state where the combination with the method exists.

  When the data contents as shown in FIG. 12 are included in the symptom-disease handling combination information 112, for example, three (generally, 1) detected by the disease state detection unit 124 from the related word / phrase state combination information 111 The coping method “combined with both of the medical condition phrases“ fever ”,“ fatigue ”,“ chill ”and one (generally one or more) personal information“ 48 years old ”received by the server control unit 102. "Administer XX" is detected from the symptom-disease coping information 112.

  As a result of such detection processing, a search result screen G2A as shown in FIG. In the search result screen G2A, compared with the search result screen G2 shown in FIG. 8, the list TB1 that displays the search results arranged in the approximate center of the screen displays the search results to which the personal information input by the user is added. It is a list TB1A to be displayed. In the list table TB1A, search results including combinations of the disease name “influenza” corresponding to the symptom information and personal information input by the user and the coping method “administer XX” are shown.

  According to such a configuration, the user can know a coping method in accordance with patient personal information such as age and sex.

  In the above embodiment, all the disease state words / phrases combined for one disease name in the disease / state disease handling combination information 112 completely match one or more disease state words / phrases detected by the disease state detection unit 124. In this case, the disease detection unit 125 detects the disease name corresponding to the one or more disease state phrases detected by the disease state detection unit 124 from the disease state disease countermeasure combination information 112, but is not limited thereto.

  Considering that it is difficult for the user to input symptom information that covers many symptoms, the user can know medical information such as the name of the disease corresponding to the symptom and how to deal with it even if the user does not input enough symptoms. It is preferable to adopt a configuration that can

  As such a configuration, for example, in the case where the combination of two or more disease states is included for each disease name in the combination treatment for disease and disease information 112, one disease name in the combination treatment for disease state and disease 112 When at least some of the pathological phrases combined with respect to one and one or more pathological phrases detected by the pathological condition detection unit 124 match, the disease detection unit 125 A configuration for detecting a disease name can be mentioned.

  However, for example, since the disease state “fever” is commonly expressed in many diseases, it is assumed that there are many disease names that are combined with the disease state phrase “fever” in the disease-state disease coping information 112. The In such a case, one pathological phrase out of a plurality of pathological phrases combined for each disease name in the pathological condition coping information 112 and one pathological phrase detected by the pathological condition detection unit 124 are obtained. If all the matching disease names are detected, it is assumed that there will be a problem that a large number of disease names are detected.

  Therefore, as a more preferable configuration, when information including a combination of two or more disease states is included for each disease name in the disease / state disease handling combination information 112, for each disease name in the disease state / reaction countermeasure combination information 112, Disease detection is performed when at least a predetermined ratio (for example, 2/3) or more of all of the combined medical condition words and one or more medical condition words detected by the medical condition detection unit 124 match. A configuration in which the unit 125 detects the one disease name is given. When such a configuration is adopted, for example, when the predetermined ratio is 2/3 and the data content as shown in FIG. 4 is included in the pathological condition coping information 112, three pathological phrases “fever” When the combination of two pathological phrases “fever” and “fatigue” among “fatigue” and “chill” is detected by the pathological condition detection unit 124, the two pathological phrases “fever” and “fatigue” The disease name “influenza” corresponding to the combination is detected.

  The predetermined ratio may be changed as appropriate according to the type and degree of fulfillment of the disease state words and phrases constituting the disease state disease combination combination information 112.

  When a plurality of disease names are detected by the disease detection unit 125, in order to narrow down the disease names to one, a question is presented to the user, and one disease name is selected according to the answer to the question. You may make it narrow down to. In order to make it possible to present such a question, for example, information related to a question for narrowing down a combination of a plurality of disease names may be stored in the storage unit 101 as appropriate.

  In the above embodiment, the medical information system 1 is realized by the server 100, the management terminal 190, and the terminals 10 to 40. However, the present invention is not limited thereto. You may make it implement | achieve the function of. At this time, the program for realizing the medical information system 1 can be in one form.

  In the above embodiment, the number of terminals 10 to 40 is four. However, the number of terminals 10 to 40 is not limited to this.

It is a figure which illustrates schematic structure of the medical information system which concerns on embodiment of this invention. It is a block diagram which shows the function structure of a medical information system. It is a figure which illustrates the data content contained in related word phrase medical condition combination information. It is a figure which illustrates the data contents contained in pathological condition disease handling combination information. It is a figure for demonstrating the process in a natural sentence analysis part. It is a figure which shows the operation | movement flow of a search process. It is a figure which illustrates a symptom input screen. It is a figure which illustrates a search result screen. It is a figure which illustrates an information provision screen. It is a figure which illustrates the disease-and-disease handling combination information after new combination information addition. It is a figure which illustrates the new information input screen which concerns on a modification. It is a figure which illustrates the pathological condition disease handling combination information which concerns on a modification. It is a figure which illustrates the search result screen concerning a modification.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Medical information system 10-40 Terminal 11,101,191 Storage part 12 Terminal control part 13 Display part 14 Operation part 100 Server 101a Analysis data 101b Thesaurus information 102 Server control part 110 Medical information database 111 Related phrase disease state combination information 112 Disease state Disease coping information 121 Machine learning unit 122 Natural sentence analysis unit 123 Phrase substitution unit 124 Disease state detection unit 125 Disease detection unit 126 Coping detection unit 127 Data update unit 190 Management terminal NT Network line PG1 to PG3 Program

Claims (12)

A database system,
Storage means for storing phrase combination information obtained by combining a plurality of disease state-related words / phrases related to a disease state for each disease state / phrase included in the plurality of disease state words / phrases respectively indicating a disease state;
In response to information input by the user, symptom reception means for receiving symptom information related to a sentence describing the symptom,
Recognizing means for recognizing a combination of two or more medical condition related phrases having a predetermined relationship from the symptom information by a predetermined language analysis process;
Pathological condition detecting means for detecting one or more pathological phrases corresponding to a combination of the two or more pathological condition-related phrases from the phrase combination information;
A database system characterized by comprising: The database system according to claim 1, wherein
An acquisition means for acquiring information on a plurality of medical condition-related phrases by machine learning using text data provided from an external device;
The recognition means is
A database system characterized by recognizing a combination of the two or more disease state related phrases using information of the plurality of disease state related phrases acquired by the acquisition means. The database system according to claim 1 or 2, wherein
The predetermined relationship is
A database system characterized by including the relationship between subject and predicate. The database system according to any one of claims 1 to 3, wherein
The storage means
Information associated with one or more synonym phrases that are different from the predetermined representative phrase and having the same meaning with respect to the predetermined representative phrase, and different from and similar to the predetermined representative phrase Storing thesaurus information for storing at least one piece of information associated with one or more similar words having the meaning of
The medical condition detection means
A combination of two or more disease state-related words after replacement in the word combination information after replacing at least one of the two or more disease state words / phrases with the corresponding representative word / phrase using the thesaurus information. A database system, characterized in that it detects one or more medical conditions that are combined with. The database system according to any one of claims 1 to 4, wherein
The word combination information is
For each disease name included in a plurality of disease names, including disease disease condition combination information combining one or more disease condition words,
The database system is
A disease name detection means for detecting a disease name corresponding to one or more disease state words detected by the disease condition detection means from the disease condition combination information;
An output means for outputting a disease name detected by the disease name detection means in a manner perceivable by a user;
A database system further comprising: The database system according to claim 5, wherein
The disease state combination information is
Contains information that combines two or more disease terms for one disease name,
The disease name detection means
The one disease name is detected when one or more medical condition words detected by the medical condition detection means match at least a part of the two or more medical condition words. . The database system according to claim 6, wherein
The disease name detection means
The database is characterized in that the one disease name is detected when one or more medical condition words detected by the medical condition detection means match at least a predetermined ratio of the two or more medical condition words. system. The database system according to any one of claims 5 to 7,
The word combination information is
For each disease name, including disease coping information that combines one or more coping methods,
The database system is
Coping method detection means for detecting one or more coping methods corresponding to the disease name detected by the disease name detection means from the disease coping combination information;
Further comprising
The output means is
A database system, characterized in that the coping method detected by the coping method detection means is output in a manner that can be perceived by the user. The database system according to any one of claims 1 to 4, wherein
An output means for outputting one or more medical condition words detected by the medical condition detection means in a manner that a user can perceive;
In response to an input operation by a user, information update means for adding information that is a combination of the one or more medical condition words output by the output means, the coping method, and result information to the word combination information;
A database system further comprising: The database system according to any one of claims 5 to 8,
The disease treatment combination information is
For each of the disease names, including information that combines one or more personal information and one or more coping methods,
The database system is
A personal information receiving means for receiving one or more personal information in response to an input operation by the user;
Further comprising
The coping method detection means,
Detecting, from the disease coping information, one or more coping methods combined with both the disease name detected by the disease name detecting unit and one or more personal information received by the personal information receiving unit. A featured database system.   The program which makes the said database system function as a database system in any one of Claims 1-10 by being performed by the computer contained in a database system. An information search method in a database system,
(a) for each pathological phrase included in a plurality of pathological phrases each indicating a pathological condition, storing phrase combination information obtained by combining a plurality of pathological phrases related to the pathological condition in the storage unit;
(b) in response to information input by the user, receiving symptom information relating to a sentence describing the symptom;
(c) recognizing a combination of two or more medical condition-related phrases having a predetermined relationship from the symptom information by a predetermined language analysis process;
(d) detecting one or more medical condition words corresponding to a combination of the two or more medical condition related phrases from the phrase combination information;
A method for retrieving information in a database system, comprising:

Images