JP2008276364A - Information processing system and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide technology capable of easily obtaining more proper information that agrees with an intention and fits to a situation. <P>SOLUTION: First and second element association information including a plurality of combinations respectively showing combinations of a plurality of elements respectively belonging to a plurality of items wherein one or more elements belong to each the item included in the plurality of items are stored. List display wherein the plurality of elements are associated between the respective items is visually output based on element association information of all or a part of pieces of the first element association information. When two or more elements are designated in the list display, the list display wherein the plurality of elements are associated between the respective items is visually output based on element association information including element accordance combination information of the second element association information when the second element association information includes the element accordance combination information showing the combination of at least the partial element of the designated two or more element. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to information processing technology.

  When various reports such as reports are created using the knowledge accumulated in the past, it has conventionally been necessary to search past reports by some method and refer to the contents.

  And the need to use effectively the information accumulated in the past is very high, and various database constructions and search techniques have been proposed (for example, Patent Documents 1 to 3).

  For example, the technique of Patent Document 1 dynamically classifies and sorts information items of search results according to the contents of search sentences, and outputs the search results in a format that is easy for the user to refer to. The technique of Patent Document 2 cuts out a necessary part from a server storing an ontology and transmits it to a client. Furthermore, the technique of Patent Document 3 obtains one search result from two or more ontology dictionaries.

JP 2005-182280 A JP 2006-11739 A JP 2006-227803 A

  However, Patent Documents 1 to 3 propose a technique for obtaining a search result from a search condition, a technique for preparing information to be searched, and a technique for presenting a search result. Within the range of stored information, only narrowed down information that matches the search conditions could be obtained.

  Therefore, for example, there is no opportunity to determine whether the obtained search result matches the user's intention or whether the search condition is set correctly. Furthermore, when there is no information that matches the search condition, knowledge accumulated in the past could not be used effectively. In other words, it is not always easy for the user to obtain information that is more appropriate and consistent with the intention according to the situation.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a technique capable of easily obtaining information that is more appropriate and consistent with the intention in accordance with the situation.

  In order to solve the above-mentioned problem, the invention of claim 1 is characterized in that one or more elements belong to each item included in a plurality of items, and a plurality of combinations each indicate a combination of a plurality of elements belonging to each of the plurality of items. Storage means for storing the first and second element related information configured respectively, and the first element related information, or a part of the first element related information Based on the element related information, display means for visually outputting a first list display of network information in which a plurality of elements are associated in a network between each item, and in response to a user action, the first list Designating means for designating two or more elements included in the list display, wherein the display means shows the combination of at least some of the two or more elements as the second element related information. Match combination information In this case, based on the second element related information or a part of the element related information including the element matching combination information of the second element related information, a plurality of elements are associated with each other in a network form. The second list display of the received network information is visually output.

  The invention according to claim 2 is the information processing system according to claim 1, wherein the designation means designates three or more elements included in the first list display in response to a user's operation. And the display means includes partial mismatch combination information indicating a combination of elements in which the second element-related information includes at least some of the three or more elements and does not include the remaining elements. If included, the second list display is visibly output based on the second element related information or a part of the element related information including the partial mismatch combination information of the second element related information. It is characterized by doing.

  The invention according to claim 3 is the information processing system according to claim 1, wherein the display means includes the all-match combination information in which the second element related information indicates a combination of the two or more elements. If included, the second list display is visibly output based on the second element-related information or a part of the element-related information including the all-matching combination information of the second element-related information. It is characterized by that.

  The invention according to claim 4 is the information processing system according to claim 3, wherein the display means includes the first element related information not including the all matching combination information, and the second element related information. When the element related information includes the all-matching combination information, based on the second element-related information or a part of the element-related information including the all-matching combination information of the second element-related information, The second list display is output visually.

  The invention according to claim 5 is the information processing system according to claim 4, wherein the second list display is visibly output by the display means in response to a user operation. The information processing apparatus further includes an adding unit that adds the all-match combination information to the first element related information.

  The invention according to claim 6 is the information processing system according to claim 5, wherein when the adding unit recognizes the right to update information, the all-matching combination is included in the first element related information. It is characterized by adding information.

  The invention according to claim 7 is the information processing system according to any one of claims 4 to 6, wherein the first element-related information includes a predetermined element and one or more first detailed elements. And the second element related information includes information in which the predetermined element is associated with one or more second detailed elements, and the information processing system includes the display means. In a state where the first and second list displays are visibly output, a visual display of detailed elements associated with the predetermined element with respect to the first list display in response to a user action Requesting means for requesting, wherein, in response to the request by the requesting means, the second detail element associated with the predetermined element is made visible based on the second element related information. It is characterized by being output to.

  The invention according to claim 8 is the information processing system according to any one of claims 1 to 7, wherein the first and / or second list display is visually output by the display means. In response, in response to the user's action, for each item included in the first and / or second list display, one or more elements are configured to form sentence information according to a predetermined sentence model. And generating means for generating new sentence information according to the predetermined sentence model.

  The invention according to claim 9 is the information processing system according to claim 8, wherein, when the generation unit recognizes a title related to generation of sentence information, based on the one or more sentence components, It is characterized in that new sentence information is generated in accordance with a predetermined sentence model.

  An invention according to claim 10 is the information processing system according to any one of claims 1 to 9, wherein at least one of the first and second element-related information includes a plurality of information. By performing predetermined information processing including language processing on sentence information, one or more elements are extracted and generated for each item constituting each piece of sentence information.

  The invention according to claim 11 is the information processing system according to any one of claims 1 to 10, wherein at least one of the first and second element related information is RDF. It contains the described information.

  The invention of claim 12 is a program for causing the information processing system to function as the information processing system according to any one of claims 1 to 11 by being executed by a computer included in the information processing system. is there.

  According to the invention of any one of claims 1 to 11, when a plurality of elements are specified in a list display in which a plurality of elements are associated in a network form with respect to one element related information, a plurality of specified elements are specified. When combination information indicating a combination of at least some of the elements is included in other element-related information, a list display in which a plurality of elements are associated with each other in relation to other element-related information is visible Therefore, not only the narrowed down information that matches the specified condition but also judgment materials for obtaining more appropriate knowledge in accordance with the situation are provided to the user. As a result, the user can easily obtain information that is more appropriate and conforms to the intention according to the situation.

  In addition, according to the invention described in any one of claims 2 to 4, the user can obtain information suggesting selection of information in accordance with the situation.

  In addition, according to the fifth aspect of the present invention, it is possible to enhance the element related information.

  According to the sixth aspect of the present invention, since users who can add information to the element related information are limited, the reliability of the data content of the element related information can be ensured.

  According to the seventh aspect of the present invention, detailed element information can be obtained from element-related information including associations between elements that are presumed to more match the user's intention.

  Further, according to the invention described in claim 8, it is possible to easily create a sentence by effectively using past knowledge.

  In addition, according to the invention described in claim 9, it is possible to limit users who can generate sentence information using a list display of element related information.

  Further, according to the invention described in claim 10, the element related information can be easily acquired.

  According to the invention described in claim 11, element related information and network information can be easily described by using RDF.

  Further, according to the twelfth aspect of the invention, the same effects as those of the first to eleventh aspects of the invention can be obtained.

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

<First Embodiment>
<Outline of information processing system>
FIG. 1 is a diagram showing an overview of an information processing system 1 according to an embodiment of the present invention.

  The information processing system 1 is a system for managing and processing medical information in a hospital, for example, and a server 100 and terminals 10 to 30 are connected to a network line LN such as a LAN so that data can be transmitted and received between them. It is configured. In addition, the storage unit 200 is connected to the server 100 so that data can be transmitted and received.

  The storage unit 200 stores a medical information database (DB) 210 that stores medical information (hereinafter also referred to as “medical information”) related to a large number of patients who are medical targets.

  The medical information DB 210 includes a list (examination list) in which names of a large number of patients are listed, and corresponds to a large number of image data obtained by imaging using radiation for each patient and each image data. Report data indicating the interpretation report to be stored is stored in association with the names of patients listed in the examination list. In the medical treatment information DB 210, attribute information indicating information related to ordering from a doctor in charge, patient, or examination is stored in association with the names of patients listed in the examination list.

  The server 100 writes / reads various information to / from the medical information DB 210 and performs information processing using various information stored in the medical information DB 210. This information processing is appropriately managed by a data management function installed in the terminals 10-30.

  The terminals 10 to 30 are each configured with a personal computer (personal computer) or the like, for example. Here, for example, the terminal 10 is provided in an examination room of a hospital, the terminal 20 is provided in a radiographing room that performs radiography, and the terminal 30 is provided in a room that interprets image data obtained by radiography and inputs an interpretation report. It is assumed that is provided. First, when an order is entered from the terminal 10 to the terminal 20, image data of the affected area is acquired by MR (Magnetic Resonance Imaging), CT (Computed Tomography), etc., and the affected area is stored in the medical information DB 210 from the terminal 20 via the server 100. Are stored. Thereafter, the doctor inputs an interpretation report on the terminal 30 while referring to a screen on which image data stored in the medical information DB 210 is visually output. In the case of a relatively small clinic, the doctors who use the terminals 10 and 30 are the same person. In a relatively large hospital, the doctor in charge uses the terminal 10 while the interpreting doctor uses the terminal 30. Such a usage pattern is assumed. The person who uses the terminal 20 includes a radiologist.

  The server 100 uses the interpretation report (existing interpretation report) already stored in the medical treatment information DB 210 to generate information (support information) that supports the input of a new interpretation report, and necessary information from this support information. Are extracted and provided to the terminal 30 in the form of a template display. Note that a function for supporting the input of a new interpretation report in the server 100 is hereinafter also referred to as an “input support function”.

  As described above, the information processing system 1 functions as a database system that extracts and presents information from the viewpoint of using support information, and functions as a reporting system from the viewpoint of inputting an interpretation report.

  FIG. 2 is a diagram illustrating a state where a new interpretation report is input in the terminal 30. As shown in FIG. 2, in the terminal 30, the doctor inputs an interpretation report while appropriately referring to the screens (image display screens) M <b> 1 and C <b> 1 that visually output the patient image data stored in the medical information DB 210. A new interpretation report is input on the screen 401.

<Input support function>
The input support function uses a large number of interpretation reports stored in the medical information DB 210 as past knowledge, extracts elements necessary for inputting a new interpretation report from the many interpretation reports, and uses RDF (Resource Description Framework) and the like. By using and structuring, it has a function to generate support information and present it appropriately.

  Hereinafter, the input support function of the information processing system 1 will be described.

  FIG. 3 is a block diagram showing a functional configuration related to input of a new interpretation report using the input support function and the input support function in the information processing system 1. Note that the functional configuration shown in FIG. 3 is realized by executing a program stored in a storage unit such as a hard disk on the server 100 and the terminal 30 with a CPU or the like. In addition, various data temporarily generated in various information processing is temporarily stored as appropriate in RAM and the like built in the server 100 and the terminal 30, respectively.

  The operations realized by the input support function mainly include an operation for generating support information (support information generation operation), an operation related to patient identification (patient specification operation), and a new interpretation report input based on the support information. Has an operation (input support operation).

  Hereinafter, the support information generation operation, the patient identification operation, and the input support operation will be sequentially described with reference to FIG.

<Support information generation operation>
The data reading unit 101, from the medical treatment information DB 210, information indicating an interpretation report (interpretation report information), attribute information corresponding to the interpretation report information (for example, information relating to ordering, patient attributes, and examination attributes), Information indicating an inspection list (inspection list information) is read. Then, the data reading unit 101 sends the interpretation report information, the attribute information, and the inspection list information to the inspected determination unit 102.

  The examination completion determination unit 102 refers to the examination list information and determines whether or not an interpretation report has been created by writing findings and the like for each patient's interpretation report. The examination list information includes information indicating whether or not an interpretation report has already been created. Then, in the support information generation operation, the inspected determination unit 102 sends the created interpretation report information, that is, inspected interpretation report information, to the data construction unit 103 together with the attribute information. Note that the examination completion determination unit 102 sends uncompleted interpretation report information, that is, unexamined interpretation report information, to the terminal 30 together with attribute information in a patient specifying operation described later.

  The data construction unit 103 identifies and extracts necessary elements from finding sentences described in natural sentences included in the interpretation report, and also extracts various elements included in the attribute information, and performs structuring using RDF. Do. The structure of the elements of the interpretation report is realized by using machine learning.

  Here, a function of machine learning (machine learning function), a function of identifying an element (identification function), and a function of structuring an interpretation report (structured function) in the data construction unit 103 will be described.

  For example, a learning corpus or the like is given as teacher data to learn information serving as a structuring reference. The learning corpus includes a large amount of text data conforming to the format of the finding sentence (sentence model) included in the interpretation report. The sentence model indicates the structure of a finding sentence, for example, a part → findings (feature-conclusion) → diagnosis (diagnosis-conclusion). In the learning corpus, for example, the classification item name of each element constituting the sentence model of the finding sentence is tagged to each word. Examples of element classification items include “part”, “findings”, “diagnosis content (hereinafter abbreviated as“ diagnosis ”), and the like.

  Specifically, in the learning corpus, for example, an element classification item name “part” is tagged to a phrase such as “frontal lobe”, “temporal lobe”, etc., and “high signal range” “low signal” The element classification item name “findings” is tagged for words such as “region” ..., and the element classification item name “diagnosis” is tagged for words such as “cerebral infarction”, “stroke”, etc. It is attached.

  In addition, when such a learning corpus is given to the data construction unit 103 from the outside, in order to create a new interpretation report, which will be described later, data indicating a typical sentence model that specifies the conclusion to some extent is also external. Given by.

  The machine learning function of the data construction unit 103 extracts a word from the learning corpus and stores it for each classification item of the corresponding element. That is, in the data construction unit 103, teacher data including a learning corpus is used as a learning material, and words belonging to each element classification item are learned and stored for each element classification item with reference to the teacher data. Here, the data that is learned and stored is used as model data (model data) that indicates what element classification items should be decomposed into each element constituting the existing interpretation report.

  The identification function of the data construction unit 103 is based on the model data learned as described above, and for the interpretation report input to the data construction unit 103, the classification items of elements and the words and phrases actually used are displayed. Identify.

  When the machine learning method described above is used, element classification items are identified only for elements (here, phrases) given in advance in the teacher data. However, if the following machine learning method is used, Element classification items are also identified for elements that were not given in the teacher data.

  For example, the machine learning function of the data construction unit 103 decomposes the learning corpus into morphemes by morpheme analysis, and for each morpheme, the morpheme itself, the morpheme part of speech, the morpheme utilization form, the front and rear (for example, two before and after) morpheme A pattern (appearance pattern) in which a morpheme belonging to a certain classification item appears is learned using information or the like. Then, by the identification function of the data construction unit 103, according to the appearance pattern, the element classification items are recognized even for elements that are not given in advance (here, words).

  More specifically, for example, in the machine learning function of the data construction unit 103, sentences in the form of “along **” (various words are entered in the “**” part) frequently appear in the learning corpus. In this case, when a word indicating a part frequently appears in the “**” part, an appearance pattern in which a word indicating the part is included in the “**” part is learned. If this appearance pattern is used, the phrase “pituitary gland” is extracted from the phrase “pituitary gland / along” as a word indicating the part from the context of the word for the existing interpretation report. The machine learning can be realized by using a so-called SVM (Support Vector Machine), which contributes to improving the accuracy of natural language processing.

  Then, by the structuring function of the data construction unit 103, based on the information identified by the identification function, the various information included in the interpretation report information and the attribute information is decomposed into words (elements) belonging to each element classification item. By describing in RDF, information related to the interpretation report is structured.

  FIG. 4 is a diagram exemplifying data (hereinafter also referred to as “single report structured data”) in which elements constituting one interpretation report (including elements constituting attribute information of an existing interpretation report) are structured. .

  As shown in FIG. 4, the element classification items “part”, “findings”, and “diagnosis” are attributed to the interpretation report file “A-0001” by RDF, and the attribute value “ The frontal lobe, “high signal”, and “tumor” are described in association with each other. In addition, “head (SKUL)” “T1”, “Taro Taro”, “Hanako Patent”, “MR” with respect to the attribute items “examination part”, “imaging condition”, “patient”, “interpretation doctor”, and “modality”. ”And“ sex ”and“ age ”as attribute items, and attribute values“ M ”and“ 53 ”are described in association with each attribute. Yes. Note that the attribute items (attribute items) to be associated in the single report structured data are not limited to those shown in FIG. 4, and may include other attribute items included in the attribute information, for example.

  In the data construction unit 103, the machine learning function, the identification function, and the structuring function generate single report structured data as shown in FIG. 4 for each of a large number of examined interpretation report information stored in the medical information DB 210. To do.

  Then, in the data construction unit 103, a plurality of attribute values associated with one file in each single report structured data are arranged for each attribute item, reassociated with each other, and described in RDF. For all single report structured data. As a result of this processing, a plurality of attribute values (here, words) are listed for each attribute item included in the plurality of attribute items, and the attribute values are associated with each other in a network form between the attribute items. Information (hereinafter also referred to as “network information”) is generated. The network information is stored in a predetermined storage unit in the server 100, thereby being used as information (support information) that supports input of an interpretation report, and a database (support information DB) 110 is constructed.

  FIG. 5 is a diagram showing an example of the data structure of the support information DB 110 constructed by structuring the elements constituting the single report structured information, that is, the elements related to a large number of existing interpretation report information. In FIG. 5, network information regarding the examination site “head (SKULL)” is illustrated. Further, in FIG. 5, related words are connected with lines, and in order to avoid complication of the figure, the word is appropriately described as “XX” or the like, and the line indicating the association is also compared with the figure. The upper part is illustrated as an example, and the other lines are omitted.

  Further, the data construction unit 103 generates information (combination information) indicating a combination of a plurality of elements (here, phrases) associated with one file in each single report structured information when generating network information. Are also stored in the support information DB 110. For example, information indicating a combination of phrases such as “head-MR-M-T1-frontal lobe-high signal-tumor” is also stored. The network information and the combination information are information indicating an association between one or more words belonging to each item among the items, and are hereinafter collectively referred to as “element related network information” as appropriate. Here, when element-related network information is generated from a large number of single report structured information, at least one element belongs to each item included in a plurality of attribute items, and the element-related network information includes a plurality of attribute items. In other words, the information includes a plurality of types of combination information each indicating a combination of a plurality of elements belonging to each.

  From another point of view, the data structure of the support information DB 110 illustrated in FIG. 5 indicates that when a plurality of elements belong to a predetermined item, the network information for each element belonging to the predetermined item And combination information. For example, a plurality of elements “CR”, “CT”, and “MR” belong to the item “modality”, and it can be said that each element “CR”, “CT”, and “MR” includes network information and combination information. . That is, the support information DB 110 indicates, for each element belonging to a predetermined item, at least one element belongs to each item included in the plurality of items and a combination of a plurality of elements respectively belonging to the plurality of items. It has information (hereinafter also referred to as “element-related information”) that includes a plurality of types of combination information. In the following, the element related information for the element “CR” is “CR element related information”, the element related information for the element “CT” is “CT element related information”, and the element related information for the element “MR” is “ Also referred to as “MR element related information”.

  In the network information stored in the support information DB 110 generated as described above, what words are described in the existing interpretation report is described in a manner in which a plurality of words are associated with each other between the attribute items. Indicated. Therefore, it is effective to output this network information visually and use it when inputting a new interpretation report. In particular, it is effective to provide each attribute item as an item of an element (input element item) and a candidate (option) for inputting a plurality of words and phrases belonging to each attribute item in the form of a template display.

  However, in the network information stored in the support information DB 110, if there are many synonyms (for example, “T2 weighted image” and “T2W1”, etc.) for terms belonging to each attribute item, there are too many choices, and the designation of the choices It becomes difficult. Therefore, it is preferable to replace the synonym with one representative word when detecting the word by the identification function of the data construction unit 103. In addition, if the attribute item includes “Conclusion” etc., there are too many choices if there are too many expressions such as “Consideration” and “Think”, and it is difficult to specify the choice. Become. In view of this, it is preferable to replace the expression shake with one representative expression when detecting the conclusion by the identification function of the data construction unit 103. For example, replacement with representative words and expressions can be realized by including in the teacher data a data table in which a plurality of words and expressions are associated with typical words and expressions. Regarding the shake of the expression, the shake of the expression may be normalized and replaced with the most frequently used ones, or all the synonyms may be replaced with the most frequently used ones.

  Moreover, in an interpretation report, a phrase in which a major classification, a minor classification, a prefix (prefix), and a suffix (suffix) are combined is often used. For example, the phrase “white matter in the right frontal lobe” is a combination of the major classification “frontal lobe”, the minor classification “white matter”, and the prefix “right”. Thus, phrases (for example, “right frontal lobe”, “left frontal lobe”, “frontal lobe white matter”, “frontal lobe cortex”, “subfrontal cortex white matter”, etc.) are composed of combinations of major classification, minor classification, prefix, and suffix. As a result, the number of options increases too much, making it difficult to specify the options.

  Therefore, a phrase (for example, a part phrase) belonging to a predetermined attribute item (for example, the attribute item “part”) in the network information is replaced with a broadly classified term of the phrase. In addition to the network information, as shown in FIG. 6, the RDF description includes detailed elements such as a minor term, a prefix, and a suffix compounded to form a phrase for a major term. The information (detailed information) associated with each other is constructed and included in the support information DB 110. The detailed information mentioned here forms an element associated with a detailed element (element with detailed information) and structured data (detailed information structured data) associated with a detailed element (detailed element). .

  Specifically, in the detailed information structured data shown in FIG. 6, the detailed elements “none” and “right” belonging to the item “prefix” of the detailed element are compared with the broad term “frontal lobe” which is an element with detailed information. “Left” and “both” are associated with each other, and detail elements “gray matter”, “cortex”, “subcortical”, and “subcortical white matter” belonging to the detail element item “subcategory” are associated.

  In order to replace a phrase belonging to a predetermined attribute item with a broad classification term or to construct detailed information structured data as shown in FIG. 6, the elements constituting the phrase are divided into four element items “major classification”. It needs to be broken down into “minor classification”, “prefix” and “suffix”. The process of decomposing the constituent elements constituting the phrase into four-element items is realized with reference to a model that lists language elements belonging to the respective item items. This model is obtained by machine learning using teacher data including a corpus in which a large number of phrases and detailed elements are listed, for example. In this corpus, item names are tagged to a large number of language elements constituting a phrase.

  By the way, for example, a phrase belonging to a certain item, such as a part phrase, also corresponds to a detailed element for a broadly classified term that is an element belonging to a certain item. Therefore, in the “detailed elements” described below, not only the detailed elements that constitute a phrase such as a subcategory term, prefix, and suffix, but also a phrase that is the result of combining multiple detailed elements. included. In addition, since the network information, the combination information, and the detailed information constitute information indicating an association between a plurality of elements (here, words / phrases) belonging to each item among the items, the “detailed element related network information” is appropriately described below. Called.

  FIG. 7 is a flowchart showing an operation flow of the support information generation operation.

  This operation flow is realized by executing a program for the input support function in the server 100. First, the data reading unit 101 reads interpretation report information from the medical treatment information DB 210 together with attribute information (step ST1). Next, the image construction report information is structured by the data construction unit 103 (step ST2). Further, attribute information is added to the information structured in step ST2, and single report structured data as shown in FIG. 4 is generated (step ST3). Then, it is determined whether or not single report structured data has been generated for all the interpretation report information stored in the medical treatment information DB 210 (step ST4), and single report structured data has not been generated for all the interpretation report information. In step ST1, the next interpretation report information and the like are read to generate single report structured data. On the other hand, when single report structured data is generated for all the interpretation reports, detailed element related network information is generated based on the single report structured data for all the interpretation report information (step ST6). finish. At this time, detailed element related network information is memorize | stored in the memory | storage part in the server 100, and support information DB110 is constructed | assembled.

  By the way, the interpretation report information stored in the medical treatment information DB 210 increases every time a new interpretation report is created in response to an input from the terminal 30. It is effective if the interpretation report information accumulated over time can be used as past knowledge. In particular, when an unprecedented finding sentence or the like is included in a newly accumulated interpretation report, it is more effective because the past knowledge further evolves. This point will be described later.

<Patient specific action>
Next, a patient specifying operation realized by the input support function when a new interpretation report is created by displaying a screen (interpretation report input screen) for inputting a new interpretation report on the terminal 30 will be described. Note that data relating to various screens visibly output on the display unit 33 of the terminal 30 is input from the server 100, and a display control unit (not shown) in the terminal 30 displays a screen on the display unit 33 based on the input data. Realize the display.

  The data reading unit 101 reads the examination list from the medical information DB 210 and transfers it to the task management unit 31 of the terminal 30. The task management unit 31 displays the examination list on the interpretation report input screen displayed on the display unit 33.

  FIG. 8 is a diagram illustrating an interpretation report input screen G1.

  As shown in FIG. 8, the interpretation report input screen G1 mainly inputs and creates a request content (ordering information) display area A1, an examination list display area A2, an examination information display area A3, and an interpretation report. An area for performing (report creation area) A4, a display area A5 for finding text, an area for writing a memo (memo area) A6, an area for attaching a representative image (image attachment area) A7, and an icon for inputting a command. It is composed of an enumerated area (command area) A8.

  Then, for example, an examination list as shown in FIG. 9 is displayed in the display area A2 of the interpretation report input screen G1. In the examination list, examinations with examination IDs are listed in order, and for each examination, whether the patient name, gender, examination ID, and interpretation report creation are unprocessed (examined) or already processed (unexamined). A state indicating is described. As described above, when the doctor appropriately operates the mouse or the like of the terminal 30 in the state where the examination list is displayed in the display area A2, a desired patient is designated from the examination list, and information indicating the designated patient name is displayed in the task list. The data is output to the management unit 31. At this time, the task management unit 31 identifies a task related to creation of a new interpretation report corresponding to the designated patient name.

  Further, under the control of the task management unit 31, the data reading unit 101 reads the attribute information related to the ordering, patient attribute, and examination attribute corresponding to the designated patient name from the medical treatment information DB 210 and sends it to the terminal 30. . At this time, on the interpretation report input screen G1, for example, the ordering information as shown in FIG. 10 is displayed in the display area A1 and the display area A3 based on the information related to the provided ordering, patient attributes, and examination attributes. The attribute information related to the patient attributes and examination attributes as shown in FIG. 11 is displayed.

<Input support operation>
In the input support operation, the network information generated as described above is visually output in the form of a template display in the report creation area A4. In this template display, choices of elements (here, phrases) are appropriately designated. Then, a new interpretation report is input.

  However, if a plurality of elements listed in each attribute item are used as options, and the entire network information (whole network information) is simply provided in the form shown in FIG. 5, the number of options listed is too large, It becomes difficult to select an option. Therefore, the information processing system 1 extracts and displays a part of network information (partial network information) from the whole network information, for example, by narrowing down elements for some attribute items in the whole network information. In the template display for displaying a list of partial network information, the number of options listed is limited to some extent, so that the options are easy to see and the options can be easily selected.

  In addition, for example, for a large classification term (for example, “frontal lobe”) belonging to a predetermined item (for example, the item “part”), various prefixes, suffixes, and minor classification terms are added to various phrases. Constitute. For this reason, a large number of phrases relating to one major classification term belong to a predetermined item. However, if you list all of a large number of phrases (for example, “right frontal lobe”, “left frontal lobe”, “frontal white matter”, “frontal cortex”, “subfrontal cortex white matter”, etc.), there are many options listed, Selection is difficult. Therefore, in the list display in the information processing system 1, first, for a plurality of phrases related to the same major classification term, the major classification term is shown as a representative, and in response to a predetermined instruction, the major classification term is related. Detailed information about multiple phrases is displayed. Thus, in the list display, one element (here, term) representing a plurality of elements (here, phrases) is displayed for the time being, and in response to a predetermined instruction, a plurality of elements related to one element are displayed. Displays elements of. With such a configuration, it becomes easy to input detailed information while clarifying the skeleton (minimum essential items) of the finding sentence.

  Here, a method for extracting a part of network information from the entire network information will be described.

  FIG. 12 is a diagram illustrating a template display (extraction condition determination template display) T1 for determining an extraction condition for extracting desired part of network information from the entire network information.

  When a patient is specified by the above-described patient specifying operation, for example, in the interpretation report input screen G1 (FIG. 8), the ordering information (FIG. 10) is displayed in the display area A1 by the function of the support information search unit 120. The attribute information (FIG. 11) relating to the patient and examination is displayed in the report creation area A4 as the extraction condition determination template display T1.

  In the extraction condition determination template display T1 (FIG. 12), for each of the attribute items “examination part”, “modality”, and “gender” of all the attribute items in the entire network information, a word (option) is selected one by one. Thus, a combination of options can be specified as an extraction condition. This extraction condition is given to the support information search unit 120 in response to input of information from the terminal 30.

  Specifically, in the extraction condition determination template display T1 shown in FIG. 12, one attribute value (option) is selected from each of the lists PL1 to PL3 by the mouse pointer MP that operates in response to the mouse operation on the terminal 30. Selected. For each attribute item, select one option at a time, place the mouse pointer MP on the condition determination button DB, and press (left-click) the left mouse button (option) for the three attribute items. An extraction condition for designating a combination of is determined.

  The support information search unit 120 performs a search for the support information DB 110 in response to the specification of the extraction condition from the terminal 30, and extracts partial network information corresponding to the extraction condition from the entire network information in a report creation area. Extracted as information (hereinafter also referred to as “display network information”) for list display as a template display in A4. At this time, the support information search unit 120 refers to the combination information stored in the support information DB 110, extracts partial network information satisfying the extraction condition from the entire network information, and extracts the partial information extracted from the combination information. Some combination information corresponding to the network information is also extracted. Note that the partial network information and the partial combination information are collectively referred to as “partially related information” as appropriate, and the partial network information and the partial combination information are also appropriately referred to as “network information” and “combination information”, respectively. Abbreviated.

  Then, a template display in which the partial network information extracted as described above is visually displayed as a list is provided. This template display supports the input of a finding sentence of a new interpretation report, and is hereinafter also referred to as “input support template display”.

  FIG. 13 is a diagram illustrating a display example of the input support template display T2. Note that various inputs and designations in the input support template display T2 are performed based on signals output from the report input unit 32 in response to various input information such as a mouse operation or voice by the doctor at the terminal 30.

  And in this input support template display T2, a visual list display of some network information, support for creating a finding sentence by specifying an element included in the list display, and presentation of reference information regarding the specification of the element, Is done. The list display of some of the network information, the support for creating the observation text, and the presentation of the reference information will be described in order.

○ List display of some network information:
In FIG. 13, when the element “CR” belonging to the item “modality” is included in the extraction condition, a list display (also referred to as “element list display”) V1 of partial network information extracted from the entire network information is displayed. The presented input support template display T2 is shown.

  As shown in FIG. 13, a predetermined item (here, item) included in the extraction condition for extracting partial network information related to the element list display V1 is displayed in the upper left area of the input support template display T2. A display element (here, “modality: CR”) M1 indicating “modality”) and an element (here, “CR”) belonging to the predetermined item is displayed. The element list display V1 is displayed in the area on the right side of the display element M1, and a plurality of items “parts” to which the elements listed in the element list display V1 belong and are input targets above the element list display V1. Display elements (item display elements) respectively indicating “findings” and “diagnosis” are sequentially displayed from left to right.

  In the element list display V1, display elements (here, phrases) F1 to F1 that respectively indicate a plurality of elements belonging to the items “parts”, “findings”, and “diagnosis” in association with the item display elements “parts”, “findings”, and “diagnosis”. F3 is displayed spatially sequentially from left to right. The types of items listed in the input support template display T2 and the display order thereof are preset, and information indicating the types of items (item type information) and information indicating the display order (item order information) are appropriately set. What is necessary is just to include in support information DB110.

  More specifically, in the input support template display T2, one or more (here, three) words (here, “pulmonary apex”, “lower lung field”, “bronch”) F1 belonging to the item “part” are displayed in the item display. They are listed in spatial order from top to bottom directly below the element “part”. Also, one or more (here three) phrases (here, “string-like shadow”, “infiltrate shadow”, “nodule shadow”) F2 belonging to the item “findings” are directly below the item display element “findings” from top to bottom. Are enumerated spatially in order. Further, one or more (here, three) phrases (here, “change in terminology”, “post-inflammation change”, and “emphysema”) F3 belonging to the item “diagnosis” are displayed directly from above to below the item display element “diagnosis”. It is enumerated spatially toward. In other words, the one or more words F1 to F3 are displayed in a manner in which the relationship with the item display element indicating the subordinate item can be visually recognized.

  Furthermore, with respect to the elements listed in the element list display V1, the elements (hereinafter referred to as “recognized relations”) between the elements in the order of being adjacent to each other are partially recognized in the network information. It is also given as “related display element”. For example, for the three items “part”, “findings”, and “diagnosis”, based on partial network information extracted in accordance with the extraction condition, the words associated with each other are related display elements (here, It is displayed in a state connected by a solid line. That is, the element list display V1 is displayed in a state where a plurality of elements are associated with each other by being connected in a network form with solid lines.

  In this way, some network information that has been narrowed down to a certain extent is visually displayed as a list. Therefore, the user can create an interpretation report while referring to a list display of elements so that past knowledge can be understood at a glance.

○ Support for creating findings:
In the input support template display T2 shown in FIG. 13, for each of the three items “part”, “findings”, and “diagnosis”, when the mouse pointer MP is placed on one word from a plurality of words F1 to F3 and left-clicked, Words corresponding to each item constituting the finding sentence of the new interpretation report are specified. At this time, the doctor who is the user refers to the sentence sentence model MD of the finding sentence displayed at the lower part of the input support template display T2. Information indicating the sentence model MD may be acquired from the outside together with teacher data and stored in the storage unit 200, for example.

  Specifically, the sentence model MD indicates a model of a finding sentence of an interpretation report such as “recognize [blank field W2] in [blank field W1] and doubt [blank field W3]”. In the blank columns W1 to W3, words and phrases that are elements belonging to the three items “part”, “findings”, and “diagnosis” (elements constituting a sentence, hereinafter also referred to as “sentence constituent elements”) apply. That is, according to the phrase designation by the doctor who is the user, one phrase (for example, “pulmonary apex”) of the plurality of phrases (options) F1 is applied to the blank column W1, and among the plurality of phrases (options) F2 Is assigned to the blank column W2, and one of the plurality of phrases (options) F3 (eg, “postoperative change”) is assigned to the blank column W3.

  In addition, in the element list display V1 shown in FIG. 13, a phrase with a predetermined display element “>” such as the phrase “pulmonary apex>” indicates a broadly classified term representing a plurality of phrases. ing. For such major classification terms, various phrases (such as “right lung apex”, “left lung apex”, “bilateral apex”) with detailed elements such as prefixes, suffixes, and minor classification terms added. In some cases, it is desirable to input such detailed elements.

  Therefore, here, when the right mouse button is pressed (right-clicked) with the mouse pointer MP placed on a word with a predetermined display element “>”, a command list CL as shown in FIG. Appear. At this time, if “input detailed information” is specified in the command list CL, the report input unit 32 sends a signal requesting the support information search unit 120 to display detailed elements. Then, the support information search unit 120 reads detailed information structured data corresponding to the phrase specified by the mouse pointer MP from the support information DB 110, and based on this detailed information structured data, a pop-up format as shown in FIG. List PLa is displayed.

  In the list PLa, information about each element related to the “prefix” is listed in order from the top. Specifically, a plurality of detail elements such as “none”, “right”, “left”, and “both sides” are listed for the prefix, and a plurality of detail elements are respectively corresponding to each detail element on the left side of the plurality of detail elements. Radio buttons are arranged. When the user operates the mouse to move the mouse pointer MP to a desired radio button and left-click, the radio button changes from a white circle state to a black circle state, and a detailed element with a black circle radio button attached to the left is displayed. It is specified. Furthermore, when the “OK” button is pressed with the mouse in a state where the detailed elements are designated using radio buttons as appropriate, the designation of the detailed elements is confirmed. At this time, if the detailed element is not specified, the term of the broad classification is specified as a sentence constituent element for the item “part”. On the other hand, if the detailed element is designated, the phrase in which the designated detailed element is added to the major classification term is replaced with the major classification term in the element list display V1.

  For example, when the detailed element “right” related to the prefix is specified in the list PLa, the phrase “right lung apex” in which the prefix “right” is added to the broadly classified term “pulmonary apex”. "Part" is replaced from the broadly classified term "pulmonary apex" belonging to the item "part" in the element list display V1. Further, in the list PLa, when the “Cancel” button is pressed with the mouse, the major classification term belonging to the item “part” of the element list display V1 is not replaced.

  In FIG. 16, for the three items “part”, “finding”, and “diagnosis”, three words “pulmonary apex”, “string-like shadow”, and “postoperative change” are designated, and the density of the display element indicating the designated word is shown. A state in which is inverted is shown. In this state, when the mouse is appropriately operated and the clear button CB is pressed by the mouse pointer MP, the designation of words and phrases for the three items “part”, “finding”, and “diagnosis” is canceled and applied to the blank fields W1 to W3. The deleted phrase is deleted. On the other hand, as shown in FIG. 16, in the input support template display T2, the mouse is appropriately operated in a state where the words specified for the three items are applied to the blank columns W1 to W3 of the sentence model MD, When the input button IB is pressed by the mouse pointer MP, a finding sentence based on the sentence model MD and the words assigned to the blank fields W1 to W3 is generated and displayed in the display area A5.

  However, in response to the pressing of the input button IB, the report composing unit 131 recognizes whether or not the user has a right of use authorized to generate the text information of the finding sentence. When it is recognized that there is a right of use, sentence information of a new finding sentence along the sentence model MD is generated based on the sentence constituent elements assigned to the blank columns W1 to W3 of the sentence model MD. The As for the right of use, for example, table-like information in which user identification information (ID, password, etc.) input when logging in at the terminal 30 and information indicating the right of use is associated is stored in the storage unit of the server 100. The information is stored and the report composition unit 131 may recognize the presence of the right of use by referring to the table-like information. If the right of use is set in this way, users who can generate text information using the element list display V1 or the like can be limited.

  At this time, reference information is presented as a material for determining whether or not the phrase designated by the doctor who is the user is the phrase to be input.

○ Presentation of reference information:
The element list display V1 is extracted from the support information DB 110 (here, CR element related information) according to the extraction condition including one element (here, “CR”) belonging to a predetermined item (here, “modality”). It is presented based on some network information. Then, the user designates each element belonging to each item while referring to the association between the elements in the element list display V1. However, since the element list display V1 presents information under limited conditions, the element most suitable for the situation even if the elements are designated along the association between elements in the element list display V1. The combination of is not always specified.

  Therefore, at least a part of the three elements specified in the element list display V1 is searched for another element related information different from the element related information from which the partial network information related to the element list display V1 is extracted. If there is a combination of elements including the above elements but not including the remaining elements, partial network information including the combination of the elements is extracted from another element related information. Based on the extracted partial network information, a list display of network information in which a plurality of elements are associated with each item is visually presented.

  For example, the partial network information related to the element list display V1 shown in FIG. 13 is extracted from the CR element related information included in the support information DB 110. Then, when three words “pulmonary apex”, “cortical shadow”, and “postoperative change” are designated for the three items “part”, “findings”, and “diagnosis” in the element list display V1, the three words Phrases that include the words “pulmonary apex” and “cortical shadow” that belong to the two items “part” and “findings”, and that do not contain the word “postoperative change” that belongs to one item “diagnosis” among the three words The support information retrieval unit 120 retrieves whether or not information indicating the combination (hereinafter also referred to as “partial mismatch combination information”) exists in the CT and MR element related information. If partial mismatch combination information exists in the CT and MR element related information, partial network information including the partial mismatch combination information is extracted from the CT and MR element related information. At this time, for example, an extraction condition (hereinafter referred to as an element that belongs to the modality in the extraction condition specified in the extraction condition determination template display T1 is replaced with an element that belongs to the modality related to the element-related information including the partial mismatch combination information. Partial network information (hereinafter also referred to as “reference partial network information”) matching the “replacement extraction condition”) is extracted from the support information DB 110.

  Then, based on this reference partial network information, a list display (reference element list display) of network information in which a plurality of elements are associated among the respective items is presented on the display unit 33. Here, when the number of items and the number of elements indicated by the CT and MR element related information are small, all of the element related information (for example, CT element related information) including partially mismatched combination information is extracted and visible. It is good also as a reference element list display by displaying automatically. The partially mismatched combination information includes information including two elements of the three elements specified in the element list display V1, that is, information in which at least a combination of elements is matched (hereinafter referred to as “element matched combination information”). Can also be said.

  FIG. 17 is a diagram illustrating an input support template display T2a on which reference information including the reference element list display V2 is presented.

  The input support template display T2a is different from the input support template display T2 in display elements (in this case, “modality”) included in the replacement extraction condition and display elements (in this case, “CT”) belonging to the predetermined item. Here, “modality: CT”) M2 and a reference element list display V2 shown in the area to the right of the display element M2 are added.

  In the reference element list display V2, display elements (here, phrases) f1 respectively indicating a plurality of elements belonging to the items “parts”, “findings”, and “diagnosis” in association with the item display elements “parts”, “findings”, and “diagnosis”. ... F3 are displayed in a spatial sequence from left to right. The input support template display T2a is displayed according to item type information and item order information included in the support information DB 110 and the like, similarly to the input support template display T2.

  More specifically, one or more (here, three) words (here, “pulmonary apex”, “lung field”, “upper lobe”) f1 belonging to the item “part” are directly below the item display element “part”. They are enumerated sequentially from top to bottom. Further, one or more (here, three) phrases (here, “string-like shadow”, “shadow”, “nodule shadow”) f2 belonging to the item “findings” are directly below the item display element “findings” from top to bottom. It is enumerated spatially toward. Further, one or more (here, three) words (here, “change in terminology”, “post-inflammation change”, “pneumonia”) f3 belonging to the item “diagnosis” are displayed directly from above to below the item display element “diagnosis”. It is enumerated spatially toward. In other words, the one or more words f1 to f3 are displayed in a manner in which the relationship with the item display element indicating the subordinate item can be visually recognized.

  Further, for the elements listed in the reference element list display V2, the relations between the elements that are related in the reference partial network information can be visually recognized between the items in the adjacent order. It is given as a display element. For example, for the three items “part”, “findings”, and “diagnosis”, based on the reference partial network information extracted in accordance with the replacement extraction condition, the words associated with each other are related display elements (here Then, it is displayed in a state connected by a solid line. That is, the reference element list display V2 is presented in a state where a plurality of elements are associated with each other in a network form with a solid line.

  In addition, in the reference element list display V2, the shading of the display elements indicating the three elements constituting the partially mismatched combination information (here, the elements “pulmonary apex”, “string-like shadow”, and “post-inflammation change”) are inverted. Shown in state.

  In this way, in the input support template display T2a, the element list display V1 related to the modality “CR” is presented with the combination of elements “postoperative change” for “pulmonary apex” and “cortical shadow”. On the other hand, when a combination of these elements is designated, in the reference element list display V2 related to the modality “CT”, a different element of “post-inflammation change” with respect to “pulmonary apex” and “cortical shadow” Reference information in which a combination exists is presented. By seeing the reference information, the doctor who is the user can determine whether or not it is correct that the combination of elements “pulmonary apex”, “cortical shadow”, and “postoperative change” is input in the element list display V1. . For example, it can be determined whether or not it is appropriate to make a diagnosis of “postoperative change” for a combination of “pulmonary apex” and “cord shadow”.

  Then, the user may refer to the reference information, for example, keep the designation of the element as it is, or change the designation of the element “post-operative change” to the designation of the element “post-inflammation change”. good. The designation of the element is performed not only by aligning the mouse pointer MP with the element enumerated in the element list display V1, but also by aligning the mouse pointer MP with the element enumerated in the reference element list display V2. You can also. In this way, the creation of a more appropriate finding sentence is promoted by referring to the elements constituting the finding sentence created based on different conditions (in this case, the inspection results of different inspection apparatuses). From another viewpoint, when it is not certain whether the combination of the elements shown in the element list display V1 is appropriate for the situation, information that suggests selection of information that is more suitable for the situation ( Here, the reference element list display V2) can be obtained.

  In the input support template displays T2 and T2a, the words are applied to the blank columns W1 to W3 of the sentence model MD, and the report input unit 32 is appropriately operated while the finding sentence is displayed in the display area A5. When the command “Registration of Findings” in the input area A8 is designated, the report composition unit 131 registers the finding sentence displayed in the display area A5 as a new finding sentence.

  At this time, the information of the newly generated finding sentence is additionally registered in the medical care information DB 210 by the data writing unit 132. For example, information indicating a finding sentence composed of natural sentences such as “recognize a chordal shadow at the apex of the lungs and suspect a postoperative change” is generated and additionally registered in the medical care information DB 210. That is, new interpretation report information (new interpretation report information) including a new finding sentence along a predetermined sentence model MD is generated.

  Here, one sentence component is specified for each item. However, the present invention is not limited to this. For example, at least one sentence component such as two or more sentence elements is specified for each item. May be specified. Therefore, the user operates the report input unit 32 while the element list display V1 is visibly displayed on the terminal 30 and the reference element list display V2 is also visibly displayed. For each item included in the display V1 and / or the reference element list display V2, by designating one or more elements as sentence constituent elements, new finding sentence information in accordance with a predetermined sentence model is generated. . For this reason, it is possible to easily create a sentence by effectively using past knowledge.

  Note that the report composing unit 131 relates to elements (here, words / phrases) allocated to the blank fields W1 to W3 of the sentence model MD, elements (herein, words / phrases) constituting the extraction condition, and an interpretation report to be input. New single report structured information may be generated by associating the patient attribute and the elements constituting the examination attribute with the file name of the interpretation report. In such a configuration, the new single report structured information is used as data for reinforcing the information stored in the support information DB 110 by the function of the data construction unit 103. That is, each time new single report structured information is input, the information is added to the entire network information stored in the support information DB 110, and the combination information is also updated. With the update of the entire network information and the like, lines and the like for associating words also increase in the element list display of partial network information extracted according to the extraction condition.

○ Operation flow of input support operation:
FIG. 18 is a flowchart showing the operation flow of the input support operation. Here, when a predetermined operation is performed on the terminal 30, the input support operation is started, the extraction condition determination template display T1 is displayed in the report creation area A4 of the interpretation report input screen G1, and the process proceeds to step ST11 in FIG.

  In step ST11, in the extraction condition determination template display T1, an extraction condition for narrowing down elements for the items “examination part”, “modality”, and “gender” is input in response to a mouse operation by the user.

  In step ST12, partial network information is extracted from the entire network information based on the extraction condition input in step ST11, and an input support template display T2 including an element list display V1 of the partial network information is displayed in the report creation area A4. Is displayed.

  In step ST13, in the input support template display T2, in response to a mouse operation by the user, a sentence component is specified for each item, and a memo or the like is input as necessary. Note that the input of the memo is executed in response to a keyboard input by generating a cursor in the memo area A6, for example.

  In step ST14, the report composition unit 131 determines whether or not there is an input instruction. Here, as long as the input button IB is not pressed by a user's mouse operation or the like, the processes in steps ST13 and ST14 are repeated. If the input button IB is pressed, it is determined that there is an input instruction and the process proceeds to step ST15.

  In step ST <b> 15, the report composition unit 131 determines whether or not there is a right to use the report input. For example, if it is determined that there is a right of use based on the user identification information input at the time of login, the process proceeds to step ST16, and if it is determined that there is no right of use, this operation flow ends.

  In step ST16, the report composition unit 131 generates sentence information of the finding sentence. Here, the sentence information of the finding sentence is generated based on the sentence component designated for each item and the sentence model.

  In step ST <b> 17, the support information search unit 120 determines whether there is partial mismatch combination information with the support information DB 110 as a search target. Here, in the support information DB 110, element related information (for example, CT, MR element related information) different from the element related information (for example, CR element related information) from which part of the network information is extracted in step ST12. The search target includes at least some of the elements specified in step ST13 (for example, elements belonging to the items “parts” and “findings”), but the remaining elements (for example, elements belonging to the item “diagnosis”) ). If there is partial mismatch combination information, the process proceeds to step ST18. If there is partial mismatch combination information, the process proceeds to step ST19.

  In step ST18, reference partial network information including partial mismatch combination information is extracted from another element related information (for example, CT, MR element related information) by the support information search unit 120, and the reference partial network information is included in this reference partial network information. Based reference element list display V2 is output visually.

  In step ST19, in response to the input from the report input unit 32, the support information search unit 120 determines whether the element designation for each item has been changed. Here, if there is no change in the element designation, the process proceeds to step ST20, and if the element designation is changed, the process returns to step ST13.

  In step ST20, it is determined whether or not a registration instruction has been given by the report composition unit 131. Here, unless the command “finding registration” in the command input area A8 is designated by the user's mouse operation or the like, the processes in steps ST19 and ST20 are repeated. When the command “finding finding” is designated, a registration instruction is issued. As a result, the process proceeds to step ST21.

  In step ST21, a new interpretation report including the text information of the finding sentence generated in step ST16 is generated and recorded in the medical care information DB 210. Then, this operation flow ends.

  As described above, in the information processing system 1 according to the first embodiment, a plurality of elements are included in the element list display V1 in which a plurality of elements related to one element related information (for example, CR element related information) are associated in a network form. It is specified. At this time, when the combination information indicating the combination of at least some of the plurality of specified elements is included in other element related information (for example, CT, MR element related information), other element related A reference element list display V2 in which a plurality of elements related to information are associated in a network form is visually output. For this reason, not only the narrowed down information that matches the specified condition but also reference information as a judgment material for obtaining more appropriate knowledge in accordance with the situation is provided to the user. As a result, the user can easily obtain information that is more appropriate and conforms to the intention according to the situation.

  In particular, in the information processing system 1 according to the first embodiment, a combination of elements that includes a part of the plurality of elements specified in the element list display V1 but does not include the remaining elements (that is, a partial mismatch). Combination information) is included in other element related information (for example, CT, MR element related information), reference partial network information including partially mismatched combination information is extracted from the other element related information, A reference element list display V2 in which elements are associated in a network form is visually output. For this reason, the user can obtain information that suggests selection of information in accordance with the situation.

Second Embodiment
In the information processing system 1 according to the first embodiment, the element list display V1 initially presented searches for information that does not partially match the combination of elements specified by the user, and the reference element list display V2 including the information is presented. It was. On the other hand, in the information processing system 1A according to the second embodiment, when the user specifies a combination of elements that are not associated in the element list display VA1 that is initially presented, information that matches the combination of the elements is displayed. The search and reference element list display VA2 including the information is presented.

  The information processing system 1A according to the second embodiment has substantially the same functions and configuration as the information processing system 1 according to the first embodiment, but is a reference partial network that is visually output as reference information. The information extraction process is different, and further different in that a function for updating information is added to add the combination of elements indicated by the reference information to the element related information related to the element list display VA1 presented initially. Accordingly, in the following description, in the information processing system 1A according to the second embodiment, the same parts as those of the information processing system 1 according to the first embodiment are denoted by the same reference numerals, and the description thereof is omitted. Differences from the information processing system 1 according to the embodiment will be mainly described.

  FIG. 19 is a diagram showing a display example of the input support template display TA2 according to the second embodiment. Note that various inputs and designations in the input support template display TA2 are performed based on signals output from the report input unit 32 in response to various input information such as operation of a mouse by a doctor at the terminal 30 and voice.

  In this input support template display TA2, a visual list display of a part of network information, support for creation of a finding sentence by designating an element included in the list display, presentation of reference information on element designation, Related information is updated. The list display of the partial network information, the support for creating the finding sentence, the presentation of the reference information, and the update of the information will be sequentially described.

○ List display of some network information:
In FIG. 19, when the element “CR” belonging to the item “modality” is included in the extraction condition, a list display (element list display) VA1 of partial network information extracted from the entire network information is presented. A support template display TA2 is shown.

  In the input support template display TA2, a predetermined item (here, item “modality”) included in the extraction condition for extracting partial network information related to the element list display VA1 and this A display element (here, “modality: CR”) MA1 indicating an element (here, “CR”) belonging to a predetermined item is displayed. The element list display VA1 is displayed in the area on the right side of the display element MA1, and a plurality of items “findings” to which the elements listed in the element list display VA1 belong and are input targets above the element list display VA1. Display elements (item display elements) each indicating “diagnosis” are displayed spatially sequentially from left to right.

  In the element list display VA1, display elements (here, phrases) F11 and F12 respectively indicating a plurality of elements belonging to the items “findings” and “diagnosis” in correspondence with the item display elements “findings” and “diagnosis” are displayed from left to right. The images are displayed sequentially in space. The input support template display TA2 is displayed according to item type information and item order information included in the support information DB 110 and the like, similarly to the input support template display T2 according to the first embodiment.

  More specifically, in the input support template display TA2, one or more (here, two) words (here, “linear shadow” and “infiltrating shadow”) F11 belonging to the item “finding” are included in the item display element “finding”. Are enumerated spatially sequentially from top to bottom. In addition, one or more (here, two) phrases (here, “atelectric lung” and “pulmonary emphysema”) F12 belonging to the item “diagnosis” are spatially sequentially from top to bottom directly below the item display element “diagnosis”. Are listed. That is, the one or more words F11 and F12 are displayed in a manner in which the relationship with the item display element indicating the subordinate item can be visually recognized.

  Further, for the elements listed in the element list display VA1, elements that are visually recognizable among the elements that are partially related to each other in the network information between the items in the adjacent order (relation display). Element). For example, with respect to the two items “findings” and “diagnosis”, based on partial network information extracted in accordance with the extraction conditions, the words associated with each other are connected by related display elements (here, solid lines). It is displayed in the displayed state. That is, the element list display VA1 is displayed in a state where a plurality of elements are associated with each other in a network form with solid lines.

○ Support for creating findings:
In the input support template display TA2 shown in FIG. 19, for each of the two items “findings” and “diagnosis”, when the mouse pointer MP is moved to one word from the plurality of words F11 and F12 and left-clicked, a new interpretation is performed. Words corresponding to each item constituting the finding sentence of the report are specified. And when designating a phrase with respect to each item, the doctor who is a user refers to the sentence sentence model MDb of the finding sentence displayed on the lower part of the input assistance template display TA2. Information indicating the sentence model MDb may be acquired from the outside together with teacher data or the like and stored in the storage unit 200, for example.

  Specifically, the sentence model MDb indicates a finding sentence model of an interpretation report such as “[Blank column W11] is recognized and [Blank column W12] is suspected.”. In the blank columns W11 and W12, phrases that are elements (sentence constituent elements) belonging to the two items “findings” and “diagnosis” are respectively applied. That is, according to the phrase designation by the doctor who is the user, one phrase (for example, “linear shadow”) of the plurality of phrases (options) F11 is applied to the blank column W11, and among the plurality of phrases (options) F12. Is applied to the blank column W12.

  In addition, in the element list display VA1 shown in FIG. 19, a phrase with a display element “>” such as the phrase “linear shadow>” indicates a broadly classified term representing a plurality of phrases. . Then, when the mouse pointer MP is placed on a word with the display element “>” and the right button of the mouse is pressed (right click), a command list CL as shown in FIG. 20 appears. At this time, if “input detailed information” is specified in the command list CL, the report input unit 32 sends a signal requesting the support information search unit 120 to display detailed elements. Then, the support information search unit 120 reads detailed information structured data (FIG. 21) corresponding to the phrase designated by the mouse pointer MP from the support information DB 110, and based on this detailed information structured data, shown in FIG. Such a pop-up list PLb is displayed.

  The detailed information structured data shown in FIG. 21 is data in which modifiers “none” and “large” are associated with the broadly classified term “linear shadow”.

  Then, in the list PLb shown in FIG. 22, information about each element related to “modifier” is listed in order from the top. Specifically, a plurality of detailed elements such as “none” and “large” are listed for the modifier, and a plurality of radio buttons are arranged on the left side of the plurality of detailed elements so as to correspond to the respective detailed elements. . When the user operates the mouse to move the mouse pointer MP to a desired radio button and left-click, the radio button changes from a white circle state to a black circle state, and a detailed element with a black circle radio button attached to the left is displayed. It is specified. Furthermore, when the “OK” button is pressed with the mouse in a state where the detailed elements are designated using radio buttons as appropriate, the designation of the detailed elements is confirmed. At this time, if the detailed element is not designated, the broad classification term is designated as the sentence constituent element for the item “findings”. On the other hand, if the detailed element is specified, the phrase in which the specified detailed element is added to the major classification term is replaced with the major classification term in the element list display VA1.

  For example, in the list PLb, when the detail element “large” relating to the modifier is designated, the phrase “large linear” in which the modifier “large” is added to the broad term “linear shadow” "Shadow" is replaced with the broad term "linear shadow" belonging to the item "findings" in the element list display VA1. In addition, when the “Cancel” button is pressed with the mouse in the list PLb, the major classification terms belonging to the item “findings” in the element list display VA1 are not replaced.

  FIG. 23 shows a state in which two words “line shadow” and “atelectasis” are designated for the two items “findings” and “diagnosis”, and the shades of display elements indicating the designated words are reversed. ing. In this state, when the mouse is appropriately operated and the clear button CB is pressed by the mouse pointer MP, the designation of the words for the two items “findings” and “diagnosis” is canceled, and the words applied to the blank fields W11 and W12 are changed. Erased. On the other hand, as shown in FIG. 23, in the input support template display TA2, the mouse is appropriately operated in a state where the phrases specified for the two items are applied to the blank columns W11 and W12 of the sentence model MDb, When the input button IB is pressed by the mouse pointer MP, a finding sentence based on the phrase applied to the sentence model MDb and the blank fields W11 and W12 is displayed in the display area A5.

  However, in the same manner as the information processing system 1 according to the first embodiment, in response to the pressing of the input button IB, does the report component 131 have a right of use for which the user is permitted to generate the sentence text information? Whether or not is recognized. When it is recognized that there is a right of use, sentence information of a new finding sentence along the sentence model MDb is generated based on the sentence components allocated to the blank columns W11 and W12 of the sentence model MDb. The

  At this time, reference information is presented as a material for determining whether or not the phrase designated by the doctor who is the user is the phrase to be input.

○ Presentation of reference information:
The element list display VA1 is extracted from the support information DB 110 (here, CR element related information) according to the extraction condition including one element (here, “CR”) belonging to a predetermined item (here, “modality”). It is presented based on some network information. Then, the user designates each element belonging to each item while referring to the association between the elements in the element list display VA1. However, since the element list display VA1 presents information under limited conditions, the element list display VA1 is most suitable for the situation only by specifying the elements along the association between elements in the element list display VA1. It is not always possible to specify a combination of elements. For this reason, it is also assumed that the user thinks that a combination of elements that are not associated in the element list display VA1 should be specified. However, it is assumed that the user feels uneasy that selecting a combination of elements not shown in the element list display VA1 is contrary to past knowledge.

  Therefore, when the element-related information from which the partial network information related to the element list display VA1 is extracted does not include information indicating the combination of elements specified in the element list display VA1, the one related to the element list display VA1. If there is information indicating a combination of all elements specified in the element list display VA1 with another element related information different from the element related information from which the part network information is extracted as a search target, the combination of the elements is included. Extract some network information from other element related information. Based on the partial network information, a list of network information in which a plurality of elements are associated with each other is visually presented.

  Specifically, the partial network information related to the element list display VA1 illustrated in FIG. 19 is extracted from the CR element related information included in the support information DB 110. Then, in the element list display VA1, for the two items “findings” and “diagnosis”, when two words “linear shadows” and “atelectasis” that are not associated with each other are designated, all designated (here, The support information search unit 120 searches whether information indicating a combination of two words (hereinafter also referred to as “all matching combination information”) exists in the CT and MR element related information. If all matching combination information exists in the CT and MR element related information, partial network information including the all matching combination information is extracted from the CT and MR element related information. At this time, for example, an extraction condition (hereinafter referred to as “the element belonging to the modality related to the element related information including all matching combination information” is replaced with an element belonging to the modality among the extraction conditions specified in the extraction condition determination template display T1. Partial network information (reference partial network information) that matches the “replacement extraction condition” is extracted from the support information DB 110.

  Then, based on this reference partial network information, a list display (reference element list display) of network information in which a plurality of elements are associated among the respective items is presented on the display unit 33. In addition, when the number of items and the number of elements indicated by the CT and MR element related information are small, all of the element related information including all matching combination information is extracted and output visually, so that the reference element list The display VA2 may be used.

  FIG. 24 is a diagram illustrating an input support template display TA2a on which reference information including the reference element list display V2 is displayed.

  As shown in FIG. 24, the input support template display TA2a is different from the input support template display TA2 in a predetermined item (here, “modality”) included in the replacement extraction condition and an element (here, in this case). A display element (in this case, “modality: CT”) MA2 indicating “CT”) and a reference element list display VA2 shown in the area on the right side of the display element MA2 are added.

  In the reference element list display VA2, display elements (herein, phrases) f11 and f12 respectively indicating a plurality of elements belonging to the items “findings” and “diagnosis” corresponding to the item display elements “findings” and “diagnosis” are displayed from the left. It is displayed in spatial order toward the right. The input support template display TA2a is displayed according to the item type information and the item order information included in the support information DB 110, similarly to the input support template display TA2.

  More specifically, one or more (here, two) phrases (here, “linear shadows” and “nodule shadows”) f11 belonging to the item “findings” are directly below the item display element “findings” from top to bottom. It is enumerated spatially toward. Furthermore, one or more (here, two) phrases (here, “atelectric lung” and “pneumonia”) f12 belonging to the item “diagnosis” are spatially sequentially from top to bottom directly below the item display element “diagnosis”. Are listed. That is, the one or more words f11 and f12 are displayed in a manner in which the relationship with the item display element indicating the subordinate item can be visually recognized.

  Furthermore, with respect to the elements listed in the reference element list display VA2, the relations between the elements related in the reference partial network information can be visually recognized between the items in the adjacent order. It is given as a display element. For example, for the two items “findings” and “diagnosis”, based on the reference partial network information extracted in accordance with the replacement extraction condition, the words associated with each other are related display elements (here, solid lines). ) Are displayed in a connected state. That is, the element list display VA2 is displayed in a state where a plurality of elements are associated with each other in a network form with solid lines.

  Further, in the element list display VA2, the display elements indicating the two elements (here, the elements “linear shadow” and “atelectasis”) indicated by the all-match combination information are displayed in an inverted state.

  In this way, in the input support template display TA2a, the element list display VA1 related to the modality “CR” specifies the combination of the element “linear shadow” and the element “atelectasis” that are not associated with each other. In this case, in the element list display VA2 related to the modality “CT”, reference information in which a combination of the element “linear shadow” and the element “atelectasis” exists is presented. By seeing the reference information, the doctor who is the user can determine whether or not it was correct to input the combination of the element “linear shadow” and the element “atelectasis” on the element list display VA1.

  Then, the user may maintain the element designation with confidence by referring to the reference information, or may appropriately change the element designation. In addition, the designation of the element is performed not only by aligning the mouse pointer MP with the element enumerated in the element list display VA1, but also by aligning the mouse pointer MP with the element enumerated in the reference element list display VA2. You can also. In this way, the creation of a more appropriate finding sentence is promoted by referring to the elements constituting the finding sentence created based on different conditions (in this case, the inspection results of different inspection apparatuses). From another point of view, when the user really wants to select a combination of elements that are not shown in the element list display VA1, the user can select information that is more suitable for the situation (here, the reference element list display VA2). ) Can be obtained.

  At this time, the detailed information structured data related to the element related information including the combination of elements specified by the user is more detailed than the detailed information structured data related to the element related information not including the combination of elements specified by the user. However, there is a high possibility that the detailed element that the user wants to input is included. Therefore, when the reference element list display VA2 is displayed, details of the element list display VA1 and the reference element list display VA2 relating to the phrase with “>” such as the phrase “linear shadow>” or the like. The information structured data is changed to detailed information structured data related to the reference element list display VA2.

  Specifically, the element-related information (for example, CR element-related information) from which the element list display VA1 is extracted corresponds to a predetermined element (for example, “linear shadow”) and one or more detailed elements in FIG. The element-related information (for example, CT element-related information) from which the reference element list display VA2 is extracted while having the detailed information structured data as shown is a predetermined element (for example, “linear shadow”) and one or more details. It has detailed information structured data as shown in FIG. 25 associated with elements. Then, in a state where the element list display VA1 and the reference element list display VA2 are visibly output on the display unit 33, a predetermined element (for example, the element list display VA1) is responded to the operation on the report input unit 32 by the user. Display of detailed elements associated with “Linear Shadow”) is required. At this time, in response to the detailed element display request, the reference element list display VA2 is determined based on the detailed information structured data (FIG. 25) included in the extracted element related information (for example, CT element related information). One or more detailed elements associated with the element (for example, “linear shadow”) are displayed on the display unit 33.

  More specifically, for example, when the mouse pointer MP is placed on the display element “linear shadow>” included in the element list display VA1 and the right button of the mouse is pressed (right click), as shown in FIG. A command list CL appears. At this time, if “input detailed information” is specified in the command list CL, the report input unit 32 sends a signal requesting the support information search unit 120 to display detailed elements. Then, the support information search unit 120 reads the detailed information structured data (FIG. 25) corresponding to the word designated by the mouse pointer MP from the support information DB 110, and based on this detailed information structured data, shown in FIG. Such a pop-up list PLc is displayed. The list PLc shown in FIG. 26 is different from the list PLb shown in FIG. 22 (here, the detailed elements are increased). And by such operation | movement, the detailed element information can be obtained from the element relevant information including the association between the elements presumed to more match the user's intention.

  Similarly to the information processing system 1 according to the first embodiment, in the terminal 30, the element list display V1 is visibly displayed, and the reference element list display V2 is also visibly displayed. By operating the report input unit 32, the user designates one or more elements as sentence constituent elements for each item included in the element list display V1 and / or the reference element list display V2. New finding information along the model is generated.

  More specifically, in the input support template displays TA2 and TA2a, the report input unit 32 is displayed in a state in which words are applied to the blank columns W11 and W12 of the sentence model MDb and the observation sentence is displayed in the display area A5. When appropriately operated and the command “finding registration” in the command input area A8 is designated, the report constructing unit 131 registers the finding sentence displayed in the display area A5 as a new finding sentence.

○ Information update:
The input support template display TA2b shown in FIG. 24 includes element related information (for example, CR element related information) related to the element list display VA1 based on element related information (for example, CT element related information) related to the reference element list display VA2. ) Is displayed in a state where information (update proposal information) for suggesting updating is displayed.

  As shown in FIG. 24, update proposal information is displayed below the display element MA2 and the reference element list display VA2 in the input support template display TA2a. This update proposal information includes the display element MA3 indicating the same content as the display element MA1 (here, “modality: CR”), and all-match combination information (here, the reference element list display VA2 with respect to the element list display VA1). Then, information that proposes to add an element (information indicating a combination of the element “linear shadow” and the element “atelectasis”) is included.

  More specifically, for example, with respect to the element list display VA1, a special related display element (in this case, a thick line) indicating the association between the element “linear shadow” and the element “atelectasis” related to the all-match combination information A list display of elements (hereinafter also referred to as “updated element list display”) VA3 to which a display element (here, “registration?”) Suggesting registration of a new relationship is attached is shown.

  In this way, in a state where the element list display VA1, the reference element list display VA2, and the updated element list display VA3 are displayed, the update button RB is moved by the mouse pointer MP in response to an operation by the user in the report input unit 32. When pressed, the function of the data construction unit 103 adds all matching combination information to the element related information (for example, CR element related information) related to the element list display VA1. At this time, the data content of the support information DB 110 is updated to include the combination of elements indicated by the updated element list display VA3, and the element related information can be enhanced.

  However, if such combination information of elements is updated by a person who does not have a high level of knowledge, there is a concern that the reliability of the data contents constituting the support information DB 110 may decrease. Therefore, in response to pressing of the update button RB, the data construction unit 103 recognizes whether or not the user has the authority (update authority) that is permitted to update the support information DB 110. When it is recognized that there is an update right, all matching combination information is added to the element related information (for example, CR element related information) related to the element list display VA1, and the data content of the support information DB 110 Is updated.

  As for the right of update, as with the right of use related to the input of the finding sentence, for example, the user identification information (ID, password, etc.) input when logging in at the terminal 30 is associated with the information indicating the right of update. The table information may be stored in the storage unit of the server 100, and the report configuration unit 131 may recognize the presence of the update right by referring to the table information. If such an update right source is used, the number of users who can add information to the element related information is limited. Therefore, the reliability of the data contents constituting the element related information can be ensured.

○ Operation flow of input support operation:
27 and 28 are flowcharts showing the operation flow of the input support operation. Here, when a predetermined operation is performed on the terminal 30, an input support operation is started, and the extraction condition determination template display T1 (FIG. 12) is displayed in the report creation area A4 of the interpretation report input screen G1, and the steps of FIG. Proceed to SP11.

  In steps SP11 to SP16, processing similar to that in steps ST11 to ST16 in FIG. 18 is performed.

  In step SP17, the support information search unit 120 uses the support information DB 110 as a search target, and includes all matching combination information including all elements specified in step SP13 (for example, elements belonging to the items “part” and “findings”). It is determined whether or not. Here, in the support information DB 110, different element related information (for example, CT, MR element related) different from the element related information (for example, CR element related information) from which the partial network information extracted in step SP12 is extracted. Information). If there is all-match combination information, the process proceeds to step SP18. If there is no all-match combination information, the process proceeds to step SP31 in FIG.

  In step SP18, the support information search unit 120 extracts reference partial network information including all matching combination information from other element related information (for example, CT, MR element related information), and is based on this reference partial network information. The reference element list display VA2 is visually output. At this time, the update proposal information is also output visually. That is, the input support template display TA2a as shown in FIG.

  In step SP19, the support information search unit 120 changes the detailed information structured data related to the element list display VA1 to the detailed information structured data related to the reference element list display VA2.

  In step SP20, the data construction unit 103 determines whether there is an instruction (update instruction) to update the data contents of the support information DB 110. Here, if the update button RB is not pressed by a user's mouse operation or the like, it is determined that there is no update instruction, and the process proceeds to step SP31 in FIG. On the other hand, if the update button RB is pressed, it is determined that there is an update instruction, and the process proceeds to step SP21.

  In step SP21, the data construction unit 103 determines whether or not there is an update right that is permitted to update the support information DB 110. For example, if it is determined that there is an update right based on the user identification information input at the time of login, the process proceeds to step SP22, and if it is determined that there is no update right, the process proceeds to step SP31 in FIG.

  In step SP22, all matching combination information is added to the element related information (for example, CR element related information) related to the element list display VA1, and the data content of the support information DB 110 is updated.

  Next, in step SP31 of FIG. 28, in response to the input from the report input unit 32, the support information search unit 120 determines whether or not the element designation for each item has been changed. Here, if there is no change in the element designation, the process proceeds to step SP32. If the element designation is changed in response to the operation of the report input unit 32, the process returns to step SP13.

  In step SP32, it is determined whether or not there is a registration instruction by the report composition unit 131. Here, unless the command “finding registration” in the command input area A8 is designated by the user's mouse operation or the like, the processing in steps SP31 and SP32 is repeated. When the command “finding finding” is designated, a registration instruction is issued. As a result, the process proceeds to step SP33.

  In step SP33, a new interpretation report including the text information of the finding sentence generated in step SP16 is generated and recorded in the medical care information DB 210. Then, this operation flow ends.

  As described above, in the information processing system 1A according to the second embodiment, a plurality of elements are designated in the element list display VA1 in which a plurality of elements are associated with each other with respect to one element related information (for example, CR element related information). Is done. At this time, when the combination information of all the specified elements is included in other element related information (for example, CT, MR element related information), a plurality of elements are associated in a network form with respect to the other element related information. The reference element list display VA2 is output visually. For this reason, for example, when it is desired to actually select a combination of elements not shown in the element list display VA1 presented initially, the reference element list display VA2 indicating the combination of elements to be actually selected is presented. Therefore, not only the narrowed down information that matches the specified condition, but also reference information as a judgment material for obtaining more appropriate knowledge in accordance with the situation is provided to the user. As a result, the user can easily obtain information that is more appropriate and conforms to the intention according to the situation.

<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 embodiment, it is assumed that the support information DB 110 is composed of a plurality of element related information (for example, CR, CT, MR element related information), and element list display V1, VA1 related to one element related information When multiple elements are specified in, partial mismatch combination information that includes some of the specified elements but does not include some elements, or all matches that indicate combinations of all specified elements When the combination information is included in other element related information (for example, CT, MR element related information), reference element list display V2, VA2 related to partial network information including partial mismatch combination information and full match combination information Is output visually, but is not limited to this. For example, even if a support information DB different from the support information DB 110 is prepared, the support information DB 110 and the support information DB different from each other are regarded as different element related information, and the same processing is performed, the same effect as the above embodiment Can be obtained.

  Here, a specific example will be described.

  FIG. 29 is a diagram illustrating a schematic configuration of an information processing system 1B according to a modification.

  First, the information processing system 1B has the same configuration as the information processing systems 1 and 1A according to the above-described embodiment (for the server 100 and the terminals 10 to 30 and the server 100 connected to be able to transmit and receive data via the network line LN). A storage unit 200) connected to be able to transmit and receive data is provided. Further, the server 100 is connected to an external server 100B through a network line NTW constituted by an Internet line or the like so as to be able to transmit and receive data.

  For example, the server 100B is installed in a larger hospital than the hospital in which the server 100 is provided. A number of terminals are connected to the server 100 via a network line LN such as a LAN provided in the hospital so that data can be transmitted and received, and the storage unit 200B is also connected so as to be able to transmit and receive data. The storage unit 200B stores a medical information DB 210B in which medical information of a large hospital having a data configuration similar to that of the medical information DB 210 is stored. Further, the server 100B has the same function as the server 100, and a support information DB 110B corresponding to the medical information DB 210B is constructed. That is, the support information DB 110 related to the medical information of the small hospital is built in the server 100, and the support information DB 110B related to the medical information of the large hospital is built in the server 100B. Here, the support information DB 110B has more information in proportion to the scale of the hospital than the support information DB 110.

  In such an information processing system 1B, when a doctor who is a user in the terminal 30 inputs a finding sentence along the element list display related to the partial network information extracted from the support information DB 110, the element list display When a combination of non-elements is specified, first, as in the second embodiment, if there is information (all-match combination information) indicating the combination of the specified elements with the support information DB 110 as a search target, the all-match combination information Reference partial network information including is extracted and a reference element list display is presented. At this time, if there is no all-match combination information in the support information DB 110, another support information DB 110B having more complete information is searched, and if there is all-match combination information, the reference information including the all-match combination information is included. Section network information is extracted and a reference element list display is presented. An example of the input support template display TB2a presented at this time is shown in FIG.

  The input support template display TB2a shown in FIG. 30 is obtained by adding that reference information for a large hospital is displayed to the input support template display TA2a shown in FIG.

  In this way, even in an organization that does not have a certain amount of information, it is possible to obtain more appropriate knowledge in accordance with the situation by using the information of another organization that has a certain amount of information through the Internet. Reference information as a judgment material for obtaining is provided to the user.

  In the first embodiment, the finding sentence is created by designating elements belonging to the three items included in the element list display V1. However, the present invention is not limited to this. For example, a blank column is included in the sentence model. When there are four or more, a finding sentence may be created by designating elements belonging to four or more items included in the element list display. More generally, by specifying an element for each of three or more items included in the element list display, three or more elements may be specified to create a finding sentence.

  In such a configuration, another element-related information different from the element-related information from which the partial network information related to the element list display is extracted is searched, and among the three or more elements specified in the element list display, If there is a combination of elements that includes at least some elements, but does not contain the remaining elements, the partial network information including the combination of the elements is extracted from other element related information, and based on the partial network information, A list display of network information in which a plurality of elements are associated with each item may be presented visually. Note that the items to which some of the elements belong and the items to which the remaining elements belong are appropriately set in advance, and information indicating the items may be stored in the server 100 and referred to during search. .

  In the second embodiment, the finding sentence is created by designating elements belonging to the two items included in the element list display VA1, but the present invention is not limited to this. For example, there is a blank column in the sentence model. When there are three or more, a finding sentence may be created by designating elements belonging to three or more items included in the element list display. More generally, by specifying an element for each of two or more items included in the element list display, two or more elements may be specified to create a finding sentence.

  In such a configuration, when the element-related information from which the partial network information related to the element list display is extracted does not include the combination of the two elements specified in the element list display, If there is a combination of all two or more elements specified in the element list display with another element related information different from the element related information from which the part network information is extracted as a search target, the combination of the elements is included. The partial network information may be extracted from other element related information, and based on this partial network information, a list display of network information in which a plurality of elements are associated with each other may be presented visually.

  In addition, in the above embodiment, all the element related information and the two or more element related information are items that constitute each piece of text information by performing predetermined information processing including language processing for a plurality of finding sentences. Although one or more elements are extracted and generated every time, the present invention is not limited to this. For example, at least a part of the element related information or at least one element related information of the two or more element related information is subjected to predetermined information processing including language processing for a plurality of finding sentences, Even when adopting a configuration in which one or more elements are extracted and generated for each item constituting each sentence information, the element related information can be easily obtained.

  In the above embodiment, all of the detailed element related network information and two or more detailed element related network information are described in RDF. However, the present invention is not limited to this. For example, at least one of the detailed element related network information A configuration may be adopted in which at least one piece of detailed element related network information of two or more detailed element related network information is easily described by RDF.

  In the sentence model shown in the above embodiment, the conclusion is fixed. However, the present invention is not limited to this. For example, for the conclusion, one word / phrase is selected from a plurality of options (for example, “Accept”, “Do not accept”, etc.). You may make it selectable, and you may enable it to selectively specify also about other words and phrases which comprise natural sentences, such as an adjective and an adverb.

  In the above embodiment, the single report structured information is generated directly on the basis of the elements specified in the input support template display. However, the natural text newly input to the medical care information DB 210 is used. Single report structured information may be generated using a certain finding sentence or the like. However, it is preferable to generate the single report structured information directly based on the designated element or the like because the amount of calculation is reduced.

  In the above embodiment, the explanation has been given by taking as an example a system that creates an interpretation report by interpreting an image obtained using radiation. However, the present invention is not limited to this. For example, other medical care such as a nursing report and an incident report The present invention can also be applied to a system for creating a document for use. Furthermore, the present invention can also be applied to the creation of reports in fields other than medical, such as sales reports. Further, the present invention can be applied to fields other than reports, such as fields that present recommended trips and products. The invention can be applied.

  In the above embodiment, the finding in the interpretation report is in the form of a natural sentence such as “recognize a chordal shadow at the apex of the lungs and suspect postoperative changes.” However, the present invention is not limited to this. The format of the finding sentence is such that “site: lung apex”, “finding: cord-like shadow”, “diagnosis: postoperative change”, and the like are structured with elements associated with each attribute item in advance. Also good. When such a finding sentence format is adopted, the functional configuration of the information processing systems 1, 1 </ b> A, 1 </ b> B is simplified and the time required for information processing is reduced by partially omitting the functional configuration of the terminal 30. Is done.

  In the above embodiment, the element list display is presented based on a part of the network information of the element related information. However, the present invention is not limited to this, and if the number of items or the number of elements is small, the element related information is displayed. May be presented based on all of the above.

  Note that the functional configuration of the information processing systems 1 and 1A according to the above embodiment may be realized by executing one program by a control unit configured by a CPU or the like of one device, or a plurality of Even if a program is executed by each control unit configured by a CPU of a single device and a plurality of devices cooperate to realize the functional configuration of the information processing systems 1 and 1A according to the above embodiment. good.

1 is a diagram illustrating a schematic configuration of an information processing system according to a first embodiment. It is a figure which illustrates the state which inputs a new interpretation report in a terminal. It is a block diagram which shows the function structure which concerns on the input of a new interpretation report. It is a figure which illustrates single report structured data. It is a figure which illustrates the data structure of support information DB. It is a figure which illustrates detailed information structured data. It is a flowchart which shows the operation | movement flow of support information generation operation | movement. It is a figure which illustrates an interpretation report input screen. It is a figure which illustrates an inspection list. It is a figure which illustrates ordering information. It is a figure which illustrates the attribute information which concerns on a patient attribute and a test | inspection attribute. It is a figure which illustrates extraction condition determination template display. It is a figure which illustrates an input support template display. It is a figure which illustrates the list of commands. It is a figure which illustrates the list | wrist of a popup format. It is a figure which illustrates an input support template display. It is a figure which illustrates the input support template display containing reference information. It is a flowchart which shows the operation | movement flow of input assistance operation | movement. It is a figure which illustrates the input assistance template display which concerns on 2nd Embodiment. It is a figure which illustrates the list of commands. It is a figure which illustrates detailed information structured data. It is a figure which illustrates the list | wrist of a popup format. It is a figure which illustrates an input support template display. It is a figure which illustrates an input support template display. It is a figure which illustrates detailed information structured data. It is a figure which illustrates the list | wrist of a popup format. It is a flowchart which shows the operation | movement flow of input assistance operation | movement. It is a flowchart which shows the operation | movement flow of input assistance operation | movement. It is a figure which shows schematic structure of the information processing system which concerns on a modification. It is a figure which illustrates an input support template display.

Explanation of symbols

1, 1A, 1B Information processing system 10-30 Terminal 31 Task management unit 32 Report input unit 33 Display unit 100, 100B Server 101 Data reading unit 102 Checked determination unit 103 Data construction unit 110, 110B Support information DB
120 Support Information Search Unit 131 Report Configuration Unit 132 Data Writing Unit 200, 200B Storage Unit 210, 210B Medical Information DB

Claims (12)

1st and 2nd which each comprised one or more elements to each item contained in a plurality of items, and each comprised a plurality of kinds of combination information respectively indicating a combination of a plurality of elements belonging to the plurality of items. Storage means for storing the element-related information of
A first list display of network information in which a plurality of elements are associated with each other in a network form based on the first element related information or a part of the element related information included in the first element related information Display means for visually outputting
In response to a user action, designating means for designating two or more elements included in the first list display;
The display means is
When the second element-related information includes element matching combination information indicating a combination of at least some of the two or more elements, the second element-related information or the second element-related information A second list display of network information in which a plurality of elements are associated with each other in a network form based on a part of the element related information including the element matching combination information. Information processing system. The information processing system according to claim 1,
The designation means is
In response to a user action, specify three or more elements included in the first list display;
The display means is
When the second element related information includes partial mismatch combination information indicating a combination of elements including at least a part of the three or more elements and not including the remaining elements, the second element related information The second list display is visibly output based on element-related information or a part of element-related information including the partial mismatch combination information of the second element-related information. Processing system. The information processing system according to claim 1,
The display means is
When the second element related information includes all matching combination information indicating a combination of the two or more elements, the second element related information or the all matching combination information in the second element related information An information processing system characterized in that the second list display is visibly output based on a part of the element related information including: The information processing system according to claim 3,
The display means is
When the first element related information does not include the all-match combination information and the second element-related information includes the all-match combination information, the second element-related information or the second element-related information An information processing system characterized in that the second list display is visibly output based on a part of the element related information including the all-match combination information in the element related information. The information processing system according to claim 4,
An adding means for adding the all-match combination information to the first element related information in response to a user action in a state where the second list display is visually output by the display means;
An information processing system further comprising: The information processing system according to claim 5,
The additional means is
An information processing system characterized by adding the all-match combination information to the first element-related information when recognizing a title related to information update. An information processing system according to any one of claims 4 to 6,
The first element related information is
Information associated with a predetermined element and one or more first detail elements;
The second element related information is
Information associated with the predetermined element and one or more second detail elements;
The information processing system is
In a state where the first and second list displays are visibly output on the display means, in response to a user action, detailed elements associated with the predetermined elements with respect to the first list display are displayed. Request means for requesting a visual indication,
Further comprising
The display means is
In response to a request from the request means, the information processing system visually outputs the second detailed element associated with the predetermined element based on the second element related information. An information processing system according to any one of claims 1 to 7,
Each item included in the first and / or second list display in response to a user action in a state where the first and / or second list display is visually output by the display means. Generating means for determining one or more elements as one or more sentence constituent elements constituting sentence information along a predetermined sentence model, and generating new sentence information along the predetermined sentence model;
An information processing system further comprising: The information processing system according to claim 8,
The generating means
An information processing system for generating new sentence information in accordance with the predetermined sentence model based on the one or more sentence constituent elements when a title relating to generation of sentence information is recognized. An information processing system according to any one of claims 1 to 9,
Information of at least one of the first and second element related information is
An information processing system in which one or more elements are extracted and generated for each item constituting each piece of sentence information by performing predetermined information processing including language processing on a plurality of pieces of sentence information. An information processing system according to any one of claims 1 to 10,
Information of at least one of the first and second element related information is
An information processing system including information described in RDF.   The program which makes the said information processing system function as an information processing system in any one of Claims 1-11 by being performed by the computer contained in an information processing system.

Images