JP2011018111A - Information processing system and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information processing technology that displays, in a classified manner, a plurality of document data including natural texts according to the contents of the natural texts.SOLUTION: An information processing system includes: a database for storing a plurality of structured document data in which two or more elements constituting the natural texts are associated with item names, respectively; a query creating means for creating a search query including the information in which one or more item names and one or more elements related to the one or more item names are associated with each other; an extraction means for extracting the plurality of structured document data corresponding to the search query from the database; a classification means for classifying the plurality of structured document data extracted by the extraction means based on a classification condition indicating the one or more item names as reference of classification; and an output means which visually outputs information related to the plurality of structured document data extracted by the extraction means based on the classification result by the classification means.

Description

  The present invention relates to an information processing system and a program.

  In recent years, in medical institutions such as hospitals, an electronic medical record system for electronically storing patient medical information and the like has been introduced. In this electronic medical record system, a large amount of information can be stored without causing an increase in information storage space. However, this electronic medical record system is required to display a large amount of information in an easy-to-see manner. For example, when displaying the chart information of one patient, an aspect is desired in which only the necessary information can be handled while grasping the overall configuration of the medical information.

  In response to such a request, information on individual charts is hierarchized, and information related to items selected according to the operation of the doctor among a plurality of items constituting the chart is selectively expanded and displayed. A technique has been proposed (for example, Patent Document 1). In addition, a technique has been proposed in which, when referring to a report of a test result relating to a single patient, contents relating to the selected classification item are extracted and displayed in accordance with the selection of a report classification item by a doctor (for example, Patent Document 2).

  In addition, according to software and the like related to electronic mail, it is generally possible to display a large number of mails in a grouped state according to the sender, subject, etc.

JP-A-9-305671 JP 2002-92159 A

  By the way, in a medical information system that enables storage and browsing of medical information such as an electronic medical record system, the past medical information that has been accumulated is used to browse a list in which diagnoses are listed and to perform the same diagnosis. It would be useful if different cases could be compared.

  However, for example, with respect to so-called interpretation reports, since information related to diagnosis based on the judgment of a doctor is generally described in a natural sentence, in the techniques of Patent Documents 1 and 2, a plurality of interpretation reports are diagnosed. Cannot be displayed in a classified and grouped state as a reference. Also, a plurality of electronic mails cannot be displayed in a state where the contents of the mail body are classified and grouped.

  Such a problem is common in general applications in which a plurality of documents including natural sentences are classified and displayed according to the contents of the natural sentences.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide an information processing technique capable of classifying and displaying a plurality of document data including a natural sentence according to the content of the natural sentence. And

  In order to solve the above problem, an information processing system according to a first aspect includes a database that stores a plurality of structured document data in which two or more elements constituting a natural sentence are each associated with an item name, and a user In response to the operation, a query generation means for generating a search query including information in which one or more item names and one or more elements related to the one or more item names are associated, and from the database to the search query Extraction means for extracting a plurality of corresponding structured document data, and a classification for classifying the plurality of structured document data extracted by the extraction means on the basis of a classification condition indicating one or more item names serving as classification criteria And output means for visually outputting information on a plurality of structured document data extracted by the extraction means based on the classification result by the classification means. That.

  An information processing system according to a second aspect is the information processing system according to the first aspect, wherein each of the structured document data is described using RDF.

  The information processing system according to the third aspect is the information processing system according to the second aspect, and SPARQL is used as the language of the search query.

  An information processing system according to a fourth aspect is the information processing system according to any one of the first to third aspects, wherein the search query relates to the first item name and the first item name. It includes information in which one or more elements are associated and the second item name and one or more elements related to the second item name are associated.

  An information processing system according to a fifth aspect is the information processing system according to any one of the first to fourth aspects, wherein the search query relates to one or more item names and the one or more item names. It includes information in which one or more elements and detailed elements indicating detailed information relating to the one or more elements are associated with each other.

  An information processing system according to a sixth aspect is the information processing system according to any one of the first to fifth aspects, wherein the query generation unit is responsive to a user operation after the classification process by the classification unit. Thus, information relating to one or more other item names different from the one or more item names and one or more elements related to the one or more other item names is added to the search query. In this way, a modified search query is generated, and the extraction unit extracts a plurality of structured documents corresponding to the modified search query from a plurality of structured document data or the database already extracted by the extraction unit. The document data is extracted again, and the classification unit classifies the plurality of structured document data extracted again by the extraction unit based on the classification condition.

  An information processing system according to a seventh aspect is the information processing system according to any one of the first to fifth aspects, wherein the query generation means responds to a user action after the classification processing by the classification means. A detailed element indicating detailed information related to the one or more elements in information associated with the one or more item names included in the search query and one or more elements related to the one or more item names Are further associated with each other to generate a search query after the change, and the extraction unit can correspond to the search query after the change from the plurality of structured document data or the database already extracted by the extraction unit. The structured document data is extracted again, and the classifying unit classifies the plurality of structured document data extracted again by the extracting unit based on the classification condition.

  An information processing system according to an eighth aspect is the information processing system according to any one of the first to seventh aspects, wherein the classification condition includes two or more item names serving as classification criteria, and the 2 The priorities relating to the above item names are indicated, and the classification means converts the plurality of structured document data extracted by the extraction means into two or more item names serving as a reference for the classification based on the priorities. Classify hierarchically using such elements.

  An information processing system according to a ninth aspect is the information processing system according to the eighth aspect, wherein the output means uses the classification means to output information related to a plurality of structured document data extracted by the extraction means. A visual state in which information subordinate to one or more elements related to at least one layer is visually output according to a user's action when visually outputting based on a hierarchical classification result; and Set to one of the invisible states that are not visually output.

  An information processing system according to a tenth aspect is the information processing system according to the eighth or ninth aspect, wherein the output means outputs information related to a plurality of structured document data extracted by the extraction means. When visually outputting based on the hierarchical classification result by the classification means, display elements each indicating the number of structured document data classified for each element related to at least one hierarchy are displayed as classification destinations. Visually output in a form corresponding to the display element indicating each element.

  An information processing system according to an eleventh aspect is the information processing system according to any one of the eighth to tenth aspects, wherein the output means outputs a plurality of structured document data extracted by the extraction means. When the information is visually output based on the hierarchical classification result by the classification means, the information related to one or more display item names among the information corresponding to the two or more structured document data is stored in the table. In addition to being visually output in a form, information relating to each display item name is output visually in an ascending or descending order according to a predetermined rule in accordance with a user's operation.

  An information processing system according to a twelfth aspect is input in response to a user's operation and a database that stores a plurality of structured document data in which two or more elements constituting a natural sentence are each associated with an item name. A plurality of structured document data based on a classification condition indicating one or more item names serving as classification criteria, and a plurality of the plurality of structured document data based on a classification result by the classification unit. Output means for visually outputting information related to the structured document data.

  An information processing system according to a thirteenth aspect is the information processing system according to the twelfth aspect, wherein each structured document data is described using RDF.

  An information processing system according to a fourteenth aspect is the information processing system according to the twelfth or thirteenth aspect, wherein the classification condition includes two or more item names serving as a reference for classification, and the two or more item names The classification means classifies the plurality of structured document data hierarchically using elements related to two or more item names serving as the classification criteria based on the priority. .

  An information processing system according to a fifteenth aspect is the information processing system according to the fourteenth aspect, wherein the output means converts the information related to the plurality of structured document data into a hierarchical classification result by the classification means. When the information is visually output based on the user's operation, the information depending on the one or more elements related to at least one layer is visually output, and the information is visually output. Set to either non-visible state.

  An information processing system according to a sixteenth aspect is the information processing system according to the fourteenth or fifteenth aspect, wherein the output means hierarchically classifies information relating to the plurality of structured document data by the classification means. When visually outputting based on the classification result, a display element indicating the number of structured document data classified for each element related to at least one hierarchy, and a display element indicating each element of the classification destination Visually output in a form corresponding to.

  An information processing system according to a seventeenth aspect is the information processing system according to any one of the fourteenth to sixteenth aspects, wherein the output means assigns information related to the plurality of structured document data to the classification means. Visually output information related to one or more display item names in information corresponding to two or more structured document data in the form of a table when outputting visually based on the hierarchical classification result by In addition, according to the operation of the user, the information related to each display item name is visually output in an ascending or descending order according to a predetermined rule.

  An information processing system according to an eighteenth aspect is the information processing system according to any one of the eighth to eleventh and fourteenth to seventeenth aspects, wherein each structured document data has a plurality of item names. Such elements include data hierarchically related according to a predetermined dependency relationship, and the dependency relationship of elements hierarchically classified based on the priority by the classification means is different from the predetermined dependency relationship .

  An information processing system according to a nineteenth aspect is the information processing system according to any one of the first to eighteenth aspects, wherein two or more elements are recognized from a natural sentence, and each of the two or more elements is recognized. The apparatus further includes data generation means for generating the structured document data stored in the database by associating item names.

  An information processing system according to a twentieth aspect is the information processing system according to any one of the first to eighteenth aspects, and for two or more item names relating to a plurality of elements constituting a natural sentence, The apparatus further includes data generation means for generating each of the structured document data stored in the database by designating one or more elements.

  An information processing system according to a twenty-first aspect is the information processing system according to any one of the first to twentieth aspects, and serves as a reference for classification included in the classification condition in accordance with a user action. Setting means for setting the above item names is further provided.

  An information processing system according to a twenty-second aspect is the information processing system according to any one of the first to twenty-first aspects, wherein a plurality of structured document data stored in the database is included in a medical field report. Such structured document data is included.

  An information processing system according to a twenty-third aspect is the information processing system according to any one of the first to twenty-second aspects, in which item names relating to two or more elements constituting each of the natural sentences are the shooting conditions , At least one item name of basic site, basic findings, diagnosis, and treatment.

  A program according to a twenty-fourth aspect is executed by a control unit included in the information processing system, thereby causing the information processing system to function as the information processing system according to any one of the first to twenty-third aspects.

  With the information processing system according to any of the first to eleventh and eighteenth to twenty-third aspects, a plurality of structured document data corresponding to the generated search query is extracted, and the plurality of structured document data is Since it is visually output after being classified based on the classification condition, a plurality of document data including the natural sentence can be classified and displayed according to the content of the natural sentence.

  With the information processing system according to any of the fourth and fifth aspects, it is possible to narrow down the displayed information by increasing the search conditions.

  In the information processing system according to any of the sixth and seventh aspects, after confirming the contents of the plurality of structured document data extracted once, the displayed information is further narrowed down by increasing the search condition. be able to.

  The information processing system according to any of the eighth and fourteenth aspects displays information related to a plurality of structured document data in an easily understandable manner, so that the user can recognize desired information in a short time. can do.

  In accordance with the information processing system according to any of the ninth, eleventh, fifteenth, and seventeenth aspects, information related to a plurality of structured document data is displayed in a form that can be easily grasped visually according to the user's intention. Therefore, the user can recognize desired information in a short time.

  The number of structured document data classified into each element is indicated by the information processing system according to any of the tenth and sixteenth aspects, so that it is possible to easily grasp the elements for which only a small number of structured document data is classified. It becomes possible to do.

  In the information processing system according to any of the twelfth to twenty-third aspects, a plurality of structured document data stored in the database is visually output after being classified based on the classification condition. Can be classified and displayed according to the contents of the natural sentence.

  According to the information processing system according to the nineteenth aspect, when a natural sentence is input, each structured document data stored in the database is generated. Therefore, a database storing a plurality of structured document data can be easily created. Can be built.

  According to the information processing system according to the twentieth aspect, a structure for generating structured document data by analyzing a natural sentence becomes unnecessary, and thus a database in which a plurality of structured document data is stored can be constructed in a short time. can do.

  According to the information processing system in the twenty-first aspect, since a plurality of structured document data is classified and output based on the classification condition intended by the user, the user can efficiently output desired information. You can browse well.

  According to the information processing system concerning the 22nd mode, a user can grasp information on a medical field easily.

  According to the information processing system according to the twenty-third aspect, it is possible to easily grasp information about item names such as imaging conditions, basic parts, basic findings, diagnosis, and treatment.

  According to the program concerning the 24th mode, the same effect as the information processing system concerning the 1st to 23rd modes can be acquired.

It is a figure showing composition of an information processing system concerning one embodiment. It is a figure which shows the functional structure which concerns on the information processing system which concerns on one Embodiment. It is a figure which shows the functional structure which concerns on report input and construction of structured information DB. It is a figure for demonstrating the structuring process of a finding sentence. It is a figure for demonstrating the structuring process of a finding sentence. It is a figure which shows the specific example of structured report data. It is a figure which illustrates an inspection list screen. It is a figure which illustrates a report input screen. It is a figure which shows the example of a display which concerns on the detail of a test | inspection. It is a figure which shows the example of a display which concerns on today's memo. It is a figure which shows the example of a display which concerns on a patient memo. It is a figure which shows the example of a display which concerns on a summary phrase input. It is a figure which shows the example of a display concerning frequent phrase input. It is a figure which shows the example of a display of an input assistance template. It is a figure which shows the functional structure which concerns on a report search. It is a figure which shows the example of a display of a search condition input screen. It is a figure which shows the example of a display of the detailed information input screen which concerns on search conditions. It is a figure which shows the example of a display of the detailed information input screen which concerns on search conditions. It is a figure which shows the example of a display of a search result display screen. It is a figure which shows the example of a display of a search result display screen. It is a figure which shows the example of a display of a display pattern edit screen. It is a figure which shows the example of a display of a display pattern edit screen. It is a flowchart which shows the operation | movement flow of report search. It is a flowchart which shows the operation | movement flow of report search. It is a figure which shows the functional structure which concerns on the report reference of one modification. It is a figure which shows the example of a display which concerns on the input assistance template of one modification. It is a figure which shows the example of a display of the report reference screen which concerns on one modification. It is a flowchart which shows the operation | movement flow of the report reference which concerns on one modification. It is a flowchart which shows the operation | movement flow of the report reference which concerns on one modification.

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

<(1) Overall configuration of information processing system>
FIG. 1 is a diagram illustrating an overall configuration of an information processing system 1 according to an embodiment.

  The information processing system 1 includes a medical information server 100, a CR (Computed Radiography) apparatus 201, an ultrasonic diagnostic apparatus (US apparatus) 202, an endoscope apparatus (ES apparatus) 203, and terminals 301 to 303. The medical information server 100 is connected to the CR device 201, the US device 202, and the ES device 203 in a manner capable of transmitting and receiving data, and the medical information server 100 is connected to each other via the network line NTW. The terminals 301 to 303 are connected in such a manner that data can be transmitted and received.

  The medical information server 100 includes a medical information database (DB) 151 that stores medical information (hereinafter also referred to as “medical information”) related to a large number of patients who are medical targets. The medical care information DB 151 includes a list (examination list) in which a large number of examinations are listed.

  In the medical information DB 151, a large number of image data obtained by the examinations listed in the examination list, and report data indicating interpretation reports corresponding to the many image data, and the examination IDs listed in the examination list are displayed. Stored in association. The medical treatment information DB 151 stores attribute information indicating information related to the patient and the examination in association with the examination IDs listed in the examination list.

  Here, the image data stored in the medical treatment information DB 151 includes image data obtained by at least one inspection device such as the CR device 201, the US device 202, and the ES device 203. Note that the medical treatment information DB 151 is a database (report DB) in which a large number of report data is stored, focusing on the fact that a large number of report data is stored.

  In the medical information server 100, various information is written into the medical information DB 151 and various information is read from the medical information DB 151, and information processing based on the various information stored in the medical information DB 151 is performed.

  The terminals 301 to 303 are configured by, for example, a general personal computer (personal computer), and the terminals 301 to 303 have the same configuration. The terminals 301 to 303 are used by, for example, a radiology interpreting doctor.

  In each of the terminals 301 to 303, it is possible to input reports (herein, interpretation reports) for various examinations while visually displaying image data relating to the various examinations stored in the medical treatment information DB 151. Data related to the interpretation report (report data) is stored in the medical information DB 151 in association with image data to be interpreted.

  Each terminal 301 to 303 can use various information stored in the medical information server 100 when inputting an interpretation report.

  For example, in the medical information server 100, a plurality of past report data stored in the medical information DB 151 is structured, and a database (structured information) in which the plurality of structured report data (structured report data) is stored. DB) 152 is constructed. At this time, a table (term table) 153 in which terms used in a plurality of interpretation reports are listed for each item name is also constructed. Then, structured report data that matches the search conditions from the terminal 301 (302, 303) is extracted from the structured information DB 152, and the structured report data is visually output at the terminal 301 (302, 303).

  In addition, for example, the medical information server 100 provides each terminal 301 to 303 with one or more candidates (options) of terms for each item name of the elements constituting the interpretation report according to the information stored in the term table 153. . At this time, in each of the terminals 301 to 303, one or more term candidates are listed for each item name on the template that supports the input of the interpretation report.

  In other words, a user such as an interpretation doctor can search and browse past interpretation reports, and create an interpretation report by specifying a desired term from candidate terms on the template while referring to past information. can do.

  A function related to input of an interpretation report (report input), a function related to construction of the structured information DB 152, and a function of search of structured report data (report search) will be described later.

  FIG. 2 is a block diagram illustrating a functional configuration according to the information processing system 1 according to the embodiment.

  As shown in FIG. 2, the medical information server 100 has a general computer configuration in which a CPU 110 and a memory 150 are connected to a bus line 101. Note that the memory 150 appropriately includes a storage unit such as a RAM, a ROM, and a hard disk. The bus line 101 is connected to the network line NTW via an interface (I / F) 160. In addition, various functions and operations of the medical information server 100 are realized by the CPU 110 operating according to a program stored in the memory 150.

  The terminals 301 to 303 have the same configuration as each other, and each have a general computer configuration in which the CPU 310 and the memory 350 are connected to the bus line 601. The memory 350 appropriately includes a storage unit such as a RAM, a ROM, and a hard disk. The bus line 601 is connected to the network line NTW via an interface (I / F) 360, and is also connected to the display unit 330 and the operation unit 340 via an interface (not shown) as appropriate. The display unit 330 is configured by, for example, a liquid crystal display or a CRT, and the operation unit 340 has a keyboard, a mouse, and the like, and receives input of various signals in response to operations by the user.

  The terminals 301 to 303 operate according to a program stored in the memory 350 by the CPU 310, whereby various functions and operations in the terminals 301 to 303 are realized. As described above, since the terminals 301 to 303 have the same configuration, the terminal 301 will be described as a representative example as appropriate in this specification.

  Hereinafter, each function of (i) construction of the structured information DB 152 and report input and (ii) report search in the information processing system 1 will be sequentially described.

<(2) Construction of structured information DB and report input>
<(2-1) Functional configuration related to construction of structured information DB and report input>
FIG. 3 is a block diagram showing a functional configuration related to the construction of the structured information DB and the report input in the information processing system 1. Note that the functional configuration shown in FIG. 3 is realized by executing programs stored in a storage unit such as a hard disk on the medical information server 100 and the terminal 301 by a CPU or the like. In addition, various data temporarily generated in various information processing executed with the functional configuration is temporarily stored as appropriate in RAM or the like built in the medical information server 100 and the terminal 301, respectively.

  As illustrated in FIG. 3, the medical information server 100 includes a data reading unit 111, an interpretation completion determination unit 112, a structured processing unit 113, a support information acquisition unit 114, and a data writing unit 115 as functional configurations. Have. The medical information server 100 further includes a medical information DB 151, a structured information DB 152, and a term table 153 stored in the memory 150. In addition, the terminal 301 includes a report generation unit 311, a task management unit 312, and a display control unit 313 as functional configurations.

<(2-1-1) Construction of structured information DB>
The data reading unit 111 receives report data indicating an interpretation report, attribute information corresponding to the report data (for example, information related to patient attributes, examination attributes, and the like), and information (examination of examination) from the medical care information DB 151. List information). Then, the data reading unit 111 sends report data, attribute information, and examination list information to the interpretation completion determination unit 112.

  The interpretation completion determination unit 112 refers to the examination list information to determine whether or not an interpretation report has been completed by writing an observation sentence or the like for the interpretation report of each examination. Note that the examination list information includes information indicating whether or not an interpretation report has been completed by writing a finding sentence or the like. Then, the interpretation completion determination unit 112 sends the created report data to the structuring processing unit 113 together with the attribute information.

  The structured processing unit 113 extracts necessary elements (here, terms and the like) from the finding sentence that is a natural sentence included in the interpretation report, and also extracts various elements included in the attribute information, and RDF (Resource Description Framework) Structuring using. The structuring of elements of the interpretation report is realized by using machine learning.

  Here, the structuring of the machine learning and the interpretation report in the structuring processing unit 113 will be described.

  For example, the structuring processing unit 113 learns information serving as a structuring reference by providing a learning corpus or the like as teacher data.

  The learning corpus includes a large amount of text data according to the finding sentence format (sentence model) of the interpretation report. This sentence model is, for example, the structure of the interpretation sentence of the interpretation report, such as imaging conditions (imaging conditions-conclusion)-> basic part (basic part-conclusion)-> basic findings (feature-conclusion)-> diagnosis (diagnosis-conclusion) Indicates. In the learning corpus, for example, item names related to each element constituting the sentence sentence sentence model are tagged to each term or the like.

  Examples of element item names include “category”, “imaging condition”, “basic part”, “basic finding”, and “diagnosis”.

  In the learning corpus, for example, the item name “category” of the element is tagged to terms such as “bone”, “pleura (margin)”,. The element item name “imaging condition” is tagged to terms such as “frontal image”, “side image”, “lateral position image”,. The item name “basic part” of the element is tagged for terms such as “the whole lung field”, “upper lung field”, and so on. The element name “basic findings” is tagged for terms such as “uniform shadow”, “striped shadow”,. The item name “diagnosis” is tagged for terms such as “pulmonary edema”, “old pulmonary tuberculosis”,.

  Note that when the learning corpus is given to the structuring processing unit 113 from the outside, data indicating a typical sentence model in which the conclusion (conclusion) is specified to some extent is also given from the outside.

  The machine learning function of the structuring processing unit 113 extracts terms and the like from the learning corpus and stores them for each corresponding item name. That is, the structuring processing unit 113 uses the teacher data including the learning corpus as the learning material, learns and stores the term and the like related to each item name for each item name with reference to the teacher data. Here, the data that is learned and stored is used as model data (model data) that indicates how each element constituting an existing interpretation report is decomposed into elements according to what item names.

  Further, the structuring processing unit 113 uses the well-known SVM (Support Vector Machine) as a reference and interprets the interpretation report input to the structuring processing unit 113 as a target while using the model data learned as described above as a reference. Recognizes names and actual terms used.

  If the machine learning method described above is used, item names can be recognized only for elements (here, terms, etc.) given in advance in teacher data, but the following so-called SVM (Support Vector Machine) is used. By using the machine learning method used, it is possible to recognize item names for elements that were not given in advance by teacher data. For example, the learning corpus is decomposed into morphemes by morpheme analysis by machine learning using SVM, and the morpheme itself, the morpheme part of speech, the morpheme utilization form, the front and rear (for example, two before and after) morpheme information, etc. A part of model data is generated by learning a pattern (appearance pattern) in which a morpheme related to a certain item name appears using.

  The structured processing unit 113 interprets various information included in the report data and attribute information based on the model data by decomposing into terms (elements) for each item name and using RDF. Information related to the report is structured. Here, in particular, when two or more terms are recognized from the finding sentence related to the report data, structured item report data is generated by associating corresponding item names with the two or more elements. .

  4 and 5 are diagrams for specifically explaining a process (structuring process) for structuring a finding sentence written in a natural sentence.

  When the observation text shown in FIG. 4 (“Nodular shadow and ground glass shadow are recognized in the upper lung field in the front image. Suspected pneumonia”) shown in FIG. 4 is input to the structuring processing unit 113. As shown in the figure, the term “imaging condition” is connected to the term “frontal image” and “de” is recognized, and the item name “basic part” is connected to the term “upper lung field” and “ni” is recognized. The item name “basic findings” is connected to the terms “nodule shadow” and “ground glass shadow” and “recognizes”, and the item name “diagnosis” is connected to the term “pneumonia” and “suspect”. Is done.

  As described above, the finding sentence recognized by the structuring processing unit 113 as to the term and the conclusion associated with each item name is described in RDF in the structuring processing unit 113 as shown in FIG. This completes the structuring process. Here, the structuring process is performed in a form including not only the information related to the report data but also elements included in the attribute information corresponding to the report data. Here, examples of the elements included in the attribute information include a patient ID, a patient name, a pseudonym, sex, age, an image ID, an image storage location, an examination ID, and an examination date. In this way, structured report data as structured document data is generated. As shown in FIG. 6, each structured report data includes data in which elements related to a plurality of item names are hierarchically related according to a predetermined dependency relationship. Note that the item names constituting the structured report data are not limited to those shown in FIG. 6, and the types of item names may be adjusted.

  In this way, the structured sentence 113 is structured by the structured processing unit 113 as the data generation means so that two or more elements constituting the natural sentence can be identified by item. Structured report data is generated.

  Then, in the structured processing unit 113, structured report data as shown in FIG. 6 is generated for each of a large number of already-interpreted report data stored in the medical information DB 151. The structured information DB 152 is constructed by accumulating a large number of structured report data generated in this manner in the memory 150.

  Here, when the structured information DB 152 is constructed, a table (term table) 153 in which elements (terms and the like) related to the item names are listed for each item name in the memory 150 is generated.

  Information constituting the structured information DB 152 and the term table 153 is distinguishable for each combination of terms relating to the examination site (for example, CHEST and HEAD) and the modality (for example, CR, ES, US, etc.). It is stored in a manner that can be distinguished for each term related to the category.

  If there are many synonyms (for example, “T2 weighted image” and “T2W1”, etc.), the number of terms stored in the term table 153 increases too much. Therefore, synonyms are replaced with one representative term when the structuring processing unit 113 generates the term table 153. Such replacement of a phrase with a representative term is realized by including a table in which a plurality of terms and a representative term are associated with each other in the teacher data.

  In the interpretation report, for example, terms such as a small classification, a prefix (prefix), and a suffix (suffix) are often added to the item names “basic part” and “basic finding”. For example, “white matter of the right frontal lobe” is a combination of the term “frontal lobe” of the basic site, the term “white matter” of the minor classification, and the term “right” of the prefix. Therefore, the structured processing unit 113 relates to the detailed item names “subcategory”, “prefix”, and “suffix” with respect to the item names “basic part” and “basic finding” from the finding sentence that is a natural sentence. Terms (hereinafter also referred to as “detail elements”) are recognized and structured report data is generated in a manner that also includes such detail elements. The term table 153 is also generated in a manner including detailed elements.

  By the way, the interpretation report stored in the medical treatment information DB 151 increases every time a new interpretation report is generated according to the input from the terminals 301 to 303. It is effective if the interpretation reports 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, since past knowledge further evolves, the useful value is high. The addition of a new interpretation report will be described later. For example, for each item name of each element constituting the finding sentence, past knowledge evolves in response to designation of the element to be actually input.

<(2-1-2) Report input>
<(2-1-2-1) Designation of examination subject to report input>
FIG. 7 is a diagram exemplifying a screen (examination list screen) T0 used for the purpose of designating examinations to be created as interpretation reports when the interpretation reports are created in the respective terminals 301 to 303.

  The examination list screen T0 is displayed on the display unit 330 under the control of the display control unit 313 when a predetermined operation is performed after the predetermined software is started in each of the terminals 301 to 303. For example, examination list information stored in the medical information DB 151 is input to the terminal 301 via the data reading unit 111 in accordance with an operation of the operation unit 340 by an interpreting doctor who is a user of the terminal 301. Then, the inspection list information is visibly output on the display unit 330 under the control of the display control unit 313, thereby realizing the display of the inspection list screen T0.

  As shown in FIG. 7, the examination list screen T0 includes information on examinations (specifically, patient ID, patient name, date of birth, age, sex, state, examination ID, examination date, examination site, Modality, number of images) is displayed in a list. In this examination list screen T0, the examination with the state “uninterpreted” corresponds to the examination for which no interpretation report has been created. In addition, in the examination list screen T0, the thick frame cursor CS1 is moved up and down in accordance with the operation of the operation unit 340 by the radiogram interpreter, and the determination button (for example, the return key) of the operation unit 340 is set in accordance with the desired examination. ) Is designated, one inspection surrounded by the thick frame cursor CS1 is designated as an inspection to be interpreted (interpretation target inspection).

  As described above, when the interpretation target inspection is designated, the screen displayed on the display unit 330 is a screen on which the image data related to the interpretation target inspection is greatly visibly output from the inspection list screen T0 (inspection image display screen). Transition to. When a button for requesting display of a screen for inputting a report is pressed after an image is visually recognized by the user on the inspection image display screen, the screen displayed on the display unit 330 is changed to the report input screen T1. Transition to FIG. At this time, the task management unit 312 identifies a task relating to the creation of a new interpretation report corresponding to the interpretation target examination, and various information and data relating to the interpretation target examination are stored in the medical information DB 151 via the data reading unit 111. Is read.

<(2-1-2-2) Report input screen>
FIG. 8 is a diagram illustrating a report input screen T1. The report input screen T1 is mainly composed of areas A1 to A9.

  Details of the examination related to the interpretation target examination (for example, FIG. 9) are displayed in the area A1, today's memo related to the interpretation target examination (for example, FIG. 10) is displayed in the area A2, and the area A3 is displayed in the area A3. A memo (for example, FIG. 11) relating to the patient involved in the interpretation target examination is displayed, and a list of phrases (summary phrases) indicating an outline generally used in the interpretation report (for example, FIG. 12) is displayed in the area A4. The list (for example, FIG. 13) of fixed phrases (frequent phrases) frequently used in the interpretation report is displayed in the area A5.

  In the area A6, for example, an image related to the interpretation target examination is displayed, and information obtained from the term table 153 or the like is visually presented in the form of a template in order to support the creation of the interpretation report. In the area A7, finding sentences constituting the interpretation report are entered according to the operation of the interpreting physician.

  In the area A8, icons for inputting commands (here, a registration button, a viewer button, and a clear button) are listed. The registration button is a button for confirming and newly registering the findings and the contents of the attached image constituting the interpretation report created on the report input screen T1. The viewer button is a button for transitioning the screen displayed on the display unit 330 to the inspection image display screen. The clear button is a button for clearing the findings and the contents of the attached image that constitute the image interpretation report created on the report input screen T1. A representative image is attached to the area A9 in accordance with the operation of the interpretation doctor.

  Next, an operation for inputting a finding sentence using the display content in the area A6 will be described. When creating an interpretation report, the interpretation doctor visually confirms various images in the examination image display, enters a finding sentence, and attaches a representative image. For this reason, the image interpretation doctor visually operates various images on the examination image display screen and then performs various operations on the operation unit 340 to change the display screen of the display unit 330 to the report input screen T1, thereby inputting an image interpretation report. A template (input support template) for assisting the user is displayed on the display unit 330.

  FIG. 14 is a diagram illustrating a display example of the input support template displayed in the area A6. This input support template corresponds to the combination of the examination part and the modality related to the interpretation target examination. In FIG. 14, as an example, an input support template corresponding to the combination of the examination site “CHEST” and the modality “CR” is shown. Note that various inputs and designations in the input support template are performed under the control of the report generation unit 311 based on a signal input in response to an operation of the operation unit 340 by the interpretation doctor.

  As shown in FIG. 14, in the input support template, category designation buttons SL1 to SL6 are displayed at the top, option enumeration areas A11 to A14 are displayed at the center, and term addition buttons Ba11 to Ba14 are displayed at the bottom. Text boxes Bx11 to Bx14, knot lists DL11 to DL14, search mode button BR, input button BT11, and reset button BT12 are displayed.

  The category designation buttons SL1 to SL6 are buttons for selectively designating category terms related to the interpretation report to be input. In general, from the term table 153, one or more terms in the category corresponding to the examination site and modality related to the interpretation target examination are recognized, so that the one or more terms can be selectively specified. A button is shown. In the input support template shown in FIG. 14, the term of the category related to the interpretation report to be input is obtained by selectively pressing any one of the six category designation buttons SL1 to SL6 with the mouse pointer MP. Is selectively specified.

  In the input support template shown in FIG. 14, one term is designated from among the six terms “bone”, “pleura (margin)”, “pleura”, “lung”, “mediastinum”, and “soft part” related to the category. Here, when a term of a desired category is designated, the data of the term relating to each item name classified with respect to the term of the designated category is obtained from the term table 153 by the support information acquisition unit 114 from the terminal 301. Sent to. Based on the term data relating to each item name obtained at this time, visible output in the option listing areas A11 to A14 is realized in the display unit 330 under the control of the display control unit 313.

  The option enumeration areas A11 to A14 are large areas that occupy the central portion of the input support template, and are sequentially arranged in this order from left to right. Specifically, a plurality of term options (for example, a front image, a side image, and a lateral position image) related to the item name “imaging condition” are listed in order from the top in the option listing area A11. In the option enumeration area A12, options (for example, the entire lung field, the upper lung field, the middle lung field,...) Relating to the item name “basic part” are listed in order from the top. In the option enumeration area A13, options of a plurality of terms relating to the item name “basic findings” (for example, uniform shadow, striped shadow, granular shadow,...) Are listed in order from the top. In the option listing area A14, a plurality of term options (for example, pulmonary edema, old pulmonary tuberculosis, cancerous lymphangiopathy,...) Related to the item name “diagnosis” are listed in order from the top. That is, in the option enumeration areas A11 to A14, a plurality of term options relating to the item names “imaging condition”, “basic part”, “basic findings”, and “diagnosis” are presented in a manner that can be distinguished for each item.

  In the option enumeration area A11, when a check mark is entered in the check box CB provided to the left of the option of a desired term by the mouse pointer MP in accordance with the operation of the operation unit 340 by the interpretation doctor, the item name “imaging condition” ”Is designated as a desired term option. At this time, the designated term is displayed in the text box Bx11 as a term relating to the item name “imaging condition”.

  In the option listing area A12, when a check mark is entered in a check box CB provided to the left of a desired term option by the mouse pointer MP in accordance with the operation of the operation unit 340 by the interpretation doctor, the item name “basic part” ”Is designated as a desired term option. At this time, the designated term is displayed in the text box Bx12 as the term relating to the item name “basic part”.

  In the option enumeration area A13, when a check mark is entered in a check box CB provided to the left of a desired term option by the mouse pointer MP in accordance with the operation of the operation unit 340 by the interpretation doctor, the item name “basic finding” ”Is designated as a desired term option. At this time, the designated term is displayed in the text box Bx13 as the term relating to the item name “basic findings”.

  In the option enumeration area A14, when a check mark is entered in the check box CB provided to the left of a desired term option by the mouse pointer MP in accordance with the operation of the operation unit 340 by the interpretation doctor, the item name “diagnosis” is displayed. For this, the desired term option is designated. At this time, the designated term is displayed in the text box Bx14 as the term relating to the item name “diagnosis”.

  In the option enumeration areas A11 to A14, when a detailed element designation button PS provided on the right side of each term option is pressed by the mouse pointer MP, a modification such as a prefix or a suffix that modifies the designated term. A screen on which detailed elements including words can be input is displayed on the display unit 330. Here, the input prefix and suffix are displayed in the corresponding text boxes Bx11 to Bx14 in a manner that modifies the specified term. For example, for the item name “basic part”, the prefix “left”, “right”, “upper part”, “lower part”, etc. can be entered, and for the item name “basic findings”, the modifier “pale” "Dark" can be input. In addition, with respect to the choice of the term with the modifier icon PI, when the modifier icon PI is pressed by the mouse pointer MP, the specified term is modified in the same manner as when the detailed element specification button PS is pressed. A screen capable of inputting detailed elements including modifiers such as prefixes and suffixes is displayed on the display unit 330.

  In each of the text boxes Bx11 to Bx14, an arbitrary term can be directly entered by causing a cursor to appear and inputting with a keyboard or the like.

  The term addition button Ba11 is a button for registering the term entered in the text box Bx11 as a term relating to the item name “imaging condition” in the term table 153. The term addition button Ba12 is a button for registering the term entered in the text box Bx12 as a term relating to the item name “basic part” in the term table 153. The term addition button Ba13 is a button for registering the term entered in the text box Bx13 as a term relating to the item name “basic findings” in the term table 153. The term addition button Ba14 is a button for registering the term entered in the text box Bx14 as a term relating to “diagnosis” entered in the term table 153.

  The knot list DL11 is a list for selectively designating knot words related to the item name “shooting condition”. The knot list DL12 is a list that selectively specifies knot words related to the item name “basic part”. The knot list DL13 is a list for selectively designating knot words related to the item name “basic findings”. The knot list DL14 is a list that selectively specifies knot words related to the item name “diagnosis”. Each knot list DL11 to DL14 has a so-called pull-down list format in which one knot word can be selectively specified while being pulled down by the mouse pointer MP.

  The input button BT11 is a button for instructing input of a finding sentence. When the input button BT11 is pressed with the mouse pointer MP in the state where terms are entered in the four text boxes Bx11 to Bx14 related to the four item names “imaging condition”, “basic part”, “basic findings”, and “diagnosis”, respectively. A finding sentence is generated based on the terms entered in the four text boxes Bx11 to Bx14 and the conclusions specified in the conclusion lists DL11 to DL14. Here, the finding sentence is generated based on the sentence model related to the finding sentence of the interpretation report under the control of the report generating unit 311. The finding sentence generated here is automatically entered in the area A7 (FIG. 8) of the report input screen T1. It is also possible to directly enter an arbitrary finding sentence by causing the cursor to appear in the area A7 and inputting from a keyboard or the like.

  The reset button BT12 is a button for inputting an instruction to reset the input state in the input support template. When the reset button BT12 is pressed with the mouse pointer MP, the terms entered in the text boxes Bx11 to Bx14 are deleted, and the knot words specified in the knot lists DL11 to DL14 are in a default state.

  The search mode button BR is a button for shifting to a mode for searching an interpretation report (search mode) in order to refer to a past interpretation report when entering a finding sentence. Here, when the search mode button BR is pressed with the mouse pointer MP, the search mode is entered. The report search in the search mode will be further described later.

  By the way, when the finding sentence is written in the area A7 and the image is attached to the area A9, when the registration button in the area A8 is pressed with the mouse pointer MP, the contents of the interpretation report are confirmed. Then, the data writing unit 115 stores report data relating to the new interpretation report in the medical treatment information DB 151. At this time, the screen displayed on the display unit 330 transitions from the report input screen T1 to the examination list screen T0 (FIG. 7).

  Here, when a new interpretation report is generated by entering a finding sentence in the area A7 based on the terms entered in the text boxes Bx11 to Bx14, the interpretation report is already structured. It becomes. That is, in the report generation unit 311 as a data generation unit, one or more terms are specified for each of two or more item names related to a plurality of elements constituting a natural sentence, and the structured information DB 152 accumulates them. Each structured report data to be generated is generated. In such a case, the data writing unit 115 updates the stored contents of the structured information DB 152 and the term table 153 based on each structured report data generated by the report generating unit 311.

  On the other hand, when new report data is generated by directly entering the finding text in the area A7, the new report data is stored in the medical information DB 151 by the data writing unit 115. Then, the structured processing unit 113 generates structured report data by structuring the new report data, and updates the storage contents of the structured information DB 152 and the term table 153. In such a process, when a finding sentence is input, each structured report data stored in the structured information DB 152 is automatically generated. For this reason, the structured information DB 152 in which a plurality of structured report data is accumulated is easily constructed.

  However, from the viewpoint of constructing the structured information DB 152 in which a plurality of structured report data is accumulated in a short time, it is preferable that each report data is generated by the report generation unit 311. This is because such a configuration eliminates the need for generating structured report data by analyzing natural sentences.

  Note that when the clear button in the area A8 is pressed with the mouse pointer MP, the contents of the information input and attached on the report input screen T1 are all cleared.

<(3) Report search>
When the search mode button BR is pressed with the mouse pointer MP on the report input screen T1, the terminal 301 is set to the search mode. Hereinafter, functions and operations related to report search in this search mode will be described in sequence.

<(3-1) Functional configuration related to report search>
FIG. 15 is a diagram illustrating a functional configuration related to report search in the information processing system 1. The functional configuration shown in FIG. 15 is realized by executing programs stored in a storage unit such as a hard disk by the CPU or the like in the medical information server 100 and the terminal 301, respectively. In addition, various data temporarily generated in various information processing executed with the functional configuration is temporarily stored as appropriate in RAM or the like built in the medical information server 100 and the terminal 301, respectively.

  As illustrated in FIG. 15, the medical information server 100 includes a report extraction unit 116 as a functional configuration. The terminal 301 includes a query generation unit 314, an information classification unit 315, and a classification condition setting unit 316 as functional configurations. Note that the terminal 301 includes the display control unit 313 as described above, and the display control unit 313 controls display of various screens related to report search.

<(3-1-1) Enter search conditions>
The query generation unit 314 generates a query (search query) related to the search condition in response to various signals input according to the operation of the operation unit 340 by the user. Here, the search query includes information indicating a search condition in which one or more item names are associated with one or more terms related to the one or more item names. Here, since the structured information DB 152 is described using RDF, examples of the language used for the search query include a so-called SPARQL (sparkle) language. When the structured information DB 152 is described in the RDB (Relational database) format, a language such as a so-called SQL (Structured Query Language) can be cited as a language used for the search query.

  FIG. 16 is a diagram illustrating a display example of the search condition input screen T2 displayed on the display unit 330 of the terminal 301. The search condition input screen T2 is a screen for the interpretation doctor to input the search conditions. When the search mode button BR is pressed with the mouse pointer MP on the report input screen T1, the search condition input screen T2 is displayed in the area A6 of the report input screen T1. Is done.

  As shown in FIG. 16, on the search condition input screen T2, a plurality of preset item names “modality”, “examination part”, “category”, “imaging condition”, “basic part”, “basic finding”, and “diagnosis” You can enter a term for each. Specifically, the term “modality” is entered in the text box Bx1, the term “test site” is entered in the text box Bx2, and the term “category” is entered in the text box Bx3. In the text box Bx4, the term “shooting condition” is entered. Further, the term “basic part” is entered in the text box Bx5, the term “basic findings” is entered in the text box Bx6, and the term “diagnosis” is entered in the text box Bx7.

  In each text box Bx1 to Bx7, for example, an arbitrary term is entered by an input operation using the keyboard of the operation unit 340 by an interpreting doctor. However, for the item names “modality” and “examination part”, in the initial state where the search condition input screen T2 is displayed, the terms related to the interpretation target examination are entered in the text boxes Bx1 and Bx2. The item name “category” is specified on the report input screen T1 in the initial state where the search condition input screen T2 is displayed, as in the text boxes Bx1 and Bx2 of the item names “modality” and “examination part”. Thus, the currently selected term is entered in the text box Bx3. Furthermore, the item names “imaging condition”, “basic part”, “basic finding”, and “diagnosis” correspond in the initial state in which the search condition input screen T2 is displayed if terms are specified on the report input screen T1. Terms specified in the item text boxes Bx4 to Bx7 are entered.

  Further, on the search condition input screen T2, for terms relating to the item names “basic part” and “basic findings”, it is possible to input detailed elements (modifiers, etc.) that qualify terms such as prefixes and suffixes. Yes.

  For example, when the detail button Db1 on the search condition input screen T2 is pressed with the mouse pointer MP in the state where the term relating to the item name “basic part” is described in the text box Bx5, the screen (detailed information) shown in FIG. Input screen) T3a is displayed in area A6 of report input screen T1. On the detailed information input screen T3a, it is possible to specify the detailed elements (here, prefix and suffix) that qualify the terms described in the text box Bx5 by using the prefix list DL31 and suffix list DL32 in the pull-down list format. it can.

  Further, for example, when the detail button Db2 of the search condition input screen T2 is pressed with the mouse pointer MP in the state where the term relating to the item name “basic findings” is described in the text box Bx6, the screen ( Detailed information input screen T3b is displayed in area A6 of report input screen T1. On the detailed information input screen T3b, a detailed element (in this case, a prefix) that modifies the term described in the text box Bx6 can be specified by the prefix list DL33 in the pull-down list format.

  On the detailed information input screens T3a and T3b, when the cancel button BT32 is pressed with the mouse pointer MP, the display of the detailed information input screens T3a and T3b is simply ended. On the other hand, when the decision button BT31 is pressed with the mouse pointer MP, the modifiers specified in the prefix lists DL31 and DL33 and the suffix list DB32 are added to the terms originally entered in the text boxes Bx5 and Bx6. Are described in text boxes Bx5 and Bx6. In the text boxes Bx1 to Bx7, two or more terms may be entered at intervals.

  In the search condition input screen T2, when the cancel button BT22 is pressed with the mouse pointer MP, the display of the search condition input screen T2 is terminated, and the state returns to the state where the input support template shown in FIG. 14 is displayed in the area A6. . On the other hand, when the search execution button BT21 is pressed with the mouse pointer MP, the query generation unit 314 generates a search query based on the conditions entered in the search condition input screen T2.

  By inputting such search conditions, item names of one or more item names of a plurality of item names “modality”, “examination part”, “category”, “imaging condition”, “basic part”, “basic finding”, and “diagnosis” are displayed. And a search query including information in which one or more terms relating to the item name are associated with each other. Here, when terms are specified for more than two item names, the contents of the search condition increase. In such a case, the information displayed in the search result display described later is narrowed down.

  For one or more and two or less item names corresponding to some item names, one or more detailed elements (modifiers, etc.) can be added to one or more terms. Accordingly, the search query may include information in which one or more item names and detailed elements indicating detailed information related to one or more elements belonging to the one or more item names are associated. In such a case, the information displayed in the search result display described later is narrowed down.

<(3-1-2) Search processing and search result classification display>
In response to the input of the search query from the query generation unit 314, the report extraction unit 116 extracts one or more structured report data corresponding to the search query from the structured information DB 152. For example, as shown in FIG. 16, the term “CR” is associated with the item name “modality”, the term “CHEST” is associated with the item name “examination site”, and the item name “category” is associated with the item name “category”. The term “lung” is associated, the term “imaging condition” is associated with the term “front view”, and the item name “basic findings” is associated with the term “ground glass shadow”. If included, structured report data that satisfies the conditions relating to these pieces of information is extracted from the structured information DB 152.

  When a plurality of structured report data is extracted by the report extraction unit 116, the information classifying unit 315 classifies the plurality of structured report data based on the classification condition. Here, the classification condition is defined by display pattern information (display pattern information), which will be described later, and includes one or more item names relating to terms constituting the finding sentence, and serves as a reference for classification of a plurality of structured report data. . The display pattern information is stored in the memory 350. Then, the display control unit 313 performs control so that the plurality of structured report data extracted by the report extraction unit 116 is visually output on the display unit 330 based on the classification result by the information classification unit 315.

  FIG. 19 is a diagram illustrating a display example of the search result display screen T4 displayed on the terminal 301. The search result display screen T4 is a screen on which search results are displayed, and is a screen on which a plurality of structured report data extracted by the report extracting unit 116 and classified by the information classifying unit 315 are visually output.

  As shown in FIG. 19, a pull-down list (display pattern list) DL41 for selectively setting a display pattern and a mode for editing a display pattern (display pattern editing) are displayed at the top of the search result display screen T4. Button (pattern edit button) BT41 for shifting to (mode). In addition, a list (search result list) L40 indicating search results is shown in the center of the search result display screen T4. In addition, a button BT42 for requesting re-input of search conditions is shown at the bottom of the search result display screen T4.

  The display pattern is set to a standard pattern in the initial state. Then, the display pattern is set to a desired pattern by operating the display pattern list DL41 using the mouse pointer MP. Here, the display pattern information includes an item name (tree display item name) in which elements (here, terms) are hierarchically displayed in a tree structure format in the search result list L40, and an item in which elements are displayed in a list format. Name (list display item name) is specified.

  For example, when the search result display screen T4 shown in FIG. 19 is displayed on the display unit 330, item names “basic part” and “diagnosis” are defined as tree display item names in the standard pattern set as the display pattern. Then, item names “findings”, “examination ID”, “examination date”, “patient ID”, and “patient name” are defined as list display item names. As for tree display item names, when two or more item names are determined, a priority order related to the two or more item names is determined. FIG. 19 shows a display example when the priority of tree display item names is set in the order of item names “basic part” and “diagnosis”.

  Therefore, here, the tree display item name in the display pattern information and the information related to the priority order function as classification conditions. For this reason, in the information classification unit 315, a plurality of structured items extracted by the report extraction unit 116 based on tree display item names and their priorities among patterns (for example, standard patterns) set as display patterns. Report data is classified hierarchically in the form of a tree structure for terms relating to two or more item names defined as tree display item names.

  For example, one or more groups (upper groups) are formed by classifying a plurality of structured report data for each term on the basis of including a term relating to the item name “basic part”. Further, the structured report data constituting each upper group is classified for each term on the basis that it includes the term related to the item name “diagnosis”, so that one or more groups (lower groups) can be obtained. It is formed.

  In the search result display screen T4 shown in FIG. 19, the search result list L40 includes an area A41 indicating a tree structure of terms related to the tree display item names, and areas A42 to A4 in which elements related to the list display item names are sequentially listed. It is comprised by A46.

  In the region A41, for example, as shown in FIG. 19, the terms “bronchial asthma” and “pneumonia” related to the item name “diagnosis” are subordinate to the term “upper lung field” related to the item name “basic site”. Is displayed. Note that the dependency relationship of terms hierarchically classified by the information classification unit 315 based on the priority order related to the tree display item name is different from the predetermined dependency relationship in each structured report data shown in FIG. The information related to the structured report data including the term “upper lung field” relating to the item name “basic part” and including the terms “bronchial asthma” and “pneumonia” relating to the item name “diagnosis” Is indicated to be subordinate to a display element indicating each corresponding term.

  Specifically, information relating to one structured report data is shown for each line from the area A42 to the area A46. Here, the observation text is shown in the area A42, the examination ID is shown in the area A43, the examination date is shown in the area A44, the patient ID is shown in the area A45, and the patient name is shown in the area A46. Accordingly, in FIG. 19, the elements related to the list display item names “findings”, “examination ID”, “examination date”, “patient ID”, and “patient name” constituting each structured report data are displayed so as to form one line. Is done. Note that the display order of the elements related to the list display item names is defined by the display pattern information.

  By such classification and hierarchical display, it is possible to easily grasp how terms related to tree display item names are used in past interpretation reports. In particular, by the hierarchical classification and display, a plurality of structured report data is displayed in a manner that is easy to visually grasp. For example, when a report search is performed according to a search condition that specifies the term “ground glass shadow” for the item name “basic findings”, the search result display screen T4 (FIG. 19) includes an interpretation report in which ground glass shadows are described as basic findings. , It is displayed in a manner that makes it clear at a glance what parts and diagnosis are included. For this reason, the image interpretation doctor can find desired information accurately and in a short time. That is, the interpreting physician can effectively and efficiently use the past knowledge constituted by a large number of interpretation reports.

  In the area A41, display elements DC1 and DC1 respectively indicate the number of structured report data classified for the terms “bronchial asthma” and “pneumonia” related to one item name “diagnosis” included in the tree display item name. DC2 is visually output in a form corresponding to the right side of the display elements of the terms “bronchial asthma” and “pneumonia”. FIG. 19 shows an example in which the number of structured report data belonging to “bronchial asthma” is two and the number of structured report data belonging to “pneumonia” is six. Such display of the number of cases enables the image interpretation doctor to easily understand the terms related to the tree display item names into which only a small number of structured report data are classified without oversight. Note that such display of the number of cases is not limited to being performed for terms related to one layer, but may be performed for terms related to one or more layers, such as terms related to two layers.

  In the area A41, for one item name “diagnosis” included in the tree display item name, the minus icons CO1 and CO2 correspond to the left side of the display elements of the terms “bronchial asthma” and “pneumonia”. Output visually. The minus icons CO1 and CO2 indicate a state in which information related to the list display item name of the structured report data is visually output in the areas A42 to A46 for each term related to the item name “diagnosis” (hereinafter referred to as “development state”). Both are also referred to as “visible state”).

  The minus icons CO1 and CO2 are changed to plus icons when operated with the mouse pointer MP in accordance with the operation of the operation unit 340 by the user. For each term related to the tree display item name in which the minus icon is changed to the plus icon, information related to the list display item name of the structured report data is not visually output to the areas A42 to A46 (hereinafter referred to as “folding”). State "or" invisible state ").

  For example, when the minus icon CO1 related to the term “bronchial asthma” shown in FIG. 19 is operated with the mouse pointer MP, it is changed to a plus icon CC1 as shown in FIG. At this time, the display mode of the search result list L40 shifts from the expanded state where the information related to the term “bronchial asthma” is visually output to the regions A42 to A46 to the folded state where the information is not output visually. Set to

  The plus icon is changed to a minus icon when operated with the mouse pointer MP in accordance with the operation of the operation unit 340 by the user. At this time, for each term related to the tree display item name in which the plus icon is changed to the minus icon, information related to the list display item name of the structured report data is not visually output to the areas A42 to 46 ( It is set so as to shift from a (folded state) to a visually output state (deployed state).

  By switching between the expanded state and the folded state according to the intention of the interpreting doctor, information related to a plurality of structured report data is displayed in a manner that is easy to visually grasp. For this reason, the image interpretation doctor can recognize desired information in a short time. Here, each term related to one tree display item name can be set to an expanded state and a collapsed state, but each term related to one or more tree display item names is expanded. What is necessary is just to be able to set to a state and a folding state.

  In the search result display screen T4, each time the display element of each list display item name in the search result list L40 is operated with the mouse pointer MP in accordance with the operation of the operation unit 340 by the user, the list display item name is displayed. The display mode of the information to which it belongs is switched between a state that is visually output in an ascending order according to a predetermined rule and a state that is visually output in a descending order according to a predetermined rule. At this time, for each structured report data, the display order is switched for the display of one line in which information related to the structured report data is displayed. However, the switching between the ascending order mode and the descending order mode is performed within the range of information of two or more structured report data belonging to the term of the lowest item name of the tree display item name.

  The predetermined rule related to the list display item name “finding sentence” includes, for example, an arrangement in the order of Japanese syllabary. The predetermined rule related to the list display item name “examination ID” includes, for example, alphabetical order. As an example of the predetermined rule for the list display item name “examination date”, for example, an arrangement in the order of schedules can be given. The predetermined rule for the list display item name “patient ID” includes, for example, an array of combinations in alphabetical order and numerical order, and the predetermined rule for the list display item name “patient name” includes, for example, five An arrangement in the order of ten letters is given.

  According to such switching between the ascending order mode and the descending order mode, information related to a plurality of structured report data is displayed in a manner that is easy to visually grasp. For this reason, the image interpretation doctor can recognize desired information in a short time. In addition, although it was possible to set the aspect of an ascending order and the aspect of a descending order about five list display item names here, respectively, it should be set as the aspect of an ascending order and the aspect of a descending order about one or more list display item names. If possible.

  In the search result display screen T4, when the setting of the display pattern type is changed in the display pattern list DL41 using the mouse pointer MP, the report extraction unit 116 corresponding to the search result performs the search result information according to the changed display pattern information. The classification of the two or more extracted structured report data is performed again by the information classification unit 315. Then, the classification result is reflected in the display contents of the search result list L40. That is, every time the setting of the display pattern type is changed, the display mode of the search result list L40 is changed according to the changed display pattern information.

<(3-1-3) Edit display pattern>
When the search result display screen T4 shown in FIGS. 19 and 20 is displayed in the area A6, when the pattern editing button BT41 is pressed with the mouse pointer MP by the operation of the operation unit 340 by the interpretation doctor, the display pattern is displayed. Is set to the display pattern editing mode. When the display pattern editing mode is set, a screen (display pattern editing screen) for editing the display pattern is displayed on the area A6.

  FIG. 21 is a diagram showing a display example of the display pattern editing screen T5 displayed on the display unit 330 of the terminal 301 (302, 303). In the state where the display pattern editing screen T5 is displayed on the display unit 330, the classification condition setting unit 316 configures the display pattern in response to various signals input in response to the operation of the operation unit 340 by the interpretation doctor 1 The above item names are set. That is, one or more item names serving as classification criteria included in the classification conditions are set.

  As shown in FIG. 21, a pull-down list format list (editing object selection list) DL51 for selecting a display pattern to be edited is provided in the upper left part of the display pattern editing screen T5. In the upper right part of the display pattern editing screen T5, a check box CB5 for designating the display pattern selected in the edit target selection list DL51 as a default display pattern is provided. An area A51 where the tree display item name is displayed and an area A52 where the list display item name is displayed are provided in the left center portion of the display pattern editing screen T5. An area A53 in which candidates (options) set as tree display item names and list display item names are provided in the right center portion of the display pattern editing screen T5. In addition, a delete button BT51, a save button BT52, an alias save button BT53, and a close button BT54 are provided at the bottom of the display pattern edit screen T5.

  In the edit target selection list DL51, only the standard pattern is displayed in the initial state, but the display pattern that is displayed each time a new display pattern is saved with another name by pressing the alias save button BT53 with the mouse pointer MP. The number of types increases.

  When the check box CB5 is checked and the save button BT52 or the alias save button BT53 is pressed with the mouse pointer MP, the type of display pattern displayed in the edit target selection list DL51 is designated as the default display pattern. Is done.

  One or more item names displayed in the area A51 are tree display item names, and the order of display from the top indicates the priority. When two or more item names are displayed in the area A51, two or more items displayed in the area A51 are displayed by performing a drag and drop operation of the display element of each item name using the mouse pointer MP. The display order of item names is changed.

  In addition, when one item name among a plurality of item names listed in the area A53 is pressed with the mouse pointer MP, the one item name is highlighted. When the first additional button AD51 is pressed with the mouse pointer MP while the one item name is highlighted, the highlighted one item name is moved to the area A51. Added to the tree display item name. On the other hand, when one item name of a plurality of item names listed in the area A51 is pressed with the mouse pointer MP, the one item name is highlighted. When the first exclusion button EX51 is pressed with the mouse pointer MP while the one item name is highlighted, the highlighted one item name is moved to the area A53. Excluded from the tree display item name.

  FIG. 22 illustrates the display pattern editing screen T5 after the item name “basic findings” is added to the tree display item name in the display pattern editing screen T5 shown in FIG.

  One or more item names displayed in the area A52 are list display item names, and the order of display in order from the top indicates the priority order. When two or more item names are displayed in the area A52, two or more items displayed in the area A52 are displayed by dragging and dropping the display element of each item name using the mouse pointer MP. The display order of item names is changed.

  In addition, when one item name among a plurality of item names listed in the area A53 is pressed with the mouse pointer MP, the one item name is highlighted. When the second additional button AD52 is pressed with the mouse pointer MP while the one item name is highlighted, the highlighted one item name is moved to the area A52. Added to the list display item name. On the other hand, when one item name of a plurality of item names listed in the area A52 is pressed with the mouse pointer MP, the one item name is highlighted. When the second exclusion button EX52 is pressed with the mouse pointer MP while the one item name is highlighted, the highlighted one item name is moved to the area A53. Excluded from the list display item name.

  When the save button BT52 is pressed with the mouse pointer MP, the edit target selected in the edit target selection list DL51 is displayed according to the item names entered in the areas A51 and A52 of the display pattern edit screen T5. The pattern information is saved. That is, the display pattern information is updated. When the alias save button BT53 is pressed with the mouse pointer MP, a new display pattern name is input and a new display corresponding to the item name entered in the areas A51 and A52 of the display pattern editing screen T5 is displayed. The pattern information is saved.

  When the delete button BT51 is pressed with the mouse pointer MP, information on the display pattern selected in the edit target selection list DL51 is deleted. However, the standard pattern information is set not to be deleted. When the close button BT54 is pressed with the mouse pointer MP, the display of the display pattern editing screen T5 is terminated, and the screen displayed on the area A6 becomes the search result display screen T4.

  By editing the display pattern in this manner, a plurality of structured report data is visibly output on the search result display screen T4 after being classified based on the classification condition intended by the interpretation doctor. For this reason, it becomes possible for an image interpretation doctor to browse desired information efficiently.

  When the close button BT54 is pressed with the mouse pointer MP and the screen displayed on the area A6 transitions from the display pattern edit screen T5 to the search result display screen T4, the display pattern list of the search result display screen T4. If the display pattern selectively set in the DL 41 has been changed in the display pattern editing screen T5, two or more extracted by the report extraction unit 116 corresponding to the search result according to the changed display pattern information. The classification of the structured report data is performed again by the information classification unit 315. Then, the classification result is reflected in the display contents of the search result list L40.

<(3-1-4) Re-search>
In a state where the search result display screen T4 shown in FIG. 19 and FIG. 20 is displayed in the area A6, a button BT42 for requesting re-input of the search condition is operated by the operation of the operation unit 340 by the interpretation doctor. When pressed by MP, the search condition input screen T2 shown in FIG. 16 is displayed in the area A6 of the report input screen T1. At this time, on the search condition input screen T2, terms related to the previous search condition are entered in the text boxes Bx1 to Bx7. For this reason, in the search condition input screen T2, it is possible to add information in which the item name is associated with one or more terms belonging to the item name to the previous search condition.

  That is, the term is specified in the previous search query on the search condition input screen T2 by the operation of the operation unit 340 by the interpretation doctor after the plurality of structured report data is once classified by the information classification unit 315. One or more terms can be specified for one or more different item names different from the above item names.

  In adding such search conditions, the query generation unit 314 associates one or more different item names with one or more terms related to the one or more different item names with respect to the previous search query. Is added, and the search query after the change is generated.

  In addition, when such a search condition is added, information in which detailed elements (modifiers, etc.) indicating detailed information related to the one or more terms are appropriately added to the search condition. Is done. In addition to adding search conditions, it is of course possible to correct a part of the previous search conditions.

  Here, when the search execution button BT21 is pressed with the mouse pointer MP, the query generation unit 314 generates a search query based on the search condition changed on the search condition input screen T2. Then, another search process based on the changed search query is performed. For this second search process, search result classification, and display of the search result after classification, the same process as described above is performed.

  With such a configuration, a plurality of structured report data once extracted are visually output, and after the contents are confirmed, the interpretation doctor increases the search conditions to update the information displayed as the search results. Narrowing is performed.

<(3-2) Operations related to report search>
23 and 24 are flowcharts showing an operation flow of report search in the information processing system 1. This operation flow is realized by executing a predetermined program in the CPU 110 of the medical information server 100 and the CPU 310 of the terminals 301 to 303. Here, when the report input screen T1 (FIG. 8) is displayed on the terminal 301, when the search mode button BR is pressed with the mouse pointer MP, the terminal 301 is set to the search mode, and the operation flow is as follows. The process proceeds to step ST1 in FIG.

  In step ST1, the search condition input screen T2 is displayed on the display unit 330 by the display control unit 313.

  In step ST2, the query generation unit 314 accepts input of terms on the search condition input screen T2 or the like, that is, search conditions.

  In step ST3, the query generation unit 314 determines whether or not the report search is canceled. Here, when the cancel button BT22 is pressed with the mouse pointer MP on the search condition input screen T2, it is determined that the report search has been canceled, and this operation flow ends. On the other hand, if the report search is not canceled, the process proceeds to step ST4.

  In step ST4, the query generation unit 314 determines whether a search is started. Here, when the search execution button BT21 is pressed on the search condition input screen T2, it is determined that the start of the search is instructed, and the process proceeds to step ST5. On the other hand, if the start of search is not instructed, the process returns to step ST2.

  In step ST <b> 5, the query generation unit 314 generates a search query based on the conditions entered in the search condition input screen T <b> 2 and transmits the search query to the report extraction unit 116.

  In step ST6, the report extraction unit 116 extracts one or more structured report data corresponding to the search query from the structured information DB 152. At this time, the extracted one or more structured report data is transmitted to the information classification unit 315.

  In step ST7, if a plurality of structured report data is extracted in step ST6, the information classifying unit 315 classifies the plurality of structured report data based on the classification condition corresponding to the display pattern information. At this time, based on the tree display item name defined by the display pattern information and the priority order thereof, a plurality of structured report data is used for terms relating to two or more item names defined as tree display item names. Classified hierarchically by format. If only one structured report data is extracted in step ST6, the classification processing of the information classification unit 315 generates data in which one structured report data is simply represented hierarchically in the form of a tree structure. Is done.

  In step ST8, the search result display screen T4 in which one or more structured report data extracted by the report extraction unit 116 and classified by the information classification unit 315 is visually output by the display control unit 313 is displayed on the display unit 330. Is displayed.

  In step ST9, the information classification unit 315 determines whether a display pattern change is instructed. Here, when the display pattern list DL41 on the search result display screen T4 is operated with the mouse pointer MP to change the display pattern type display, it is determined that the display pattern change is instructed, Return to step ST7. At this time, in step ST7, the classification of one or more structured report data is performed according to the changed display pattern, and in step ST8, the one or more structured report data after classification is visually output. The result display screen T4 is displayed on the display unit 330. On the other hand, if the change of the display pattern is not instructed in step ST9, the process proceeds to step ST10.

  In step ST10, the display control unit 313 determines whether or not the search result display is to be terminated. Here, when the display element for closing the search result display screen T4 is pressed with the mouse pointer MP, the display of the search result display screen T4 is ended, and this operation flow is ended. On the other hand, if the display element for closing the search result display screen T4 is not pressed with the mouse pointer MP, the process proceeds to step ST11 in FIG.

  In step ST11, the display control unit 313 determines whether or not addition of a search condition is requested. Here, when the button BT42 for requesting re-input of the search condition on the search result display screen T4 is pressed with the mouse pointer MP, it is determined that the addition of the search condition is requested, and step ST1 in FIG. Proceed to On the other hand, if the addition of the search condition is not requested, the process proceeds to step ST12.

  In step ST12, the display control unit 313 determines whether display pattern editing is requested. Here, when the pattern edit button BT41 is pressed with the mouse pointer MP on the search result display screen T4, it is determined that display pattern editing is requested, and the process proceeds to step ST13. On the other hand, if editing of the display pattern is not requested, the process proceeds to step ST8 in FIG. At this time, the screen displayed on the entire area A6 of the display unit 330 is the search result display screen T4.

  In step ST <b> 13, the display pattern editing screen T <b> 5 is displayed on the display unit 330 by the display control unit 313.

  In step ST14, the classification condition setting unit 316 inputs information related to tree display item names, list display item names, and the like on the display pattern editing screen T5. That is, a display pattern editing operation is accepted.

  In step ST15, the classification condition setting unit 316 determines whether to save the display pattern being edited on the display pattern editing screen T5. Here, when the save button BT52 is pressed with the mouse pointer MP on the display pattern edit screen T5, it is assumed that the display pattern being edited is being saved, and the process proceeds to step ST16. When the alias saving button BT53 is pressed with the mouse pointer MP on the display pattern editing screen T5, the process proceeds to step ST16 after inputting the name of a new display pattern. On the other hand, if it is not requested to save the display pattern being edited, the process proceeds to step ST17.

  In step ST16, the classification condition setting unit 316 stores information on display patterns corresponding to the item names entered in the areas A51 and A52 of the display pattern editing screen T5 in the memory 350. Then, the process returns to step ST14.

  In step ST17, the display control unit 313 determines whether or not the display of the display pattern editing screen T5 is to be ended. Here, when the close button BT54 on the display pattern edit screen T5 is pressed with the mouse pointer MP, it is determined that the display on the display pattern edit screen T5 is ended, and the process proceeds to step ST18. On the other hand, if it is not determined that the display of the display pattern editing screen T5 is terminated, the process returns to step ST14.

  In step ST18, the display control unit 313 ends the display of the display pattern editing screen T5.

  In step ST19, the information classification unit 315 determines whether or not the currently designated display pattern has been changed by an editing operation. Here, if the display pattern specified on the search result display screen T4 has been changed by the editing operation, it is determined that the currently specified display pattern has been changed by the editing operation, and the process proceeds to step ST7 in FIG. move on. At this time, one or more structured report data are classified again based on the display pattern after being changed by the editing operation (step ST7), and the search result display screen T4 is displayed (step ST8). On the other hand, if the currently designated display pattern has not been changed by the editing operation, the process proceeds to step ST8 in FIG.

  As described above, according to the information processing system 1 according to the present embodiment, a plurality of structured report data corresponding to the generated search query is extracted, and the plurality of structured report data is based on the classification condition. Visually output after classification. For this reason, it becomes possible to classify and display a plurality of document data including a natural sentence (in this case, an interpretation report) according to the contents of the natural sentence. Thereby, for example, a doctor's request such as confirming a list of observation sentences for similar cases or comparing cases related to the same diagnosis is satisfied.

<(4) Modification>
It should be noted that the present invention is not limited to the above-described embodiment, and various changes and improvements can be made without departing from the gist of the present invention.

  For example, in the above-described embodiment, a plurality of structured report data extracted from the structured information DB 152 by the report search is classified and output visually, but the present invention is not limited to this. For example, a large number of structured report data included in the structured information DB 152 is classified and output visually without performing a report search, or a large number of structured report data included in the structured information DB 152 is a category term. They may be classified and output visually. Even with such a configuration, a plurality of structured report data stored in the structured information DB 152 is visually output after being classified based on the classification condition. For this reason, similarly to the above-described embodiment, it is possible to display a plurality of pieces of document data including natural sentences (for example, interpretation reports) classified according to the contents of the natural sentences. Hereinafter, a specific example of such a configuration will be shown and described.

  FIG. 25 is a diagram illustrating a functional configuration related to a process (report reference process) when referring to a past image interpretation report in the information processing system 1A according to a modification. Note that in the information processing system 1A according to the modified example, the report information extraction unit 116 is changed to the report reading unit 118 as compared with the information processing system 1 according to the above-described embodiment, so that the medical information server 100 is medical. The terminal 301 (302, 303) is changed to the terminal 301A (302A, 303A) by changing to the information server 100A and changing the query generation unit 314 to the reference request unit 317. Note that, in the information processing system 1A according to the modified example, the same portions as those of the information processing system 1 according to the above-described embodiment are denoted by the same reference numerals and description thereof is omitted, and different points are mainly described below. explain.

  The reference request unit 317 transmits a signal (request signal) for enabling reference to a past interpretation report to the report reading unit 118 in accordance with the operation of the operation unit 340 by the interpretation doctor.

  FIG. 26 is a diagram illustrating an input support template T6A according to a modification. The input support template T6A according to the modification is shifted to a mode (reference mode) in which the search mode button BR of the input support template T6 according to the one embodiment refers to a past interpretation report (report reference). Button (reference mode button) BRA has been changed. When the reference mode button BRA is pressed with the mouse pointer MP, a request signal is transmitted from the reference request unit 317 to the report reading unit 118.

  The report reading unit 118 reads a plurality of structured report data from the structured information DB 152 in response to a request signal from the reference request unit 317. Here, a combination of terms relating to a modality relating to an examination subject to interpretation and an examination site is distinguished and accumulated, and terms of categories selectively designated by the six category designation buttons SL1 to SL6 are distinguished and accumulated. One or more structured report data are read from the structured information DB 152. At this time, when a plurality of structured report data is read by the report reading unit 118, the plurality of structured report data is classified by the information classification unit 315 based on the classification condition defined by the display pattern information. Then, under the control of the display control unit 313, a plurality of structured report data is visually output on the display unit 330 based on the classification result classified by the information classification unit 315.

  FIG. 27 is a diagram illustrating a display example of a screen (report reference screen) T4A for referring to a past interpretation report in the information processing system 1A according to the modification. The report reference screen T4A according to the modification has substantially the same configuration as the search result display screen T4 according to the embodiment, although the screen name is different.

  FIG. 28 and FIG. 29 are flowcharts showing an operation flow of report reference in the information processing system 1A according to one modification. This operation flow is realized by executing a predetermined program in the CPU 110 of the medical information server 100A and the CPU 310 of the terminals 301A to 303A. Here, when the input support template (FIG. 26) is displayed on the terminal 301A, when the reference mode button BRA is pressed with the mouse pointer MP, the terminal 301A is set to the reference mode, and this operation flow is executed. This is started, and the process proceeds to step SP1 in FIG.

  In step SP1, the report reading unit 118 reads a plurality of structured report data related to the terms of the modality, examination site, and category specified from the structured information DB 152. At this time, the read one or more structured report data is transmitted to the information classification unit 315.

  In step SP2, if a plurality of structured report data is read out in step SP1 by the information classification unit 315, the plurality of structured report data is classified based on the classification condition corresponding to the display pattern information. . At this time, as in step ST7 of FIG. 23, two or more items in which a plurality of structured report data are defined as tree display item names based on the tree display item names defined by the display pattern information and their priorities. Terms belonging to names are classified hierarchically in the form of a tree structure. Note that if only one structured report data is read in step SP1, the classification processing of the information classification unit 315 includes data in which one structured report data is simply represented hierarchically in the form of a tree structure. Generated.

  In step SP3, a report reference screen T4A in which one or more structured report data read by the report reading unit 118 and classified by the information classifying unit 315 is visually output by the display control unit 313 is displayed on the display unit 330. Is displayed.

  In step SP4, as in step ST9 of FIG. 23, the information classification unit 315 determines whether or not an instruction to change the display pattern has been issued. Here, when the display pattern display is changed by operating the display pattern list DL41 of the report reference screen T4A with the mouse pointer MP, it is determined that the change of the display pattern is instructed, and the process proceeds to step SP2. Return. On the other hand, if the change of the display pattern is not instructed, the process proceeds to step SP5.

  In step SP5, the display control unit 313 determines whether or not the display of the report reference screen T4A is terminated. Here, when the display element for closing the report reference screen T4A is pressed with the mouse pointer MP, the display of the report reference screen T4A is ended, and this operation flow is ended. On the other hand, if the display element for closing the report reference screen T4A is not pressed with the mouse pointer MP, the process proceeds to step SP11 in FIG.

  In step SP11, as in step ST12 of FIG. 24, the display control unit 313 determines whether editing of the display pattern is requested. Here, when the pattern editing button BT41 is pressed with the mouse pointer MP on the report reference screen T4A, it is determined that the editing of the display pattern is requested, and the process proceeds to step SP12. On the other hand, if editing of the display pattern is not requested, the process proceeds to step SP3 in FIG. At this time, the screen displayed on the entire surface of the area A6 on the display unit 330 is the report reference screen T4A.

  In step SP12, the display pattern edit screen T5 is displayed on the display unit 330 by the display control unit 313 as in step ST13 of FIG.

  In step SP13, processing similar to that in step ST14 in FIG. 24 is performed.

  In step SP14, as in step ST15 of FIG. 24, the classification condition setting unit 316 determines whether or not to save the display pattern being edited on the display pattern editing screen T5. Here, when the save button BT52 is pressed with the mouse pointer MP on the display pattern edit screen T5, it is assumed that the display pattern being edited is requested to be saved, and the process proceeds to step SP15. When the alias saving button BT53 is pressed with the mouse pointer MP on the display pattern editing screen T5, the process proceeds to step SP15 after inputting the name of a new display pattern. On the other hand, if it is not requested to save the display pattern being edited, the process proceeds to step SP16.

  In step SP15, processing similar to that in step ST16 in FIG. 24 is performed. Then, the process returns to step SP13.

  In step SP16, as in step ST17 of FIG. 24, the display control unit 313 determines whether or not the display of the display pattern editing screen T5 is terminated. Here, when the close button BT54 on the display pattern edit screen T5 is pressed with the mouse pointer MP, it is determined that the display on the display pattern edit screen T5 is ended, and the process proceeds to step SP17. On the other hand, if it is not determined that the display of the display pattern editing screen T5 is terminated, the process returns to step SP13.

  In step SP17, processing similar to that in step ST18 in FIG. 24 is performed.

  In step SP18, as in step ST19 of FIG. 24, the information classification unit 315 determines whether or not the currently designated display pattern has been changed by an editing operation. If the display pattern specified on the report reference screen T4A has been changed by the editing operation, it is determined that the currently specified display pattern has been changed by the editing operation, and the process proceeds to step SP2 in FIG. . At this time, one or more structured report data are classified again based on the display pattern after being changed by the editing operation (step SP2), and the report reference screen T4A is displayed (step SP3). On the other hand, if the currently designated display pattern has not been changed by the editing operation, the process proceeds to step SP3 in FIG.

  In the above-described embodiment and modification, there are four item names for which elements (for example, terms) can be input in the input support template: “imaging condition”, “basic part”, “basic finding”, and “diagnosis”. However, it is not limited to this. For example, the item name “treatment” may be added. Examples of the element related to the item name “treatment” include “If a polyp is found, a cell is taken and a biopsy is performed”. The item name to which a plurality of elements constituting a finding sentence that is a natural sentence includes at least one item name of “imaging condition”, “basic part”, “basic finding”, “diagnosis”, and “treatment” You can do that. When such a configuration is adopted, it is possible to easily grasp information about item names such as imaging conditions, basic parts, basic findings, diagnosis, treatment, and the like.

  In the above-described embodiment and modification, in the input support template, the choices of terms related to the item names are listed based on the term table 153, but the present invention is not limited to this. For example, based on structured report data stored in the structured information DB 152, the relationship of terms between item names may be indicated by visible display elements (for example, line segments) or the like.

  In the above-described embodiment and modification, the information processing system 1 (1A) is configured to include the medical information server 100 (100A) and the terminals 301 to 303 (301A to 303A). Absent. For example, the functions of the medical information server 100 (100A) and the terminals 301 to 303 (301A to 303A) may be realized in one personal computer.

  In the above-described embodiment and modification, the user's interpretation doctor inputs various information and instructions by operating the operation unit 340. However, the present invention is not limited to this. For example, the user's utterance Various information and instructions may be input in accordance with various user operations such as.

  In the above-described embodiment and modification, the input, search, and reference targets are interpretation reports. However, the present invention is not limited to this. For example, it may be a general report in the medical field such as an incident report or a diagnostic report. That is, the plurality of structured report data stored in the structured information DB 152 may be structured report data related to a general medical field report. Even with such a configuration, the user can easily grasp information in the medical field.

  Furthermore, the object of input, search, and reference may be a general document such as a business report or various reports.

  It should be noted that all or a part of the above-described embodiment and modification may of course be combined as appropriate as long as no contradiction occurs.

1, 1A Information processing system 100, 100A Medical information server 110, 310 CPU
DESCRIPTION OF SYMBOLS 113 Structured process part 114 Support information acquisition part 115 Data writing part 116 Report extraction part 118 Report reading part 150 Memory 301-303, 301A-303A Terminal 311 Report generation part 313 Display control part 314 Query generation part 315 Information classification part 316 Classification condition setting unit 317 Reference request unit 330 Display unit 340 Operation unit 350 Memory

Claims (24)

A database for storing a plurality of structured document data in which two or more elements constituting a natural sentence are respectively associated with item names;
Query generation means for generating a search query including information in which one or more item names and one or more elements related to the one or more item names are associated according to a user's operation;
Extraction means for extracting a plurality of structured document data corresponding to the search query from the database;
Classification means for classifying a plurality of structured document data extracted by the extraction means based on a classification condition indicating one or more item names serving as classification criteria;
An output means for visually outputting information on a plurality of structured document data extracted by the extraction means based on a classification result by the classification means;
An information processing system comprising: The information processing system according to claim 1,
Each of the structured document data is
An information processing system characterized by being described using RDF. The information processing system according to claim 2,
An information processing system, wherein SPARQL is used as a language of the search query. An information processing system according to any one of claims 1 to 3,
The search query is
Information in which the first item name is associated with one or more elements related to the first item name, and the second item name is associated with one or more elements related to the second item name. An information processing system including: An information processing system according to any one of claims 1 to 4,
The search query is
An information processing system including information in which one or more item names, one or more elements related to the one or more item names, and detailed elements indicating detailed information related to the one or more elements are associated with each other . An information processing system according to any one of claims 1 to 5,
The query generation means is
Depending on the user's operation after the classification process by the classification means, one or more different item names different from the one or more item names and one or more different item names are associated with the search query. By adding information associated with the above elements, you can generate a modified search query,
The extraction means comprises:
Re-extracting a plurality of structured document data corresponding to the changed search query from a plurality of structured document data already extracted by the extracting means or the database;
The classification means is
An information processing system, wherein a plurality of structured document data extracted again by the extracting means is classified based on the classification condition. An information processing system according to any one of claims 1 to 5,
The query generation means is
According to the user's operation after the classification process by the classification means, the information in which the one or more item names included in the search query and one or more elements related to the one or more item names are associated, By further associating detailed elements indicating detailed information related to one or more elements, a modified search query is generated,
The extraction means comprises:
Re-extracting a plurality of structured document data corresponding to the changed search query from a plurality of structured document data already extracted by the extracting means or the database;
The classification means is
An information processing system, wherein a plurality of structured document data extracted again by the extracting means is classified based on the classification condition. An information processing system according to any one of claims 1 to 7,
The classification condition is
Shows two or more item names that are the criteria for classification, and the priority order related to the two or more item names,
The classification means is
A plurality of structured document data extracted by the extracting unit is classified hierarchically using elements related to two or more item names serving as a reference for the classification based on the priority order. Processing system. The information processing system according to claim 8,
The output means is
When visually outputting information on a plurality of structured document data extracted by the extracting unit based on a hierarchical classification result by the classifying unit, at least one layer is selected according to a user's operation. An information processing system, wherein information dependent on one or more elements is set to one of a visible state in which the information is visually output and a non-visible state in which the information is not visually output. An information processing system according to claim 8 or claim 9, wherein
The output means is
When the information related to the plurality of structured document data extracted by the extracting means is visually output based on the hierarchical classification result by the classifying means, the information is classified for each element related to at least one hierarchy. An information processing system characterized in that display elements each indicating the number of structured document data are visually output in a form corresponding to display elements indicating respective classification destination elements. An information processing system according to any one of claims 8 to 10,
The output means is
When the information related to the plurality of structured document data extracted by the extracting unit is visually output based on the hierarchical classification result by the classifying unit, information corresponding to two or more pieces of structured document data is displayed. Among them, information related to one or more display item names is visually output in the form of a table, and the information related to each display item name is in an ascending or descending order according to a predetermined rule according to a user's operation. An information processing system characterized by visual output. A database for storing a plurality of structured document data in which two or more elements constituting a natural sentence are respectively associated with item names;
Classification means for classifying the plurality of structured document data based on a classification condition indicating one or more item names serving as a classification reference, in accordance with an instruction input in response to a user operation;
An output means for visually outputting information related to the plurality of structured document data based on a classification result by the classification means;
An information processing system comprising: An information processing system according to claim 12,
Each of the structured document data is
An information processing system characterized by being described using RDF. An information processing system according to claim 12 or claim 13,
The classification condition is
Shows two or more item names that are the criteria for classification, and the priority order related to the two or more item names,
The classification means is
An information processing system characterized in that the plurality of structured document data are classified hierarchically using elements related to two or more item names serving as a reference for the classification based on the priority order. The information processing system according to claim 14,
The output means is
When visually outputting the information related to the plurality of structured document data based on the hierarchical classification result by the classification means, the information depends on one or more elements related to at least one hierarchy according to a user's operation. The information processing system is characterized in that the information to be set is set to one of a visible state that is visually output and a non-visible state that is not visually output. The information processing system according to claim 14 or 15,
The output means is
Structured document data classified for each element related to at least one hierarchy when visually outputting information related to the plurality of structured document data based on a hierarchical classification result by the classification means An information processing system characterized in that display elements each indicating the number of items are visually output in a form corresponding to display elements indicating each element of the classification destination. The information processing system according to any one of claims 14 to 16, wherein
The output means is
When information related to the plurality of structured document data is visually output based on a hierarchical classification result by the classification unit, one or more displays among information corresponding to two or more structured document data respectively Information related to the item name is output visually in the form of a table, and information related to each display item name is output visually in an ascending or descending order according to a predetermined rule in accordance with the user's action. An information processing system characterized by this. The information processing system according to any one of claims 8 to 11 and claims 14 to 17,
Each of the structured document data is
The elements related to the plurality of item names include data hierarchically related according to a predetermined dependency relationship,
An information processing system, wherein a dependency relationship of elements classified hierarchically by the classification unit based on the priority order is different from the predetermined dependency relationship. The information processing system according to any one of claims 1 to 18, wherein
Data generating means for recognizing two or more elements from a natural sentence and associating an item name with each of the two or more elements to generate each structured document data stored in the database;
An information processing system further comprising: The information processing system according to any one of claims 1 to 18, wherein
Data generating means for generating each of the structured document data stored in the database by designating one or more elements for two or more item names related to a plurality of elements constituting a natural sentence;
An information processing system further comprising: An information processing system according to any one of claims 1 to 20,
A setting means for setting one or more item names serving as classification criteria included in the classification condition according to a user's operation;
An information processing system further comprising: The information processing system according to any one of claims 1 to 21,
A plurality of structured document data stored in the database is
An information processing system comprising structured document data relating to a report in the medical field. The information processing system according to any one of claims 1 to 22,
Item names related to two or more elements constituting each natural sentence are as follows:
An information processing system comprising at least one item name of imaging conditions, basic sites, basic findings, diagnosis, and treatment.   24. A program that, when executed by a control unit included in an information processing system, causes the information processing system to function as the information processing system according to any one of claims 1 to 23.

Images