JP7240707B2 - Information processing system - Google Patents

Description

The present invention relates to an information processing system for structuring freely input text information.

When analyzing freely input text information, it is conceivable to structure the text information by executing natural language analysis processing.

For example, systems for entering text information include electronic medical records entered by doctors when they examine patients, and nursing records by nurses. These systems (medical information systems) mainly consist of information related to insurance claims (names of diseases covered by insurance, prescriptions/treatments for insurance claims, laboratory findings, hospital admission/discharge history, medical examination history, etc.) and records of medical care and nursing care for patients. It consists of a medical information record. Since the information on the former insurance claim is uniform, it is often done mainly by selection from a pull-down menu or the like, or by selection from a master, and the information is sufficiently structured.

On the other hand, most of the medical information records, which are records of medical care and nursing care for patients, are entered as text information, for example, in the free description section of electronic medical charts. The text information freely entered as medical information records includes, for example, symptoms and chief complaints, history of present illness (why the patient came to see a doctor), medical history, family history, oral history, allergy history, physical findings, vital signs, interpretation of laboratory findings, There is a progress after coming to the hospital. All of these are very important information for clinical and medical research.

In other words, in the current medical information system, information related to insurance claims is well structured, but important information for clinical and medical research is only freely entered as text information. not structured.

Non-Patent Document 1 and Non-Patent Document 2 show an example of a conventional electronic medical record system.

Shimadzu Corporation, "Introduction of integrated electronic medical record system "SimCLINIC T3"", [online], Internet <URL: http://www.med.shimadzu.co.jp/application/info/09.html> FUJITSU LIMITED, ”Electronic medical record system for medium-sized hospitals FUJITSU Healthcare Solution HOPE EGMAIN-LX”, [online], Internet <URL: http://www.fujitsu.com/jp/solutions/industry/healthcare/products/ egmainlx/>

In Japan, it is common to have medical services covered by health insurance. In medical treatment covered by health insurance, the names of diseases covered by insurance corresponding to the treatments and prescriptions are recorded, and medical fees are billed based on the treatments, prescriptions, and names of diseases covered by insurance. This medical fee claim is made based on the "ordering" information (insurance claim information) in the electronic medical record. Therefore, there is an entry field for recording the name of the disease covered by insurance in the "ordering" information of the electronic medical record, and the doctor must record the name of the disease covered by insurance there. As a result, the name of the patient's disease that is reliably recorded on the electronic medical record is the name of the disease covered by insurance.

However, this insurance disease name is far from the "disease name (clinical disease name) that accurately reflects the patient's condition" (the name of the disease included in the medical record that the doctor thinks is optimal for the patient's condition). There are many.

For example, when a doctor prescribes an analgesic to an outpatient, it is covered by insurance regardless of the patient's actual disease name (clinical disease name). Phenomena such as "arrhythmia" being recorded as an insurance disease name have occurred in all the patients who underwent examinations, regardless of the clinical disease name.

In addition, in the case of a patient with a long course, the insurance disease names corresponding to each treatment and examination are entered in the electronic chart, so the insurance disease names may exceed 100. Therefore, it is not easy to immediately determine which is the clinical disease name, which is the actual disease name of the patient, just by looking at the list of disease names insured in the electronic medical record.

Due to this reality, in the medical information system, as described above, information related to insurance claims, such as names of insured diseases, prescriptions and treatments for insurance claims, and laboratory findings, are recorded in a structured manner. Looking at structured information about insurance claims is of little use from the point of view of medical research and the actual clinical practice of health care workers.

In the medical information system, important information from the viewpoint of medical research and actual clinical work is described in the medical information record, which is free-input text information. Therefore, from the viewpoint of medical research and actual clinical work, it is important to structure and manage free-input text information so that it can be used as a database for information retrieval and statistical information processing. preferable.

However, conventional medical information systems are not suitable for retrieving or processing input textual information because text is only freely entered in the free entry fields of electronic charts and nursing records.

Therefore, when freely entered text information is used as a target for medical research or clinical work, conventionally, a format is prepared for each research, such as ``people taking a certain drug, people not taking it'', ``people not A method is used in which individual information such as "persons with a history of a specific disease" is read back through medical records (including electronic medical records) and nursing records to confirm the content, and then transcribed into an online format.

Such work has traditionally been addressed by hiring research assistants for the purpose of conducting medical research or building a national registry, or by having junior physicians work outside of clinical practice. However, only a limited number of medical institutions are able to employ research assistants and employ young doctors outside of clinical work, and they are only able to structure text information related to medical care limited to specific research themes. Furthermore, it is a heavy burden for young doctors to perform such work.

These challenges are not limited to medical information systems such as electronic medical charts and nursing records, but are required in all computer systems where there is a desire to structure textual input information. .

Therefore, one method is to improve the technology of natural language analysis processing by using artificial intelligence (AI) or the like while keeping the input method as it is.

However, the accuracy of automatically structuring text information entered by humans is still far from high due to the complexity of the language. In particular, if the grammar is not used correctly in the input text information, the analysis accuracy is significantly reduced. In clinical settings that are extremely busy and require prompt responses, it is often the case that the free entry fields of electronic medical charts are not entered using correct grammar, but are entered for the time being. In addition, even if there is time to spare, the free entry fields in the electronic medical record may be written by the doctor only as memos. In many cases, the information is not described to the extent that a third party can understand it.

For example, in the free entry field of an electronic medical record, text information such as that shown in Fig. 13 is entered, and when attempting to structure the medical history name and related information in association with each other, natural language analysis processing technology It is not easy to structure even if the

When analyzing freely input text information in this way, no matter how advanced the technology of natural language analysis processing is, it is not easy to structure the input text information with high accuracy.

In view of the above problems, the present inventor has invented an information processing system that allows a person who inputs freely text information to input the text information while being conscious of the structuring of the text information.

A first invention is an information processing system functioning on a computer having an arithmetic device and a display device, wherein the arithmetic device performs text information input reception processing for accepting free input of text information in a free input field of an electronic medical record. a structuring processing unit that generates structured information in a table format for each predetermined information type based on the text information extracted from the text information that has received the free input using natural language analysis processing ; a display processing unit for displaying the text information and the structured information on the display device , wherein the structuring processing unit detects a predetermined information type from the text information that has received the free input. Target information corresponding to the information type of the text information extracted by using one or more of the above natural language analysis processing of dependency analysis, context analysis, or machine learning by a learning model using a neural network The extracted target information is stored in a table format as structured information, the display processing unit displays the text information in a first display area of the display device, and converts the structured information to the display. The information processing system displays the structured information in a second display area of a device, and displays the structured information in the second display area according to input of text information in the first display area.

By using the present invention, a person who inputs text information will input text information while viewing how it is displayed as structured information. Therefore, when unnatural structured information is displayed, it can be recognized and corrected. Also, when inputting text information, it is possible to learn how to properly display structured information without reading manuals, etc., as experience, so that texts that are easy to structure can be learned from the beginning. Be careful to enter information. As a result, the user's input naturally becomes an input that is easy to structure, so that even if the natural language analysis processing technique is not high, it is possible to obtain appropriately structured information.
By displaying the freely input text information and the structured information based on it in different display areas, the visibility can be improved.

For example, in a medical institution to which the present inventor belongs, when an attempt was made to structure text information in a free input column of an electronic medical record, text information was input in various ways depending on the input person. accuracy was not high. However, when the present invention is used, the input of text information to the free entry field of the electronic medical record by the doctor itself is improved, and structured information with accuracy close to 100% can be obtained.

In the above-described invention, the display processing section can be configured as an information processing system capable of switching whether or not the second display area is displayed.

The second display area is displayed so that the user who freely inputs the text information can easily recognize the structuring of the input text information. Therefore, even if the input person is proficient and has input text information that is naturally easy to structure, it may be troublesome to display it, and the range of display such as the first display area is narrowed. end up Therefore, it is preferable that the presence/absence of display in the second display area can be switched.

In the above-described invention, when the second display area is not displayed and it is determined that the structured information satisfies a predetermined condition, the display processing unit displays the second display area. It can be configured as an information processing system having a display area.

As the person who inputs text information becomes more proficient, he or she will input text information that is easier to structure, so the purpose can be achieved without displaying the second display area. However, for some reason, such as when text information is input after a period of time has passed, or when an inexperienced person inputs anew, it is not always possible to input text information that is easy to structure. . In that case, structuring may not be performed properly. Therefore, when certain conditions are met, such as the possibility of an erroneous determination of structured information, the second display area is automatically displayed again, raising awareness of the input of text information that is easy to structure. can be done again.

In the above-described invention, the display processing unit changes the display form or adds a mark to the text information reflected in the structured information among the freely input text information and displays it in the first display area. It can be configured like an information processing system to display.

In the above invention, the display processing unit changes the display form of the text information reflected in the structured information among the freely input text information and displays it in the first display area, and changes the display form. It can be configured as an information processing system that displays structured information based on text information in the second display area in association with the display mode.

This allows the user to visually check which part of the freely entered text information has been reflected as structured information. Therefore, when inputting text information with awareness of structuring, it becomes easier for the inputting person to be aware of where to pay attention, and it is possible to further encourage the inputting person's awareness. In particular, by associating the display form of the text information in the first display area with the structured information in the second display area, the visibility of the association is improved.

The first invention can be realized by causing a computer to read and execute the program of the invention. That is, a computer is provided with a text information input reception processing unit that receives free input of text information in a free input field of an electronic medical record, and based on the text information extracted from the text information that has received the free input using natural language analysis processing , An information processing program that functions as a structuring processing unit that generates structured information in a table format for each predetermined type of information, and a display processing unit that displays the text information and the structuring information on a display device of the computer. The structuring processing unit extracts text information by detecting a predetermined information type from the text information for which the free input has been accepted, and applies target information corresponding to the information type to dependency analysis, context Analysis or machine learning using a learning model using a neural network is extracted using one or more of the natural language analysis processing, the extracted target information is stored in a table format as structured information, and the display processing The unit displays the text information in a first display area of the display device, displays the structured information in a second display area of the display device, and inputs text information to the first display area. The information processing program causes the structured information to be displayed in the second display area in response to the second display area.

By using the information processing system of the present invention, a person who inputs text that can be freely input will recognize how the text information is structured when inputting the text information. . Therefore, since the awareness of the structuring of the text information is made, when automatically inputting the text information, the input with the structuring in mind will be kept in mind. As a result, the freely input text information itself becomes an input suitable for structuring, and the accuracy of structuring can be improved.

1 is a block diagram schematically showing an example of overall processing functions of an information processing system of the present invention; FIG. 1 is a conceptual diagram schematically showing an example of a hardware configuration of a computer on which an information processing system of the present invention functions; FIG. FIG. 10 is a diagram showing an example of text information entered in a free entry field of an electronic medical chart and structured information obtained as a result of structuring processing for the text information; FIG. 10 is a diagram showing an example of a spelling variation dictionary for disease names; FIG. 10 is a diagram showing an example of a spelling variation dictionary for drug names; FIG. 4 is a diagram showing an example of a display relationship between a first display area and a second display area; It is a figure which shows an example of a structured information storage part. 4 is a flow chart showing an example of a processing process of the information processing system of the present invention; FIG. 10 is a diagram showing an example of text information when an input person inputs text information that can be easily structured using the present invention, and an example of structured information that has undergone structuring processing according to the present invention; FIG. 11 is a block diagram showing an example of overall processing functions of an information processing system according to a second embodiment; FIG. 10 is a diagram showing an example of changing the display form of text information reflected in structured information among the text information displayed in the first display area, such as character color, thickness, and font; FIG. 10 is a diagram showing an example of a case where text information reflected in structured information among text information displayed in the first display area and the display form of the corresponding structured information are changed so as to correspond to each other; It is a figure which shows an example when text information is input into the free entry column of an electronic medical record as usual.

An example of the overall processing functions of the information processing system 1 of the present invention is shown in the block diagram of FIG. FIG. 2 shows an example of the hardware configuration of a computer that implements the information processing system 1 of the present invention.

The information processing system 1 is a computer that executes the processing of the present invention. The computer used in the information processing system 1 includes an arithmetic unit 70 such as a CPU for executing arithmetic processing of programs, a storage device 71 such as a RAM or hard disk for storing information, a display device 72 such as a display (screen), and a keyboard. , an input device 73 such as a pointing device (mouse, numeric keypad, etc.), and a communication device 74 for transmitting/receiving the processing result of the arithmetic unit 70 and the information stored in the storage device 71 via a network such as the Internet or a LAN. there is Each function (each means) implemented on a computer is executed by reading the means (program, module, etc.) for executing the process into the arithmetic unit 70 . When each function uses the information stored in the storage device 71 in its processing, the corresponding information is read from the storage device 71 and the read information is appropriately used for the processing in the arithmetic device 70 . Further, although the information processing system 1 in FIG. 1 is implemented by one computer, the functions may be distributed among a plurality of computers. Computers include various information processing devices such as servers, personal computers, and workstations. Also, a so-called cloud format may be used. For example, a computer used by an inputter to input text information and a computer that accepts the input text information and executes each process in the present invention may be separated.

If the computer has a touch panel display, the display device 72 and the input device 73 may be integrated. Touch panel displays are often used in portable communication terminals such as tablet computers and smartphones, but are not limited to these. The touch panel display is a device in which the functions of the display device 72 and the input device 73 are integrated in that input can be performed directly on the display using a predetermined input device (such as a pen for touch panel) or a finger.

The computer may also be capable of voice input. In this case, the voice uttered by the input person is collected by a predetermined microphone (sound collecting device), converted into text information, and the text is input to the corresponding portion.

Each means in the present invention is only logically distinguished in its function, and may physically or practically form the same area. Moreover, the processing described in the present invention is merely an example, and the processing can be changed as appropriate.

The information processing system 1 has a text information input reception processing unit 11 , a text information storage unit 12 , a structuring processing unit 13 , a matching dictionary storage unit 14 , a display processing unit 15 and a structured information storage unit 16 .

The text information input reception processing unit 11 receives text information freely input by a person who inputs text information, and stores the text information in the text information storage unit 12 . For example, if the information processing system 1 is a medical information system, it accepts the input of text information written in the free entry field of the electronic chart. Free input means that the input person freely inputs text information, instead of selecting information from a pull-down menu or inputting text into an input field in which the type of information to be input is predetermined.

The text information storage unit 12 stores text information whose input is received by the text information input reception processing unit 11 . For example, if the information processing system 1 is a medical information system, it stores medical information records such as text information written in free entry fields of electronic medical charts. It should be noted that information relating to insurance claims may be stored regardless of medical information records.

The structuring processing unit 13 monitors the input of the text information received by the text information input reception processing unit 11, performs natural language analysis processing such as dependency analysis and context analysis, and refers to the matching dictionary storage unit 14 described later. , freely input text information is structured and stored in the structured information storage unit 16 as structured information. Various techniques can be used for structuring processing of text information in the structuring processing unit 13 . At this time, the notation fluctuation processing may be executed by referring to the collation dictionary storage unit 14, which will be described later. The notation fluctuation processing is a process of unifying the notation into a standard notation when there are multiple notations for the same event. As a computer system used for natural language analysis processing, for example, LUIS (Language Understanding) of Microsoft Azure provided by Microsoft Corporation can be used.

The structuring of text information is to standardize the content for each predetermined information type based on freely input text information. For example, it is held in table format. One process for structuring text information is as follows. Based on freely input text information, tags attached to predetermined text information units such as sentences, clauses, and paragraphs are retrieved by referring to the matching dictionary storage unit 14 or by using natural language analysis processing such as context analysis. , and text information with standardization tags (classification of text information for each information type). Since the target information to be extracted is determined for each information type, natural language analysis processing is used to extract the predetermined target information to be extracted from the text information of each information type. stored in a standardized table as standardized information.

When the information processing system 1 of the present invention is used in a medical information system, the information types include "current medical history", "pre-existing medical history", "oral medication", "physical findings", "progress after visiting the hospital", and the like. The corresponding target information is "symptoms" for the information type "current medical history", "disease name" as a past history for the information type "previous history", "drug name" for the information type "oral medicine", and "drug name" for the information type "medicine". "Progress after visit" includes "disease name" as a diagnosis name. The target information also includes positive/negative expressions and additional information (remarks column) for symptoms and disease names. A table is generated for each information type, and the target information is stored in this table. For example, in the table of the information type "history of present illness", "symptoms" and positive/negative expressions corresponding to them are stored in association with each other. The type of information to be provided and how to set the target information for each type of information can be arbitrarily set depending on the field in which the information processing system of the present invention is used. In the records of general doctors and nurses, there is an information type set that is unified to some extent.

The structuring processing unit 13 refers to the tag indicating the type of information stored in the tag pattern dictionary of the matching dictionary storage unit 14 in the text information that has been input, and performs natural language analysis such as contextual analysis on the input text information. When it is detected that there is an information type by the analysis process, a table corresponding to the information type, for example, a table of the information type "present illness history", a table of the information type "previous history", a table of the information type "oral medicine", a table of the information type " It is determined whether or not a table of "physical findings" and a table of information type "post-visit progress" have already been generated. Then, if a table corresponding to the detected information type has not been generated, that table is generated. Target information is extracted using natural language analysis processing for each text information of each information type, and the target information is sorted and stored in the generated information type table. On the other hand, if a table corresponding to the detected information type already exists, the target information is extracted using natural language analysis processing for each text information of the information type, and the extracted target information is stored in the table corresponding to the detected information type. Sort and store information. Whether or not there is a table corresponding to an information type can be determined by setting the correspondence between the information type and the table in advance, and determining whether or not the table has been generated based on the correspondence.

With respect to the detected target information, a collation dictionary storage unit 14, which will be described later, is referenced to determine whether or not the target information contains an expression indicating a symptom name, a disease name, or a drug name. The unit 14 specifies a matching character string, adds or changes a predetermined standard notation or standard code to the detected target information in the matching dictionary storage unit 14, and executes processing to unify the notation. may

For example, when detecting "headache" as target information using natural language analysis processing technology for text information of the information type "current illness history", the reference dictionary storage unit 14 is referred to and the standard symptom name is " "Headache" is changed to "headache", and "+" is determined as the positive/negative expression, and "headache" and "+" are stored in the table of the information type "present illness history" in association with each other. Similarly, when "aspirin" and "statin" are detected as target information using natural language analysis processing technology for the text information of the information type "oral medicine", the collation dictionary storage unit 14 is referred to, and the standard drug "Biaspirin" and "Statin" are set as the names, and their codes (therapeutic efficacy classification codes) are added, and stored in association with the table of the information type "oral medicine".

The structuring processing unit 13 changes the target information in each information type extracted from the input text information as described above to a standard notation, adds a code, and distributes it to each information type table. Store.

For example, when text information as shown in FIG. 3A is entered in the free entry field of the electronic medical chart, the structuring processing unit 13 uses "S:" as a tag indicating the information type and the information type "current illness history , the information type 'oral medicine' is detected from 'internal medication:', the information type 'physical findings' is detected from 'O:', and the information type 'progress after visit' is detected from 'A/P:'. Also, the information type "previous history" is detected by contextual analysis of the text information "after catheter treatment for myocardial infarction." Then, the text information between the detected information type and the next information type is cut out as the text information of the first detected information type (including the case where it is physically cut out and specified as a processing target). That is, the information type "current illness history" is detected by detecting "S:", and the information type "previous history" is detected by contextual analysis of the text information "after catheter treatment for myocardial infarction." Then, the text information between the information type "history of current illness" and the information type "pre-existing medical history" is divided as text information corresponding to the information type "history of current illness". FIG. 3B schematically shows a state in which text information is cut out for each information type.

Target information is extracted using natural language analysis processing from the text information corresponding to the segmented information type "current illness history." Since the target information is associated with each information type, for example, the target information "symptoms" and its positive/negative expressions in the information type "current illness history" are extracted. Then, the positive/negative expressions are associated with the extracted "symptoms" and stored in a table.

By sorting the target information for each divided information type and storing it in the table as structured information, a table for each information type can be created as shown in FIG. 3(c).

The processing in the structuring processing unit 13 is not limited to the processing described above, and can be realized by various natural language analysis processing. For example, using natural language analysis processing by deep learning, which executes machine learning by referring to a learning model in which the weighting coefficients between neurons in each layer of a neural network consisting of a large number of intermediate layers are optimized. good too. Also, the natural language analysis processing may be executed by AI (artificial intelligence) using deep learning or machine learning or by AI not using them.

The collation dictionary storage unit 14 stores a notation fluctuation dictionary for judging fluctuations in notation of symptom names, disease names, drug names, etc., a dictionary of pattern tables such as negative expressions and ambiguous expressions, and types of input text information (for example, medical A dictionary of frequently appearing abbreviations and specific tag patterns (for example, "Chief Complaint:", "A/P", etc.) is stored according to the information record).

The notation variation dictionary for symptom names, disease names, and drug names stores, for example, standard notations, codes, and notation patterns for symptom names, disease names, and drug names. Fig. 4 shows an example of a notation variation dictionary for disease names. showing. Also, FIG. 5 shows an example of a notation variation dictionary for drug names, showing a case where standard drug names, generic names, efficacy classification codes, and notation patterns are associated and stored.

The collation dictionary storage unit 14 is not limited to the above, and may appropriately include dictionaries necessary for executing structuring processing based on text information.

The display processing unit 15 displays the text information received by the text information input reception processing unit 11 and the structured information corresponding to the text information. Specifically, an area (first display area 101) that accepts free input and a display area ( A second display area 102) is displayed. Freely input text information is displayed in the first display area 101 , and structured information based on the text information is displayed in the second display area 102 .

For example, if the information processing system 1 is a medical information system, the first display area 101 is the free entry field of the electronic medical record, and the second display area is the structured information based on the text information entered in that area. It becomes the display area 102 . When text information is input to the first display area 101, the structuring processing unit 13 monitors it, performs structuring processing by executing natural language analysis processing, and displays text information in the second display area 102. Display structured information sequentially. That is, text information input to the first display area 101 is received by the text information input reception processing unit 11, the input is monitored by the structuring processing unit 13, and predetermined conditions are determined based on the input text information. If it is determined that the conditions are satisfied, a table is generated according to the conditions. Then, the generated table and information stored therein are displayed in the second display area 102 .

The first display area 101 and the second display area 102 are preferably different display areas, but they may be displayed on the same display area. It is preferable that the freely input text information and the corresponding structured information are displayed at the same time. In the second display area 102, structured information 103a indicating the standard symptom name and its presence/absence, structured information 103b indicating the standardized medical history name and corresponding information (remarks), standardized information role name and its corresponding Structured information 103c indicating the therapeutic classification code and structured information 103d indicating the standardized diagnosis name and information (remarks) corresponding thereto are displayed.

The display processing unit 15 can sequentially display the structured information resulting from the structuring processing in the structuring processing unit 13 in the second display area 102 in response to the input of text information to the first display area 101 . Preferably, the text information stored in the text information storage unit 12 is extracted and displayed in the first display area 101, and the extracted text information is subjected to structuring processing in the structuring processing unit 13 at a predetermined timing. may be executed to display the structured information in the second display area 102 .

Note that the display processing unit 15 displays structured information based on the text information displayed in the first display area 101 in the second display area 102, but the text for each information type extracted during the structuring process is displayed in the second display area 102. Part or all of the information may also be displayed in the first display area 101, the second display area 102, or another display area (not shown). That is, the text information in FIG. 3(b) may be displayed.

The structured information storage unit 16 stores a table (structured information) of text information structured by the structuring processing unit 13 . For example, it may be a database or other storage format. FIG. 7 shows an example of the structured information storage unit 16. As shown in FIG. FIG. 7(a) is structured information indicating standard symptom names and their presence/absence, FIG. 7(b) is structured information indicating standardized medical history names and corresponding information (remarks), and FIG. ) is structured information indicating the standardized information role name and its corresponding pharmacological classification code, and FIG. 7(d) is structured information indicating the standardized diagnosis name and its corresponding information (remarks).

Next, an example of the processing process of the information processing system 1 of the present invention will be described with reference to the flowchart of FIG. In the following description, the information processing system 1 is a medical information system, and the case in which the information on the content of medical treatment (medical information record) is input to the electronic chart is shown. The information processing system 1 may be a computer system other than the medical information system as long as it is a computer system that allows free input of text information.

Patients who have been transported to hospitals or other medical institutions by ambulance or the like are treated by doctors. When providing medical care, the doctor asks the patient or his/her attendant for attribute information such as the patient's name, as well as current symptoms, medical history, oral medications that are taken on a daily basis, and other information. Text is entered in the free entry field (S100). In addition, the doctor inputs into the electronic medical record information on treatment and medication given to the patient by the doctor and information on medical care such as the name of the disease diagnosed. Information entered as text in the free entry field of the electronic chart is received by the text information input reception processing unit 11 and stored in the text information storage unit 12 . Then, the display processing unit 15 sequentially displays them in the first display area 101 of the medical information system.

In parallel with this, the structuring processing unit 13 monitors the input text information (S110) and executes structuring processing (S120).

Specifically, the structuring processing unit 13 refers to the tag indicating the type of information stored in the tag pattern dictionary of the matching dictionary storage unit 14 in the text information that has been input, and the context for the input text information. When an information type is detected by a natural language analysis processing technique such as analysis, it is determined whether or not a table corresponding to the information type has already been generated. Then, if a table corresponding to the detected information type has not been generated, that table is generated.

Then, the text information is cut out for each information type, the target information is extracted from the text information of each information type using natural language analysis processing, and the target information is sorted and stored in the generated information type table.

On the other hand, if there is already a table corresponding to the detected information type, the target information is extracted using natural language analysis processing from the text information extracted for each information type, and the table corresponding to the detected information type is: Sort and store the extracted target information.

Before storing the target information in the table, the matching dictionary storage unit 14 is referred to determine whether there is a matching character string in the symptom name, disease name, drug name, etc. If there is a matching character string, converts the detected target information into standard notation or code, or adds to the detected target information.

By executing the processing as described above, the text information entered in the free entry field can be sequentially converted into structured information.

Then, the display processing unit 15 displays the text information received by the text information input reception processing unit 11 in the first display area 101, and displays the structure obtained by executing the structuring processing by the structuring processing unit 13 described above. change information is displayed in the second display area 102 (S130). An example of this is shown in FIG.

By executing the process described above, the person who enters text information can enter text information while viewing how it is displayed as structured information, and repeat the process. By doing so, they will try to input text information that is naturally easy to structure. For this reason, conventionally, text information was input as shown in FIG. 13, but even if the content is the same, users will naturally try to input text information that is easy to structure as shown in FIG.

In the above embodiment, the display processing unit 15 displays the first display area 101 and the second display area 102, but the second display area 102 is displayed by the operation of the person who inputs the text information. It may be possible to switch whether to perform or not.

Further, the information processing system 1 may include a structured information determination processing unit 17 that monitors structured information that has undergone structuring processing. An example of the overall processing functions of the information processing system 1 in this case is shown in the block diagram of FIG.

The structured information determination processing unit 17 determines whether the structured information stored in the structured information storage unit 16 has an erroneous determination. For example, if the target information in structured information is a character string with more than a specified number of characters or less than a specified number of characters, or if it is information that is not related to the information type, the possibility of misjudgment of structured information is reduced. If it is determined that there is, a predetermined control process is executed.

Whether or not the information is related to the information type can be determined, for example, by referring to the text information input to the first display area 101 (the text information stored in the text information storage unit 12), and checking the text information of the information type. There is a method of judging whether or not it is included in the extracted information. In addition, information types and their corresponding important information are searched by a search engine, and whether search results exist or not, information types and their related words are stored in advance in association with predetermined dictionary information, and are stored as one of the important information. It can be determined by a method such as whether or not the relevant word is included in part or all. Other methods can also be used to determine whether the information is related to the information type.

It is preferable that the structured information determination processing unit 17 is executed particularly when the second display area 102 is not displayed. In this case, the predetermined control processing includes causing the display processing unit 15 to display the second display area 102, displaying a predetermined warning message, and the like.

In the above-described embodiment, the display processing unit 15 may further display the correspondence relationship between freely input text information and structured information structured based thereon. That is, for the text information reflected in the structured information among the text information displayed in the first display area 101, the display form such as character color, thickness, font, and decoration can be changed, or a mark can be added. good. An example of display in this case is shown in FIG. In Fig. 11, structured information (table) showing standard symptom names and their presence or absence based on "headache", "runny nose", "throat pain", "fever MAX 38 degrees", and "no diarrhea and vomiting" is shown. Based on "after catheter treatment for myocardial infarction", structured information (table) showing standardized medical history names and corresponding information (remarks) is generated. Structured information (table) indicating standardized drug names and therapeutic classification codes is generated based on "(statins)", and standardized diagnosis names and corresponding information (table) are generated based on "cold?" ing. Therefore, the display form is changed for each of these text information. In addition, in FIG. 11, text information is shown in bold and underlined.

Further, when the display form of the text information is changed, the information stored in each table of the structured information displayed in the second display area 102 can also be displayed in the corresponding display form such as color, thickness, font, and decoration. may be displayed to make it easier to grasp the correspondence relationship between the freely input text information and the structured information structured based on the text information. An example of display in this case is shown in FIG. In FIG. 12, each piece of text information and the table have the same display form for convenience of explanation, but the display form of the table and the corresponding text information may be different for each table.

In each of the above-described embodiments, the input person may directly input corrections, deletions, changes, additions, etc. to the structured information displayed in the second display area 102 . This input is stored in the structured information storage unit 16 .

By using the information processing system 1 of the present invention, a person who inputs freely text information can input text information while recognizing how it is structured. Become. Therefore, since the awareness of the structuring of the text information is made, when automatically inputting the text information, the input with the structuring in mind will be kept in mind. As a result, the freely input text information itself becomes an input suitable for structuring, and the accuracy of structuring can be improved.

1: information processing system 11: text information input acceptance processing unit 12: text information storage unit 13: structuring processing unit 14: matching dictionary storage unit 15: display processing unit 16: structured information storage unit 17: structured information determination processing Unit 70: Arithmetic device 71: Storage device 72: Display device 73: Input device 74: Communication device 100: Screen 101: First display area 102: Second display area 103: Structuring displayed in the second display area information

Claims (6)

An information processing system functioning on a computer having an arithmetic device and a display device,
The computing device is
a text information input reception processing unit that receives free input of text information in a free input field of an electronic medical chart ;
a structuring processing unit that converts the text information into table format structured information for each predetermined information type based on the text information extracted using natural language analysis processing from the text information that has received the free input;
a display processing unit for displaying the text information and the structured information on the display device;
The structuring processing unit
For the text information extracted by detecting a predetermined information type from the text information for which the free input is accepted, the target information corresponding to the information type is subjected to dependency analysis, context analysis, or a learning model using a neural network. extracted using one or more of the above natural language analysis processes of machine learning, and storing the extracted target information in a table format as structured information,
The display processing unit
displaying the text information in a first display area of the display device;
displaying the structured information in a second display area of the display device;
displaying the structured information in the second display area in response to input of text information in the first display area;
An information processing system characterized by: The display processing unit
The presence or absence of display of the second display area can be switched.
The information processing system according to claim 1, characterized by: The display processing unit
displaying the second display area when it is determined that the structured information satisfies a predetermined condition when the second display area is not displayed;
3. The information processing system according to claim 1, comprising: The display processing unit
displaying the text information reflected in the structured information among the freely input text information in the first display area by changing the display form or adding a mark;
4. The information processing system according to any one of claims 1 to 3, characterized by: The display processing unit
changing the display form of the text information reflected in the structured information among the freely input text information and displaying it in the first display area;
displaying in the second display area in correspondence with the display form of structured information based on the text information whose display form is changed;
5. The information processing system according to any one of claims 1 to 4, characterized by: the computer,
A text information input reception processing unit that receives free input of text information in the free input field of the electronic medical record ,
A structuring processing unit that generates structured information in a table format for each predetermined type of information based on the text information extracted from the text information for which the free input is accepted using natural language analysis processing ;
An information processing program that functions as a display processing unit that displays the text information and the structured information on a display device of the computer,
The structuring processing unit
For the text information extracted by detecting a predetermined information type from the text information for which the free input is accepted, the target information corresponding to the information type is subjected to dependency analysis, context analysis, or a learning model using a neural network. extracted using one or more of the above natural language analysis processes of machine learning, and storing the extracted target information in a table format as structured information,
The display processing unit
displaying the text information in a first display area of the display device;
displaying the structured information in a second display area of the display device;
displaying the structured information in the second display area in response to input of text information in the first display area;
An information processing program characterized by:

Images