Medical document creation support system, medical document creation support method, program

Abstract

This technology provides medical document creation support that is intuitive for users such as doctors and allows output in the user's preferred style of expression. [Solution] The medical document creation support system 2 includes a template acquisition unit 20 that acquires a template 10 which includes text that can be arbitrarily set by the user, keywords written in a manner indistinguishable from the text, and output manner commands that indicate the output manner of first information corresponding to the keywords in the patient's medical record 30; an identification unit 22 that can identify the keywords and output manner commands included in the acquired template 10; a document generation unit 23 that can perform the following: inputting a prompt to an extraction system 4 which includes an instruction to extract the first information from the patient's medical record 20; obtaining second information which is obtained by converting the first information into the output manner indicated by the output manner command; and generating a medical document which includes the second information at the position corresponding to the keywords in the template 10.

Description

Technical Field

[0001] This disclosure relates to a technology for assisting in the creation of medical documents.

Background Art

[0002] Doctors involved in a patient's medical care have the task of creating medical documents such as medical information provision documents and discharge summaries. A medical information provision document describes a patient's medical history, symptoms, test results, and treatment process so that doctors at the receiving medical institution can quickly and accurately understand the patient's medical history and current medical condition. A discharge summary describes a patient's medical history, the course of hospitalization, the state at the time of discharge, and future treatment plans in order to be useful when communicating with medical staff other than the attending physician at the medical institution to which the attending physician belongs after discharge.

[0003] Medical documents such as medical information provision documents and discharge summaries are created by doctors while referring to a patient's medical records including the patient's medical treatment records and test results. A system for assisting in the creation of medical documents is desired.

[0004] Patent Document 1 discloses a system using an artificial intelligence engine trained to input a patient's medical record and information for creating a referral letter and output any of a suspected disease name, affected area, necessity for detailed examination, and necessity for surgery.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] A system that performs specific learning on a model called an artificial intelligence engine so that specific information is output, as in Patent Document 1, requires appropriate learning according to its operation.

[0007] When using templates that can be freely configured by the user (a physician), some systems may require SQL statements to be included in the template to extract necessary information from medical records, or the settings for extracting necessary information from medical records may need to be configured separately from the template. However, there is a need for a system that can be used intuitively even by physicians who are not familiar with SQL or other information and communication technologies.

[0008] Furthermore, there is a demand to be able to format the information extracted from medical records in a way that is preferred by the doctor.

[0009] This disclosure provides medical document creation support technology that is intuitive for users such as doctors and allows output in the user's preferred style of expression. [Means for solving the problem]

[0010] The medical document creation support system of this disclosure includes: a template acquisition unit that acquires a template that can be set by the user and includes text that can be arbitrarily set by the user, keywords written in a manner indistinguishable from the text, and an output manner command that indicates the output manner of first information corresponding to the keywords in the patient's medical record; an identification unit that can identify the keywords and the output manner command included in the acquired template; a document generation unit that can perform the following: inputting a prompt to an extraction system that includes an instruction to extract the first information corresponding to the keywords from the patient's medical record; obtaining second information obtained by converting the first information corresponding to the keywords in the patient's medical record into the output manner indicated by the output manner command; and generating a medical document including the second information at the position corresponding to the keywords in the template. [Effects of the Invention]

[0011] According to this disclosure, when creating medical documents, users will be able to use the system intuitively and output documents in a style that suits their preferences. [Brief explanation of the drawing]

[0012] [Figure 1] This is a block diagram showing the medical document creation support system of the first embodiment. [Figure 2] This figure shows an example of a template. [Figure 3] This is a diagram showing an example of a medical record. [Figure 4] This figure shows an example of a prompt in the first embodiment and an example of output from the extraction system based on that prompt. [Figure 5] This figure shows an example of a generated medical document. [Figure 6] This diagram illustrates the types and meanings of output mode commands and the processing of output mode commands for the first embodiment and its modified examples. [Figure 7] This diagram illustrates the types and meanings of output mode commands and the processing of output mode commands for the first embodiment and its modified examples. [Modes for carrying out the invention]

[0013] [First Embodiment] Hereinafter, a first embodiment of this disclosure will be described with reference to the drawings.

[0014] Figure 1 is a block diagram showing the medical document creation support system 2 of the first embodiment. As shown in Figure 1, the medical document creation support system 2 of the first embodiment includes a template acquisition unit 20, a medical record acquisition unit 21, a specification unit 22, and a document generation unit 23. Each of these units (20-23) is realized by one or more processors of a computer reading a predetermined program. The medical document creation support system 2 can use an extraction system 4 to generate medical documents based on a template 10 and medical records 30 and provide them to users such as doctors.

[0015] Extraction System 4 is a machine learning model, such as an LLM (Large Language Model), that can respond with output corresponding to a prompt from the patient's medical record 30 based on the prompt, and is also called a generative AI capable of generating new content (such as text). Extraction System 4 can be provided by, for example, Microsoft Azure OpenAI, Google GCP Gemini / MedLM, Anthropic Claude, etc.

[0016] The template acquisition unit 20 acquires the template 10 from the template storage unit 11, such as a database, which stores the templates 10. The acquired template 10 may be one specified by the user or one of the default templates associated with the user. The template 10 is a template for medical documents, and is individually created in advance for each predetermined unit, such as each type of document or each medical department in a medical institution, and stored in the template storage unit 11.

[0017] Figure 2 shows an example of template 10. As shown in Figure 2, template 10 includes text that can be arbitrarily set by the user (medical professionals such as doctors), keywords that can be distinguished from the text, and output mode instructions. Template 10 does not include the patient's medical record 30. In Figure 2, for illustrative purposes, keywords are shown in light gray and output mode instructions are shown in dark gray. Examples of the text include greetings, names of medical departments, names of doctors, and preferred expressions in medical documents. The keyword is a word used to extract information from the medical record 30 and can be arbitrarily set by the user. For example, in order to extract "abdominal pain" from the medical record 30 containing the text "The patient complains of abdominal pain", any keyword such as "medical condition", "chief complaint", "symptom" can be set. Even if an arbitrary keyword is set, there is a high possibility that the extraction system 4 interprets the keyword and extracts information related to the keyword (hereinafter sometimes referred to as the first information corresponding to the keyword). As a result, intuitive use of the system becomes possible. The first information means information obtained by instructing the extraction system 4 to extract information corresponding to the keyword without specifying an instruction for the output format (output mode). In many cases, the information recorded in the patient's medical record 30 is directly extracted.

[0018] The output mode command indicates the output mode of the first information corresponding to the keyword in the patient's medical record 30 and can be used by the user to specify the output mode. The output mode command can also be said to be a command to convert the first information into the second information in the output mode indicated by the output mode command. The output mode command can instruct various output modes. Details will be described later.

[0019] In order to distinguish keywords, output mode commands, and text, in the first embodiment, keywords and output mode commands can be described in a distinguishable manner by being surrounded by a pair of first identifiers in template 10. Examples of the pair of first identifiers include, but are not limited to, the start tag “{” and the end tag “}”. Also, in the first embodiment, an output mode command can be described in a distinguishable manner from a keyword by being surrounded by a pair of second identifiers within the range surrounded by the pair of first identifiers (start tag “{” and end tag “}”). Similar to the pair of first identifiers, examples of the pair of second identifiers include, but are not limited to, the start tag “{” and the end tag “}”. A keyword can be described in a distinguishable manner from an output mode command by not being surrounded by the pair of second identifiers within the range surrounded by the pair of first identifiers (start tag “{” and end tag “}”). Since the pair of second identifiers are within the pair of first identifiers, the so-called first identifiers and second identifiers have a nested structure.

[0020] An example of template 10 shown in FIG. 2 includes eight keywords: “admission date”, “surgery date”, “affected eye”, “disease name”, “disease name”, “surgical procedure”, “progress”, and “discharge date”. These eight keywords in template 10 and the text other than the output mode commands associated with the keywords are not used for the prompt. In the example shown in Figure 2, the keyword "admission date" is associated with the output format instruction "Please format it as yyyy / mm / dd" by the paired first identifier and paired second identifier. In the example shown in Figure 2, the keyword "surgery date" is associated with the output format instruction "Please format it as yyyy / mm / dd" by the paired first identifier and paired second identifier. In the example shown in Figure 2, the keywords "surgical eye," "surgical procedure," and "progress" are not associated with any output format instructions. In the example shown in Figure 2, the keyword "disease name" associated with the output format instruction "Please express in Japanese," and the keyword "disease name" associated with the output format instruction "Please express in English," are included in template 10. In the example shown in Figure 2, the keyword "discharge date" is associated with the output format instruction "Please format it as mm / dd" by the paired first identifier and paired second identifier.

[0021] The medical record acquisition unit 21 shown in Figure 1 acquires medical records 30 from a medical record storage unit 31, such as a database (DB), which stores medical records 30. The medical records 30 to be acquired may be those specified by the user or those designated as default by the user. The medical records 30 are information that includes text representing the patient's medical records, test results, etc. For example, as illustrated in Figure 3, the medical records 30 include text. Figure 3 is a diagram showing an example of a medical record 30. Note that the execution order of the process by which the template acquisition unit 20 acquires the template 10 and the process by which the medical record acquisition unit 21 acquires the medical record 30 is not in any particular order.

[0022] The identification unit 22 shown in Figure 1 identifies one or more keywords and output mode commands associated with those keywords included in the template 10. In the first embodiment, it identifies that the portion enclosed by tags (a pair of first identifiers) is either a keyword or an output mode command. Furthermore, within the range enclosed by the pair of first identifiers, it identifies whether it is a keyword or an output mode command depending on whether it is enclosed by tags (a pair of second identifiers). This identification can be done using regular expressions or a parser. From the template 10 shown in Figure 2, eight keywords are identified: "admission date," "surgery date," "surgical eye," "disease name," "disease name," "surgical procedure," "progress," and "discharge date." In addition, if there are output mode commands for the eight keywords, those output mode commands are identified.

[0023] The document generation unit 23 shown in Figure 1 is capable of generating prompts and inputting prompts into the extraction system 4. A prompt includes an instruction to extract first information corresponding to a keyword from the patient's medical record 30. A prompt may include the data to be extracted (the patient's medical record 30). A prompt may include an instruction to specify the output format (output method). A prompt is written in a format acceptable to the extraction system 4. Figure 4 shows an example of a prompt in the first embodiment and an example of output from the extraction system 4 based on that prompt. The example shown at the top of Figure 4 includes two prompts. The first prompt includes an instruction section that instructs the extraction of first information corresponding to the keyword "Extract and display the following items from the following medical record: admission date, surgery date, eye operated on, disease name, surgical procedure, progress, discharge date." The first prompt also includes instructions to specify the output format, such as "However, the admission date should be in the format yyyy / mm / dd," "However, the surgery date should be in the format yyyy / mm / dd," "However, the disease name should be expressed in Japanese," and "However, the discharge date should be in the format mm / dd." The document generation unit 23 generates these based on the output format instructions in the template 10. The second prompt includes an instruction section that instructs the extraction of first information corresponding to the keyword "Extract the following item from the following medical record: Disease name." The second prompt also includes an instruction to specify the output format, such as "However, the disease name should be expressed in English." Figure 4 omits the illustration of the input data (patient's medical record 30).

[0024] When the document generation unit 23 shown in Figure 1 executes the process of inputting a prompt to the extraction system 4, the document generation unit 23 receives output from the extraction system 4 corresponding to the prompt. The lower part of Figure 4 shows the output from the extraction system 4 obtained using a prompt (see upper part of Figure 4) that instructs the system to extract information from the medical record 30 exemplified in Figure 3. As shown in the lower part of Figure 4, some output is obtained for eight keywords.

[0025] As shown in the lower part of Figure 4, for the eight keywords for which there is an output mode instruction, the output mode is as indicated by the output mode instruction. For example, in the medical record 30 shown in Figure 3, the first information corresponding to the surgery date is "2024-05-07", and the first information "2024-05-07" is converted to the second information "May 06, 2024" according to the output mode instruction. For example, in the medical record 30 shown in Figure 3, the first information corresponding to the hospitalization date is "2024-05-06", and the first information "2024-05-06" is converted to the second information "May 06, 2024" according to the output mode instruction. For example, in the medical record 30 shown in Figure 3, the first information corresponding to the disease name "glaucoma" is converted (translated) to the second information "glaucoma" and "Glaucoma" according to the output mode instructions for each of the two keywords. For example, in the medical record 30 shown in Figure 3, the first piece of information corresponding to the discharge date is "2024-05-10". The first piece of information "2024-05-10" is converted to the second piece of information "May 10th" according to the output method command. In other words, in the first embodiment, the document generation unit 23 shown in Figure 1 uses the extraction system 4 to perform a process to obtain second information, which is obtained by converting first information corresponding to keywords in the patient's medical record 30 into an output format indicated by the output format command. Furthermore, the document generation unit 23 can use the extraction system 4 to perform a process to acquire first information corresponding to keywords in the patient's medical record 30, and thereafter perform a process to convert the acquired first information into second information that will be the output format indicated by the output format command.

[0026] The document generation unit 23 shown in Figure 1 generates a medical document based on the template 10 and the output obtained from the extraction system 4. Specifically, the document generation unit 23 performs a process to generate a medical document by including second information at the positions corresponding to keywords in the template 10. Figure 5 shows an example of a generated medical document. In Figure 5, the second information is shown in light gray. As shown in Figure 5, the keywords, output mode commands, and identifiers in the template 10 shown in Figure 2 are replaced with second information obtained by the extraction system 4.

[0027] According to this system, users such as doctors simply write their preferred text into a template 10, which serves as a model for medical documents, and add keywords in a distinguishable way (using identification tags in the first embodiment) in the parts where they need to refer to medical records 30. The system then automatically generates medical documents by extracting primary information corresponding to the keywords from medical records 30 and supplementing it in the template 10. Users only need to review and modify the automatically generated documents, rather than creating them from scratch, making the creation of medical documents much easier. Furthermore, by associating output mode commands with keywords and describing them in template 10, it becomes possible to automatically generate medical documents by converting the first information corresponding to the keywords extracted from the medical record 30 into second information of the preferred output mode.

[0028] <Types, meanings, and processing of output mode commands for the first embodiment and modified examples> FIG. 6 and FIG. 7 are explanatory diagrams regarding the types and meanings of output mode commands and the processing of output mode commands for the first embodiment and the modification. As shown in FIG. 6, the first type of output mode command instructs the output mode of a numerical value, a date, or a time. The first type of output mode command shown in FIG. 6 instructs the output mode of a date as described above. The date is specified in the notation of "yyyy year mm month dd day". yyyy indicates the 4-digit AD year, mm indicates the month in 2 digits, and dd indicates the day in 2 digits. In the first embodiment, as shown in FIG. 6, the process of extracting the first information corresponding to the keyword and the process of converting it into the second information are realized by a prompt, but it is not limited thereto. For example, in the modification shown in FIG. 6, the process of extracting the first information is realized by a prompt, and the process of converting the first information into the second information is executed by the document generation unit 23. The conversion process depends on the programming language. For example, the to_date function is used. In the above first type of output mode command, an example of the specification of the date format is given, but it is not limited thereto. For example, if the output mode command for a numerical value is "\000,000", the first information "12345" can be converted into the second information "\12,345" (the format with a ¥ mark and commas). For example, if the output mode command for a numerical value is "00.00", the first information "12.34567" can be converted into the second information "12.35" (display up to the second decimal place). For example, if the output mode command for a date with time or a date-free time includes "HH hour", the first information "2024-12-01 22:12:34" can be converted into the second information "22 hours" (display of only the time).

[0029] The second type of output command shown in Figure 6 instructs the system to translate the first information corresponding to the keyword into a specified language. The second type of output command described above instructs the system to express the information in Japanese or in English (translate it into English). In the first embodiment, as shown in Figure 6, the process of extracting the first information corresponding to the keyword and the process of converting (translating) it into the second information are implemented in a single prompt, but the system is not limited to this. For example, in the modified example shown in Figure 6, the process of extracting the first information is implemented in the first prompt, and the process of translating the first information (glaucoma) obtained in the first prompt into English is implemented in the second prompt. Alternatively, the document generation unit 23 may execute the process of converting to the second information without using a prompt. In that case, a translation function that translates the Japanese term "glaucoma" into English, such as TRANSLATE("glaucoma","jp","en"), may be used.

[0030] The third type of output command shown in Figure 6 instructs the conversion of the first information corresponding to the keyword into a number or character. The third type of output command shown in Figure 6 instructs the conversion of the first information ("4000 pieces / μl") into a number. The second information converted into a number becomes "4000". In the first embodiment, as shown in Figure 6, the process of extracting the first information corresponding to the keyword and the process of converting it to the second information are implemented by prompts, but this is not limited to this. For example, in the modified example shown in Figure 6, the process of extracting the first information is implemented by prompts, and the process of converting the first information to the second information is executed by the document generation unit 23. The conversion process depends on the programming language, but for example, it can be exemplified by performing a process of extracting the numerical part from a string using the substring function and then performing a conversion process to a number using the number.parse function. The above third type of output instruction provides an example of a conversion process to a numerical value, but is not limited to this. For example, a conversion to a character may also be instructed. Conversion to a character can be used, for example, when using a DPC code (diagnostic group classification code) which combines numerical values, letters, and characters, to convert numerical values ​​into character information and then combine them to represent the code.

[0031] The fourth type of output command shown in Figure 6 instructs the output after conditional branching according to the first information corresponding to the keyword. An example of template 10 shown in Figure 6 is "if{second surgery}?{perform the second surgery,}else?{do not perform the second surgery,}". This means that conditional branching is performed according to the first information corresponding to the keyword "second surgery". If the first information corresponding to the keyword "second surgery" exists, it outputs "perform the second surgery," and if the first information corresponding to the keyword "second surgery" does not exist, it outputs "do not perform the second surgery,". Further keywords may be embedded in the template that results from the conditional branching.

[0032] The fifth type of output command shown in Figure 7 instructs the system to summarize the first information corresponding to the keyword. The fifth type of output command shown in Figure 7 instructs the system to summarize the first information corresponding to the keyword "progress" ("2024-05-07 No abnormalities 2024-05-08 No abnormalities 2024-05-09 No abnormalities 2024-05-10 No abnormalities" (see Figure 5)) within 100 characters. The 100-character limit is optional, and the number of characters can also be changed. In the first embodiment, as shown in Figure 7, the process of extracting the first information corresponding to the keyword and the process of converting (summarizing) it into the second information are implemented in a single prompt, but the system is not limited to this. For example, in the modified example shown in Figure 7, the process of extracting the first information is implemented in the first prompt, and the process of summarizing the first information obtained in the first prompt ("No abnormalities No abnormalities No abnormalities") within 100 characters is implemented in the second prompt. This allows the first piece of information, "No abnormalities, No abnormalities, No abnormalities, No abnormalities," to be summarized into the second piece of information, "No abnormalities."

[0033] The sixth type of output command shown in Figure 7 instructs the output of the option corresponding to a keyword from among multiple options. The sixth type of output command shown in Figure 7 instructs the system to select and output the corresponding option from among multiple options (○, △, ×) based on the first information corresponding to the keyword "result". For example, if the first information is a positive term such as "good", ○ is output; if the first information is a negative term such as "bad", × is output; and if the first information is a neutral term such as "neither good nor bad", △ is output. In the first embodiment, as shown in Figure 7, the process of extracting the first information corresponding to the keyword and the process of converting it to the second information are realized in a single prompt, but the system is not limited to this. For example, in the modified example shown in Figure 7, the process of extracting the first information is realized in the first prompt, and the process of selecting and outputting one of the options according to the first information obtained in the first prompt is realized in the second prompt. The extraction system 4 has the criteria for which of the multiple options the first information will be, and the extraction system 4 makes the selection according to that criterion.

[0034] The seventh type output command shown in Figure 7 instructs the calculation of information corresponding to multiple keywords. The seventh type output command shown in Figure 7 instructs the system to calculate the BMI by performing calculations using the first information corresponding to the keyword (weight) and the first information corresponding to the keyword (height), and to output the BMI. In the first embodiment, as shown in Figure 7, the process of extracting the first information corresponding to the keyword and the process of converting it to the second information are implemented in a single prompt, but the system is not limited to this. For example, in the modified example shown in Figure 7, the process of extracting the first information is implemented in the first prompt, and the process of instructing calculations using the first information obtained in the first prompt is implemented in the second prompt.

[0035] [Alternative Embodiment] (A) In the first embodiment, the output mode command is described in a way that distinguishes it by being enclosed by a pair of second identifiers within a range enclosed by a pair of first identifiers, and the keyword is described in a way that distinguishes it by being not enclosed by a pair of second identifiers within a range enclosed by a pair of first identifiers, but is not limited thereto. For example, the output mode command may be described in a way that distinguishes it by being not enclosed by a pair of second identifiers within a range enclosed by a pair of first identifiers, and the keyword may be described in a way that distinguishes it by being enclosed by a pair of second identifiers within a range enclosed by a pair of first identifiers.

[0036] (B) In the first embodiment, the first identifier and the second identifier may be the same identifier, but are not limited thereto. The first identifier and the second identifier may be different.

[0037] (C) In the above embodiment, if the medical document creation support system 2 can call the extraction system 4, the medical document creation support system 2 may or may not have the extraction system 4.

[0038] (D) In ​​the above embodiment, the document generation unit 23 replaces keywords in the template 10 with output corresponding to keywords obtained from the extraction system 4, for example, by replacing "disease name" with "glaucoma," but is not limited to this. For example, the keyword may be left in place without being replaced and deleted, such as replacing "disease name" with "disease name glaucoma," so that both the keyword and the output corresponding to the keyword are included in the medical document. As an example, the output may be written after the keyword, or before the keyword. In other words, including the mode in which the keyword is replaced and deleted, and the mode in which the keyword is left in place, the document generation unit 23 may perform a process to include the output corresponding to the keyword obtained from the extraction system 4 at the position in the template 10 corresponding to the keyword (such as before or after the keyword).

[0039] (E) In the above embodiment, the document generation unit 23 replaces and deletes identifiers that make keywords distinguishable, but is not limited to this, and identifiers may be left in place.

[0040] (F) In the above embodiment, there are cases in which one prompt is input to the extraction system 4 for one template, and cases in which multiple prompts are input to the extraction system 4 for one template 10, but the examples in the above embodiment are not limited to these cases. The number of prompts for one template 10 can be changed as appropriate.

[0041] [1] As described above in the embodiment, the medical document creation support system 2 may also include: a template acquisition unit 20 that acquires a template 10 which includes text that can be arbitrarily set by the user, keywords written in a manner indistinguishable from the text, and output manner commands that indicate the output manner of first information corresponding to the keywords in the patient's medical record 30, and which can be set by the user; an identification unit 22 that can identify the keywords and output manner commands included in the acquired template 10; a document generation unit 23 that can perform the following: inputting a prompt to the extraction system 4 that includes an instruction to extract first information corresponding to the keywords from the patient's medical record 30; obtaining second information obtained by converting the first information corresponding to the keywords in the patient's medical record 30 into the output manner indicated by the output manner command; and generating a medical document including the second information at the position corresponding to the keywords in the template 10. With this configuration, users such as doctors can simply write their preferred text in the template 10, which serves as a model for medical documents, and clearly indicate keywords in the sections where they need to refer to the medical record 30. This allows for the automatic generation of medical documents that supplement the primary information corresponding to the keywords from the medical record 30, making the creation of medical documents much easier. Furthermore, by associating output mode commands with keywords and describing them in template 10, it becomes possible to automatically generate medical documents by converting the first information corresponding to the keywords extracted from the medical record 30 into second information of the preferred output mode. Therefore, it becomes intuitively usable by users such as doctors, and allows for output in the expression preferred by the user.

[0042] [2] As in the medical document creation support system 2 described in [1] above, keywords and output mode commands may be described in a way that distinguishes them by being enclosed in a pair of first identifiers in the template 10, and either the keyword or the output mode command may be described in a way that distinguishes them by being enclosed in a pair of second identifiers within the range enclosed by the pair of first identifiers, and the other of the keyword or the output mode command may be described in a way that distinguishes them by not being enclosed in a pair of second identifiers within the range enclosed by the pair of first identifiers. This configuration allows keywords and output mode commands to be associated simply by nesting the first and second identifiers, making it easier for users such as doctors to use the system.

[0043] [3] As in the medical document creation support system 2 described in [1] or [2] above, the output mode command may specify the output mode as a number, date, or time. This makes it possible to output numbers, dates, or times in the desired output format.

[0044] [4] As in the medical document creation support system 2 described in any of [1] to [3] above, the output mode command may instruct the system to translate the first information corresponding to the keyword into a specified language. This makes it possible to automatically generate medical documents using a different language from the medical record 30.

[0045] [5] As in the medical document creation support system 2 described in any of [1] to [4] above, the output mode command may instruct the conversion of the first information corresponding to the keyword into numerical or character values. This makes it possible to automatically generate medical documents using desired numerical values ​​or characters.

[0046] [6] As in the medical document creation support system 2 described in any of [1] to [5] above, the output mode command may instruct the output after conditional branching according to the first information corresponding to the keyword. This makes it possible to automatically generate medical documents using desired expressions based on conditional branching.

[0047] [7] As in the medical document creation support system 2 described in any of [1] to [6] above, the output mode command may be an instruction to summarize the first information corresponding to the keyword. This makes it possible to automatically generate medical documents using summaries of medical records 30.

[0048] [8] As in the medical document creation support system 2 described in any of [1] to [7] above, the output mode command may instruct the output of the option corresponding to the keyword from among multiple options. This makes it possible to automatically generate standardized medical documents using multiple options.

[0049] [9] As in the medical document creation support system 2 described in any of the above [1] to [8], the output mode command may instruct the calculation of information corresponding to multiple keywords. This makes it possible to automatically generate medical documents using numerical calculations.

[0050]

[10] As in the embodiment described above, the medical document creation support method may also include: obtaining a user-configurable template 10 which includes text that can be arbitrarily set by the user, keywords written in a manner indistinguishable from the text, and output manner commands indicating the output manner of first information corresponding to the keywords in the patient's medical record 30; identifying the keywords and output manner commands included in the obtained template 10; inputting a prompt to the extraction system 4 that includes an instruction to extract the first information corresponding to the keywords from the patient's medical record 30; obtaining second information obtained by converting the first information corresponding to the keywords in the patient's medical record 30 into the output manner indicated by the output manner command; and generating a medical document by including the second information at the position corresponding to the keywords in the template 10.

[0051]

[11] As in the above embodiment, the program may cause one or more processors to execute the method described in

[10] above.

[0052] Although embodiments of this disclosure have been described above with reference to the drawings, it should be understood that the specific configurations are not limited to these embodiments. The scope of this disclosure is indicated not only by the description of the embodiments above but also by the claims, and further includes all modifications within the meaning and scope equivalent to the claims.

[0053] The structures adopted in each of the above embodiments can be adopted in any other embodiment. The specific configuration of each part is not limited to the above embodiments, and various modifications are possible without departing from the spirit of this disclosure. [Explanation of symbols]

[0054] 2: Medical document creation support system 4: Extraction System 10: Template 20: Template acquisition section 22: Specific part 23: Document Generation Department 30: Medical Records

Claims

1. A template acquisition unit that acquires a template that can be set by the user, which includes text that can be arbitrarily set by the user, keywords that are written in a manner that can be distinguished from the text, and output mode commands that indicate the output mode of first information corresponding to the keywords in the patient's medical record, An identification unit capable of identifying the keywords and output mode commands included in the acquired template, A document generation unit capable of performing the following processes: inputting a prompt to an extraction system that includes an instruction to extract the first information corresponding to the keyword from the patient's medical record; obtaining second information obtained by converting the first information corresponding to the keyword in the patient's medical record into an output format indicated by the output format instruction; and generating a medical document by including the second information at the position corresponding to the keyword in the template; A medical document creation support system equipped with the following features.

2. The aforementioned keywords and output mode commands are described in the template in a way that allows them to be distinguished by being enclosed in a pair of first identifiers. Either the keyword or the output mode command is described in a way that distinguishes it by being enclosed by the second identifier within the range enclosed by the first identifier, The medical document creation support system according to claim 1, wherein the other of the keyword and the output mode command is described in a way that distinguishes it by being enclosed within the range enclosed by the paired first identifier and not enclosed by the paired second identifier.

3. The medical document creation support system according to claim 1 or 2, wherein the output mode command specifies an output mode of a number, date, or time.

4. The medical document creation support system according to claim 1 or 2, wherein the output mode command instructs to translate the first information corresponding to the keyword into a specified language.

5. The medical document creation support system according to claim 1 or 2, wherein the output mode command instructs to convert the first information corresponding to the keyword into numerical values ​​or characters.

6. The medical document creation support system according to claim 1 or 2, wherein the output mode command instructs the output after conditional branching according to the first information corresponding to the keyword.

7. The medical document creation support system according to claim 1 or 2, wherein the output mode command instructs a summary of the first information corresponding to the keyword.

8. The medical document creation support system according to claim 1 or 2, wherein the output mode command instructs the output of a choice corresponding to the keyword from among a plurality of choices.

9. The medical document creation support system according to claim 1 or 2, wherein the output mode command instructs the calculation of information corresponding to a plurality of keywords.

10. Obtaining a user-configurable template that includes text that can be arbitrarily set by the user, keywords described in a manner indistinguishable from the text, and output mode commands indicating the output mode of first information corresponding to the keywords in the patient's medical record, Identifying the keywords and output mode commands included in the acquired template, The process involves inputting a prompt to the extraction system instructing it to extract the first information corresponding to the keyword from the patient's medical record; obtaining second information by converting the first information corresponding to the keyword in the patient's medical record into an output format indicated by the output format command; and generating a medical document by including the second information in the position corresponding to the keyword in the template. A method for supporting the creation of medical documents, including...

11. A program that causes one or more processors to execute the method described in claim 10.

Images