Clinical chart creation support device, and clinical chart creation support method

Abstract

To appropriately perform creation of a clinical chart based on an entry format of a clinical chart.SOLUTION: A server 1 according to an aspect of the present invention comprises: a text dividing unit 15 that divides by clause a free text including information obtained from a patient; and a clause classification unit 16 that classifies the clauses obtained through the division performed by the text dividing unit 15 into any one of a plurality of sections based on an entry format of a clinical chart according to the details indicated by the clauses, and outputs a result of the classification.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a medical record creation support device and a medical record creation support method.

Background Art

[0002] Many dentists study medicine such as diagnosis and treatment methods during their studies at dental universities and the like, but often do not have many opportunities to deeply study how to write medical records. Therefore, many dentists graduate from dental universities and the like without learning how to write medical records and immediately start clinical practice.

[0003] However, providing high-quality dental care cannot be achieved only by improving treatment techniques. Recording patient information accurately and without delay in the medical record is also an important factor in providing high-quality dental care. In a dental clinic, since dentists may change, it is ideal to record a medical record that allows any dentist to understand the patient's condition as well as the attending dentist.

[0004] As a method for maintaining the quality of medical records created at a certain level, the "SOAP format", which is a medical record entry format in medical institutions, is known. The SOAP format is a method of entering information obtained from a patient through examination and the like into four items: S (subjective), O (objective), A (assessment), and P (plan).

[0005] "S" indicates subjective information, and the content of the chief complaint by the patient and the like are classified into S. "O" indicates objective information, and objective information obtained from examinations and tests is classified into O. "A" indicates an evaluation, and comprehensive evaluations such as the doctor's diagnosis and the results of analysis or interpretation based on the content of S and O are classified into A. "P" indicates a plan, and treatment policies and plans determined based on "A" are classified into P.

[0006] For example, Patent Document 1 describes an electronic medical record system in which diseases selected from a list of diseases displayed on a diagnosis screen, doctors' findings, medical treatment policies to be carried out thereafter, etc. are copied into the SOAP text, and the copied content is recorded in the medical record.

Prior Art Documents

Patent Documents

[0007]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0008] However, as described above, it is also a fact that many dentists do not deeply study the way of writing medical records. Therefore, inexperienced dentists may be troubled by how to create a medical record when outputting the content obtained through diagnosis as a record in the medical record. In addition, such dentists often classify and record the content obtained through diagnosis into incorrect SOAPs. Such problems commonly exist not only among dentists but also among all medical staff who record medical records.

[0009] The present invention has been made in consideration of the above situation, and an object of the present invention is to enable appropriate creation of a medical record based on the entry format of the medical record.

Means for Solving the Problems

[0010] A medical record creation support device according to an aspect of the present invention includes a text splitting unit that splits a free text including information obtained from a patient into segments for each clause, and each of the clauses split by the text splitting unit is classified into any one of a plurality of categories based on the entry format of the medical record according to the content indicated by the clause, and a clause classification unit that outputs the classification result.

Effects of the Invention

[0011] According to at least one aspect of the present invention, a medical record can be appropriately created based on the medical record entry format. Problems, configurations, and effects other than those described above will be clarified by the description of the following embodiments.

Brief Description of the Drawings

[0012]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Best Mode for Carrying Out the Invention

[0013] Hereinafter, examples of embodiments for carrying out the present invention (hereinafter referred to as "embodiments") will be described with reference to the accompanying drawings. Various numerical values and the like in the embodiments of the present invention are merely examples, and the present invention is not limited to the embodiments described below. Also, in this specification and the drawings, the same reference numerals will be given to the same components or components having substantially the same functions, and duplicate explanations will be omitted.

[0014] <Schematic Configuration of Medical Record Creation Support System> First, with reference to FIG. 1, the configuration of a medical record creation support system 100 according to an embodiment of the present invention will be described. FIG. 1 is a diagram showing a schematic configuration example of the medical record creation support system 100.

[0015] As shown in FIG. 1, the medical record creation support system includes a server 1 (an example of a medical record creation support device) and terminal devices 2-1 to 2-n (n is a natural number of 2 or more). The server 1 and each of the terminal devices 2-1 to 2-n are communicably connected via a network N.

[0016] The server 1 is placed in, for example, a cloud environment, generates a sentence for medical record transcription from free text (hereinafter simply referred to as "text") transmitted from a plurality of terminal devices 2, and transmits the generated sentence to the terminal devices 2. More specifically, the server 1 divides the text transmitted from the plurality of terminal devices 2 into clauses, classifies the clauses into any of the categories of SOAP. Then, the server 1 selects clauses for transcription into the medical record from among the classified clauses, forms the selected clauses into a sentence, and transmits it to the terminal devices 2.

[0017] The terminal device 2 is a device composed of a PC (Personal Computer) installed in a dental clinic and a mobile terminal such as a smartphone. Although only two terminal devices 2-1 and 2-n are shown as the terminal device 2 in FIG. 1, it is assumed that a plurality of other terminal devices 2 are also connected to the server 1. In addition, when it is not necessary to distinguish the terminal devices 2-1 to 2-n from each other in the following description, these are collectively referred to as the terminal device 2.

[0018] <Configuration of the control system of the medical record creation support system> Next, with reference to FIG. 2, the configuration of the control systems of the server 1 and the terminal device 2 that constitute the medical record creation support system 100 will be described. FIG. 2 is a block diagram showing a configuration example of the control systems of the server 1 and the terminal device 2.

[0019] [Server] As shown in FIG. 2, the server 1 includes a control unit 10, a storage unit 11, an input unit 12, a display unit 13, a communication I / F (Interface) unit 14, a text segmentation unit 15, a clause classification unit 16, a clause selection unit 17, and a text formatting unit 18, which are respectively connected to a bus B1. The control unit 10 is an arithmetic device composed of a CPU (Central Processing Unit) 101, a ROM (Read Only Memory) 102, and a RAM (Random Access Memory) 103.

[0020] The CPU 101 reads the program code of the software that realizes each function of the server 1 according to the present embodiment from the ROM 102, expands it in the RAM 103, and executes it. Note that the server 1 may include a processing device such as an MPU (Micro-Processing Unit) instead of the CPU 101. Variables, parameters, etc. generated during the arithmetic processing by the CPU 101 are temporarily written into the RAM 103.

[0021] As the storage unit 11, for example, an HDD (Hard Disk Drive), an SSD (Solid State Drive), a flexible disk, an optical disk, a magneto-optical disk, a CD-ROM, a CD-R, a non-volatile memory card, etc. can be used. In the storage unit 11, in addition to the OS (Operating System) and various parameters, a program for operating the server 1 is stored.

[0022] Note that the program for operating the server 1 may be stored in the ROM 102. The program is stored in the form of computer-readable program code. The CPU 101 sequentially executes operations according to the program code. That is, the ROM 102 or the storage unit 11 is an example of a computer-readable non-transitory recording medium storing a program to be executed by a computer.

[0023] The input unit 12 is composed of, for example, a mouse, a keyboard, etc., and generates an operation signal according to an operation by the user and supplies it to the CPU 101. The display unit 13 is a monitor composed of, for example, an LCD (Liquid Crystal Display), etc. Note that the input unit 12 and the display unit 13 may be integrally configured as a touch panel. Also, the server 1 may be configured not to include the input unit 12 and the display unit 13.

[0024] The communication I / F 14 is composed of a communication device or the like that controls communication performed with the terminal device 2.

[0025] The text splitting unit 15 splits the text transmitted from the terminal device 2 into a plurality of clauses based on information such as punctuation marks, line breaks, spaces, etc. in the text. The text splitting unit 15 is composed of, for example, a program for performing clause splitting, an AI (Artificial Intelligence) model that has performed machine learning, a language model such as an LLM (Large Language Models).

[0026] For an AI model that has undergone machine learning, a learning model or the like that has learned correspondence information between various sentences and each clause that makes up the sentence can be used as learning data. Also, when the text segmentation unit 15 is configured by a language model such as an LLM, by inputting a prompt such as "Please segment the following sentence into clauses" into the language model, the input text can be segmented into clauses.

[0027] The clause classification unit 16 is configured by an AI model or the like that has been fine-tuned by learning medical records created in the SOAP format by experienced dentists or the like. It is considered that in the medical records created by experienced dentists or the like, each clause included in the text and the SOAP category into which this clause should be classified are appropriately associated. That is, the clause classification unit 16 learns the correspondence information between each clause included in the text and the appropriate SOAP category into which the clause should be classified. Then, the clause classification unit 16 classifies each clause segmented by the text segmentation unit 15 into each SOAP category based on the learned content.

[0028] Note that the clause classification unit 16 does not classify each clause segmented by the text segmentation unit 15 into only one SOAP category, but classifies it into multiple categories according to the degree of confidence in the classification. Then, the clause classification unit 16 outputs to the terminal device 2 the category with the highest confidence in which the clause is classified and the confidence information for classifying the clause into this category. In addition, the clause classification unit 16 also outputs to the terminal device 2 other categories in which the same clause is classified and the confidence information for classification into each category. Note that when there is a clause that cannot be classified into any SOAP category, the clause classification unit 16 also outputs information indicating that the clause is unclassified.

[0029] The clause selection unit 17 selects clauses to be transcribed into the medical record from among the clauses classified by the clause classification unit 16. Specifically, the clause selection unit 17 deletes clauses that are not appropriate for transcription into the medical record from among the clauses classified by the clause classification unit 16. Phrases that are not suitable for transcription into the medical record include, for example, phrases for which the difference between the minimum and maximum confidence levels associated with each category into which the phrase is classified is less than or equal to a predetermined threshold, or phrases that are not classified into any of the SOAP categories.

[0030] For example, phrases that have little relevance to the content of the examination, etc., have a low confidence level regarding which SOAP category they are classified into, so the difference between the minimum and maximum confidence levels becomes small. Therefore, the phrase selection unit 17 excludes such phrases from the targets for transcription into the medical record.

[0031] The sentence formatting unit 18 formats each phrase classified into any of the SOAP categories into a natural sentence suitable for entry in the medical record. Note that the sentence formatted by the sentence formatting unit 18 is a sentence from which unnecessary phrases have been deleted by the phrase selection unit 17. When the sentence formatting unit 18 is configured by an LLM, the sentence formatting unit 18 can, for example, input phrases, SOAP classification results, and prompts into the LLM to format them into a natural sentence suitable for transcription into the medical record. For example, it is possible to configure the prompt with text such as "You are a dentist. You will now create a medical record in the SOAP format. Please format the sentence so that it becomes natural Japanese based on the input phrase and the SOAP classification information."

[0032] The classification results of the phrases into each SOAP category by the phrase classification unit 16 and the sentence formatted by the sentence formatting unit 18 are transmitted to the terminal device 2 and displayed on the screen of the terminal device 2 (see FIG. 9). A user, such as a dentist, can check the display content on the screen of the terminal device 2 and correct the correspondence between the phrases and the classification results into each SOAP category, or correct the formatted sentence.

[0033] [Terminal device] Next, with reference to FIG. 2 as well, the configuration of the control system of the terminal device 2 will be described. The terminal device 2 includes a control unit 20, a storage unit 21, an input unit 22, a display unit 23, a communication I / F unit 24, a display control unit 25, and a medical record transfer unit 26, which are respectively connected to the bus B2. The control unit 20 is an arithmetic unit composed of a CPU 201, a ROM 202, and a RAM 203.

[0034] The CPU 201 reads the program code of the software that realizes each function according to the present embodiment from the ROM 202, expands it in the RAM 203, and executes it. Alternatively, the CPU 201 directly reads the program code from the ROM 202 and executes it as it is. Note that the terminal device 2 may include a processing device such as an MPU instead of the CPU 201. Variables, parameters, etc. generated during the arithmetic processing by the CPU 201 are temporarily written into the RAM 203.

[0035] As the storage unit 21, for example, an HDD, an SSD, a flexible disk, an optical disk, a magneto-optical disk, a CD-ROM, a CD-R, a non-volatile memory card, etc. can be used. In addition to the OS and various parameters, programs for operating the terminal device 2, medical records, etc. are also stored in this storage unit 21.

[0036] Note that the program for operating the terminal device 2 may be stored in the ROM 202. The program is stored in the form of program code readable by a computer. The CPU 201 sequentially executes operations according to the program code. That is, the ROM 202 or the storage unit 21 is used as an example of a computer-readable non-transitory recording medium that stores a program executed by a computer.

[0037] The input unit 22 is composed of, for example, a mouse, a keyboard, a microphone, etc., and generates an operation signal according to the input by the user and supplies it to the CPU 201. Information obtained from the patient, such as the content of the chief complaint of the patient obtained by interview, the result of oral examination, and the result of X-ray examination, is input into the input unit 22 in text form by the user. Note that the information obtained from the patient may be input by voice via the microphone.

[0038] The display unit 23 is, for example, a monitor composed of an LCD or the like. On the display unit 23, for example, a medical record creation support screen Sc (see FIG. 8 etc.) for supporting the creation of a medical record is displayed. Note that the input unit 22 and the display unit 23 may be integrally configured as a touch panel.

[0039] The communication I / F 24 is composed of a communication device or the like that controls the communication performed with the server 1. The display control unit 25 performs control to display a medical record creation support screen Sc etc. on the display unit 23. The medical record transcription unit 26 transcribes the formatted text transmitted from the server 1 into the medical record.

[0040] <Example of text input to the terminal device> Next, an example of text input to the terminal device 2 by a user who is a dentist or the like, that is, an example of input of information obtained from the patient by examination or the like, will be described. FIG. 3 is a diagram showing an example of text input to the terminal device 2.

[0041] FIG. 3 shows input examples of three texts #1 to #3. In FIG. 3, a solid underline is drawn for clauses to be classified as S (subjective information), a one-dot chain underline is drawn for clauses to be classified as O (objective information), a two-dot chain underline is drawn for clauses to be classified as A (assessment), and a dashed underline is drawn for clauses to be classified as P (plan). Note that the information on the classification of each clause into the SOAP categories may be shown by color-coding or the like.

[0042] For example, in Example #1, the clause "I want teeth cleaning" should be classified as "S", and the clauses up to "Gingival redness (+) (abbrev.) Plaque: little" should be classified as "O".

[0043] Also, in Example #2, the clause "Calculus deposition is observed" should be classified as "O". Since the clause "Do not increase the side brushing with brushing" is difficult to classify into any category of SOAP, there is no underline indicating the type of the classified category. The clause "The oral condition has been improving" should be classified as "A", and the clause "Check brushing" should be classified as "P". The clause "I want to treat dental caries" should be classified as "S".

[0044] <Example of learning data for clause classification section> Next, the learning data of the clause classification section 16 will be described. FIG. 4 is a diagram showing an example of the learning data of the clause classification section 16. In the example shown in FIG. 4, clauses such as "I want teeth cleaning" associated with the category "S", "Periodontal examination + scaling" associated with the category "P", and "Observation of calculus deposition" associated with the category "O" are included in the learning data.

[0045] By creating learning data based on the medical records created by experienced dentists, the clause classification section 16 learned with the learning data can appropriately classify the input clauses into the correct (to be classified) categories of SOAP.

[0046] <Example of correspondence between clauses classified by clause classification section and confidence level of classification> Next, examples of the clauses classified by the clause classification section 16 and the confidence level of classification into categories will be described. FIG. 5 is a diagram showing an example of the correspondence between the clauses classified by the clause classification section 16 and the confidence level of classification.

[0047] In the topmost text field of FIG. 5, a clause "showing an improving trend due to the effect of brushing" is shown, and in the column of classification results, for each category of SOAP, it is shown that S is classified with a confidence level of "0.2", O is "0.05", A is "0.7", and P is "0.05". In the example shown in FIG. 5, the confidence level of classification is indicated by the Softmax function, and according to the Softmax function, the total of the confidence levels is always 1.0. Note that the confidence level of classification into a category may be indicated by a method other than the Softmax function.

[0048] In the text (clause) shown at the topmost of FIG. 5, since the confidence level of "0.7" for classification into the category of "A" is the highest, the clause classification unit 16 classifies the clause "showing an improving trend due to the effect of brushing" into the category of "A". Then, the clause classification unit 16 outputs the classification result of the clause and the information on the confidence level of classification in each category into which the clause is classified.

[0049] In the second example from the bottom of FIG. 5, it is shown that the classification results of the clause "It seems that a trip to Okinawa was made" for each category are "S: 0.3, O: 0.2, A: 0.25, P: 0.25". Since the clause "It seems that a trip to Okinawa was made" is a clause that is difficult to classify (has a weak relationship) into any category of SOAP, the confidence level values are also similar values in all of SOAP. Therefore, the difference between the maximum value of the confidence level, which is "0.3", and the minimum value, which is "0.2", also becomes small. Note that in FIG. 5, the texts (clauses) in the second to fifth paragraphs and the seventh paragraph can be considered in the same way.

[0050] In this embodiment, the clause selection unit 17 deletes a clause in which the difference between the maximum value and the minimum value of the confidence level is equal to or less than a predetermined value. As a result, clauses that are difficult to classify into each section of SOAP will not be transferred to the medical record. Note that as an initial value, a value such as 0.1 can be set for the "predetermined value", and it is preferable that this value can be adjusted by the user while viewing the classification result by the clause selection unit 17. For example, a range such as 0.0 to 1.0 can be set as the adjustment range by the user. The medical record creation support system 100 may transfer the information indicated by the clauses that are difficult to classify into each section of SOAP to the medical record as patient supplementary information or the like.

[0051] <Example of text formatting by the text formatting unit> Next, an example of formatting the text by the text formatting unit 18 will be described. FIG. 6 is a diagram showing an example of a text formatted by the text formatting unit 18 using clauses. In Example #1 of FIG. 6, there is shown a text "The patient visited the hospital on the basis of a request for tooth cleaning. As findings, gingival redness (+), calculus (+), no pus, and the depth of the pocket was mild in 90.9%, moderate in 8.0%, severe in 1.1%, and plaque was confirmed to be low."

[0052] This text is generated based on the text of Example #1 in FIG. 3. The types of underlines given to the text shown in FIG. 6 indicate the types of each section of SOAP into which the clauses are classified, similar to those shown in FIGS. 3 to 5. And the text without an underline is the part added by the text formatting unit 18. The text formatting unit 18 can generate a text that is natural in Japanese and suitable for entry in the medical record, as shown in Example #1 to Example #3 of FIG. 6, using the clauses classified into each section of SOAP.

[0053] <Example of the configuration and screen transition of the medical record creation support screen> Next, the configuration of the medical record creation support screen Sc displayed on the display unit 23 of the terminal device 2 and an example of the screen transition of the medical record creation support screen Sc will be described. First, referring to FIG. 7, the configuration of the medical record creation support screen Sc will be described. FIG. 7 is a diagram showing an example configuration of the medical record creation support screen Sc.

[0054] The medical record creation support screen Sc shown in FIG. 7 is a screen displayed when creating a medical record for patient Mr. X. At the top of the medical record creation support screen Sc, menus such as "File", "Edit", and "View" are displayed. Below the menu display area, three areas are provided in order from the left of the screen: the oral cavity information display area Ar1, the medical record creation area Ar2, and the patient information display area Ar3.

[0055] Information regarding the patient's oral cavity is shown in the oral cavity information display area Ar1. Information related to the creation of the medical record is displayed in the medical record creation area Ar2. Information regarding the patient, such as the patient's age, gender, and medical history, is displayed in the patient information display area Ar3. At the top of the medical record creation area Ar2, menus such as "Date Specify" and "Search" are displayed, and an AI menu Ma is provided at the right end of the menu. This AI menu Ma is a menu for executing the medical record creation support process according to this embodiment. When this AI menu Ma is selected by the user, the medical record creation screen Sc1 shown in FIG. 8 below pops up and is displayed on the screen of the medical record creation support screen Sc.

[0056] FIG. 8 is a diagram showing an example configuration of the medical record creation screen Sc1. A text input area Ar11 is provided on the medical record creation screen Sc1 shown in FIG. 8, and a classification button Bn1 is provided on the right side of the text input area Ar11. In the example shown in FIG. 8, information obtained from the patient through examination or the like is input into the text input area Ar11. The text input into the text input area Ar11 is the same as the text shown in Example #3 of FIG. 3.

[0057] The classification button Bn1 is a button for instructing the classification of the text input in the text input area Ar11 into each section of SOAP. When the classification button Bn1 is pressed, the text input in the text input area Ar11 is segmented into clauses by the text segmentation unit 15 and then classified into one of the sections of SOAP by the clause classification unit 16.

[0058] Figure 9 is a diagram showing an example of the display of the classification result on the medical record creation screen Sc1. On the medical record creation screen Sc1 shown in Figure 9, a classification result display area Ar12 and a sentence display area Ar13 are provided below the text input area Ar11.

[0059] In the classification result display area Ar12, the sections of SOAP, the clauses associated with the sections, and the information on the confidence level of the classification of each clause into each section of SOAP are displayed. By checking the content displayed in the classification result display area Ar12, the user can check the classification result of the text they input into each section of SOAP.

[0060] When the display part of the section of SOAP is selected by an operation such as tapping or clicking, as shown in Figure 9, the classification change instruction part Sc11 is displayed in the form of a balloon. In the classification change instruction part Sc11, each button corresponding to each section of SOAP and a "×" button are displayed. The user can change the section to which the clause is classified by pressing any button of SOAP. The "×" button is a button for inputting an instruction of "not classifiable".

[0061] On the right side of the text input area Ar11, a formatting button Bn2 and a transcription button Bn3 are arranged. The formatting button Bn2 is a button for instructing the formatting of the text using the classified clauses displayed in the classification result display area Ar12. When the formatting button Bn2 is pressed, text formatting by the text formatting unit 18 is performed, and the formatted text is displayed in the sentence display area Ar13. By checking the text displayed in the text display area Ar13, the user can determine whether the content of the text they entered is appropriately reflected in the text, or whether the generated text is suitable for transcription into the medical record, etc.

[0062] The transcription button Bn3 is a button that instructs to directly transcribe into the medical record those classified into the SOAP category among the classification results displayed in the classification result display area Ar12. An example of transcription into the medical record when the transcription button Bn3 is pressed will be described in detail with reference to the following Fig. 10.

[0063] Note that the text displayed in the text display area Ar13 can be modified by the user. The transcription button Bn4 arranged on the right side of the text display area Ar13 is a button that instructs to transcribe the text displayed in the text display area Ar13 into the medical record. Therefore, the user can also transcribe into the medical record the modified text obtained by modifying the text formatted by the text formatting unit 18 by modifying the text displayed in the text display area Ar13 and pressing the transcription button Bn4.

[0064] Next, an example of the display of the medical record after the transcription button Bn3 or the transcription button Bn4 is pressed on the medical record creation screen Sc1 will be described. Fig. 10 is a diagram showing an example of the display of the medical record when the text is displayed in association with each category of SOAP. The medical record creation support screen Sc shown in Fig. 10 is displayed when the transcription button Bn3 is pressed on the medical record creation screen Sc1 shown in Fig. 9. In the area surrounded by the thick frame in the medical record creation area Ar2 of the medical record creation support screen Sc, the classification results shown in the classification result display area Ar12 of the medical record creation screen Sc1 shown in Fig. 9 are displayed in association with each category of SOAP.

[0065] Therefore, according to the display example shown in FIG. 10, a user who checks the contents of the medical record can appropriately grasp the contents described in the medical record while checking the correspondence with each section of SOAP. If input fields corresponding to each section of SOAP are provided on the medical record system side (not shown), it is also possible to allocate and transfer the classification results shown in the classification result display area Ar12 of the medical record creation screen Sc1 shown in FIG. 9 to the input fields of each section. Further, if there are no such input fields on the medical record system side, the classification results for each section of SOAP can be transferred in a list, one line for each.

[0066] FIG. 11 is a diagram showing a display example of a medical record when only text is displayed. According to the display example shown in FIG. 11, a user who checks the contents of the medical record can easily understand the contents described in the medical record by natural text.

[0067] <Medical Record Creation Support Processing by Medical Record Creation Support System> Next, the medical record creation support processing by the medical record creation support system 100 according to the present embodiment will be described. FIG. 12 is a flowchart showing an example of the procedure of the medical record creation support processing by the medical record creation support system 100.

[0068] First, the input unit 22 of the terminal device 2 receives text input from the user (step SA1). This text is input by a dentist or the like without considering classification into SOAP or the like of information obtained from the patient through examination or the like. The text input to the terminal device 2 in step S1 is transmitted to the server 1 via the communication I / F unit 24 (see FIG. 2).

[0069] Next, the text splitting unit 15 of the server 1 splits the text transmitted from the terminal device 2 into sentence units (step SB1). Then, the text splitting unit 15 inputs each sentence obtained by splitting to the sentence classification unit 16 (step SB2). The clause classification unit 16 classifies each of the input clauses into any of the SOAP categories (step SB3). After the process of step SB3, the clause classification unit 16 outputs the clause and the information on the confidence level of the classification of the clause into the category. At this time, the clause classification unit 16 outputs not only the information on the category with the highest accuracy in which the clause is classified, but also the information on the categories with the second or lower confidence level and the information on the confidence level of the classification into each category.

[0070] Next, the clause classification unit 16 determines whether the clause to be classified is the last clause (step SB4). If it is determined in step SB4 that it is not the last clause (step SB4 is NO), the clause classification unit 16 changes the clause to be classified into each category of SOAP to the next clause and returns to step SB2.

[0071] On the other hand, if it is determined in step SB4 that it is the last clause (step SB4 is YES), the clause classification unit 16 returns the correspondence information between the classification result of the clause and the classification confidence level to the terminal device 2 via the communication I / F unit 14 (see FIG. 2) (step SB5). Then, the clause selection unit 17 excludes the clauses whose difference between the minimum value and the maximum value of the classification confidence level is less than or equal to a predetermined value from the candidates for formatting into the text for medical record transcription (step SB6).

[0072] On the other hand, the correspondence information between the classification result of the clause and the classification confidence level returned from the server 1 in step SB5 is received by the terminal device 2. Then, the display control unit 25 (see FIG. 2) of the terminal device 2 displays the correspondence information between the classification result of the clause and the classification confidence level on the medical record creation screen Sc1 (see FIG. 8) (step SA2). By performing the process of step SA2, a user such as a dentist can confirm, via the display content of the medical record creation screen Sc1, into which category of SOAP and with what confidence level each clause included in the text created by the user is classified.

[0073] Next, the user determines whether the classification result is appropriate (step SA3). If it is determined in step SA3 that the classification result is inappropriate (step SA3 is NO), the user corrects the classification. Then, the user transmits the corrected classification and the clause to the server 1 via the communication I / F unit 24 of the terminal device 2 (step SA4).

[0074] The clause classification unit 16 of the server 1 inputs the clause from which the clause that was not subject to sentence formatting in step SB6 has been excluded, the category into which the clause has been classified, and the prompt to the sentence formatting unit 18 (step SB7). Also, in step SB7, when information on the corrected classification is transmitted from the terminal device 2 in step SA4, the clause classification unit 16 inputs the information on the corrected classification to the sentence formatting unit 18.

[0075] Next, the sentence formatting unit 18 formats the sentence based on the clause input in step SB7, the category into which the clause has been classified, and the prompt, and returns the formatted sentence to the terminal device 2 via the communication I / F unit 14 (step SB8). Then, the medical record transcription unit 26 of the terminal device 2 transcribes the formatted sentence returned from the server 1 into the medical record (step SA5). After the processing of step SA5, the medical record creation support process by the medical record creation support system 100 ends.

[0076] In the medical record creation support system 100 according to the above-described embodiment, the text input by a user such as a dentist is divided into clauses, and the divided clauses are appropriately classified into the categories of SOAP. Therefore, according to the present embodiment, even a dentist or the like who is inexperienced in creating medical records and has little experience can appropriately create a medical record in accordance with the medical record entry format (SOAP format) based on the classification result.

[0077] In addition, in the medical record creation support system 100 according to the above-described embodiment, the text input to the terminal device 2 is automatically classified into the SOAP categories. Therefore, according to the present embodiment, it is possible to prevent the creation time of the medical record from being prolonged due to a dentist or the like being confused about the classification into SOAP.

[0078] In addition, in the above-described embodiment, the clause classification unit 16 outputs, as the classification result of the clause into the SOAP categories, not only the category with the highest classification confidence but also the information on the category with low confidence, together with the confidence. Then, this information is displayed on the medical record creation support screen Sc. Therefore, the user can confirm via the medical record creation support screen Sc whether the clauses corresponding to each SOAP category are appropriately included in the text created by the user, and whether the SOAP category assumed by the user was correct, and the like.

[0079] That is, the medical record creation support system 100 according to the present embodiment gives the user a hint for creating a sentence that further improves the classification accuracy for each SOAP category. Therefore, according to the present embodiment, it is possible to improve the medical record creation technique of a user such as a dentist who uses the medical record creation support system 100.

[0080] In addition, in the medical record creation support system 100 according to the above-described embodiment, the classification result of each SOAP category by the clause classification unit 16 is once presented to the user via the medical record creation support screen Sc (see FIG. 9), and the user's change of the category is accepted. Therefore, according to the present embodiment, even when the classification by the clause classification unit 16 is incorrect, the user can appropriately correct the correspondence between the clause and the SOAP category.

[0081] In addition, in the above-described embodiment, the clause selection unit 17 excludes from the output classification result the clauses in which the difference between the minimum value and the maximum value of the classification confidence for each SOAP category is equal to or less than a predetermined value. Therefore, according to the present embodiment, it is possible to prevent clauses that have little relevance to any of the SOAP categories and are not very related to the diagnosis content from being transcribed into the medical record.

[0082] In the above-described embodiment, the sentence formatting unit 18 constituted by a large language model or the like formats the sentence for transcription into the medical record. Therefore, according to the present embodiment, a medical record expressed in a natural sentence is generated.

[0083] In the above-described embodiment, an example is given in which the sentence is formatted by the sentence formatting unit 18 after the classification by the clause classification unit 16, but the present invention is not limited to this. For example, the sentence formatted by the sentence formatting unit 18 may be input to the clause classification unit 16.

[0084] Also, the clause classification unit 16 may be constituted by a large language model. In this case, for example, by inputting a prompt such as "Please answer with one of SOAP as the classification that can be read from this clause" to the clause classification unit constituted by the large language model, the clause can be classified into the SOAP categories. Then, the sentence formatting unit 18 formats one sentence corresponding to the category after classification using the clauses classified into any one of the SOAP categories.

[0085] By configuring the clause classification unit 16 and the sentence formatting unit 18 in this way, for example, when there are a plurality of pieces of information corresponding to each category of SOAP in the text input by a user such as a dentist, or when the text is not input in the order of S, O, A, P, each sentence corresponding to each classification of SOAP can be obtained as an output.

[0086] Also, the medical record creation support system 100 according to the present invention may be configured not to include the sentence formatting unit 18. Alternatively, it may be configured such that the user can determine whether or not to perform sentence formatting by the sentence formatting unit 18.

[0087] In the above-described embodiments, an example where the medical record entry format is in the SOAP format has been given. However, the present invention is not limited to this. The medical record entry format may be other formats such as SOAPIE (Subject Object Assessment Plan Intervention Evaluation).

[0088] In the above-described embodiments, an example where the text splitting unit 15 and the clause classification unit 16 are separately provided in the server 1 has been given. However, the present invention is not limited to this. The clause classification unit 16 may be configured to also perform text splitting.

[0089] In the above-described embodiments, an example where the content learned by the clause classification unit 16 is used only for creating medical records has been given. However, the present invention is not limited to this. For example, the medical record creation support system 100 according to the present invention may generate an educational screen for creating medical records based on the content learned by the clause classification unit 16 and display it on the display unit 23.

[0090] Note that the above-described embodiments have described the configurations of the devices (server, terminal device) and the system (medical record creation support system) in detail and specifically for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described.

[0091] Also, the control lines or information lines shown by solid lines in FIGS. 1 and 2 show those considered necessary for explanation, and not necessarily all the control lines and information lines are shown on the product. In practice, it may be considered that almost all the components are interconnected.

[0092] Also, in this specification, the processing steps for describing time-series processing include not only the processing that is performed in time series in the described order, but also the processing that is not necessarily processed in time series, and can be executed in parallel or individually (for example, parallel processing or object-based processing).

Explanation of Reference Numerals

[0093] 1... Server, 2... Terminal device, 2-1... Terminal device, 2-n... Terminal device, 15... Text segmentation unit, 16... Clause classification unit, 17... Clause selection unit, 18... Sentence formatting unit, 25... Display control unit, 26... Medical record transfer unit, 100... Medical record creation support system, Sc1... Medical record creation screen

Claims

1. A text splitting unit that splits a free text including information obtained from a patient into segments for each clause, A clause classification unit that classifies each of the clauses split by the text splitting unit into any one of a plurality of categories based on the entry format of a medical record according to the content indicated by the clause, and outputs a classification result, A medical record creation support device.

2. The clause classification unit is configured by a learning model that has learned correspondence information between the clauses included in the text and the categories into which the clauses are to be classified The medical record creation support device according to claim 1.

3. The clause classification unit classifies the clauses split by the text splitting unit into all of the categories that can be assumed as classification targets, and outputs the correspondence information between the clauses and the categories and the information on the confidence level of the classification into the categories as the classification result The medical record creation support device according to claim 2.

4. Further comprising a clause selection unit that excludes from the classification result the clauses that are difficult to classify into a specific one of the categories The medical record creation support device according to claim 3.

5. The clauses that are difficult to classify into the specific category are clauses in which the difference between the minimum value and the maximum value of the confidence level of classification when the clause classification unit classifies into each of the plurality of categories is equal to or less than a predetermined value, or clauses for which classification into the specific category is impossible The medical record creation support device according to claim 4.

6. Further comprising a sentence formatting unit that formats a sentence for entry into a medical record based on the classification result, and outputs the formatted sentence and information on the category into which each segment included in the sentence is classified The medical record creation support device according to claim 5.

7. The sentence formatting unit is configured by a large language model The medical record creation support device according to claim 6.

8. A display control unit that causes a display device to display a screen that displays the classification result and that enables correction of the correspondence relationship between the clauses and the categories in the classification result The medical record creation support device according to claim 7.

9. Comprising a medical record transcription unit that transcribes the classification result output from the clause classification unit or the sentence output from the sentence formatting unit into a medical record The medical record creation support device according to claim 8.

10. A text splitting procedure that splits a free text including information obtained from a patient into segments for each clause, A medical record creation support method, comprising: a clause classification procedure for classifying each of the clauses divided by the text division procedure into any one of a plurality of categories based on the medical record entry format according to the content indicated by the clause, and outputting a classification result. Medical record creation support method.

Images