Program, information processing device, method, and system

Abstract

To provide a program, an information processing device, a method, and a system for saving the labor needed for recording electronic medical charts having been inputted based on voice data.SOLUTION: In an electronic medical chart system 1, a terminal device includes a memory 15 and a storage unit 16 for storing structured data of electronic medical chart templates. The structured data is composed of the input items and input contents of an electronic medical chart template that are associated with each other, and the input contents include a selective input content for selecting from options and / or a free input content that can be freely entered. A program stored in the storage unit causes a processor 19 to execute the steps of: receiving input of the input items of the electronic medical chart template and speech data including input contents corresponding to the input items; and specifying the recording content to be recorded to electronic medical chart template data on the basis of the selective input content and / or the free input content of the electronic medical chart template included in the speech data.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to a program, an information processing device, a method, and a system. [Background technology]

[0002] Electronic medical records are known in which a doctor electronically records the contents of a medical interview with a patient and the results of the interview, and further electronically records the history of medical treatments performed on the patient. The contents of the electronic medical record are sometimes created according to an electronic medical record template.

[0003] A technique related to the above-mentioned technique is disclosed in Patent Document 1.

[0004] Patent Document 1 discloses a technology relating to a medical support device. In the medical support device, an input item display means displays input items on a display. An input item selection means selects one of the input items. A voice recognition means performs voice recognition of the input voice using a selected dictionary, and extracts word candidates for the voice. A word candidate display means displays the extracted word candidates on a display. A selection operation receiving means receives a selection operation of one word candidate from the word candidates. A memory control means stores the one selected word candidate in a memory means as an answer to the one selected input item. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2013-156844 A Summary of the Invention [Problem to be solved by the invention]

[0006] In the technology described in Patent Document 1, speech recognition processing is performed using a specialized dictionary in the medical field, but template input in the medical field often involves homonyms with different spellings, and even at present there are certain limitations to the accuracy of speech recognition processing, so doctors and other medical personnel may need to make corrections to the content entered into the electronic medical record template as a result of the speech recognition processing.

[0007] Therefore, the present disclosure has been made to solve the above problem, and its purpose is to provide a technology that reduces the labor required for recording into an electronic medical record template input based on voice data. [Means for solving the problem]

[0008] A program for operating a computer having a processor and a memory. The memory stores structured data of an electronic medical record template, the structured data being data in which input items of the electronic medical record template are associated with input contents, and the input contents include at least one of multiple-choice input contents for selecting an option and free input contents allowing free description. The program causes the processor to execute a first step of receiving input of speech data from a user, the speech data including input items of the electronic medical record template and input contents corresponding to the input items, and a second step of specifying record contents to be recorded in the electronic medical record template data based on at least one of the multiple-choice input contents or free input contents of the electronic medical record template included in the speech data. Effect of the Invention

[0009] According to the present disclosure, it is possible to reduce the labor required for recording input data into an electronic medical record template based on voice data. [Brief description of the drawings]

[0010] [Figure 1] 1 is a diagram showing an overall configuration of a system according to an embodiment; [Diagram 2]FIG. 2 is a diagram illustrating a functional configuration of a terminal device according to an embodiment. [Diagram 3] FIG. 2 is a diagram illustrating a functional configuration of a server according to an embodiment. [Figure 4] FIG. 2 is a diagram showing a data structure of an electronic medical record database according to one embodiment. [Diagram 5] 11 is a flowchart illustrating an example of a processing flow in a system according to an embodiment. [Figure 6] 13 is a flowchart showing another example of the processing flow in the system according to an embodiment. [Figure 7] 13 is a flowchart showing yet another example of the processing flow in the system according to an embodiment. [Figure 8] FIG. 11 is a schematic diagram illustrating an example of a screen displayed on a terminal device according to an embodiment. [Figure 9] FIG. 11 is a schematic diagram illustrating another example of a screen displayed on a terminal device according to an embodiment. [Figure 10] FIG. 11 is a schematic diagram illustrating yet another example of a screen displayed on the terminal device according to an embodiment. [Figure 11] FIG. 11 is a schematic diagram illustrating yet another example of a screen displayed on the terminal device according to an embodiment. [Figure 12] FIG. 11 is a schematic diagram illustrating yet another example of a screen displayed on the terminal device according to an embodiment. [Figure 13] FIG. 11 is a schematic diagram illustrating yet another example of a screen displayed on the terminal device according to an embodiment. [Figure 14] FIG. 11 is a diagram illustrating a procedure for generating a letter of introduction by a system according to an embodiment. [Figure 15] FIG. 11 is a diagram illustrating a procedure for generating a letter of introduction by a system according to an embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, an embodiment of the present disclosure will be described with reference to the drawings. In all the drawings explaining the embodiment, the same reference numerals are given to common components, and repeated explanations are omitted. Note that the following embodiment does not unduly limit the contents of the present disclosure described in the claims. In addition, not all of the components shown in the embodiment are essential components of the present disclosure. In addition, each figure is a schematic diagram and is not necessarily illustrated strictly.

[0012] In the following description, a "processor" refers to one or more processors. The at least one processor is typically a microprocessor such as a CPU (Central Processing Unit), but may be another type of processor such as a GPU (Graphics Processing Unit). The at least one processor may be a single-core or multi-core.

[0013] Furthermore, the at least one processor may be a processor in the broad sense, such as a hardware circuit (for example, a field-programmable gate array (FPGA) or an application specific integrated circuit (ASIC)) that performs part or all of the processing.

[0014] In the following explanation, information that gives an output for an input may be described using an expression such as "xxx table", but this information may be data of any structure or a learning model such as a neural network that generates an output for an input. Therefore, the "xxx table" may be called "xxx information".

[0015] Furthermore, in the following description, the configuration of each table is an example, and one table may be divided into two or more tables, or all or part of two or more tables may be one table.

[0016] In addition, in the following explanation, the processing may be described with the "program" as the subject, but since the program is executed by a processor to perform a specified processing step by appropriately using a memory unit and / or an interface unit, etc., the subject of the processing may be the processor (or a device such as a controller having the processor).

[0017] The program may be installed in a device such as a computer, or may be, for example, in a program distribution server or a computer-readable (e.g., non-transitory) recording medium. In the following description, two or more programs may be realized as one program, or one program may be realized as two or more programs.

[0018] In the following description, identification numbers are used as identification information for various objects, but other types of identification information (for example, identifiers including kanji characters, alphabetic characters, or codes) may be used instead.

[0019] In addition, in the following description, when describing elements of the same type without distinguishing between them, reference signs (or common signs among the reference signs) may be used, and when describing elements of the same type with distinction between them, the identification numbers (or reference signs) of the elements may be used.

[0020] In the following description, the control lines and information lines are those that are considered necessary for the description, and not all control lines and information lines in the product are necessarily shown. All components may be connected to each other.

[0021] <0 System Overview> The system according to the present disclosure is a system for recording the contents of an electronic medical record based on an electronic medical record template by using voice recognition. In this specification, the contents of the electronic medical record are generated based on the electronic medical record template.

[0022] An electronic medical record template is structured data having input fields and input contents associated with these input fields. Here, structured data is data that is predefined before being placed in storage and formatted to have a certain prescribed structure. In contrast, unstructured data is data that is stored in plain text and is not processed until it is used. The input fields of an electronic medical record are defined based on this electronic medical record template. The system includes a system that assists in the input of template input fields using an electronic medical record template input assistance system that recreates the input fields of an electronic medical record template into a different form, such as a web form.

[0023] The input items correspond to each item in the electronic medical record, and are relatively short contents using medically specified terms so that medical professionals can identify which item it is. The input content is the content entered by a doctor or medical professional into the input item associated with this input content. The input content may be in multiple choice format or open-ended question format, and the question format may vary. If the input content is in multiple choice format, one of the options (there may be only one option), and if the input content is open-ended, free text is entered. Note that if the input content is in multiple choice format, the options include medically specified terms. The electronic medical record template data is the content entered by a doctor or medical professional based on the electronic medical record template, and is the specific content of the input content of the electronic medical record template.

[0024] In medical practice, doctors and medical staff need to input a large amount of data into electronic medical records based on electronic medical record templates, which are structured data, some of which are multiple choice format, and some of which are free response format.

[0025] The input items and input contents of the electronic medical record template are created on the assumption that they will be entered / modified / added by medical professionals and viewed by medical professionals. This means that the input items and input contents are based on medical knowledge and need to be medically accurate. However, the amount of information that medical professionals, including doctors, need to record in the electronic medical record is huge, and the effort required is enormous. For example, at the admission and discharge support center of a certain medical facility, there is about six pages of input content, and it takes about 20 minutes per patient to find which item in the vertically long profile column or assessment sheet to enter the content into and enter it there, which is the reason for overtime work for nurses and medical office staff.

[0026] From this perspective, it is conceivable to use voice recognition to record the contents of electronic medical records. However, even at present, the accuracy of voice recognition processing is not sufficient. Furthermore, in order to improve the accuracy of voice recognition, customization for each medical institution would be necessary, but as the amount of work required increases, costs rise and this puts a strain on the hospital's management. In the manufacturing industry, a method known as mass customization is used to design manufacturing processes that assume products will be customized, thereby increasing the unit price of the product, but there are no known precedents for such design in the field of voice input of electronic medical record templates.

[0027] Therefore, in the system according to the present disclosure, when recording the contents of an electronic medical record based on an electronic medical record template, the input items and input contents of the electronic medical record template are identified using medically designated terms included in speech data uttered by a medical professional as a key, it is identified which input item the speech data relates to and which input content item or option the speech data relates to, and the speech data is subjected to speech recognition for the identified input items and input contents, thereby identifying the contents of the electronic medical record. By adopting such a configuration, it is possible to efficiently and accurately record the contents of the electronic medical record using speech recognition technology.

[0028] The following three points are important to improve the accuracy of speech recognition. First, collect past data from each medical facility that has been input using templates, perform machine learning on the words contained in the past data, and create speech recognition specialized for that template input to improve accuracy. Second, create a voice input guide UI (user interface) that guides the user to speak the words that have already been learned when inputting the content of the speech. Specifically, a list of medically designated terms and examples of their input contents and options or input examples are displayed on the screen of the user, and when inputting the speech, the user is guided to read the words displayed on the screen as much as possible, thereby improving accuracy. Third, use on-site data to convert homonyms to the same spelling as much as possible, and guide more detailed kanji conversion if necessary.

[0029] By sharing the electronic medical record template created in this way, which comes with a voice recognition engine specialized for template input and a template input guide UI, among multiple medical facilities, it will be possible to use highly accurate voice recognition throughout Japan.

[0030] As a type of electronic medical record template, there are profile information sheets and assessment sheets that nurses enter. Therefore, the present invention also implies use for entering data for items such as profile information sheets and assessment sheets.

[0031] However, the accuracy of voice recognition is not 100%, just like with human operators. The fact that it is easier to check whether the input is correct also has a significant impact on user ease of use.

[0032] <One embodiment> <1 Overall system configuration> FIG. 1 is a diagram showing the overall configuration of an electronic medical record system 1 according to the present embodiment. As shown in FIG. 1, the electronic medical record system 1 includes a plurality of terminal devices (terminal devices 10a and 10b are shown in FIG. 1. Hereinafter, they may be collectively referred to as "terminal devices 10") and a server 20. The terminal devices 10 and the server 20 are connected to each other so as to be able to communicate with each other via a network 80. The network 80 is configured as a wired or wireless network. In this embodiment, the server 20 is a server having a function as a Web server (including a cloud server), and exchanges information with the terminal device 10 through Web pages. In addition, a Web page browser for viewing Web pages is installed in the terminal device 10, but a dedicated application for providing the services of the server 20 may be installed and configured to be viewable through the dedicated application.

[0033] The terminal device 10 is realized by a desktop PC (Personal Computer), a laptop PC, etc. Alternatively, the terminal device 10 may be a mobile terminal such as a tablet compatible with a mobile communication system, or a smartphone.

[0034] The terminal device 10 is a device operated by a medical professional or an administrator of the electronic medical record system 1. Here, the concept of a medical professional includes doctors, nurses, medical technicians, etc. In the following explanation, except when a distinction is made between medical professionals and the administrator of the system 1, the medical professional is considered to include the administrator of the system 1.

[0035] A medical worker uses the terminal device 10 to record the contents of an electronic medical record based on an electronic medical record template. At this time, the medical worker inputs speech data into the terminal device 10 and instructs the input / correction / addition. The terminal device 10 performs voice recognition on the speech data to obtain voice recognition data, and inputs / corrects / adds to the record contents based on this voice recognition data. The medical worker then gives an instruction to record the input contents that have been input / corrected / added as the record contents of the electronic medical record.

[0036] Furthermore, medical staff can create / modify / add to an electronic medical chart template using the terminal device 10. In this case, medical staff modify / add / delete the input items and input contents of the electronic medical chart template (delete here includes not only deleting the input contents, but also combining multiple input contents into one input content to reduce the overall number of input contents), and generate / modify / add / delete options for the input contents corresponding to the input items.

[0037] The terminal device 10 is communicatively connected to the server 20 via a network 80. The terminal device 10 is connected to the network 80 by communicating with communication devices such as a wireless base station 81 compatible with communication standards such as 4G, 5G, and LTE (Long Term Evolution), and a wireless LAN router 82 compatible with wireless LAN (Local Area Network) standards such as IEEE (Institute of Electrical and Electronics Engineers) 802.11. As shown in FIG. 1, the terminal device 10 includes a communication IF (Interface) 12, an input device 13, an output device 14, a memory 15, a storage unit 16, and a processor 19.

[0038] The communication IF 12 is an interface for inputting and outputting signals so that the terminal device 10 can communicate with an external device. The input device 13 is an input device (e.g., a keyboard, a touch panel, a touch pad, a pointing device such as a mouse, etc.) for receiving an input operation from a user. The output device 14 is an output device (a display, a speaker, etc.) for presenting information to a user. The memory 15 is for temporarily storing a program and data processed by the program, etc., and is a volatile memory such as a DRAM (Dynamic Random Access Memory). The storage unit 16 is a storage device for saving data, and is, for example, a flash memory or a HDD (Hard Disc Drive). The processor 19 is hardware for executing an instruction set described in a program, and is composed of an arithmetic unit, a register, a peripheral circuit, etc.

[0039] The server 20 is managed by an administrator of the electronic medical record system 1 of this embodiment, and the stored contents are modified / added / deleted as appropriate by medical staff who are users of the terminal device 10. The server 20 is an electronic medical record device, and in a medical facility, medical staff view the input items and input contents of the electronic medical record via the terminal device 10 and modify / add to the input contents. The server 20 also accepts editing operations of the electronic medical record template performed by the medical staff via the terminal device 10, and modifies / adds / deletes the electronic medical record template based on the editing operations.

[0040] The server 20 is a computer connected to a network 80. The server 20 includes a communication IF 22, an input / output IF 23, a memory 25, a storage 26, and a processor 29.

[0041] The communication IF 22 is an interface for inputting and outputting signals so that the server 20 can communicate with an external device. The input / output IF 23 functions as an interface with an input device for receiving an input operation from a user and an output device for presenting information to a user. The memory 25 is for temporarily storing programs and data processed by the programs, etc., and is a volatile memory such as a DRAM (Dynamic Random Access Memory). The storage 26 is a storage device for saving data, such as a flash memory or a HDD (Hard Disc Drive). The processor 29 is hardware for executing an instruction set described in a program, and is composed of an arithmetic unit, a register, a peripheral circuit, etc.

[0042] <1.1 Functional configuration of the terminal device 10> Fig. 2 is a block diagram showing an example of a functional configuration of the terminal device 10 shown in Fig. 1. The terminal device 10 shown in Fig. 2 is realized by, for example, a PC, a mobile terminal, or a wearable terminal. As shown in Fig. 2, the terminal device 10 includes a communication unit 120, an input device 13, an output device 14, an audio processing unit 17, a microphone 171, a speaker 172, a storage unit 180, and a control unit 190. The blocks included in the terminal device 10 are electrically connected by, for example, a bus or the like.

[0043] The communication unit 120 performs processing such as modulation and demodulation processing for the terminal device 10 to communicate with other devices. The communication unit 120 performs transmission processing on a signal generated by the control unit 190 and transmits the signal to the outside (for example, the server 20). The communication unit 120 performs reception processing on a signal received from the outside and outputs the signal to the control unit 190.

[0044] The input device 13 is a device for inputting instructions or information by a user who operates the terminal device 10. The input device 13 may be realized by, for example, a keyboard, a mouse, a reader, or the like. When the terminal device 10 is a mobile terminal or the like, the input device 13 is realized by, for example, a touch-sensitive device 131 in which an instruction is input by touching an operation surface. The input device 13 converts an instruction input by a user into an electrical signal, and outputs the electrical signal to the control unit 190. The input device 13 may include, for example, a receiving port that receives an electrical signal input from an external input device.

[0045] The output device 14 is a device for presenting information to a user who operates the terminal device 10. The output device 14 is realized, for example, by a display 141 or the like. The display 141 displays data according to the control of the control unit 190. The display 141 is realized, for example, by an LCD (Liquid Crystal Display) or an organic EL (Electro-Luminescence) display or the like.

[0046] The audio processing unit 17 performs, for example, digital-analog conversion processing of an audio signal. The audio processing unit 17 converts a signal provided from the microphone 171 into a digital signal and provides the converted signal to the control unit 190. The audio processing unit 17 also provides the audio signal to the speaker 172. The audio processing unit 17 is realized, for example, by a processor for audio processing. The microphone 171 accepts audio input and provides an audio signal corresponding to the audio input to the audio processing unit 17. The speaker 172 converts the audio signal provided from the audio processing unit 17 into audio and outputs the audio to the outside of the terminal device 10.

[0047] The storage unit 180 is realized by, for example, the memory 15 and the storage unit 16, and stores data and programs used by the terminal device 10. The storage unit 180 stores, for example, an electronic medical record template 182, speech data 183, voice recognition data 184, medically specified terminology data 185, teacher data 186, a learning model 187, and electronic medical record template data 188.

[0048] The electronic medical record template 182 is a template used by a medical professional to record the contents of an electronic medical record using the electronic medical record template 182, and is an electronic medical record template 2024 acquired from the storage unit 202 of the server 20 by an electronic medical record template acquisition unit 195 described later. The electronic medical record template 182 may be a part of the electronic medical record template 2024. That is, although details will be described later, the electronic medical record template 2024 stored in the storage unit 202 of the server 20 is all electronic medical record templates managed by the server 20 (i.e., the electronic medical record device), but the electronic medical record template to be modified / added to the terminal device 10 described later may be an electronic medical record template used by a medical professional when recording the contents of a record in a specific medical facility, sometimes in a specific medical office of a specific medical facility.

[0049] The speech data 183 is data of speech input by a medical professional recorded via the microphone 171 of the voice processing unit 17.

[0050] The voice recognition data 184 is voice recognition data obtained as a result of the voice recognition unit 196 of the control unit 190 performing voice recognition based on the speech data 183. As an example, the voice recognition data 184 may be data containing a mixture of kanji and hiragana, or may be data of a pair of kanji and its pronunciation in hiragana or katakana.

[0051] The medically designated term data 185 is medically designated term data used by the input item specification unit 197 of the control unit 190 to specify the input item related to the input content to be corrected / added when correcting / adding the input content of the electronic medical record based on the voice recognition data 184. The medically designated term is not limited to electronic medical records, but is generally used by medical professionals when writing in medical records, and includes at least so-called medical terms, and further includes specialized terms used in input items. Preferably, the medically designated term data 185 has a synonym dictionary of these medically designated terms, etc., and is used to identify and maintain the linkage of items when correcting and updating. Note that medical terms are terms used in medical institutions to accurately describe medical terms, such as "medical history," "history of current illness," "medication history," "social history," "disease name," and "medication name." In addition, there are synonyms for "history of current illness," such as "current symptom" and "HPI," and the linkage may be induced or confirmed using the synonyms when linking templates.

[0052] The learning model 187 is a learning model used when the voice recognition unit 196 performs voice recognition based on the speech data 183 uttered by a medical worker or the like and acquired by the microphone 171, and generates the voice recognition data 184. That is, the learning model 187 uses the speech data 183 as input data and outputs the voice recognition data 184.

[0053] The learning model 187 is obtained by making the machine learning model perform machine learning according to the model learning program (not shown) based on the teacher data 186. The teacher data 186 is generally used when performing voice recognition by machine learning, and consists of a pair of speech data of many speakers and text data corresponding to the speech data and correctly recognized by voice. In this case, the terminal device 10 may have a plurality of learning models 187 and teacher data 186. In particular, since a plurality of types of electronic medical record templates 182 may be used in this embodiment, it is preferable to appropriately select the learning model 187 (and thus the teacher data 186 that is the premise thereof) depending on the electronic medical record template, so that the storage unit 180 preferably has a plurality of learning models 187 and teacher data 186 corresponding to the learning model 187.

[0054] The learning model 187 according to this embodiment is, for example, a parameterized composite function in which a plurality of functions are combined. The parameterized composite function is defined by a combination of a plurality of adjustable functions and parameters. The prediction model according to this embodiment may be any parameterized composite function that satisfies the above requirements, but is assumed to be a multi-layer network model (hereinafter referred to as a multi-layer network). A prediction model using a multi-layer network has an input layer, an output layer, and at least one intermediate layer or hidden layer provided between the input layer and the output layer. The prediction model is expected to be used as a program module that is a part of artificial intelligence software.

[0055] As the multi-layered network according to the present embodiment, for example, a deep neural network (DNN) that is a multi-layered neural network that is the subject of deep learning may be used. As the DNN, for example, a convolution neural network (CNN) that targets images may be used.

[0056] In addition, the above is merely an example of a prediction model, and the prediction model may have other configurations. For example, the prediction model may be a rule-based model described by a function in which the chief complaint information and the environmental information are variables and each variable is assigned a coefficient derived from past performance.

[0057] Preferably, the teacher data 186 includes previously stored speech data 183 and the electronic medical chart template 2024. In other words, it is preferable that the teacher data 186 is not only teacher data for general speech recognition, but also actual data that has already been incorporated as an electronic medical chart template and is suitable for speech recognition for the electronic medical chart template 182.

[0058] Preferably, teacher data 186 includes medically designated terminology data 185. That is, learning model 187 is a model trained using medically designated terminology data 185. This makes it possible to improve the accuracy of speech recognition by speech recognition unit 196 based on speech data 183 including medically designated terms from medical professionals.

[0059] In addition, the teacher data 186 may include electronic medical record template data 2023 stored in the past, which naturally includes personal information such as the patient's name, date of birth, contact information, and very rare diseases. The Personal Information Protection Act strictly defines the management method for such personal information. Therefore, using such restricted electronic medical record template data 2023 for voice recognition processing requires strict handling and carries the risk of personal information leakage. Therefore, in the system 1 according to the present disclosure, at least a part of the actual electronic medical record template data 2023 is anonymized or modified, thereby protecting personal information and improving the convenience of data utilization.

[0060] Specifically, as the teacher data 186, the actual electronic medical record template data 2023 is modified by increasing or decreasing some of the numbers by a predetermined value, or by replacing the input contents of the same input items as those of the electronic medical record template data 2023 of another patient. In addition, personal information is made anonymous by masking proper nouns. At the same time, morphological analysis is performed on words, and very rare words are masked.

[0061] It is also preferable to include data on the relationship between kanji and its pronunciation as the teacher data 186. By using a learning model 187 that has been machine-learned using such teacher data 186, when the input item specification unit 197, which will be described later, specifies the correction content, the pronunciation can be input as the speech data 183, and an instruction to replace the pronunciation with another kanji having a similar sound can be given.

[0062] The electronic medical chart template data 188 is data in which specific input contents of the electronic medical chart template 182 are input as a result of the voice recognition process by the terminal device 10 .

[0063] The control unit 190 is realized by the processor 19 reading the application program 181 stored in the storage unit 180 and executing instructions included in the application program 181. The control unit 190 controls the operation of the terminal device 10. The control unit 190 operates according to the application program 181 stored in the storage unit 180, thereby fulfilling the functions of an operation reception unit 191, a transmission / reception unit 192, a data processing unit 193, a presentation control unit 194, an electronic medical chart template acquisition unit 195, a voice recognition unit 196, an input item identification unit 197, a record content recording unit 198, and an electronic medical chart template data transmission unit 199.

[0064] The operation reception unit 191 performs processing for receiving instructions or information input from the input device 13. Specifically, for example, the operation reception unit 191 receives information based on instructions input from a keyboard, a mouse, or the like.

[0065] Furthermore, the operation reception unit 191 receives voice instructions input from the microphone 171. Specifically, for example, the operation reception unit 191 receives a voice signal input from the microphone 171 and converted into a digital signal by the voice processing unit 17. For example, the operation reception unit 191 analyzes the received voice signal and extracts a predetermined noun, thereby acquiring an instruction from the user.

[0066] The transmitting / receiving unit 192 performs processing for the terminal device 10 to transmit and receive data to and from an external device such as the server 20 in accordance with a communication protocol. Specifically, for example, the transmitting / receiving unit 192 transmits the business content input by the user to the server 20. In addition, the transmitting / receiving unit 192 receives information about the user from the server 20.

[0067] The data processing unit 193 performs calculations on the data that the terminal device 10 accepts as input in accordance with the application program 181, and performs processing to output the calculation results to the memory 15 or the like.

[0068] The presentation control unit 194 controls the output device 14 to present the information provided by the server 20 to the user. Specifically, for example, the presentation control unit 194 causes the display 141 to display the information transmitted from the server 20. In addition, the presentation control unit 194 causes the speaker 172 to output the information transmitted from the server 20.

[0069] The electronic medical chart template acquisition unit 195 acquires the electronic medical chart template 2024 stored in the storage unit 202 of the server 20 , and stores it in the storage unit 180 as the electronic medical chart template 182 .

[0070] Based on speech data 183 input by a medical worker via microphone 171, speech recognition unit 196 inputs this speech data 183 into learning model 187, thereby obtaining speech recognition data 184 as an output result.

[0071] At this time, it is preferable that the voice recognition unit 196 performs voice recognition on the voice data 183 corresponding to the input items and input contents of the electronic medical record template 182 contained in the voice data 183, determines which input item the voice data 183 relates to, and based on this determination result, selects a learning model 187 suitable for the input items and input contents that are the determined results from the multiple learning models 187 stored in the memory unit 180, and performs voice recognition processing based on the selected learning model 187.

[0072] When performing voice recognition processing based on the speech data 183, the voice recognition unit 196 records the time when the voice recognition processing on the speech data 183 was performed, that is, the playback point of the speech data 183. Then, the voice recognition unit 196, together with the presentation control unit 194, presents the playback point of the speech data 183 in a state that is visible to the medical staff operating the terminal device 10 via the display 141 of the output device 14. The display mode of the playback point is arbitrary, and one example is a display mode in the form of a seek bar. In addition, a button or the like that accepts an instruction input to start / pause playback of the speech data 183 is displayed on the display screen of the playback point, and when there is an operation input of the button from the medical staff who is the operator of the terminal device 10, playback of the speech data 183 is started / paused. Accordingly, the voice recognition processing of the voice recognition unit 196 is started / paused.

[0073] Here, the input items and input contents of the electronic medical record template 182 include the name of the hospital and the name of the patient's family. For example, the name of the patient's family should ideally be written in kanji, but it is not necessarily wrong to write it in katakana. As a result of the voice recognition process by the voice recognition unit 196, it is difficult to identify the correct kanji writing, for example, to identify the kanji of the same sound but different writing such as Watanabe Akira-san and Watanabe Akira-san. Furthermore, it is time-consuming to correct the misrecognition. For this reason, it is preferable that the voice recognition unit 196 limit the writing to katakana or hiragana for the part determined to correspond to the name in the utterance data 183. At the same time, a separate UI displays candidates for what kanji to change the katakana writing to, and prompts the user to correct it by clicking, etc., thereby making it possible to input to change the katakana writing to a specific kanji writing. Note that if a kanji is uniquely identified among the past correct answer data, it may be displayed after conversion in advance. In addition, past correction contents may be stored and corrected automatically. In addition, instructions for correction may be given by voice, such as "Kanji conversion, Akira is the first character of 'Showa' (Showa)" after a specific wake-up word.

[0074] In addition, both tooth extraction and suture removal are medical terms pronounced "basshi," but the former is a word commonly used in dentistry and the latter in surgery. Both can be read as "basshi" during learning, and a function can be provided that prompts the user to select corrections for each department terminal or automatically makes corrections based on the department.

[0075] The input item specifying unit 197 refers to the voice recognition data 184 which is the output result of the voice recognition unit 196, and collates this voice recognition data 184 with the medically designated terminology data 185 to specify the input items of the electronic medical chart template included in the voice recognition data 184. Then, the input item specifying unit 197 specifies the input contents to be input / corrected / added from the voice recognition data 184 based on the speech data 183 spoken by the medical staff following the specified input item. In other words, the input item identification unit 197 recognizes that the input items and input contents in the speech data 183 (and thus the voice recognition data 184) are spoken continuously (continuous here means that there is a time interval that allows the speaker, the medical professional, to objectively recognize that the input items and the input contents that correspond to these input items and that have been entered / corrected / added are spoken in a continuous sequence. In other words, not only cases where there is no time interval at all, but also cases where it is objectively recognized as being spoken in a continuous sequence can fall into the category of continuous), and based on this continuity, the input contents that the speaker, the medical professional, has instructed to be entered / corrected / added are extracted from the voice recognition data 184, and candidates for correction / addition to the input / content are extracted based on this voice recognition data 184.

[0076] Furthermore, the input item specifying unit 197 may extract multiple candidates for the input item included in the utterance data 183 (voice recognition data 184) based on the above-mentioned continuity, and present to the operator of the terminal device 10 (i.e., the medical staff) which input item the correction / addition instruction is for. After this, the recorded content recording unit 198 accepts the input item selection instruction from the operator of the terminal device 10, and confirms the correction / addition content. As an example, the voice recognition results may be searched using a heuristic search algorithm or the like to search for kanji notation candidates, and if there are multiple candidates, they may be displayed to encourage kanji conversion.

[0077] Here, the speech data 183 of the medical worker who is the speaker is considered to have a series of a noun input item, a particle such as "wa" connected to this input item, and an input content to be corrected / added that is uttered following this particle. The input item specification unit 197 presumes that the speech data 183 uttered before this particle corresponds to the input item using a specific particle (for example, "wa") as a key, specifies the input item, and determines that the speech data 183 uttered following the particle (for example, "wa") uttered following the specified input item is input content associated with this input item and is input content uttered by the speaker to input / correct / add, and specifies the input content to be corrected / added based on the voice recognition data 184.

[0078] Furthermore, since the input item identification unit 197 can identify the input content from the continuity between the input item and the particle, if the input content can be expressed by a certain number of options (i.e., if the input content is in the form of options), it may prepare candidates (options) for the input content in advance and present the candidates for the input content to the operator (the medical professional who is the speaker) and ask him or her to select one of the candidates for the input content.

[0079] In this embodiment, an electronic medical record template 182 is stored in the memory unit 180 of the terminal device 10, and the input content is in the form of multiple choice in the electronic medical record template 182, and the content (description) of the options themselves is also specified. Therefore, the input item identification unit 197 can easily identify the option corresponding to the voice recognition data 184 related to the input content by using the medically specified terminology included in the voice recognition data 184.

[0080] During voice input, the input item identification unit 197 may change the screen display so that when the user speaks a medically designated term, a list of options is visible, and when the medically designated term is identified, the screen display may change so that the list of options is visible, the options or the medically designated term may be highlighted, or the screen may scroll to a location where the options are visible.

[0081] Furthermore, the input item identification unit 197 records the voice during voice input, and when an item name is clicked on the UI, playback is started from the start of the voice used during the corresponding voice input, making it possible to check whether the voice has been input correctly. If the voice is recognized as a medically specified term during voice playback, it may be possible to scroll to that item, for example, so that it is possible to check whether the input is correct.

[0082] Thereafter, the record content recording unit 198 accepts an instruction to select the input content from the operator of the terminal device 10, confirms the input / correction / addition content to the electronic medical chart template 182, i.e., the record content, and generates the electronic medical chart template data 188.

[0083] The operation of the above-mentioned input item identification unit 197 can also be realized as the operation of the speech recognition unit 196 by providing patterns (e.g., associations between input items, particles, and input contents, and candidates for the input contents) in the teacher data 186 and training the learning model 187 based on this teacher data 186.

[0084] Furthermore, if the input items of the electronic medical record template 2024 stored in the memory unit 202 of the server 20 are assigned an identifier, for example a number, for identifying each input item, and this identifier is also included in the electronic medical record template 182, the input item identification unit 197, upon determining that the voice recognition data 184 contains an identifier of the input item, may identify the input item to be input / corrected / added based on this identifier, and may further determine that the input / correction / addition content of the input content associated with the identified input item is included in the voice recognition data 184, and identify the correction / addition content of the input content.

[0085] Furthermore, information (not shown) that specifies and limits the types of words and / or characters that can be input as the input contents may be stored in the storage unit 180. In this case, when specifying the input contents from the voice recognition data 184, the input item specifying unit 197 may determine whether the text data included in the voice recognition data 184 matches the specified and limited information, and if it matches, may specify it as the input of the input contents to be input / corrected / added. As an example, if the input item is "patient contact information", the input contents must be a numeric string. Therefore, the part of the voice recognition data 184 that is a numeric string may be specified as the input contents. Alternatively, the voice recognition unit 196 may perform voice recognition processing based on such information, determining that if the input item is "patient contact information", the input contents uttered following this input item must be a numeric string.

[0086] Furthermore, in the case where the input contents are options, if the voice recognition data 184 contains a medically designated term included in one of the options, the input item specification unit 197 may determine that the voice recognition data 184 spoken by the medical professional who is the speaker to instruct input / correction / addition represents this option, and may specify the input contents to be input / correction / addition. Furthermore, when the input item specification unit 197 specifies an input item based on the voice recognition data 184, it may present options of the input contents associated with the specified input item. Thereafter, the record content recording unit 198 accepts a selection input from the medical professional who is the operator, and determines the input / correction / addition contents to the electronic medical record, i.e., the record contents, based on the accepted selection input.

[0087] The input item specification unit 197 may instruct the input item by the item serial number associated with the input item. For example, instead of specifying "no sleep disorder", it may specify based on the item number name, such as "item 34 is none". In this case, which item corresponds to item 34 is displayed on the display screen, and voice input is guided.

[0088] Here, the medical worker who is the speaker utters a predetermined word indicating a division of an input item or input content, for example, "line break," and as a result of voice recognition by the voice recognition unit 196, if the predetermined word is included in the voice recognition data 184, the input item identification unit 197 determines that the division of the input item or input content has been input by this word. Furthermore, the input item identification unit 197 also identifies the input item and input content for the voice recognition data 184 following this dividing word. In this way, even if the speaker makes a series of utterances, the input item and input content can be reliably identified.

[0089] The electronic medical record template 182 may also have a table structure. For example, when inputting a medical history, the input item specification unit 197 can structure the data in a table structure by speaking in a certain format. For example, if it is read as "In 2000, diagnosed with hypertension, treated at Nogaki Hospital, currently undergoing treatment. Line break. In 2010, diagnosed with dyslipidemia, treated at our hospital, currently cured," it is recognized that the medical history information follows the first "medical history." In fact, in this example, it is understood that there are two illnesses in the medical history, and that "The first illness, onset year 2000, illness name, hypertension, treating hospital, Nogaki Hospital, transcription, ongoing treatment," and "The second illness, onset year 2010, illness name, dyslipidemia, treating hospital, our hospital, transcription, cured." It would be possible to enter each of these information one by one, such as "Medical history 1 is high blood pressure, year of onset of medical history 1 is 2000, hospital where medical history 1 was treated is Nogaki Hospital," but by inputting table information by voice in the format shown above, it is possible to reduce the amount of speech and save time.

[0090] The input item specification unit 197 displays on the UI as "(Period: at age XX / Around XX years / From XX), (Diagnosis of) (Name of illness), At (Hospital name) (Treatment: Surgery / Oral medication / Inpatient treatment / Treatment for (Name of treatment)), Currently, (Transscription: Cured / Recovery / Under treatment)")", enabling the speaker to smoothly input information of the above table structure.

[0091] The voice recognition unit 196 and the input item specification unit 197 may display the progress of these operations on the display 141 during the operations of generating voice recognition data 184 based on the utterance data 183 and specifying the input items and input contents based on the voice recognition data 184. As an example, the voice recognition unit 196 and the input item specification unit 197 may display the voice recognition data 184 as text, and display the text-displayed voice recognition data 184 together with the specified input items and input contents. As an example, the text display may be displayed as a floating text box, and after the result of the voice recognition process by the voice recognition unit 196 is displayed as text as voice recognition data 184, when the input items and input contents are specified by the input item specification unit 197, the text box may be displayed so as to move to the location of the specified input item, etc.

[0092] In particular, the voice recognition unit 196 and the input item specification unit 197 may divide the same screen of the display 141 into left and right or top and bottom, display the electronic medical record template 182 on one side, and display sets of input items and input contents listed up and down or left and right on the other side. In such a display mode, since sets of input items and input contents are listed up and down or left and right, as the specification work by the input item specification unit 197 progresses sequentially, the text display related to the specified input items, etc. progresses sequentially either up and down or left and right. Then, in conjunction with the progress of the text display, the corresponding parts of the electronic medical record template 182 (i.e., the input items) may also progress sequentially up and down or left and right.

[0093] The recorded content recording unit 198 accepts input from the operator (including selection input and input of acceptance / cancellation, etc.) for the input items and input contents identified by the input item identification unit 197 and presented to the operator of the terminal device 10, inputs / modifies / adds to the input items and input contents based on the accepted input, and confirms the input of these input items, etc. Then, the recorded content recording unit 198 uses the confirmed input items and input contents to confirm the input / modification / addition contents to the electronic medical record, that is, the electronic medical record template data 188, which is the recorded contents. The recorded content recording unit 198 temporarily stores the confirmed electronic medical record template data 188 in the memory unit 180.

[0094] At this time, the recorded content recording unit 198 may present the operator with a choice of whether to correct or add to each of the input items and / or input contents identified by the input item identifying unit 197, and may input / correct / add to the input items and input contents based on a selection instruction from the operator. In particular, when an input content to be recorded already exists (i.e., an input content to be recorded already exists for the input item identified by the input item identifying unit 197), the recorded content recording unit 198 may display on the display 141 a message that confirms that an input content already exists and further whether or not this input content may be corrected / added to the medical staff operating the terminal device 10. Then, the input content may be corrected / added after waiting for an operation input from the medical staff instructing correction / addition.

[0095] Furthermore, when the input item / input content identified by the input item identifying unit 197 is not what the operator intended, the recorded content recording unit 198 may refer to the voice recognition data 184 stored in the storage unit 180 and instruct the input item identifying unit 197 to redo the specified action. Furthermore, when the input content identified by the input item identifying unit 197 is what the operator intended but the input item associated with this input content is different from what the operator intended, the recorded content recording unit 198 may refer to the voice recognition data 184 and accept an input from the operator that associates the identified input content with a different input item.

[0096] The electronic medical chart template data sending unit 199 sends the electronic medical chart template data 188 confirmed by the recorded content recording unit 198 to the server 20.

[0097] <1.2 Functional configuration of server 20> Fig. 3 is a diagram showing an example of the functional configuration of the server 20. As shown in Fig. 4, the server 20 exerts the functions of a communication unit 201, a storage unit 202, and a control unit 203.

[0098] The communication unit 201 performs processing for the server 20 to communicate with external devices.

[0099] The storage unit 202 includes, for example, an electronic medical record DB 2022, an electronic medical record template data 2023, and an electronic medical record template 2024.

[0100] The electronic medical record DB 2022 is a database for managing electronic medical record data on patients who have visited a medical facility that uses the server 20. The electronic medical record DB 2022 may manage electronic medical record data in a plurality of medical facilities. Details will be described later.

[0101] The electronic medical record template data 2023 is the electronic medical record template data 188 generated by the terminal device 10, and becomes part of the recorded contents of the electronic medical record by being imported into the electronic medical record DB 2022. The electronic medical record template data 2023 has input items and input contents associated with the input items. There is no particular limitation on the data format of the electronic medical record template data 2023, but the electronic medical record template data 2023 of this embodiment is data written in XAML (Extensible Application Markup Language) converted into JSON (JavaScript Object Notation) (JavaScript is a registered trademark) format. It is preferable that the electronic medical record template data 2023 has an identifier such as a numeric string assigned to the input item, and this identifier also constitutes the electronic medical record template data 2023.

[0102] The electronic medical record template 2024 is a template for generating the electronic medical record template data 2023. The electronic medical record template 2024 is structured data that specifies input items and input contents associated with these input items. There is no particular limitation on the data format of the electronic medical record template 2024, but the electronic medical record template 2024 of this embodiment is data written in XAML that has been converted into JSON format, similar to the electronic medical record template data 2023. Similar to the electronic medical record template data 2023, it is preferable that the electronic medical record template 2024 has an identifier such as a numeric string assigned to its input items, and this identifier also constitutes the electronic medical record template 2024.

[0103] Each electronic medical record template 2024 is associated with an identifier for identifying the respective electronic medical record template 2024. As an example, the identifier of the electronic medical record template 2024 is a numeric string of a predetermined number of digits. The identifier of the electronic medical record template 2024 is assigned by the electronic medical record template creation module 2033 described later.

[0104] The control unit 203 is realized by the processor 29 reading an application program 2021 stored in the storage unit 202 and executing instructions included in the application program 2021. The control unit 203 operates in accordance with the application program 2021 to perform functions shown as a reception control module 2031, a transmission control module 2032, an electronic medical chart template creation module 2033, and an electronic medical chart template data recording module 2034.

[0105] The reception control module 2031 controls the process in which the server 20 receives a signal from an external device in accordance with a communication protocol.

[0106] The transmission control module 2032 controls the process in which the server 20 transmits signals to external devices in accordance with a communication protocol.

[0107] By sharing an electronic medical record template created in one medical facility, which includes a voice recognition engine specialized for template input and a template input guide UI, among multiple medical facilities, it becomes possible to use highly accurate voice recognition throughout Japan. In order to achieve this sharing, the electronic medical record template creation module 2033 imports the electronic medical record template while maintaining the association between the electronic medical record template data and the voice recognition. Each item of the template is recognized programmatically by an individual identifier associated with each item, but when imported, an identifier may be automatically assigned, making it difficult to maintain the association. In such a case, a correspondence table of identifiers before and after import is created based on the fact that the location information and item information match the file exported immediately after import, and this correspondence table is used to associate the identifiers of each item of the imported template with the medically designated terms before import. This allows voice input to be performed using the identifiers of each item of the template of each medical institution. In addition, when importing, it may be possible to address this by adding a mechanism for maintaining the identifiers to the template import tool after confirming that the identifiers of each item to be added do not collide with other identifiers.

[0108] In the present embodiment, a plurality of answer candidates (candidates of input contents) of the medical interview questions may be associated with the input item. In other words, the input contents may be one answer candidate selected from the plurality of answer candidates. is stored in the storage unit 202.

[0109] The electronic medical chart template data recording module 2034 records the electronic medical chart template data 2023 in the storage unit 202 based on the electronic medical chart template data 188 acquired from the terminal device 10.

[0110] At this time, if electronic medical chart template data 2023 already exists for a specific patient (and furthermore in a specific medical office), the electronic medical chart template data recording module 2034 may display a screen for confirming with the operator of the terminal device 10 or the administrator of the server 20 whether the electronic medical chart template data 188 acquired from the terminal device 10 should be overwritten, added, or replaced to the already existing electronic medical chart template data 2023, and may request confirmation input from the operator, etc. Then, depending on the contents of the confirmation input, overwriting / addition / replacement, etc. to the electronic medical chart template data 2023 may be performed.

[0111] In particular, patient profile information, such as the patient's address, gender, height, weight, allergy information, etc., is patient-specific information and is unlikely to be changed, so it is preferable that the electronic medical record template data recording module 2034 always inputs confirmation such as overwriting the profile information.

[0112] When inputting the electronic medical record template data 2023, it is possible to initially input information by referencing the previous template input information from the electronic medical record, reducing the effort of inputting, but in that case, it may become difficult to distinguish between the parts that have been corrected as a result of voice recognition and the initial input values. By using different colors for parts that cannot be input by voice, parts that have been corrected by voice input, and parts that have not been input by voice, it is possible to show the user whether additional information is required or not.

[0113] To reduce the time and effort required to enter options, data from an electronic medical questionnaire or personal health records may be used. In that case, the structured data entered on a smartphone may be converted into a QR code (registered trademark), which may be read by a QR code reader on a terminal in the hospital's network and the data may be transferred to the hospital's network in its structured form.

[0114] In this case, there will be three inputs: information from the medical record, information from the electronic questionnaire, and information from voice recognition, and the origin of the data can be made clear to the user by changing the display, such as by changing the color of the shading, to indicate which information each is based on and whether there are any corrections. As another mechanism, information on whether an input item can be recognized by voice or not can be displayed by shading, etc.

[0115] Furthermore, the electronic medical record template data recording module 2034 may notify the medical fee calculation module (not shown) that the medical fee may be changed due to the input content, based on the content (particularly the input content) of the electronic medical record template data 2023. As an example, if the input content of the electronic medical record template data 2023 includes content indicating that the patient related to the electronic medical record template data 2023 is a dialysis patient, it may notify that the medical fee at the time of hospitalization should be changed appropriately.

[0116] Furthermore, the electronic medical record template data recording module 2034 may generate a sentence to be generated by a medical professional based on the electronic medical record template data 2023. A typical example of such processing is a referral letter or a discharge summary / admission summary provided by a medical professional to another medical facility based on the electronic medical record template data 2023. The items and contents to be written in the referral letter are based on the electronic medical record template data 2023, and the electronic medical record template data recording module 2034 generates text data to be written in the referral letter based on the electronic medical record template data 2023. At this time, the electronic medical record template data recording module 2034 may use the structured data as an input item as a prompt for a sentence generation task of a large-scale language model such as ChatGPT, and create a referral letter sentence or a summary proposal.

[0117] At that time, the electronic medical record data recording module 2034 may select from the structured data the content that is desired to be written in the referral letter or summary based on the chief complaint / referral purpose, or may automatically select using machine learning, and may use the selected structured data as an input item, a sentence template, or a prompt for a large-scale language model to create a referral letter sentence, summary, or report to a pharmaceutical company. At that time, a function for automatically selecting a template from a plurality of sentence templates based on the selected structured data may be implemented, or the structured data may be categorized and reorganized by category to create structured data based on the category. In this way, the electronic medical record template recording module 2034 can generate medical sentences such as a referral letter, a discharge summary / admission summary, and a report to a pharmaceutical company based on the electronic medical record template data 2023.

[0118] The output of a sentence generation task using a large-scale language model is often replaced with incorrect information. Changing values ​​such as the test name and test values ​​in the referral letter causes a major problem in terms of reliability, so it is necessary to check that they have not been replaced. Therefore, it is important for the electronic medical record data recording module 2034 to provide a means of easily checking that numbers, values, etc. are not misaligned as a UI.

[0119] A UI that records highlighting keywords associated with template item names and creates correspondences while highlighting the generated text, and displays which template description will be written where, is useful for confirmation. In addition, it is difficult to confirm when a number is switched, for example from 1.0 to 1.1. However, if the expression 1.0 occurs in multiple places in the same sentence, it is difficult to identify the correspondence using character matching.

[0120] Therefore, in order to prevent the numbers from being swapped, the electronic medical record template data recording module 2034 creates a prompt by changing the Na value presented in the prompt to 0.0000001 and the second K value to 0.0000002, values ​​that are less likely to conflict, and then uses that prompt to execute a sentence generation task, linking the 0.0000001 in the sentence created to the first actual Na value and the 0.0000002 to the actual K value, and simultaneously replacing the numbers, thereby generating a sentence while maintaining the correspondence.

[0121] In addition, by converting the values ​​that are difficult to collide into a format notation such as [Na value] [K value], it is possible to keep the correspondence unique. In one example, a sentence such as "The results of the electrolyte test during hospitalization were Na [Na value], K [K value]" is generated. In this way, the generated sentence can be used like a "sentence template", and from the next time onwards, the sentence creation task can be performed without generating sentences, and there is no need to recheck for discrepancies in values. It is also easy to check that the intended numbers and information are written correctly in the generated sentence.

[0122] As another example, structured information may be organized according to categories such as test values. In one example, structured data is divided into two parts, the value of the item name "Blood Sampling Result: Na" being "130 meq / l" and the value of the item name "Blood Sampling Result" being "K5.0 meq / l", and data is prepared in which both of these belong to the category of electrolyte test, and the value for the item name "Electrolyte Test Result" is structured as "Na130 meq / l, K5.0 meq / l", and the structured data is organized in one line, such as "Electrolyte Test Result: Na 130 meq / l, K5.0 meq / l", such as "The electrolyte test result on that day was [Electrolyte Test Result]." This may increase the sense of unity of the generated sentence. Template sentences such as "The electrolyte test result during hospitalization was [Electrolyte Test Result]" are possible.

[0123] Alternatively, as a text generation task using a large-scale language model, a prompt may be created that combines additional interview text for the patient, text instructing the patient to give a response to obtain response options, and structured text selected from the above template structured text, and based on the response obtained as a result of running the text generation task using the prompt, the response content may be unified in terms, and a UI may be created that allows the patient to input the response as a multiple-choice electronic questionnaire with unified terms, thereby creating a UI for the patient to input additional structured text with unified terms. In addition, when entering test values, a link to the test result report may be added to clarify the basis.

[0124] <2 Data Structure> Fig. 4 is a diagram showing the data structure of a database stored in the server 20. Note that Fig. 4 is an example, and does not exclude data that is not shown.

[0125] The database shown in Figure 4 is a relational database, which is used to manage data sets called tables, which are structured by rows and columns, by associating them with each other. In a database, a table is called a table, a column in a table is called a column, and a row in a table is called a record. In a relational database, it is possible to set relationships between tables and associate them.

[0126] Usually, a column that serves as a primary key for uniquely identifying a record is set in each table, but setting a primary key to a column is not essential. The control unit 203 of the server 20 can cause the processor 29 to add, delete, or update records in a specific table stored in the storage unit 202 according to various programs.

[0127] Fig. 4 is a diagram showing the data structure of the electronic medical record DB 2022. As shown in Fig. 4, each record of the electronic medical record DB 2022 includes, for example, an item "electronic medical record ID", an item "patient ID", an item "department ID", and an item "electronic medical record data". Each item of the electronic medical record DB 2022 is input by the electronic medical record template data recording module 2034 when the electronic medical record template data recording module 2034 generates the electronic medical record template data 2023. The information stored in the electronic medical record DB 2022 can be changed and updated as appropriate.

[0128] The item "electronic medical record ID" is an ID for identifying an electronic medical record managed by the system 1 (particularly the server 20) of this embodiment. The item "patient ID" is an ID for identifying a patient related to medical information managed by the electronic medical record identified by the item "electronic medical record ID". The item "department ID" is an ID for identifying a department related to medical information managed by the electronic medical record identified by the item "electronic medical record ID". The item "electronic medical record data" is information related to the file name of the electronic medical record template data 2023 related to the electronic medical record identified by the item "electronic medical record ID".

[0129] <3 Example of operation> An example of the operation of the terminal device 10 and the server 20 will now be described.

[0130] Fig. 5 is a flowchart showing an example of the operation of the terminal device 10. Fig. 5 is a flowchart showing an example of the operation when the operator of the terminal device 10 inputs / modifies / adds to the electronic medical chart template data 188 by voice input.

[0131] First, in step S500, the control unit 190 selects a patient related to the electronic medical chart template data 188 to be input / modified / added. Specifically, for example, the control unit 190 accepts a selection input of a patient from the operator of the terminal device 10 via the input device 13.

[0132] Next, in step S501, the control unit 190 calls up the electronic medical chart template 182 that is the basis of the electronic medical chart template data 188 related to the patient selected in step S500 from among the electronic medical chart templates 182 stored in the storage unit 180 of the terminal device 10.

[0133] Next, in step S502, the control unit 190 calls up the electronic medical chart template data 188 relating to the patient selected in step S500 from the electronic medical chart templates 182 stored in the storage unit 180 of the terminal device 10.

[0134] Next, in step S503, the control unit 190 causes the display 141 of the terminal device 10 to display an input guide for medically designated terms and the like, which is displayed as guidance for voice input by the user of the terminal device 10. Specifically, for example, the control unit 190 causes the input item identification unit 197 to display on the display 141 an input guide for medically designated terms and the like, which is displayed as guidance for voice input by the user of the terminal device 10.

[0135] Next, in step S504, the control unit 190 accepts voice input regarding the input items and input contents to be input into the electronic medical chart template data 188, following the input guide displayed in step S503, via the microphone 171 of the voice processing unit 17. Specifically, for example, the control unit 190 accepts voice input regarding the input items and input contents to be input into the electronic medical chart template data 188, following the input guide displayed in step S501, via the microphone 171 of the voice processing unit 17, using the voice recognition unit 196, and stores the voice input in the storage unit 180 as speech data 183.

[0136] Next, in step S504, the control unit 190 performs voice recognition processing on the speech data 183 accepted in step S502, and identifies the input items and input contents to be input / corrected / added from the voice recognition result. Specifically, for example, the control unit 190 performs voice recognition processing on the speech data 183 accepted in step S504 by the voice recognition unit 196 to obtain voice recognition data 184. Next, the input item identification unit 197 identifies the contents of the electronic medical record template data 188 to be input / corrected / added based on the voice recognition data 184.

[0137] Next, the control unit 190 presents the voice recognition contents, which are the input items, etc., identified in step S504, on the display 141. Specifically, for example, the control unit 190 causes the input item identification unit 197 to present the voice recognition contents, which are the input items, etc., identified in step S504, on the display 141.

[0138] Next, in step S505, the control unit 190 causes the display 141 to display conversion candidates for kanji characters based on the speech recognition result in step S504. Specifically, for example, the control unit 190 causes the display 141 to display conversion candidates for kanji characters based on the speech recognition result in step S504.

[0139] The voice recognition content is usually written in hiragana or katakana. Therefore, in step S506, the control unit 190 accepts an instruction input for converting the voice recognition content written in hiragana or the like into kanji, which is made by the user of the terminal device 10 using the input device 13. Specifically, for example, the control unit 190 accepts an instruction input for converting the voice recognition content written in hiragana or the like into kanji, which is made by the user of the terminal device 10 using the input device 13, through the input item specification unit 197.

[0140] Next, in step S507, the control unit 190 accepts the instruction input for kanji conversion given by the user of the terminal device 10 using the input device 13 in step S506, and determines the kanji conversion result based on this selection input. Specifically, for example, the control unit 190 accepts the instruction input for kanji conversion given by the user of the terminal device 10 using the input device 13 in step S506, through the input item specification unit 197, and determines the kanji conversion result based on this selection input.

[0141] In step S508, the control unit 190 confirms the contents of the electronic medical chart template data 188 with the kanji conversion result identified in step S507. Specifically, for example, the control unit 190 confirms the contents of the electronic medical chart template data 188 with the kanji conversion result identified in step S507 by the recorded content recording unit 198. Thereafter, the electronic medical chart template data sending unit 199 sends the confirmed electronic medical chart template data 188 to the server 20, and the electronic medical chart template data recording module 2034 of the server 20 stores the electronic medical chart template data 2023 sent from the terminal device 10 in the storage unit 202.

[0142] Thereafter, the control unit 203 of the server 20 imports the electronic medical chart template data 2023 input in step S506 into the electronic medical chart DB 2022. Specifically, for example, the control unit 203 imports the electronic medical chart template data 2023 transmitted from the terminal device 10 into the electronic medical chart DB 2022 by the electronic medical chart template data recording module 2034.

[0143] Fig. 6 is a flowchart showing an example of the operation of the server 20. Fig. 6 is a flowchart showing an example of the operation when an operator of the server 20 creates an electronic medical record template 2024 linked to a voice recognition engine.

[0144] First, in step S600, the control unit 203 selects an electronic medical record template that is the basis of the electronic medical record template 2024 created in Fig. 6. Specifically, for example, the control unit 203 selects an electronic medical record template that is the basis of the electronic medical record template 2024 created in Fig. 6 by the electronic medical record template creation module 2033. The electronic medical record template selected in step S600 may be the electronic medical record template 2024 stored in the storage unit 202 of the server 20, or may be stored in an external data server not shown in Fig. 1.

[0145] Next, in step S601, the control unit 203 acquires a large amount of electronic medical record template data to be used as learning data for a voice recognition engine to be linked to the electronic medical record template 2024 created in Fig. 6. Specifically, for example, the control unit 203 acquires a large amount of electronic medical record template data to be used as learning data for a voice recognition engine to be linked to the electronic medical record template 2024 created in Fig. 6 by the electronic medical record template creation module 2033. The electronic medical record template data acquired in step S601 may be the electronic medical record template data 2023 stored in the storage unit 202 of the server 20, or may be stored in an external data server not shown in Fig. 1.

[0146] The electronic medical record template data acquired in step S601 is input by a doctor or medical staff based on any of the electronic medical record templates 2024, and also includes profile information such as the patient's name. Therefore, in step S602, the control unit 203 performs anonymization processing mainly on the profile information of the electronic medical record template data acquired in step S601. Specifically, for example, the control unit 203 performs anonymization processing mainly on the profile information of the electronic medical record template data acquired in step S601 by the electronic medical record template creation module 2033.

[0147] Next, in step S603, the control unit 203 unifies homonymous kanji notations included in the electronic medical chart template data acquired in step S601. Specifically, for example, the control unit 203 unifies homonymous kanji notations included in the electronic medical chart template data acquired in step S601 using the electronic medical chart template creation module 2033.

[0148] In step S604, the control unit 203 creates correct phonetic pronunciation data based on the result of the unification of homonymous characters in step S603. Specifically, for example, the control unit 203 creates correct phonetic pronunciation data using the electronic medical record template creation module 2033 based on the result of the unification of homonymous characters in step S603.

[0149] Next, in step S605, the control unit 203 synthesizes / creates the voice correct answer data based on the voice correct reading data generated in step S604. Specifically, for example, the control unit 203 uses the electronic medical record template creation module 2033 to synthesize / create the voice correct answer data based on the voice correct reading data generated in step S604.

[0150] Then, in step S606, the control unit 203 uses the voice correct answer data created in step S605 to train the voice recognition engine (combination of teacher data and learning model). Specifically, for example, the control unit 203 uses the voice correct answer data created in step S605 by the electronic medical record template creation module 2033 to train the voice recognition engine (combination of teacher data and learning model).

[0151] Next, in step S607, the control unit 203 links the voice recognition engine trained in step S606 with the electronic medical chart template called up in step S600. Specifically, for example, the control unit 203 links the voice recognition engine trained in step S606 with the electronic medical chart template called up in step S600 by the electronic medical chart template creation module 2033. In addition, the electronic medical chart template creation module 2033 of the control unit 203 links the voice recognition input word candidates with medically specified terms / options / input items of the electronic medical chart template.

[0152] Then, in step S608, the control unit 203 displays the audio correct reading data generated in step S604 as a correct answer example in the guide of the electronic medical chart template. Specifically, for example, the control unit 203 displays the audio correct reading data generated in step S604 as a correct answer example in the guide of the electronic medical chart template by the electronic medical chart template creation module 2033. After this, the process returns to step S604, and the processes from creating the audio correct reading data to displaying the correct answer example are repeated.

[0153] 6, in the system 1 of this embodiment, the voice recognition engine is trained based on the input result of the electronic medical chart template data that has already been input, so that a voice recognition engine that is more customized than voice recognition by a general voice recognition engine, and furthermore, that has improved voice recognition accuracy in input of the electronic medical chart template, can be linked to the electronic medical chart template. As a result, by using the electronic medical chart template with the voice recognition engine linked to it, input operations using the electronic medical chart template can be performed with high accuracy.

[0154] Fig. 7 is a flowchart showing an example of the operation of the server 20. Fig. 7 is a flowchart showing an example of the operation when an operator of the server 20 exports an electronic medical chart template to be provided mainly to other medical institutions based on an already existing electronic medical chart template.

[0155] In step S700, the control unit 203 imports an electronic medical chart template that is the basis of the electronic medical chart template to be exported. Specifically, for example, the control unit 203 imports the electronic medical chart template that is the basis of the electronic medical chart template to be exported by the electronic medical chart template creation module 2033. The electronic medical chart template to be imported in step S700 may be the electronic medical chart template 2024 stored in the storage unit 202 of the server 20, or may be stored in an external data server not shown in FIG. 1.

[0156] Next, in step S701, the control unit 203 exports an electronic medical chart template to be provided mainly to other medical institutions based on the electronic medical chart template imported in step S700. Specifically, for example, the control unit 203 exports an electronic medical chart template to be provided mainly to other medical institutions based on the electronic medical chart template imported in step S700 by the electronic medical chart template creation module 2033.

[0157] Next, in step S702, the control unit 203 creates a correspondence table for the electronic medical chart template imported in step S700 and the electronic medical chart template exported in step S701 based on the matching of input items, more specifically, the matching of positions. Specifically, for example, the control unit 203 creates a correspondence table for the electronic medical chart template imported in step S700 and the electronic medical chart template exported in step S701, using the electronic medical chart template creation module 2033, based on the matching of input items, more specifically, the matching of positions.

[0158] Next, in step S703, the control unit 203 links the voice recognition input word candidates with the medically specified terms / options / input items of the electronic medical chart template using the correspondence table created in step S702. Specifically, for example, the control unit 203 links the voice recognition input word candidates with the medically specified terms / options / input items of the electronic medical chart template using the correspondence table created in step S702 by the electronic medical chart template creation module 2033. This linking operation may be performed by converting identifiers such as numbers associated with the input items of the electronic medical chart template.

[0159] This allows the electronic medical record template to be completed in step S704, which will be provided primarily to other medical institutions.

[0160] <4 Screen example> Hereinafter, examples of screens output to the terminal device 10 will be described with reference to FIGS.

[0161] FIG. 8 is a diagram showing an example of a screen displayed on the display 141 of the terminal device 10 when the operator of the terminal device 10 performs the input operation of the electronic medical chart template data 2023. As shown in FIG.

[0162] A screen 801 showing electronic medical record template data 2023, which is the input contents, is displayed on the left side of the screen 800 of the display 141 of the terminal device 10 as a result of voice recognition processing based on the input items of the electronic medical record template 2024 stored in the storage unit 202 of the server 20 and the speech data 183 from the operator. A voice input guidance screen 802 is displayed on the upper right side of the screen 800 of the display 141, and a screen 803 displaying the voice input result based on the speech data 183 is displayed on the lower right side of the screen 800. Furthermore, a screen 804 for displaying kanji conversion candidates based on the speech input is superimposed on the screen 803. Furthermore, buttons 805 and 806 for instructing the start of recording the speech data 183 and the saving of the speech data 183 are displayed on the lower part of the screen 800.

[0163] The operator of the terminal device 10 performs an operation input such as clicking these buttons 805 and 806 through the input device 13 to give an instruction to start recording the speech data 183 or an instruction to save the speech data 183 .

[0164] Fig. 9 is a diagram showing details of the screen 801 shown in Fig. 8. As already described, the screen 900 (801) is a screen showing the electronic medical chart template data 2023 which is the input contents input as a result of the voice recognition processing based on the input items of the electronic medical chart template 2024 and the speech data 183 from the operator.

[0165] A screen 900 displayed on the display 141 of the terminal device 10 displays an input item 901 of the electronic medical record template 2024 and an input content 902 associated with this input item 901. The input item 901 also displays a numeric string 903 for identifying this input item 901. As the terminal device 10 performs voice recognition processing, voice input results are sequentially input into the input content 902. Also, some items have already been input in the input content 902 of the electronic medical record template 2024, and the voice input is performed to add / correct the already input items. As shown in FIG. 9, the already input items are displayed in a specific color as "no corrections," and items that have been added / corrected by voice input are displayed in a color different from the already input items.

[0166] Fig. 10 is a diagram showing details of the screen 803 shown in Fig. 8. The screen 1000 (803) is a screen for displaying the voice input result based on the speech data 183, as already described.

[0167] An input item 1001, which is voice recognition data 184, and an input content 1002 associated with this input item 1001 are displayed on a screen 1000 of the display 141 of the terminal device 10. A numeric string 1003 for identifying this input item 1001 is also displayed on the input item 1001. As shown in Fig. 10, the voice recognition unit 196 and the input item identification unit 197 identify a portion to be input as the input content 1002 from the voice recognition data 184, and the portion identified as the input content 1002 is underlined 1004.

[0168] Fig. 11 is a diagram showing details of the screen 802 shown in Fig. 8. As already explained, the screen 1100 (802) is a guidance screen for voice input.

[0169] A screen 1100 of the display 141 of the terminal device 10 displays input items 1101 which are input items of the electronic medical record template 2024 and include medically specified terms, and examples 1102 of input contents to be entered into the input items 1101. Furthermore, if there are items already entered in the electronic medical record template 2024, the already entered items are displayed in the location of the examples 1102.

[0170] The playback of the speech data 183 may be started by the operator of the terminal device 10 selecting and inputting the input contents in FIG. 9 or the input items and display positions of the input contents shown in FIG.

[0171] FIG. 12 is a diagram showing a screen for displaying examples of kanji conversion candidates that are popped up on the screen 800 shown in FIG. 8 when the operator of the terminal device 10 is performing voice input using the screen 800. In FIG.

[0172] A screen 1200 of the display 141 of the terminal device 10 displays hiragana notation 1201 before conversion and conversion candidate examples 1202 when the voice recognition unit 196 performs kanji conversion based on the speech data 183. The operator of the terminal device 10 confirms the kanji conversion by selecting one of the conversion candidate examples 1202 using the input device 13.

[0173] Fig. 13 is a diagram showing an example of electronic medical record template data provided by another medical institution when importing an electronic medical record template. The configuration of the electronic medical record template data is similar to that shown in Fig. 9, so a detailed description will be omitted. Since it is electronic medical record template data, it even includes profile information.

[0174] FIG. 14 is a diagram for explaining the procedure for generating a letter of introduction text generated by the server 20 based on electronic medical record template data.

[0175] A list of electronic medical record template data (input contents) for a specific patient is displayed in the upper part of Fig. 14. From this list of input contents, the operator of the server 20 selects input contents necessary for generating the referral letter text and input contents not necessary (necessary / not necessary for the script).

[0176] After the selection of the input contents is completed, the server 20 generates a script to be input to the sentence generation task of the large-scale language model. The generated script is displayed in the middle of Fig. 14. As already explained, in this case, some numerical values ​​(in the illustrated example, numerical values ​​of the test results indicating the blood sampling results) are replaced with dummy values ​​in order to check whether the sentences are generated accurately when the sentence generation task is performed.

[0177] The results of inputting the script into a sentence generation task of a large-scale language model are shown in the lower part of Figure 14.

[0178] FIG. 15 is a diagram showing an example of character string replacement performed by the voice recognition unit 196 in the procedure for generating the letter of introduction text shown in FIG.

[0179] <5. Effects of one embodiment> As described above in detail, the system 1 of this embodiment allows the recorded contents of medical practice, such as an input electronic medical record, to be added to or corrected in a simple procedure. This point will be described in detail below.

[0180] The medical industry is an industry where mistakes are not tolerated. And in the medical industry, there is a high need for structured data using templates. Structured data can reduce medical errors. Electronic medical record templates make it possible to standardize business flows.

[0181] However, inputting structured data into electronic medical records is very time-consuming. For example, at the admission and discharge support center of a certain medical facility, there are six pages of input contents, and it takes about 20 minutes per patient to instruct which items the contents should be input into and input them there. With the system 1 according to this embodiment, this work can be reduced to about five minutes.

[0182] There are three reasons why using voice recognition to input structured data has not been used in the field until now: the first is that the accuracy of voice recognition is not sufficient, the second is that there are easier ways to input data than voice recognition, and the third is that it is difficult to correct mistakes.

[0183] In the system 1 according to the present embodiment, the first problem, the problem of the accuracy of the voice recognition, is solved by limiting the usage scenarios, simultaneously displaying the input item names of the template and the input content candidates related to these input items, and narrowing down the input voice patterns. In the system 1 according to the present disclosure, the WER (Word Error Rate) associated with the voice recognition is reduced from 6% to about 2%.

[0184] The second problem was solved by conducting multiple choice questions, which is an easier input method than voice recognition, before voice recognition and then displaying the results. Any parts of the input that were insufficient could then be corrected or added using voice recognition.

[0185] The third problem of making it difficult to correct mistakes is solved by displaying other candidates after inputting the text, allowing the user to select from the other candidates, and by making it easy to check what voice was used to input the text.

[0186] In particular, in the system 1 of this embodiment, since the input content in the multiple-choice format usually contains medically specified terms, the speech recognition process is performed using these medically specified terms as keys to identify at least the input content in the multiple-choice format, thereby further improving the accuracy of speech recognition.

[0187] <6 Notes> In addition, the above-described embodiments are described in detail to clearly explain the present disclosure, and are not necessarily limited to those including all of the described configurations. In addition, some of the configurations of each embodiment can be added to, deleted from, or replaced with other configurations.

[0188] In addition, the above-mentioned configurations, functions, processing units, processing means, etc. may be realized in part or in whole by hardware, for example, by designing them as integrated circuits. The present invention can also be realized by software program code that realizes the functions of the embodiments. In this case, a storage medium on which the program code is recorded is provided to a computer, and a processor included in the computer reads the program code stored in the storage medium. In this case, the program code itself read from the storage medium realizes the functions of the above-mentioned embodiments, and the program code itself and the storage medium storing it constitute the present invention. Examples of storage media for supplying such program code include flexible disks, CD-ROMs, DVD-ROMs, hard disks, SSDs, optical disks, magneto-optical disks, CD-Rs, magnetic tapes, non-volatile memory cards, ROMs, etc.

[0189] Furthermore, the program code for realizing the functions described in this embodiment can be implemented in a wide range of program or script languages, such as assembler, C / C++, perl, Shell, PHP, Java (registered trademark), and the like.

[0190] Furthermore, the program code of the software that realizes the functions of the embodiments may be distributed over a network and stored in a storage means such as a computer's hard disk or memory, or in a storage medium such as a CD-RW or CD-R, and the processor of the computer may read out and execute the program code stored in the storage means or storage medium.

[0191] The matters described in the above embodiments will be supplemented below. (Appendix 1) A program (181) for operating a computer having a processor (19) and memories (15, 16), wherein structured data of an electronic medical record template (182) is stored in the memory (15, 16), the structured data being data in which input items of the electronic medical record template (182) are associated with input contents, and the input contents include at least one of multiple-choice input contents for selecting an option or free input contents allowing free description, the program (181) causing the processor (19) to execute a first step (S504) of accepting input of utterance data (183) from a user, the utterance data (183) including the input items of the electronic medical record template (182) and input contents corresponding to the input items, and a second step (S507) of specifying record contents to be recorded in the electronic medical record template data (188) based on at least one of the multiple-choice input contents or the free input contents of the electronic medical record template (182) included in the utterance data (183). (Appendix 2) A program (181) for operating a computer having a processor (19) and memories (15, 16), wherein structured data of an electronic medical record template (182) is stored in the memory (15, 16), the structured data being data in which input items of the electronic medical record template (182) are associated with input contents, the input contents including multiple-choice input contents for selecting options, the program (181) causing the processor (19) to execute a first step (S504) of receiving input of speech data (183) from a user, the speech data (183) including the input items of the electronic medical record template (182) and input contents corresponding to the input items, and a third step (S507) of specifying record contents to be recorded in the electronic medical record template data (188) based on at least the multiple-choice input contents of the input contents of the electronic medical record template (182) included in the speech data (183). (Appendix 3) The second step (S507) includes a fourth step (S505) of performing voice recognition on the received speech data (183) to obtain voice recognition data (184) and identifying the record contents based on the voice recognition data (184), and the program (181) further executes a fifth step (S507) of at least one of correcting and adding to the input contents of the electronic medical record template (182) using the record contents identified in the fourth step (S505). The program (181) described in Appendix 1. (Appendix 4) The program (181) further causes the processor (19) to execute a sixth step of evaluating whether or not a medical fee application is possible based on the contents to be recorded in the electronic medical record template data (188) and displaying the result of the evaluation. The program (181) described in Appendix 1. (Appendix 5) In a fifth step (S507), a display manner of a portion of the record content identified in the fourth step (S505) where at least one of corrections and additions has been made is changed, and the portion where at least one of corrections and additions has been made is clearly indicated. (Appendix 6) In a fourth step (S505), when there are multiple kanji conversion candidates in the voice recognition data (184) based on the speech data (183), the kanji conversion candidates are displayed and a selection input for one of the kanji conversion candidates is accepted, thereby enabling a specific input of a kanji in the voice recognition data (184), the program (181) described in Appendix 3. (Appendix 7) In a fifth step (S507), when the record content identified in the fourth step (S505) is a person's name, at least a part of the record content is corrected and / or added to the input content of the electronic medical record template (182) to unify the spelling variation to katakana or hiragana for the person's name, the program (181) described in Appendix 3. (Appendix 8) The input contents and speech data (183) include an item number name, and in a fourth step (S505), the program (181) described in Appendix 3 identifies the recorded contents based on the item number name. (Appendix 9) The electronic medical record template (182) includes table information, and in a first step (S504), the speech data (183) includes a particle or term that identifies a column of the table in the table information, and a term that means moving to input information in the next row of the table in the table information or a term that identifies the row, and in a second step (S507), a program (181) described in Appendix 1 that identifies the recorded content for a specific row and column. (Appendix 10) The input content and the speech data (183) each contain medically specified terms, and in a fourth step (S505), the program (181) described in Appendix 3 identifies the recorded content using the medically specified terms contained in the voice recognition data (184) and the medically specified terms contained in the input content. (Appendix 11) A program (181) according to appendix 10, which in a fourth step (S505) presents a guide showing at least one of a medically designated term, an item number name, or an item name, and options and input examples for input items. (Appendix 12) The program (181) further causes the processor (19) to execute a seventh step of displaying at least one of the input items, the input contents, and the medically designated terms, item number names, or item names contained in the voice recognition data (184) on the same screen, and in the seventh step, also displays the input items of the electronic medical record template (182) related to the record contents identified in the fourth step (S505) on the same screen. The program (181) described in Appendix 11. (Appendix 13) In a seventh step, the program (181) described in appendix 12 starts playing back utterance data (183) by accepting a selection instruction for a displayed input item. (Appendix 14) A program (181) described in Appendix 12, which in a first step (S504) stores utterance data (183) in a memory (15, 16), and in a fourth step (S505) starts playback of the utterance data (183) by accepting a selection instruction for an input item displayed in the seventh step, and continuously changes the display position of the input items and input contents in conjunction with the playback of the utterance data (183). (Appendix 15) The program (181) further causes the processor (19) to execute an eighth step (S601) of accepting electronic medical chart template data (188) or personal healthcare record data in which input contents have already been entered, and in a fourth step (S505), changes and displays the input contents of the electronic medical chart template data (188) accepted in the eighth step (S601), the input items and input contents of the electronic medical chart template (182), and the voice recognition results of the voice recognition data (184). The program (181) described in Appendix 12. (Appendix 16) In a fourth step (S505), a combination of a medically specified term, which is a noun, and a particle connected to this medically specified term is identified from the speech recognition data (184), and the record content is identified based on the combination of this medically specified term and the particle. (Appendix 17) The memories (15, 16) store speech recognition engines (186, 187) for performing speech recognition, and the speech recognition engines (186, 187) include a learning model (187) trained using teacher data (186) including medically specified terms, and a program (181) described in Appendix 3. (Appendix 18) The speech recognition engine (186, 187) is a program (181) described in Appendix 17, which includes a learning model (187) learned using training data (186) including input items and input contents. (Appendix 19) The program (181) further causes the processor (19) to receive electronic medical record template data (188) into which input contents have already been input, and to perform anonymization processing on a part of the input contents of the received electronic medical record template data (188) in a ninth step (S601); generate teacher voice reading data for the voice recognition engines (186, 187) based on the input contents of the electronic medical record template data (188) received in the ninth step (S601), create voice correct answer data based on this teacher voice reading data, and perform machine learning of the learning model (187) based on this voice correct answer data. The program (181) described in Appendix 18. (Appendix 20) A program (181) according to appendix 1, in which an identifier for identifying an input item of each electronic medical record template (182) is associated with each electronic medical record template (182). (Appendix 21) 2. The program (181) according to claim 1, wherein the electronic medical record template (182) is shareable among a plurality of medical facilities, and the electronic medical record template (182) shareable among a plurality of medical facilities is associated with the same identifier. (Appendix 22) The program (181) causes the processor (19) to execute a tenth step (S700) of accepting the import of an electronic medical record template (182), and in the tenth step (S700), the program (181) described in Appendix 21 does not change the identifier when the electronic medical record template (182) is imported. (Appendix 23) The program (181) described in Appendix 3 causes the processor (19) to execute an eleventh step (S701) of accepting import of the electronic medical record template (182) and then exporting the accepted electronic medical record template (182), and a twelfth step (S702) of updating the correspondence between the voice recognition data (184) and the input items of the electronic medical record template (182) by generating a correspondence table from the consistency of positions of the input items of the electronic medical record template (182) imported and exported in the eleventh step (S701) and replacing identifiers based on the generated correspondence table. (Appendix 24) The program (181) causes the processor (19) to execute a 13th step of accepting a selection input for an input item of an electronic medical record template (182), and a 14th step of creating at least one of a referral letter, a medical summary, or a report to a pharmaceutical company by using a sentence template or by creating a prompt for a language generation model and using the language generation model, based on the input content corresponding to the input item selected in the 13th step. The program (181) described in Appendix 3. (Appendix 25) The program (181) causes the processor (19) to execute a 15th step of accepting a selection input for an input item of an electronic medical record template (182), and a 16th step of creating a prompt for a language generation model based on the input content corresponding to the input item selected in the 15th step, and using the language generation model to create at least one of a referral letter template, a medical treatment summary template, or a report to a pharmaceutical company, in which the input content of the electronic medical record template (182) and the corresponding sentences are linked. The program (181) described in Appendix 3. (Appendix 26) The information processing device (10) includes a processor (19) and memories (15, 16), in which structured data of an electronic medical record template (182) is stored in the memory (15, 16), the structured data being data in which input items of the electronic medical record template (182) are associated with input contents, and the input contents include multiple-choice input contents for selecting an option and free input contents allowing free description, and the processor (19) executes a first step (S504) of receiving input of speech data (183) from a user, in which the speech data (183) includes the input items of the electronic medical record template (182) and the input contents corresponding to these input items, and a second step (S507) of specifying record contents to be recorded in the electronic medical record template data (188) based on at least one of the multiple-choice input contents or the free input contents of the electronic medical record template (182) included in the speech data (183). (Appendix 27) The information processing device (10) includes a processor (19) and memories (15, 16), in which structured data of an electronic medical record template (182) is stored in the memory (15, 16), the structured data being data in which input items of the electronic medical record template (182) are associated with input contents, the input contents including multiple-choice input contents for selecting options, and the processor (19) executes a first step (S504) of receiving input of speech data (183) from a user, the speech data (183) including the input items of the electronic medical record template (182) and input contents corresponding to the input items, and a third step (S507) of specifying record contents to be recorded in the electronic medical record template data (188) based on at least the multiple-choice input contents of the input contents of the electronic medical record template (182) included in the speech data (183). (Appendix 28) A method executed by a computer (10) having a processor (19) and memories (15, 16), the method comprising: storing structured data of an electronic medical record template (182) in the memories (15, 16); the structured data being data in which input items and input contents of the electronic medical record template (182) are associated with each other; the input contents including multiple-choice input contents for selecting an option and free input contents allowing free description; the processor (19) executes a first step (S504) of receiving input of speech data (183) from a user, the speech data (183) including the input items of the electronic medical record template (182) and the input contents corresponding to the input items; and a second step (S507) of specifying record contents to be recorded in the electronic medical record template data (188) based on at least one of the multiple-choice input contents or the free input contents of the electronic medical record template (182) included in the speech data (183). (Appendix 29) A method executed by a computer (10) having a processor (19) and memories (15, 16), the method comprising: storing structured data of an electronic medical record template (182) in the memories (15, 16); the structured data being data in which input items of the electronic medical record template (182) are associated with input contents; the input contents including multiple-choice input contents for selecting options; and the processor (19) executing a first step (S504) of receiving input of speech data (183) from a user, the speech data (183) including the input items of the electronic medical record template (182) and input contents corresponding to the input items; and a third step (S507) of specifying record contents to be recorded in the electronic medical record template data (188) based on at least the multiple-choice input contents of the input contents of the electronic medical record template (182) included in the speech data (183). (Appendix 30) A system (1) comprising: a memory (15, 16) storing structured data of an electronic medical record template (182), the structured data being data in which input items of the electronic medical record template (182) are associated with input contents, the input contents including multiple-choice input contents for selecting an option and free input contents allowing free description; a means (196) for receiving input of speech data (183) from a user, the speech data (183) including the input items of the electronic medical record template (182) and input contents corresponding to these input items; and a means (197) for specifying record contents to be recorded in the electronic medical record template data (188) based on at least one of the multiple-choice input contents or the free input contents of the electronic medical record template (182) included in the speech data (183). (Appendix 31) A system (1) comprising: a memory (15, 16) storing structured data of an electronic medical record template (182), the structured data being data in which input items of the electronic medical record template (182) are associated with input contents, the input contents including multiple-choice input contents for selecting an option and free input contents allowing free description; a means (196) for receiving input of speech data (183) from a user, the speech data (183) including the input items of the electronic medical record template (182) and input contents corresponding to these input items; and a means (197) for specifying record contents to be recorded in the electronic medical record template data (188) based on at least the multiple-choice input contents of the input contents of the electronic medical record template (182) included in the speech data (183). [Explanation of symbols]

[0192] 1...Electronic medical record system 10, 10a, 10b...Terminal device 15...Memory 16...Storage unit 19...Processor 20...Server 25...Memory 26...Storage 29...Processor 180...Storage unit 181...Application program 182...Electronic medical record template data 183...Speech data 184...Voice recognition data 185...Medical designated terminology data 186...Teacher data 187...Learning model 190...Control unit 191...Operation reception unit 192...Transmission / reception unit 193...Data processing unit 194...Presentation control unit 195...Electronic medical record template data acquisition unit 196...Voice recognition unit 197...Input item identification unit 198...Recorded content recording unit 199...Electronic medical record template data transmission unit 201...Communication unit 202...Storage unit 203...Control unit 2021...Application program 2022...Electronic medical record DB 2023: Electronic medical record template data 2024: Electronic medical record template 2031: Reception control module 2032: Transmission control module 2033: Electronic medical record template creation module 2034: Electronic medical record template data recording module

Claims

1. A method for generating medical documents, which is performed by an information processing device comprising memory and a processor, (a) A step of obtaining structured data relating to the patient from the memory, (b) A step of assigning a uniquely identifiable, non-collisionable value or non-collisionable format notation to each of the multiple input items included in the structured data, and generating information that maintains the correspondence between the input item and the non-collisionable value or non-collisionable format notation, (c) A step of generating a template or prompt that includes the conflict-free values ​​or conflict-free format notation, having a language generation model generate text using the template or prompt as input, and obtaining a generated text that includes the conflict-free values ​​or conflict-free format notation, (d) A step of replacing the conflict-free values ​​or conflict-free format notations in the generated text with the actual values ​​or actual notations of the structured data based on the information that maintains the correspondence relationship, (e) A step of outputting the medical text after the substitution, Includes, A method for generating medical documents, wherein the aforementioned medical document is generated as a medical document that includes at least one of the following: a treatment summary, a referral letter, an inpatient summary, a discharge summary, a report to a pharmaceutical company, or a medical record.

2. A method for generating medical documents according to claim 1, A method for generating medical documents, wherein the collision-resistant value is a value that does not cause a collision in character matching with other values ​​in the structured data, and the collision-resistant format notation is a notation that does not cause a collision in character matching with other notations in the structured data.

3. In the medical document generation method described in claim 1, A medical document generation method that includes the step of outputting the document in a manner that allows for the identification of the portion that has been changed to an actual value or actual notation by the substitution with the other portion.

4. The aforementioned collision-resistant values ​​or collision-resistant format notations are (i) Minute numerical values ​​that do not conflict with other numerical values ​​in the structured data, (ii) The name of the input item enclosed in a separator, or (iii) a number having a number of decimal places equal to or greater than a specified number of digits, A method for generating a medical document according to claim 1, comprising at least one of the following.

5. The aforementioned information processing device, The processor is an information processing device that performs the medical document generation method according to any one of claims 1 to 4.

6. A program that causes a computer to execute the medical document generation method described in any one of claims 1 to 4.

7. A computer-readable recording medium on which the program described in claim 6 is recorded.

Images