Appraiser for creating opinion papers, method for creating opinion papers, and program for creating opinion papers

Abstract

To provide a medical opinion preparation device that significantly reduces the burden on attending physicians when preparing medical opinion reports. [Solution] The opinion form creation device 40 of the present invention includes an acquisition unit that acquires information regarding the state of the subject 60 obtained by a sensor 10, and an input information generation unit that generates input information for the input items of the attending physician's opinion form for the subject 60 based on the state information acquired by the acquisition unit.

Description

Technical Field

[0001] The present invention relates to an opinion letter creation device, an opinion letter creation method, and an opinion letter creation program.

Background Art

[0002] With the increase in the elderly population, the number of certified care recipients is also increasing. In the certification of care needs, the attending physician diagnoses the subject and creates an attending physician's opinion letter for the subject.

[0003] In relation to this, Patent Document 1 below discloses a technique for analyzing the electronic medical record of a subject and creating an attending physician's opinion letter. The technique of Patent Document 1 extracts the diagnosis and disease name of the subject and the treatment content from the electronic medical record, and automatically generates an attending physician's opinion letter based on this information. According to the technique of Patent Document 1, the burden on the attending physician who creates the attending physician's opinion letter can be reduced.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, the above technique only inputs information for some input items of the attending physician's opinion letter. Since it is necessary to input information that cannot be obtained from the electronic medical record, such as opinions regarding the current state of the subject, in the attending physician's opinion letter, the above technique has a problem that it cannot sufficiently reduce the burden on the attending physician who creates the attending physician's opinion letter.

[0006] The present invention has been made in view of the above problems. Therefore, an object of the present invention is to provide an opinion letter creation device, an opinion letter creation method, and an opinion letter creation program that can sufficiently reduce the burden on the attending physician who creates the attending physician's opinion letter. [Means for solving the problem]

[0007] The above objectives of the present invention are achieved by the following means.

[0008] (1) A medical opinion preparation device comprising: an acquisition unit that acquires information about the condition of a subject obtained by a sensor; and an input information generation unit that generates input information for input items of a medical opinion form for the subject based on the information about the condition acquired by the acquisition unit.

[0009] (2) The opinion statement creation apparatus according to (1) above, further comprising a calculation unit that calculates the value of each of a plurality of indicators relating to the status of the subject based on the status information acquired by the acquisition unit, and the input information generation unit generates the input information based on the values ​​of the plurality of indicators relating to the subject using a learning model that has learned the relationship between the values ​​of the plurality of indicators and the input information.

[0010] (3) The opinion paper creation device according to (2) above, wherein the acquisition unit further acquires information from the electronic medical record of the subject, and the input information generation unit generates the input information based on the values ​​of the multiple indicators and the electronic medical record information for the subject, using a learning model that has learned the relationship between the values ​​of the multiple indicators and the electronic medical record information and the input information.

[0011] (4) The opinion paper creation apparatus according to (3) above, wherein the information relating to the subject's condition includes imaging data obtained by imaging the subject with an image sensor, the opinion paper creation apparatus further includes a captioning unit that adds captions to images based on the imaging data, and the input information generation unit generates the input information based on the values ​​of the multiple indicators, the electronic medical record information, and the caption information for the subject, using a learning model that has learned the relationship between the values ​​of the multiple indicators, the electronic medical record information, and the caption information and the input information.

[0012] (5) The opinion paper creation apparatus according to (2) above, wherein the information relating to the subject's condition includes imaging data obtained by imaging the subject with an image sensor, the opinion paper creation apparatus further includes a captioning unit that adds captions to images based on the imaging data, and the input information generation unit generates the input information based on the values ​​of the multiple indicators and the caption information for the subject using a learning model that has learned the relationship between the values ​​of the multiple indicators and the caption information and the input information.

[0013] (6) The opinion form creation device according to any one of (1) to (5) above, further comprising a data generation unit that generates output data for outputting the input information generated by the input information generation unit in the format of a physician's opinion form in which the input information is entered into the corresponding input item.

[0014] (7) The opinion form creation device according to (6) above, further comprising a notification unit for notifying of any unentered input items in the attending physician's opinion form.

[0015] (8) A method for creating a medical opinion form, comprising the steps of (a) acquiring information about the condition of a subject obtained by a sensor, and (b) generating input information for input fields of a medical opinion form for the subject based on the information about the condition acquired in step (a).

[0016] (9) A program for creating a medical opinion form that causes a computer to perform the following steps: (a) a step to acquire information about the condition of a subject obtained by a sensor, and (b) a step to generate input information for input fields of a medical opinion form for the subject based on the information about the condition acquired in step (a). [Effects of the Invention]

[0017] According to the present invention, the burden on the attending physician who prepares the attending physician's opinion can be significantly reduced. [Brief explanation of the drawing]

[0018] The advantages and features provided by one or more embodiments of the present invention will be more fully understood from the following detailed description and the accompanying drawings, which are for illustrative purposes only and are not intended to define the limitations of the present invention. [Figure 1] It is a diagram showing a schematic configuration of an opinion letter creation system. [Figure 2] It is a block diagram showing a schematic configuration of a detection device. [Figure 3] It is a block diagram showing a schematic configuration of a terminal device.​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​As shown in Figure 1, the opinion document creation system 1 comprises a detection device 10, a terminal device 20, a first server device 30, and a second server device 40. The detection device 10, terminal device 20, first server device 30, and second server device 40 are configured to communicate with each other via a network 50.

[0022] The detection device 10 is installed in the living spaces of the subjects 60, such as their rooms in their homes or in their rooms within care facilities. The terminal device 20 is used, for example, by the subjects 60's attending physician or by local government officials who hold care certification review meetings. The first server device 30 is an on-premise server located on the premises of the hospital where the attending physician works, or a cloud server using a commercial cloud service. The second server device 40 is an on-premise server located on the premises of the care facility, or a cloud server using a commercial cloud service. The network 50 consists of the internet and an intranet.

[0023] <Detection device 10> Figure 2 is a block diagram showing the schematic configuration of the detection device 10. The detection device 10 is installed as a sensor box on the ceiling or upper part of the wall of the room where the subject 60 lives.

[0024] As shown in Figure 2, the detection device 10 comprises a control unit 11, a communication unit 12, a camera (image sensor) 13, a Doppler sensor 14, and a microphone (voice sensor) 15, which are interconnected by a bus.

[0025] The control unit 11 is composed of a CPU (Central Processing Unit) and memory such as RAM (Random Access Memory) and ROM (Read Only Memory), and controls each of the above parts and performs various calculation processes according to the program.

[0026] The communication unit 12 is an interface for communicating with other devices, and various wired or wireless communication interfaces are used.

[0027] Camera 13 captures images of the subject 60 from the ceiling or upper part of the wall of the living room and outputs the image data of the subject 60.

[0028] The Doppler sensor 14 transmits and receives microwaves to the subject 60 to detect the subject 60's body movements (for example, breathing) and outputs the subject 60's body movement data.

[0029] The microphone 15 captures the sound from inside the room of the subject 60 and outputs the voice data of the subject 60.

[0030] <Terminal device 20> Figure 3 is a block diagram showing the schematic configuration of the terminal device 20. The terminal device 20 is, for example, a PC (Personal Computer).

[0031] As shown in Figure 3, the terminal device 20 comprises a control unit 21, a storage unit 22, a communication unit 23, a display unit 24, and an input unit 25, which are interconnected by a bus. Note that, to avoid repetition in the explanation, the parts of the terminal device 20 that have the same functions as those of the detection device 10 will not be described.

[0032] The storage unit 22 consists of an HDD (Hard Disk Drive) or SSD (Solid State Drive) and stores various programs and data.

[0033] The display unit 24 is, for example, a liquid crystal display, which displays various information.

[0034] The input unit 25 is equipped with a keyboard, numeric keypad, mouse, etc., and accepts input of various instructions and information.

[0035] <First Server Device 30> Figure 4 is a block diagram showing the schematic configuration of the first server device 30.

[0036] As shown in Figure 4, the first server device 30 comprises a control unit 31, a storage unit 32, and a communication unit 33, which are interconnected by a bus. Note that the above-mentioned parts of the first server device 30 have the same functions as the above-mentioned parts of the detection device 10 and the terminal device 20, so their descriptions are omitted.

[0037] The first server device 30 is an electronic medical record server, and the storage unit 32 of the first server device 30 stores the electronic medical records of the subjects 60.

[0038] <Second Server Device 40> Figure 5 is a block diagram showing the schematic configuration of the second server device 40. The second server device 40 corresponds to the opinion document preparation device of the present invention.

[0039] As shown in Figure 5, the second server device 40 comprises a control unit 41, a storage unit 42, and a communication unit 43, which are interconnected by a bus. Note that the above-mentioned parts of the second server device 40 have the same functions as the above-mentioned parts of the detection device 10 and the terminal device 20, so their explanation is omitted.

[0040] Figure 6 is a diagram showing the contents of the storage unit 42 of the second server device 40. As shown in Figure 6, numerical data 100 is stored in the storage unit 42 of the second server device 40. The numerical data 100 is, for example, data showing the status of subject 60 for the past month, and includes values ​​(numerical values) of multiple indicators related to the status of subject 60.

[0041] Furthermore, the storage unit 42 of the second server device 40 stores programs corresponding to the acquisition unit 110, calculation unit 120, input information generation unit 130, data generation unit 140, and notification unit 150. The acquisition unit 110 acquires information regarding the status of the subject 60. The calculation unit 120 calculates the values ​​of several indicators related to the status of the subject 60 based on the information regarding the status of the subject 60. The input information generation unit 130 generates input information for the input fields of the attending physician's opinion form for the subject 60 based on the values ​​of several indicators for the subject 60 and the information in the subject 60's electronic medical record. The input information generation unit 130 includes a learning model that has learned the relationship between the values ​​of several indicators and the information in the electronic medical record for each subject and the input information for the attending physician's opinion form for the subject, and generates input information using this learning model. The learning model is composed of a neural network or the like. The data generation unit 140 generates output data for outputting the input information generated by the input information generation unit 130 in the format of a physician's opinion form, with the input information entered into the corresponding input items. The notification unit 150 notifies the system of any unentered input items in the physician's opinion form. The functions of each of the above units are performed by the control unit 41 executing the corresponding programs.

[0042] Furthermore, the detection device 10, terminal device 20, first server device 30, and second server device 40 may have components other than those described above, and may not have some of the components described above.

[0043] In the opinion form creation system 1 configured as described above, when a subject 60 applies for long-term care certification, the detection device 10 collects information about the subject 60's condition in order to create a physician's opinion form for the subject 60. Then, for example, based on information about the subject 60's condition over the past month and information from the subject 60's electronic medical record, a physician's opinion form for the subject 60 is created. The operation of the opinion form creation system 1 in creating the physician's opinion form will be explained below with reference to Figures 7 to 9.

[0044] First, referring to Figure 7, the operation of the second server device 40, which quantifies the state of the subject 60 based on information regarding the subject 60's state, will be explained. In this embodiment, the state of the subject 60 is quantified separately for physical function and cognitive function.

[0045] Figure 7 is a flowchart showing the procedure for the digitization process performed by the second server device 40. The process shown in the flowchart in Figure 7 is executed by the control unit 41 according to the program stored in the storage unit 42 of the second server device 40.

[0046] (Step S101) First, the second server device 40 acquires information regarding the status of the subject 60. More specifically, the second server device 40 acquires a day's worth of imaging data, motion data, and audio data of the subject 60 collected by the detection device 10.

[0047] (Step S102) Next, the second server device 40 quantifies the physical functions of the subject 60. More specifically, based on the data acquired in step S101, the second server device 40 calculates numerical values ​​for each of several indicators related to the physical functions of the subject 60.

[0048] In this embodiment, the second server device 40 estimates the joint points (joint positions) of the subject 60 from the imaging data of the subject 60 using known machine learning techniques such as OpenPose. The second server device 40 then calculates the subject 60's movement speed (walking speed) as an indicator of the subject 60's physical function, for example, from the movement trajectory of the subject 60's head position. The second server device 40 then quantifies the subject 60's movement speed on a scale of 0 to 5, with higher values ​​indicating faster movement speed. It should be noted that the higher the subject 60's physical function, the faster their movement speed tends to be.

[0049] Furthermore, the second server device 40 calculates the subject 60's daily range of movement (activity range) from the movement trajectory of the subject 60's head position, as another indicator of the subject 60's physical function. The second server device 40 then quantifies the subject 60's range of movement on a scale of 0 to 5, with higher values ​​indicating a wider range of movement. It should be noted that the higher the subject 60's physical function, the wider their range of movement tends to be.

[0050] Furthermore, the second server device 40 calculates the rotational speed of the subject 60 from the movement trajectories of the subject 60's head and shoulder positions, as another indicator of the subject 60's physical function. The second server device 40 then quantifies the rotational speed of the subject 60 on a scale of 0 to 5, with higher values ​​indicating faster rotational speed. It should be noted that the higher the subject 60's physical function, the faster their rotational speed tends to be.

[0051] (Step S103) Next, the second server device 40 quantifies the cognitive function of the subject 60. More specifically, based on the data obtained in step S101, the second server device 40 calculates a numerical value for each of several indicators related to the cognitive function of the subject 60.

[0052] In this embodiment, the second server device 40 calculates the amount of time the subject 60 is continuously asleep from the subject 60's body movement data. The second server device 40 then calculates, for example, the subject 60's continuous nighttime sleep duration as an indicator of the subject 60's cognitive function. The second server device 40 then quantifies the subject 60's continuous nighttime sleep duration on a scale of 0 to 5, with higher values ​​indicating longer continuous nighttime sleep duration. It should be noted that the higher the subject 60's cognitive function, the longer the subject 60's continuous nighttime sleep duration tends to be.

[0053] Furthermore, the second server device 40 calculates the speech rate of the subject 60 from the subject 60's voice data as another indicator of the subject 60's cognitive function. The second server device 40 then quantifies the subject 60's speech rate on a scale of 0 to 5, with higher values ​​indicating faster speech rate. It should be noted that the higher the subject 60's cognitive function, the faster their speech rate tends to be.

[0054] Furthermore, the second server device 40 transcribes the speech content of the subject 60 from the subject 60's audio data. The second server device 40 then counts the number of specific words (such as negative words) included in the subject 60's speech content as another indicator of the subject 60's cognitive function. The second server device 40 then quantifies the subject 60's speech content on a scale of 0 to 5, with a higher value indicating a smaller number of specific words. It should be noted that the higher the subject 60's cognitive function, the fewer specific words tend to be included in the subject 60's speech content.

[0055] (Step S104) The second server device 40 then stores the numerical values ​​of each indicator related to the subject 60's condition and terminates processing. More specifically, the second server device 40 stores the numerical values ​​of each of the multiple indicators related to physical function calculated in step S102 and the numerical values ​​of each of the multiple indicators related to cognitive function calculated in step S103 in the storage unit 42 and terminates processing.

[0056] As described above, according to the flowchart shown in Figure 7, the numerical values ​​of multiple indicators related to the subject's condition 60 are calculated from the daily data collected by the detection device 10 and stored in the memory unit 42. In this embodiment, multiple indicators related to the subject's physical or cognitive function are each quantified within the range of 0 to 5 and stored in the memory unit 42.

[0057] In addition, in the opinion form creation system 1 of this embodiment, the second server device 40 calculates the numerical values ​​of multiple indicators related to the condition of the subject 60 every day for 30 days prior to the creation of the attending physician's opinion form. Once 30 days' worth of data has been stored in the storage unit 42, the 30 days' worth of values ​​for each indicator are averaged, and the numerical values ​​(30-day averages) for each indicator related to the condition of the subject 60 for the most recent month are calculated. The numerical values ​​for each indicator related to the condition of the subject 60 for the most recent month are stored in the storage unit 42 as numerical data 100.

[0058] Furthermore, the indicators for the subject 60's condition are not limited to those listed above, and various indicators may be used. For example, the degree of head sway during walking of subject 60, calculated from subject 60's imaging data, may be used as an indicator of physical function. Alternatively, the walking area of ​​subject 60 (near the wall / center of the room), calculated from subject 60's imaging data, may be used as an indicator of physical function. Or, as an indicator of subject 60's cognitive function, the time subject 60 spends standing still, calculated from subject 60's imaging data, may be used, or the time period subject 60 spends walking may be used. Alternatively, as an indicator of subject 60's cognitive function, the intonation of subject 60's speech, the frequency of subject 60's self-talk, or sudden sounds occurring in subject 60's room, calculated from subject 60's voice data, may be used. In addition, the numerical range for each indicator is not limited to 0 to 5; for example, different numerical ranges may be set for each indicator.

[0059] The operation of the second server device 40, which prepares the attending physician's opinion for the subject 60, will be explained below with reference to Figures 8 and 9.

[0060] Figure 8 is a flowchart showing the procedure for creating an opinion statement, which is performed by the second server device 40. The process shown in the flowchart in Figure 8 is executed by the control unit 41 according to the program stored in the storage unit 42 of the second server device 40.

[0061] (Step S201) First, the second server device 40 reads numerical data for the subject 60. More specifically, the second server device 40 reads the values ​​(30-day averages) for the most recent month for each of several indicators related to the subject 60's condition from the storage unit 32.

[0062] (Step S202) Next, the second server device 40 acquires the electronic medical record information of the subject 60. More specifically, the second server device 40 accesses the first server device 30 and acquires the electronic medical record information of the subject 60 from the storage unit 32 of the first server device 30.

[0063] (Step S203) Next, the second server device 40 inputs numerical data and electronic medical record information into the learning model. More specifically, the second server device 40 inputs the numerical values ​​of each indicator read in step S201 and the electronic medical record information obtained in step S202 into the learning model, which has learned the relationship between the numerical values ​​of each indicator, the electronic medical record information, and the input information of the attending physician's opinion form. The learning model, having received the numerical values ​​of each indicator and the electronic medical record information for the subject 60, outputs the input information for the input items of the attending physician's opinion form for the subject 60.

[0064] (Step S204) Next, the second server device 40 creates a physician's opinion form. More specifically, the second server device 40 generates output data for outputting a physician's opinion form in which the input information output from the learning model in step S203 is entered into the corresponding input items. The output data for the physician's opinion form is configured to allow editing of the input information.

[0065] Furthermore, for formal input fields in the attending physician's opinion form, such as the applicant's name, medical institution name, and attending physician's name, the input information is extracted from the electronic medical record using a predetermined conversion table, without using a learning model, and the extracted input information is entered into the corresponding input fields.

[0066] (Step S205) Next, the second server device 40 determines whether or not there are any unentered input items in the input fields of the attending physician's opinion form. More specifically, the second server device 40 determines whether or not there are any input fields in the multiple input fields of the attending physician's opinion form from which output data was generated in step S204 that do not contain any input information. For example, if the detection device 10 has not obtained information regarding the subject 60's eating habits, the input field "eating behavior" (see Figure 9) related to dietary habits in the attending physician's opinion form will be left unentered.

[0067] If it is determined that there are no unentered input items (step S205: NO), the second server device 40 proceeds to the process in step S207. On the other hand, if it is determined that there are unentered input items (step S205: YES), the second server device 40 proceeds to the process in step S206.

[0068] (Step S206) If it is determined that there are any unentered input items (step S205: YES), the second server device 40 performs a process to highlight the unentered input items and proceeds to step S207. More specifically, the second server device 40 performs a process to change the background color of the unentered input items in the attending physician's opinion form to a different color from the background color of the other input items and proceeds to step S207.

[0069] (Step S207) Then, the second server device 40 outputs the attending physician's opinion form and terminates processing. More specifically, the second server device 40 sends the output data for the attending physician's opinion form to the terminal device 20 and terminates processing. As a result, the attending physician's opinion form, with information entered in multiple input fields, is displayed on the display unit 24 of the terminal device 20.

[0070] As described above, according to the flowchart shown in Figure 8, a physician's opinion report for subject 60 is created based on the numerical values ​​of multiple indicators related to subject 60's condition and the information in subject 60's electronic medical record, and is displayed on the display unit 24 of the terminal device 20. Users of the terminal device 20 (subject 60's physician or local government officials) can check the contents of the physician's opinion report displayed on the display unit 24 and edit it as necessary.

[0071] Figure 9 shows an example of a physician's opinion form 200 displayed on the display unit 24 of the terminal device 20. As shown in Figure 9, the physician's opinion form 200 includes formal input items such as the applicant's name and the name of the medical institution, and non-formal input items related to the applicant's level of independence in daily life, living condition, necessary care services, etc. In the opinion form creation system 1 of this embodiment, the input information for formal input items such as the applicant's name and the name of the medical institution is generated based on the information in the electronic medical record.

[0072] On the other hand, information entered into non-formal input fields regarding the applicant's level of independence in daily life, living conditions, and necessary care services is mainly generated based on information obtained from cameras, Doppler sensors, and microphones. For example, for the input field "Short-term memory" on the right side of the attending physician's opinion form 200 shown in Figure 9, the input information "○○" is generated. In this embodiment of the opinion form creation system 1, since the learning model is also trained on information from the electronic medical record, information from the electronic medical record can also be used for information entered into non-formal input fields.

[0073] With this configuration, input information for the attending physician's opinion form is generated based on information obtained from sensors such as cameras and Doppler sensors, thus reducing the burden on the attending physician who prepares the form. In particular, the burden on the attending physician, who is not a caregiving expert, is reduced when it comes to inputting information such as what kind of assistance the 60-year-old subject needs from a caregiving perspective and how to realize the life the 60-year-old subject desires.

[0074] Furthermore, according to the opinion form creation system 1 of this embodiment, the attending physician's opinion form is created based on objective information obtained from sensors such as cameras and Doppler sensors, making it possible to conduct a fair and highly accurate assessment in the long-term care certification assessment.

[0075] Furthermore, according to the opinion form creation system 1 of this embodiment, unentered input items in the attending physician's opinion form are displayed with a different background color for emphasis, making it easier for users of the terminal device 20 to notice unentered input items and preventing omissions in entering information.

[0076] (Variation 1) In the embodiment described above, input information for the attending physician's opinion for subject 60 was generated based on the numerical values ​​of multiple indicators for subject 60 and the information in the electronic medical record. However, unlike the embodiment described above, input information for the attending physician's opinion for subject 60 may be generated based solely on the numerical values ​​of multiple indicators for subject 60.

[0077] In this case, the learning model is not trained on electronic medical record information, but rather on the relationship between the numerical values ​​of multiple indicators and the input information of the attending physician's opinion form. Based on the numerical values ​​of multiple indicators for each of the 60 subjects, the input information for the attending physician's opinion form for each of the 60 subjects is generated.

[0078] (Second Embodiment) Next, a second embodiment of the present invention will be described with reference to Figure 10. This embodiment is one in which a caption (explanatory text) is added to an image based on imaging data, and a physician's opinion is created by further utilizing the information in the caption. Except for the fact that a caption is added to an image based on imaging data and the information in the caption is utilized, the configuration of the opinion creation system 1 according to this embodiment is the same as the configuration of the opinion creation system 1 according to the first embodiment, so a detailed explanation will be omitted.

[0079] Figure 10 is a diagram showing the contents of the storage unit 42 of the second server device 40 according to this embodiment. As shown in Figure 10, the storage unit 42 of the second server device 40 stores programs corresponding to the acquisition unit 110, calculation unit 120, input information generation unit 130, data generation unit 140, notification unit 150, and assignment unit 160. The acquisition unit 110, calculation unit 120, data generation unit 140, and notification unit 150 are the same as in the first embodiment, so their description is omitted. The functions of each of the above units are performed by the control unit 41 executing the corresponding program.

[0080] The captioning unit 160 adds captions to images based on the subject's imaging data 60. The captioning unit 160 includes a learning model that has learned the relationship between specific situations in images (moving images) based on the subject's imaging data for multiple subjects and the content of the captions added to the images, and adds captions to images based on the subject's imaging data 60. The captioning unit 160 analyzes the subject's imaging data 60 and adds captions such as "A staff member is putting shoes on the subject" or "The subject is walking to the toilet by themselves" to images of corresponding scenes. Note that the technique of adding captions to images using a learning model such as a neural network is a well-known technique, so a detailed explanation is omitted.

[0081] The input information generation unit 130 generates input information for the input fields of the attending physician's opinion form for the subject 60 based on the numerical values ​​of multiple indicators related to the subject 60's condition, the information from the subject 60's electronic medical record, and the information from the caption. The input information generation unit 130 includes a learning model that has learned the relationship between the numerical values ​​of multiple indicators, the information from the electronic medical record, and the information from the caption and the input information for the attending physician's opinion form for multiple subjects, and generates input information using this learning model.

[0082] According to the opinion form creation system 1 of this embodiment, configured as described above, captions are added to images based on imaging data of the subject 60. Then, input information for the input fields of the attending physician's opinion form is generated from the numerical values ​​of multiple indicators related to the subject 60's condition, the information from the subject 60's electronic medical record, and the caption information added to the image of the subject 60.

[0083] This configuration makes it possible to create more accurate physician's opinions for each of the 60 subjects.

[0084] (Modification 2) In the embodiment described above, the input information for the attending physician's opinion for subject 60 was generated based on the numerical values ​​of multiple indicators for subject 60, the information from the electronic medical record, and the caption information. However, unlike the embodiment described above, the input information for the attending physician's opinion for subject 60 may be generated based on the numerical values ​​of multiple indicators for subject 60 and the information from the caption. Alternatively, the input information for the attending physician's opinion for subject 60 may be generated based solely on the information from the caption for subject 60.

[0085] In this case, the learning model is not trained on the electronic medical record information, but rather on the relationship between the numerical values ​​and caption information of multiple indicators and the input information of the attending physician's opinion form. This learning model generates the input information for the attending physician's opinion form for the 60 subjects. Alternatively, the learning model is trained on the relationship between the caption information and the input information of the attending physician's opinion form.

[0086] The present invention is not limited to the embodiments described above, and can be modified in various ways within the scope of the claims.

[0087] For example, in the embodiment described above, if the attending physician's opinion form contains input items that have not been entered, the user of the terminal device 20 is notified of the unentered input items by changing the background color of the input items to highlight them. However, the method for notifying the user of unentered input items is not limited to changing the background color of the input items. For example, the user may be notified of unentered input items by changing the text color of the input items. Alternatively, the user may be notified of unentered input items by displaying a comment such as "Manual input is required" on the display unit 24 of the terminal device 20.

[0088] Furthermore, in the embodiments described above, the acquisition of imaging data, body movement data, and audio data of the subject 60 was explained as an example of information regarding the subject 60's state. However, information regarding the subject's state is not limited to these data. For example, vital data such as the subject's heart rate and body temperature may be acquired as information regarding the subject's state, or data related to excretion and sleep may be acquired. In addition, data regarding the subject's movement (angular velocity, acceleration) may be acquired by a wearable sensor attached to the subject's body or clothing. For example, instead of a Doppler sensor, body movement data of the subject 60 may be acquired by a wearable sensor. Also, unlike the embodiments described above, audio data of the subject 60 may not be acquired, and only imaging data and body movement data of the subject 60 may be acquired. Furthermore, the values ​​of the indicators regarding the subject 60's state are not limited to numerical values, and any values ​​such as symbols or letters may be used.

[0089] Furthermore, in the embodiment described above, information regarding the condition of the subject 60 over the past month was collected, and a physician's opinion was prepared. However, the period for collecting information to prepare the physician's opinion is not limited to one month (30 days), but may be, for example, one week (7 days).

[0090] The processing units in the flowcharts of the embodiments described above are divided according to the main processing content in order to facilitate understanding of each process. The present invention is not limited by how the processing steps are classified. Each process can be further divided into more processing steps. Also, one processing step may perform even more processes.

[0091] In the embodiments described above, the functions of each device may be implemented by other devices. For example, each function of the second server device 40 may be implemented by the detection device 10, or by other server devices. For example, the detection device 10 may calculate the values ​​of each indicator from the data of the subjects 60.

[0092] The means and methods for performing various processing tasks in the opinion document creation apparatus according to the above embodiment can be implemented by either a dedicated hardware circuit or a programmed computer. The program may be provided, for example, on a computer-readable recording medium such as a USB (Universal Serial Bus) memory or a DVD (Digital Versatile Disc)-ROM, or it may be provided online via a network such as the Internet. In this case, the program recorded on the computer-readable recording medium is usually transferred to and stored in a storage unit such as an HDD. Furthermore, the program may be provided as a standalone application software, or it may be incorporated into the software of the opinion document creation apparatus as a function of the apparatus.

[0093] While embodiments of the present invention have been described and illustrated in detail, the disclosed embodiments are for illustrative purposes only and are not limiting. The scope of the present invention should be interpreted in accordance with the language of the appended claims. [Explanation of symbols]

[0094] 1. Opinion document creation system, 10 detection devices, 11,21,31,41 Control Unit, 12, 23, 33, 43 Communications Department, 13 cameras, 14 Doppler sensors, 15 microphones, 20 terminal devices, 22,32,42 storage section, 24 Display section, 25 Input section, 30. First server device, 40 Second server device, 50 networks, 60 target individuals.

Claims

1. An acquisition unit that acquires information about the subject's condition obtained by a sensor, An input information generation unit generates input information for the input fields of the attending physician's opinion form for the subject, based on the information regarding the state acquired by the acquisition unit. An opinion paper creation device having the following features.

2. The system further includes a calculation unit that calculates the value of each of several indicators relating to the subject's condition based on the information about the condition acquired by the acquisition unit, The opinion statement creation device according to claim 1, wherein the input information generation unit generates the input information based on the values ​​of the multiple indicators for the subject, using a learning model that has learned the relationship between the values ​​of the multiple indicators and the input information.

3. The acquisition unit further acquires the electronic medical record information of the subject, The opinion statement creation device according to claim 2, wherein the input information generation unit generates the input information based on the values ​​of the multiple indicators and the electronic medical record information for the subject, using a learning model that has learned the relationship between the values ​​of the multiple indicators and the electronic medical record information and the input information.

4. The information relating to the subject's state includes imaging data obtained by imaging the subject with an image sensor. The aforementioned opinion paper creation device is The system further includes a captioning unit that adds captions to images based on the aforementioned imaging data. The opinion statement creation device according to claim 3, wherein the input information generation unit generates the input information based on the values ​​of the multiple indicators, the electronic medical record information, and the caption information for the subject, using a learning model that has learned the relationship between the values ​​of the multiple indicators, the electronic medical record information, and the caption information and the input information.

5. The information relating to the subject's state includes imaging data obtained by imaging the subject with an image sensor. The aforementioned opinion paper creation device is The system further includes a captioning unit that adds captions to images based on the aforementioned imaging data. The opinion paper creation device according to claim 2, wherein the input information generation unit generates the input information based on the values ​​of the multiple indicators and the caption information for the subject, using a learning model that has learned the relationship between the values ​​of the multiple indicators and the caption information and the input information.

6. The opinion form creation device according to any one of claims 1 to 5, further comprising a data generation unit that generates output data for outputting the input information generated by the input information generation unit in the format of a physician's opinion form in which the input information is entered into the corresponding input item.

7. The opinion form creation device according to claim 6, further comprising a notification unit for notifying of unentered input items in the attending physician's opinion form.

8. (a) A step of acquiring information about the subject's condition obtained by a sensor, Step (b) generates input information for the input fields of the attending physician's opinion form for the subject, based on the information regarding the state obtained in step (a), A method for preparing an opinion paper that includes the following.

9. Procedure (a) for obtaining information about the subject's condition obtained by a sensor, A procedure (b) to generate input information for the input fields of the attending physician's opinion form for the subject, based on the information regarding the state obtained in the procedure (a) above, A program that uses a computer to generate opinion papers.

Images