Medical support device, its operation method and operation program, and medical support system

Abstract

To provide a medical care support device capable of improving prediction accuracy by collecting the operation history of many users while avoiding the risk of leaking patient information, an operating method thereof, an operating program, and a medical care support system.SOLUTION: A medical care support device 11 includes an operation history acquisition unit 63, and a prediction execution unit 64. The operation history acquisition unit 63 acquires an operation record including a piece of operation information of the operations made on a terminal unit and a piece of user information of a user who has operated. The prediction execution unit 64 learns based on the operation record including the user information and outputs the next operation candidate for the terminal unit using the generated learned model.SELECTED DRAWING: Figure 6

Description

Technical Field

[0001] The present invention relates to a medical support device, an operating method and an operating program thereof, and a medical support system.

Background Art

[0002] In the medical field, in order for medical staff such as doctors and medical technicians to smoothly proceed with examinations and tests, etc., an integrated medical support device and a medical support system that share the medical process and medical results among medical staff or between medical departments, etc. are used. The medical support device supports medical treatment, for example, by providing a list display of the medical process and medical results for a plurality of patients to medical staff (Patent Document 1).

[0003] On the other hand, in the medical field as well, efforts are being made to improve the efficiency of the work of medical staff by using machine learning. For example, in the information processing device described in Patent Document 2, the operation history made during past examinations is analyzed and learned on an operation screen such as an electronic medical record. Then, for the medication and disease names input and operated by medical staff, the next operation is predicted based on the learning results. The predicted operation is proposed as the next operation candidate.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] In the medical support device and medical support system described in Patent Document 1 above, medical information and medical treatment results are displayed, and in the information processing device described in Patent Document 2 above, things containing patients' personal information are handled, such as performing analysis and learning from the operation screen of the electronic medical record. For this reason, in order to avoid the risk of leakage of patients' personal information and the like, in each hospital facility, there are many cases where it is operated only on the internal network of the facility.

[0006] In addition, when causing the information processing device described in Patent Document 2 above to perform analysis and learning of operation history for the medical support device and medical support system described in Patent Document 1 above, there are the following problems. For example, in machine learning for a device that recognizes a lesion or the like in a medical image, a large number of medical images can be accumulated and learned in advance. However, when learning the operation history in a medical support device and a medical support system, it is necessary to accumulate and learn the operation history when the same device, the same system, and at least medical staff of the same occupation as those of the medical support device and medical support system used within a predetermined hospital facility operate. That is, in the case of a learning result generated based on an operation history that is different in any one of the device, the system, and the occupation, it is difficult to obtain a high learning effect.

[0007] In addition, in machine learning, the prediction accuracy can be improved by accumulating more samples and appropriately continuing learning. However, considering the leakage of patients' personal information and the like, it is not possible to collect operation histories regarding a wide variety and a large number of users only by learning with a medical support device and a medical support system within one hospital facility, and the prediction accuracy cannot be improved.

[0008] Therefore, an object of the present invention is to provide a medical support device, an operating method and an operating program thereof, and a medical support system that can collect operation histories of many users and improve prediction accuracy while avoiding the risk of leakage of patient information.

Means for Solving the Problems

[0009] The medical support device of the present invention includes a processor, which acquires an operation history including operation information of an operation performed on a terminal device and user information of a user who performed the operation, and uses a learned model generated by learning based on the operation history including the user information to output the next operation candidate for the terminal device.

[0010] The user information includes at least one user attribute such as the user's occupation, the user's medical department, or the disease name of the patient in charge of the user. The processor preferably outputs the next operation candidate for the terminal device using a learned learned model generated by learning based on a plurality of operation histories having the same user attribute.

[0011] The user information includes a user ID that identifies an individual who is the user. The processor preferably outputs the next operation candidate for the terminal device using a learned learned model generated by learning based on a plurality of operation histories having the same user ID according to the amount of the plurality of operation histories having the same user ID.

[0012] The processor preferably outputs the next operation candidate for each user information using the learned model.

[0013] The operation information preferably includes at least one of the name of the operation target operated by the user, the type of the operation target, the frequency of operating the operation target, or the order of operating the operation target.

[0014] The processor preferably outputs the operation target to be operated next as the next operation candidate.

[0015] The processor preferably outputs the next operation candidate according to whether the next operation candidate has been operated or not.

[0016] The operation history preferably includes at least one of the name of the document created by the user, the type of the document created, the frequency of creating the document, or the order of creating the document.

[0017] Preferably, as the next operation candidate, the processor outputs a document to be created next.

[0018] Preferably, as the next operation candidate, the processor proposes to display a screen for editing a document to be created next.

[0019] Preferably, the processor outputs the next document according to whether the next document has been created.

[0020] The operation history includes at least the creation time of the document created by the user. Preferably, the processor uses a learned model generated by learning based on the operation history including user information to output the next document for the terminal device and the creation time of the next document.

[0021] The operation history includes medical items and a medical schedule. Preferably, the processor uses the learned model to output a document and a creation time according to the medical schedule.

[0022] The operation history includes the type of examination or treatment ordered by the user and the order time when the user ordered the examination or treatment. Preferably, the processor uses the learned model to output at least one of the types of the next examination or treatment and at least one of the order times of the next examination or treatment.

[0023] Preferably, it includes a medical support device, a terminal device, and an external server having a learned model. Based on the operation history obtained from the medical support device, the external server uses the learned model to derive the next operation candidate and outputs it to the derived medical support device.

[0024] The operating method of the medical support device of the present invention includes the steps of obtaining an operation history including operation information of an operation performed on a terminal device and user information of a user who performed the operation, and outputting the next operation candidate for the terminal device using a learned model generated by learning based on the operation history including the user information.

[0025] The operating program of the medical support device of the present invention causes a computer to realize a function of obtaining an operation history including operation information of an operation performed on a terminal device and user information of a user who performed the operation, and a function of outputting the next operation candidate for the terminal device using a learned model generated by learning based on the operation history including the user information.

Advantages of the Invention

[0026] According to the present invention, it is possible to collect the operation histories of many users while avoiding the risk of leakage of patient information and improve the prediction accuracy.

Brief Description of the Drawings

[0027]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Mode for Carrying Out the Invention

[0028] [First Embodiment] As shown in FIG. 1, the medical support system 10 is a computer system that provides medical support in a medical facility such as a hospital, and includes medical support devices 11 installed in a plurality of medical facilities A, B, ···, X, client terminals 12 installed in the same medical facilities A, B, ···, X as the medical support devices 11, a learning device 13, a network 14, and the like. Note that the medical support system 10 includes a medical information system 17 (see FIG. 2) provided in each medical facility A, B, ···, X. Also, a plurality of medical support devices 11 may be installed in each medical facility A, B, ···, X. The learning device 13 is an external server installed on the cloud.

[0029] Network 14 is a wide area network (Wide Area Network, WAN) that widely connects the medical support devices 11 and the learning device 13 placed in a plurality of medical facilities A, B, ···, X via a public line network or a dedicated line network such as the Internet.

[0030] As shown in FIG. 2, the medical support device 11 is connected to the medical information system 17 provided in the medical facility A via the network 16 installed within one medical facility A. Although not shown, the medical information system 17 is similarly provided in other medical facilities B, ···, X, and the medical support device 11, the client terminal 12, etc. are connected to the network 16 in the same manner as in FIG. 2.

[0031] The medical information system 17 includes the medical support device 11, the client terminal 12, and the server group 18, and is configured to be able to transmit and receive data to and from each other via the network 16. The network 16 is a local area network (LAN), and it is desirable to use a communication cable such as an optical fiber so that medical image data can be transferred at high speed.

[0032] The client terminal 12 (terminal device) is a terminal for receiving services (providing the functions of the medical support device 11) from the medical support device 11, and is a computer (including the case of a tablet terminal, etc.) directly operated by medical staff such as doctors, medical technicians, or nurses. The client terminal 12 is installed in various medical departments such as internal medicine or surgery, various examination departments such as radiology or clinical laboratory, the nursing center, or other necessary locations. Also, the client terminal 12 can be provided for each medical staff, and can also be shared by a plurality of medical staff. Therefore, as shown in FIG. 2, the medical information system 17 includes a plurality of client terminals 12. For example, the group G1 is "internal medicine" to which doctors A1 and A2 belong, and doctors A1 and A2 each hold a client terminal 12. Similarly, for example, the group G2 is "surgery" to which doctor B1 belongs, and there is at least one client terminal 12 in the group G2. Also, for example, the group G19 is "radiology" to which technician N1 belongs, and there is at least one client terminal 12 in the group G19.

[0033] The medical support device 11 provides a display screen including medical data (such as images themselves) and / or information indicating the location of the medical data (such as a link to an image, etc.) to the client terminal 12 in response to a request from the client terminal 12, for example. Medical data refers to images, reports, examination results, other data obtained or created during diagnosis, examination, or surgery, etc., or data obtained as a result of medical treatment, or information indicating their locations (so-called links (aliases), etc.). The medical support device 11 obtains the medical data used on the display screen from the server group 18.

[0034] The display screen provided by the medical support device 11 to the client terminal 12 refers to the data used by the client terminal 12 to form the screen of the display unit 36 (see FIG. 3) of the client terminal 12. Further, the display screen provided by the medical support device 11 to the client terminal 12 includes not only the data for full-screen display that the client terminal 12 uses to configure the display of the entire screen, but also the data for configuring the display related to a part of the screen. For example, in the present embodiment, the medical support device 11 provides the client terminal 12 with a display screen that can be displayed in a general window format on a part of the screen of the display unit 36.

[0035] Specifically, the display screen provided by the medical support device 11 to the client terminal 12 includes an initial screen 71 (see FIG. 9), a clinical flow screen 81 (see FIG. 9), a timeline screen (not shown), a layout display screen 101 (see FIG. 10), and the like. The clinical flow screen 81 is a display screen that displays, for a plurality of patients, for each patient, the identification information of the patient and a part or all of the medical process in association with each other. The identification information of the patient is, for example, the name, date of birth, age, or gender of the patient, or an ID (Identification Data) such as a unique number and / or symbol assigned to the patient (hereinafter referred to as patient ID). The medical process refers to the process or result of medical treatment that has already been performed and the medical treatment that is planned to be performed in the future. Therefore, the medical process may include not only the medical data that has already been acquired, etc., but also the medical data that is planned to be acquired, etc. The medical data that is planned to be acquired, etc. is, for example, information regarding the presence or absence of an order for a specific examination, the scheduled date and time thereof, or the type of medical data that is planned to be acquired, etc. The timeline screen is a display screen that displays, for a specific patient, a part or all of the medical process of that patient in chronological order on one screen. The layout display screen 101 is a display screen that displays, for a specific patient, a part or all of the medical process of that patient arranged vertically and horizontally (for example, arranged in a tile shape).

[0036] The medical support device 11 provides a display screen to the client terminal 12 in a description format using a markup language such as XML (Extensible Markup Language) data. The client terminal 12 displays the XML-formatted display screen using a web browser. Note that the medical support device 11 can provide the display screen to the client terminal 12 in other formats such as JSON (JavaScript (registered trademark) Object Notation) instead of XML.

[0037] The server group 18 searches for medical data corresponding to a request from the medical support device 11 and provides the medical data corresponding to the request to the medical support device 11. The server group 18 includes an electronic medical record server 21, an image server 22, a report server 23, and the like.

[0038] The electronic medical record server 21 has a medical record database 21A for storing electronic medical records. An electronic medical record is an aggregate of one or more pieces of medical data. Specifically, an electronic medical record includes, for example, medical data such as examination records, results of specimen tests, patients' vital signs, orders for examinations, treatment records, or accounting data. The electronic medical record can be input and viewed using the client terminal 12.

[0039] Note that an examination record is a record of the content and results of an interview or palpation, or a disease name. A specimen is blood or tissue collected from a patient, and a specimen test is a blood test or a biochemical test. Vital signs are data indicating the state of a patient such as the patient's pulse, blood pressure, or body temperature. An order for an examination is a request for an examination such as a specimen test, imaging using various modalities, creation of a report, treatment or surgery, or administration of medicine. A treatment record is a record of treatment, surgery, administration of medicine, or a prescription. Accounting data is data related to consultation fees, medicine fees, or hospitalization fees.

[0040] The image server 22 is a so-called PACS (Picture Archiving and Communication System) server and has an image database 22A in which inspection images are stored. The inspection images are images obtained by various image inspections such as CT (Computed tomography) inspection, MRI (Magnetic Resonance Imaging) inspection, X-ray inspection, ultrasonic inspection, or endoscopic inspection. These inspection images are recorded in a format compliant with, for example, the DICOM (Digital Imaging and Communications in Medicine) standard. The inspection images can be viewed using the client terminal 12.

[0041] The report server 23 has a report database 23A that stores reading reports. A reading report (hereinafter simply referred to as a report) is a report summarizing the reading results of inspection images obtained by image inspections. The reading of inspection images is performed by a radiologist. The report can be created and / or viewed using the client terminal 12.

[0042] The above-mentioned electronic medical records, inspection images, and reports are each accompanied by a patient ID. In addition to the patient ID, the electronic medical record is accompanied by information for identifying the medical staff who input the medical data for each piece of medical data. The inspection image is accompanied by information for identifying the medical staff (specifically, the medical technologist) who performed the inspection in addition to the patient ID. The report is accompanied by information for identifying the medical staff (specifically, the radiologist) who created it. The information for identifying the medical staff is the name of the medical staff or the ID such as a unique number and / or symbol assigned to each medical staff (hereinafter referred to as the medical staff ID).

[0043] The medical support device 11, the client terminal 12, the learning device 13, and each of the servers 21 to 23 that make up the server group 18 are configured by installing an operating system program and an application program such as a server program or a client program on a computer such as a server computer, a personal computer, or a workstation. That is, the basic configurations of the medical support device 11, the client terminal 12, the learning device 13, and each of the servers 21 to 23 that make up the server group 18 are the same, and include a CPU (Central Processing Unit), a memory, a storage, a communication unit, etc., and a connection circuit that connects these. The communication unit is a communication interface (modem, router, or LAN board, etc.) for connecting to the network 14 or the network 16. The connection circuit is, for example, a motherboard that provides a system bus and / or a data bus, etc.

[0044] As shown in FIG. 3, in addition to a CPU 31, a memory 32, a storage 33, a communication unit 34, and a connection circuit 35, the client terminal 12 includes a display unit 36 and an operation unit 37. The display unit 36 is, for example, a display using liquid crystal or the like, and has a screen for displaying at least the display screen provided by the medical support device 11. The operation unit 37 is, for example, a pointing device such as a mouse and / or an input device such as a keyboard. The display unit 36 and the operation unit 37 can constitute a so-called touch panel.

[0045] The client terminal 12 stores, in the storage 33, in addition to an operating system program and the like, an operation program 39. The operation program 39 is an application program for receiving the provision of the functions of the medical support device 11 using the client terminal 12. In the present embodiment, the operation program 39 is a program of a web browser. However, the operation program 39 can be a dedicated application program for receiving the provision of the functions of the medical support device 11. Note that the operation program 39 may include one or a plurality of gadget engines for controlling part or all of the display screen provided by the medical support device 11. A gadget engine is a subprogram that exhibits various functions by operating in association with a web browser or the like.

[0046] When the operation program 39 is started on the client terminal 12, as shown in FIG. 4, the CPU 31 of the client terminal 12 functions as a GUI (Graphical User Interface) control unit 41 and a request issuing unit 42 in cooperation with the memory 32.

[0047] The GUI control unit 41 displays the display screen provided by the medical support device 11 on the web browser on the display unit 36. The GUI control unit 41 controls the client terminal 12 in response to an operation instruction input using the operation unit 37, such as a click operation of a button with a pointer.

[0048] The request issuing unit 42 issues various processing requests (hereinafter referred to as processing requests) to the medical support device 11 in response to the operation instructions of the operation unit 37. The processing requests issued by the request issuing unit 42 are, for example, a request for distributing a display screen or a request for editing a display screen. The request issuing unit 42 transmits the processing requests to the medical support device 11 via the communication unit 34 and the network 16.

[0049] The display screen delivery request requests the diagnostic support device 11 to deliver a display screen having a specific configuration. For example, depending on the display screen delivery request, any one of the clinical flow screen 81, the layout display screen 101, etc. can be specified and received for delivery.

[0050] The display screen editing request requests the diagnostic support device 11 to edit the content such as diagnostic data to be displayed on the display screen after receiving the delivery of the display screen having a specific configuration from the diagnostic support device 11. For example, when receiving the delivery of the clinical flow screen 81, requests such as specifying or changing the list of patients to be displayed, specifying or changing the display target period of the diagnostic process, specifying or changing the diagnostic process to be the display target, or sorting (sorting) the display content are display screen editing requests.

[0051] Note that the display screen delivery request and / or editing request, etc. include information such as the medical staff ID and the address on the network of the client terminal 12. The medical staff ID is input on a login screen (not shown) to the diagnostic support system 10 (or the diagnostic support device 11).

[0052] As shown in FIG. 5, the diagnostic support device 11 includes a CPU 51, a memory 52, a storage 53, a communication unit 54, and a connection circuit 55. The diagnostic support device 11 can be provided with a display unit and / or an operation unit as necessary in the same manner as the client terminal 12, and can also attach a display unit and / or an operation unit as necessary, but in this embodiment, the diagnostic support device 11 does not have a display unit and an operation unit.

[0053] The medical support device 11 stores, in the storage 53, in addition to an operating system and the like, an operation program 59. The operation program 59 is an application program for causing the computer constituting the medical support device 11 to function as the medical support device 11. When the operation program 59 is started, as shown in FIG. 6, the CPU 51 of the medical support device 11 cooperates with the memory 52 and functions as a request reception unit 61, a display screen generation unit 62, an operation history acquisition unit 63, a prediction execution unit 64, and the like.

[0054] The request reception unit 61 receives various processing requests such as a display screen distribution request and an edit request from the client terminal 12. When the request reception unit 61 receives various processing requests, the request reception unit 61 inputs a processing instruction to each unit that executes the corresponding processing according to the content of the requested processing. For example, when there is a display screen distribution request from the client terminal 12, the request reception unit 61 inputs a generation instruction for the corresponding display screen to the display screen generation unit 62. Similarly, when there is an edit request for the display screen from the client terminal 12, the request reception unit 61 inputs an edit instruction for the corresponding display screen to the display screen generation unit 62. Note that the request reception unit 61 also receives a login request to the medical support device 11, and a login processing unit (not shown) executes login processing such as confirmation of the medical staff ID and password.

[0055] The display screen generation unit 62 generates or edits various display screens such as the clinical flow screen 81. The display screen generation unit 62 also functions as an operation proposal unit in the claims. In the present embodiment, when there is a request for distribution of a new display screen, the display screen generation unit 62 generates XML data representing the display screen, and when there is a request for editing the display screen, edits the previously created XML data according to the request content.

[0056] The display screen generation unit 62 accesses the server group 18 as necessary, and acquires information regarding the medical process or the like used for generating or editing the display screen. Note that the display screen generation unit 62 can hold part or all of the information regarding the medical process or the like acquired from the server group 18 in order to reduce the access frequency to the server group 18. When the login processing unit completes the login processing normally, the display screen generation unit 62 generates an initial screen 71 (see FIG. 9) that is first displayed after login. Also, when creating or editing the initial screen 71, the display screen generation unit 62 acquires information necessary for generating or editing the initial screen 71 from the server group 18, the client terminal 12, or a device or system that cooperates with other medical support systems 10.

[0057] The operation history acquisition unit 63 extracts, for example, information regarding an input operation on the client terminal 12 by a medical staff who is a user from among various processing requests from the client terminal 12 received by the request reception unit 61, and acquires an operation history. This operation history is accompanied by user identification information for identifying the user who uses the client terminal 12. The operation of the operation history acquisition unit 63 extracting information regarding an input operation on the client terminal 12 by the user and acquiring an operation history constitutes an operation history acquisition step.

[0058] FIG. 7 shows an example of an operation history when an input operation is performed on the client terminal 12. For example, medical facility information, date and time information, operation information, user identification information, and reference patient identification information are included in the operation history. Also, the example shown in FIG. 10 is an example of an input operation when editing an electronic medical record, an inspection image, a report, etc. when the layout display screen 101 is displayed on the client terminal 12.

[0059] Medical facility information is information about the medical facility where the medical support device 11 is installed, and includes facility ID, facility name, information on medical departments, etc. Note that, without being limited thereto, the medical facility information may include the number of registered users, address, contact information, etc. The medical facility information may be pre-stored, for example, in the storage 53 of the medical support device 11, or may be acquired from the client terminal 12 or the server group 18.

[0060] Operation information is information regarding the operations when a medical staff member who is a user inputs and operates the client terminal 12, and includes, for example, function name, operation target, operation content, operation attributes, etc. Specifically, the function name is inspection data viewing, the operation target is the file name of an endoscopic image, and the operation content includes instructions such as endoscopic image OPEN (opening the endoscopic image file), image movement, and image enlargement. Further, when the operation content is image movement, the operation attribute includes the numerical value of the coordinates corresponding to the amount of image movement, and when the operation content is image enlargement, the operation attribute includes the numerical value of the magnification ratio (display magnification) corresponding to the amount of image enlargement. Note that, without being limited thereto, in addition to endoscopic images, medical images such as X-ray images, test results such as blood tests and pathological tests, and test data such as test reports may also be operation targets, and the operation content may include the order in which the operation target is referred to, changes in the display layout input and operated by the user, etc.

[0061] The user identification information attached to the operation history identifies the user who uses the client terminal 12, such as the user ID, job type, gender, age, etc. The user ID is, for example, the number entered when logging in to the client terminal 12, etc. Information such as job type, gender, and age may be pre-stored in the storage 53 of the medical support device 11 in association with the user ID, or may be obtained from the client terminal 12 or the server group 18. Note that the user identification information attached to the operation history may include the personal information of the user (such as the name of the medical staff who is the user). In that case, when transmitting the operation history to an external server as described later, it is preferable to delete the part of the user's personal information before transmission. In the present embodiment, the form does not include the personal information of the user. Also, the user identification information may include the number of years of experience, etc.

[0062] In addition, the reference patient identification information attached to the operation history is the patient identification information included in the display screen such as the layout display screen 101 displayed when using the client terminal 12, that is, the patient ID associated with the electronic medical record, inspection image, report, etc. edited by the client terminal 12. Also, disease names, gender, age, etc. other than the patient ID may be obtained from the client terminal 12 or the server group 18. Also, the reference patient identification information may include the number of years of hospitalization, etc. Note that the reference patient identification information attached to the operation history may include the personal information of the patient (such as the name of the patient). In that case, when transmitting the operation history to an external server as described later, it is preferable to delete the part of the patient's personal information before transmission. In the present embodiment, the form does not include the personal information of the patient.

[0063] As described above, the operation history acquisition unit 63 transmits the operation history with user identification information and the like attached thereto to the learning device 13 via the network 14. The learning device 13, like the medical support device 11, has a well-known hardware configuration such as a CPU 51, a memory 52, a storage 53, a communication unit 54, and a connection circuit 55, and a well-known operating system and the like are installed, and is a high-performance computer having the function of a server.

[0064] As shown in FIG. 8, the learning device 13 functions as an acquisition unit 65, a registration unit 66, a storage unit 67, a learning unit 68, and a control unit 69 by an operating system or the like. As described above, the acquisition unit 65 acquires the operation history transmitted from the medical support devices 11 installed in a plurality of medical facilities A, B... X.

[0065] The control unit 69 controls the processing flow of the acquisition unit 65, the registration unit 66, and the learning unit 68. The registration unit 66 registers the operation history acquired by the acquisition unit 65 and the user identification information attached to the operation history in the storage unit 67. The storage unit 67 may be, for example, a part of the storage device provided in the learning device 13 or a storage device connected via the network 14.

[0066] The registration unit 66 registers the operation history as a sample for machine learning or the like by the learning unit 68. The registration unit 66 repeats the registration of the operation history from the medical support device 11 during the operation of the medical support system 10.

[0067] The learning unit 68 performs machine learning to generate a learned model that outputs the next operation candidate when any input operation is performed on the client terminal 12 by using a plurality of operation histories registered in the memory unit 67. In the present embodiment, the learning unit 68 extracts, in particular, the data to be operated on, the operation content (function) used as the input operation, or the order of the operation content used, and performs machine learning. The learning unit 68 reads out the operation history registered in the memory unit 67 and the attached user identification information, and generates a learned model, for example, from a plurality of operation histories having the same user ID, or from a plurality of operation histories of users having the same attributes. Users having the same attributes refer to users whose job type, medical department, patient's disease name, etc. included in the user identification information are the same. Alternatively, initially, a learned model may be generated from the operation histories of users having the same attributes, and when the operation histories having the same user ID are accumulated up to a predetermined number, a learned model may be generated from the plurality of operation histories having the same user ID. When a learned model is generated from the operation histories having the same user ID, it is possible to perform prediction optimized for the individual user. On the other hand, when a learning model is generated from the operation histories of users having the same attributes, there is an advantage that more operation histories can be collected as samples.

[0068] The learning device 13 transmits the learned model generated from the operation history to the medical support device 11 via the network 14. In this case, referring to the user ID attached to the operation history that is a sample of the learned model, the learned model is transmitted to the medical support device 11 that is the transmission source from which the operation history was transmitted.

[0069] The prediction execution unit 64 predicts the next operation when the client terminal 12 is input-operated. The prediction execution unit 64 can be configured by using the learned model (so-called AI (artificial intelligence) program) generated by the learning device 13 described above.

[0070] The prediction execution unit 64 configured using the learned model outputs the next operation candidate when any input operation is performed on the client terminal 12. The operation of the prediction execution unit 64 predicting the next operation candidate when the client terminal 12 is input-operated constitutes a prediction execution step. The input operation of the client terminal 12 is acquired from the request reception unit 61 or the like in the same manner as when the operation history acquisition unit 63 acquires the operation history. For example, when a learned model is generated from an example of the operation history such as that shown in FIG. 7 described above and the prediction execution unit 64 is configured from this learned model, the prediction execution unit 64 focuses on the endoscopic image as the data to be operated on. Then, for the input operation of "image OPEN" of the endoscopic image, the next operation candidate of "movement of the endoscopic image" is output as the next operation candidate. Alternatively, for the input operation of "movement of the endoscopic image", the next operation candidate of "enlargement of the endoscopic image" is output. Further, when outputting the movement of the endoscopic image as the next operation candidate, it is preferable to output with the movement amount attached, and when outputting the enlargement of the endoscopic image, it is preferable to output with the enlargement ratio attached.

[0071] In the present embodiment, the display screen generation unit 62 makes a proposal to the client terminal 12 from the next operation candidate predicted by the prediction execution unit 64. Specifically, the display screen generation unit 62 generates or edits XML data representing the display screen using the next operation candidate predicted by the prediction execution unit 64 and transmits it to the client terminal 12. The operation of the display screen generation unit 62 making a proposal to the client terminal 12 from the next operation candidate predicted by the prediction execution unit 64 constitutes an operation proposal step.

[0072] The medical support system 10 configured as described above operates as follows. First, when a medical staff logs in to the medical support system 10 using the client terminal 12, the display screen generation unit 62 generates the initial screen 71 shown in FIG. 9 based on the settings etc. for each medical staff and provides it to the client terminal 12. Thereby, the client terminal 12 displays the initial screen 71 on the screen of the display unit 36.

[0073] The initial screen 71 has, for example, three display columns: a schedule display column 72, a mail display column 73, and a list display column 74. The display contents of the schedule display column 72 and the mail display column 73 are generated by a gadget engine, which is part of the operating program 39 of the client terminal 12, obtaining information from the client terminal 12 and other devices or systems. Also, in this embodiment, the list display column 74 displays at least a part of the clinical flow screen 81. For this reason, the display screen generation unit 62 generates the initial screen 71 including the schedule display column 72 and the mail display column 73 without content, and the list display column 74 including the content of the clinical flow screen 81. The client terminal 12 uses the gadget engine to display on the screen of the display unit 36 the initial screen 71 with the contents of the schedule display column 72 and the mail display column 73 replenished.

[0074] If all of the content to be displayed does not fit within the list display column 74, a scroll bar 78 and a scroll bar 79 for transitioning (so-called scrolling) the display content of the list display column 74 are displayed in the list display column 74 or in the vicinity of the list display column 74. The scroll bar 78 is a GUI that is operated when transitioning the display content of the list display column 74 in the left-right direction to display the non-displayed part. The scroll bar 79 is a GUI that is operated when transitioning the display content of the list display column 74 in the up-down direction to display the non-displayed part. The display and control of such GUIs are performed by the GUI control unit 41.

[0075] On the above-described initial screen 71, when an operation such as a predetermined menu is performed using a GUI such as a pointer (not shown), the request issuing unit 42 issues a distribution request for the display screen. In this embodiment, in order to display the layout display screen 101 that is not displayed on the initial screen 71, an operation for displaying the layout display screen 101, for example, an input operation of selecting one of the patients displayed in the list display column 74 using a GUI is executed. Thereby, the request issuing unit 42 issues a distribution request for the layout display screen 101.

[0076] When the request issuing unit 42 issues a distribution request for a display screen, in the medical support device 11, the request receiving unit 61 receives the distribution request for the display screen, and the display screen generation unit 62 generates a display screen related to the distribution request for the display screen. In the present embodiment, the display screen generation unit 62 refers to the patient identification information (for example, patient ID) included in the list display column 74 and acquires information related to the patient. Specifically, the electronic medical record, inspection image, report, etc. with the same patient identification information as the patient identification information included in the list display column 74 are appropriately acquired from the server group 18 or the like. Then, using the information related to the patient acquired by referring to the patient identification information, the layout display screen 101 is generated.

[0077] The GUI control unit 41 of the client terminal 12 receives the distribution of the display screen generated as described above, displays this on the screen of the display unit 36 instead of the initial screen 71, or displays it in another window or the like while leaving the initial screen 71 and overlapping it.

[0078] As described above, when the display screen generation unit 62 generates a display screen related to a distribution request, before the generation of the display screen, simultaneously with (in parallel with) the generation of the display screen, or after the generation of the display screen, the prediction execution unit 64 outputs the next operation candidate for the input operation. That is, the prediction execution unit 64 outputs the next operation candidate for the input operation of displaying the layout display screen 101 (see FIG. 10) in the client terminal 12. When a learned model is generated from a plurality of operation histories including the example shown in FIG. 7 and the prediction execution unit 64 is configured from this learned model, the prediction execution unit 64 outputs the next operation candidate such as OPEN of the endoscope image for the input operation of displaying the layout display screen 101. Alternatively, when the endoscope image is included from the beginning (before the input operation) as the information for creating the layout display screen 101, the next operation candidates such as moving the endoscope image or enlarging the endoscope image are output for the input operation of image OPEN of the endoscope image.

[0079] Next, the display screen generation unit 62 makes a proposal to the client terminal 12 based on the next operation candidate predicted by the prediction execution unit 64. That is, for the input operation of displaying the layout display screen 101 shown in FIG. 10, as shown in FIG. 11, a display screen in which the endoscopic image 102 is superimposed on the layout display screen 101 is edited. Here, the endoscopic image 102 to be superimposed on the layout display screen 101 is an endoscopic image with the same patient identification information as the patient identification information obtained when creating the layout display screen 101. For example, it is the most recent endoscopic image at the time of shooting. Alternatively, an endoscopic image in which the suspected diseased area is shown most clearly using a CAD (computer-aided diagnosis) function or the like may be displayed. Also, in the case of an endoscopic image, parts that are frequently referred to by the user, such as the esophagogastric junction, duodenal bulb, anterior wall of the stomach, gastric angle, lower body of the stomach, middle body of the stomach, upper body of the stomach, etc., may be automatically laid out and displayed in the order of frequent reference.

[0080] Note that when the endoscopic image 102 is included from the beginning as information for creating the layout display screen 101, instead of displaying the endoscopic image 102, the layout may be changed, that is, a display screen in which the endoscopic image 102 is moved or the endoscopic image 102 is enlarged may be edited. Then, the display screen generation unit 62 distributes the edited display screen to the client terminal 12. Also, in this case, the prediction execution unit 64 predicts the movement amount and magnification rate of the endoscopic image 102, and it is preferable that the display screen generation unit 62 moves the endoscopic image 102 by the movement amount predicted by the prediction execution unit 64 and enlarges the endoscopic image 102 at the magnification rate predicted in the same way.

[0081] Thereafter, the GUI control unit 41 of the client terminal 12 receives the distribution of the display screen edited as described above and displays it on the screen of the display unit 36 instead of the layout display screen 101 displayed first.

[0082] As described above, in the medical support system 10 and the medical support device 11 of the present embodiment, since the operation history is transmitted to the learning device 13 as an external server to generate a learned model, it is possible to collect a sufficient number of operation histories as samples, and the accuracy of the learned model and the prediction execution unit 64 can be improved. Further, when transmitting the operation history to the learning device 13 as an external server, since a user ID that does not include personal information is attached to the operation history as user identification information, the risk of leakage of personal information can be avoided.

[0083] Note that the editing of the display screen based on the next operation candidate predicted by the prediction execution unit 64 performed by the display screen generation unit 62 is not limited to the above, and for example, as shown in FIG. 12, the display of the endoscope image 102 may be changed. In this case, as the next operation candidate predicted by the prediction execution unit 64, although it is the image OPEN of the endoscope image, if the endoscope image 102 is included from the beginning as information for creating the layout display screen 101, the display of the endoscope image 102 may be changed.

[0084] In the example shown in FIG. 12, the frame line surrounding the endoscope image 102 is thickened, and the color of the frame line 102A is changed (for convenience of illustration, instead of changing the color, hatching is applied inside the frame line). And, similar to the above embodiment, the GUI control unit 41 of the client terminal 12 displays the edited layout display screen 101 on the screen of the display unit 36. Further, not limited to this, when the endoscope image 102 is included from the beginning as information for creating the layout display screen 101, both the change in the display of the endoscope image shown in FIG. 12 and the movement of the endoscope image or the enlargement of the endoscope image shown in FIG. 11 may be performed. Also, in FIG. 11, one endoscope image 102 is displayed, but not limited to this, a plurality of endoscope images may be displayed.

[0085] In addition, as another display based on the next operation candidate predicted by the prediction execution unit 64, which is performed by the display screen generation unit 62, the operation content (function) that the user should input may be displayed. For example, when the next operation candidate predicted by the prediction execution unit 64 is the movement of the endoscope image or the magnification of the endoscope image, this may be displayed as the operation content 103 (see FIG. 12) that the user should input. Further, when the learned model has learned the order of operation contents, the operation content that the user should input next may be displayed by paying attention to the operation content that the user input last time.

[0086] [Second Embodiment] In the first embodiment described above, the learning unit 68 extracts the data of the operation target in the operation history, the function used as the input operation, and the order of the input operations and performs machine learning. However, the content of machine learning from the operation history is not limited to this. In the second embodiment, the symptoms, disease names, and examination names of the patients diagnosed by the user in the operation history are used as the operation targets, and when it is a predetermined symptom, disease name, or examination name, machine learning may be performed on what examination data was referred to, etc. Note that the configurations of the medical support system 10 and the medical support device 11 are the same as those in the first embodiment described above.

[0087] In the operation history shown in FIG. 13, on the left side is a list of the symptoms, disease names, and examinations of the patient as the operation target, and on the right side is a list of the examination data referred to by the medical staff as the user when the examinations included in the operation target were performed. The referred examination data differs depending on the occupation of the user. Similar to the first embodiment, the medical support device 11 of this embodiment transmits the operation history shown in FIG. 13 to the learning device 13 with the user identification information attached.

[0088] In this embodiment, the learning unit 68 of the learning device 13 uses the symptoms, disease names, and examination names of the patients treated by the user (such as in medical treatment) as the operation targets, and for predetermined symptoms, disease names, and examination names, it performs machine learning on what examination data was referred to. Specifically, the learning unit 68 generates a learned model that outputs the names of the examination data with high reference frequencies for each user occupation for predetermined symptoms, disease names, and examination names. The learned model generated by the learning unit 68 is transmitted to the medical support device 11 and constitutes the prediction execution unit 64 of the medical support device 11 in the same manner as in the first embodiment.

[0089] When the learned model is generated as described above, the medical support system 10 operates as follows. Note that until the medical staff logs in to the medical support system 10 using the client terminal 12 and the layout display screen 101 is displayed, it is the same as in the first embodiment. Then, the prediction execution unit 64 extracts symptoms, disease names, and examination names as operation targets from data such as the electronic medical record, examination images, and reports included in the layout display screen 101.

[0090] Then, the prediction execution unit 64 outputs the name of the examination data that the user will refer to next from the extracted symptoms, disease names, and examination names. For example, when a learned model is generated from an example of the operation history such as that shown in FIG. 13 described above, and the prediction execution unit 64 is constituted from this learned model, focusing on the symptoms, disease names, and examination names, when the user's occupation is an endoscopist, the endoscope image is predicted as the name of the examination data.

[0091] The display screen generation unit 62 makes a proposal to the client terminal 12 from the next operation candidate predicted by the prediction execution unit 64. That is, for the input operation of displaying the layout display screen 101, the display screen of the inspection data (for example, an endoscopic image) to be referred to next is edited to replace the layout display screen 101 or to be superimposed on the layout display screen 101. Then, the display screen generation unit 62 distributes the edited display screen to the client terminal 12. The GUI control unit 41 of the client terminal 12 receives the distribution of the display screen edited as described above and displays it on the screen of the display unit 36 instead of the layout display screen 101 initially displayed. By the above operations, similar to the first embodiment, it is possible to avoid the risk of leakage of personal information, and it is possible to improve the prediction accuracy of the learned model and the prediction execution unit 64.

[0092] [Third Embodiment] Examples of the operation history learned by the learning unit 68 are not limited to those shown in the first and second embodiments above. For example, in the operation history, a written document created by the user is used as the operation target, and the type and creation frequency of the operation target written document, or the order in which the written documents are created, etc. are extracted from the operation history to create a learned model. Note that the configurations of the medical support system 10 and the medical support device 11 are the same as those in the first embodiment.

[0093] FIG. 14 is an example of the operation history used in this embodiment. In this operation history, the left column is the user's occupation, and the right column is the name of the written document created by the user corresponding to the left column. Similar to the first embodiment, the medical support device 11 of this embodiment transmits the operation history shown in FIG. 14 to the learning device 13 with user identification information attached. Note that in FIG. 14, a referral letter is described as the name of the written document created by the nurse, but this is written by the doctor on behalf of the nurse, and it is a document that needs to be finally confirmed and signed by the doctor. Similarly, for a medical certificate, a prescription, etc., it is permitted for staff other than the doctor to act on behalf of the doctor as an assistant on the condition that the doctor finally confirms and signs. There are cases where a nurse or a medical clerk acts on behalf of the doctor.

[0094] In this embodiment, the learning unit 68 of the learning device 13 targets the name of the document created by the user, and machine-learns the type and creation frequency of the created document, or the order in which the created document is created. Specifically, the learning unit 68 generates a learned model that outputs the names of created documents with high creation frequencies for each occupation of the user by machine learning. The learned model generated by the learning unit 68 is transmitted to the medical support device 11 and is used in the prediction execution unit 64 of the medical support device 11 in the same manner as in the first embodiment.

[0095] When the learned model is generated as described above, the medical support system 10 operates as follows. Note that until the medical staff logs in to the medical support system 10 using the client terminal 12 and the layout display screen 101 is displayed, it is the same as in the first embodiment. Then, the prediction execution unit 64 extracts the occupation of the user from the various processing requests based on the logged-in user ID. Then, the prediction execution unit 64 outputs the name of the created document that the user is likely to create next from the extracted occupation of the user. For example, when a learned model is generated from an example of an operation history such as that shown in FIG. 14 described above and the prediction execution unit 64 is configured from this learned model, focusing on the occupation of the user, when the occupation of the user is a radiologist, a general X-ray reading report is predicted as the name of the created document that is likely to be created next. Also, when the learned model learns the order in which created documents are created for each occupation of the user, it may be possible to predict the name of the created document that is likely to be created next by focusing on the document created by the user last time.

[0096] The display screen generation unit 62 proposes to the client terminal 12 from the next operation candidates predicted by the prediction execution unit 64. That is, for an input operation indicating that a user of a predetermined occupation has logged in, a created document that is likely to be created next (for example, a general X-ray reading report) is set as the document to be created next, and the display screen is edited to replace the layout display screen 101 or to be superimposed on the layout display screen 101. Then, the display screen generation unit 62 distributes the edited display screen to the client terminal 12.

[0097] The GUI control unit 41 of the client terminal 12 receives the distribution of the display screen edited as described above, and displays it on the screen of the display unit 36, replacing the layout display screen 101 that was initially displayed. For a document set as the document to be created next by the display screen generation unit 62, if the user has already created it, it is not necessary to display the created document. Through the above operations, similar to the first embodiment, it is possible to avoid the risk of personal information leakage and improve the prediction accuracy of the learned model and the prediction execution unit 64.

[0098] [Fourth Embodiment] In the above third embodiment, the prediction execution unit 64 predicts the name of the document to be created next for each occupation of the user. However, the prediction by the prediction execution unit 64 is not limited to this, and it may predict and propose the name of the document that is likely to be created according to the inspection execution status and document creation status for each user or each occupation of the user.

[0099] FIG. 15 is an example of the operation history used in this embodiment. In this operation history, the left column shows the occupation of the user, the middle column shows the name of the document created by the user corresponding to the left column, and the right column shows the creation time when the user corresponding to the left column created the document corresponding to the middle column. Note that as the creation time included in the operation history, a more detailed time or time zone may be obtained. For example, the time may be limited, such as within a few hours after the endoscopy examination, and multiple creation times may be obtained for one document. Also, regarding the creation time, regardless of the medical items such as after the examination or after the diagnosis, the time or time zone when the user created the document may be obtained.

[0100] In the present embodiment, the learning unit 68 of the learning device 13 targets the document name created by the user and performs machine learning on the creation time of the created document. Specifically, the learning unit 68 generates a learned model that outputs the creation time when the created document is frequently created for each occupation of the user. The learned model generated by the learning unit 68 is transmitted to the medical support device 11 and constitutes the prediction execution unit 64 of the medical support device 11 in the same manner as in the first embodiment.

[0101] When the learned model is generated as described above, the medical support system 10 operates as follows. The prediction execution unit 64 outputs the document that is likely to be created and the creation time when the document is likely to be created for each occupation of the user extracted from various processing requests. For example, when a learned model is generated from an example of the operation history as shown in FIG. 15 described above and the prediction execution unit 64 is configured from this learned model, focusing on the occupation of the user, when the occupation of the user is a radiologist, a general X-ray imaging report is predicted as the document name that is likely to be created, and after general X-ray imaging is predicted as the creation time when the document is likely to be created.

[0102] The display screen generation unit 62 makes a proposal to the client terminal 12 from the next operation candidate predicted by the prediction execution unit 64. In this case, after acquiring the prediction by the prediction execution unit 64, the display screen generation unit 62 accesses the server group 18 and also acquires the creation status of whether the predicted document has been created or not. As shown in FIG. 16, the display screen generation unit 62 sets the creation timing (for example, after general X-ray imaging) with a high possibility of creation predicted by the prediction execution unit 64 as the creation timing at which the user should create a document, and at that creation timing, performs a display 105 to prompt the creation of the document to be created next (for example, a general X-ray reading report). Also, in this case, the document with a high possibility of being created next is set as the document to be created next. Note that the creation timing here is not limited to after or before any medical treatment, nor is it limited to a specific time such as hours and minutes, but also includes time zones such as morning and afternoon, dates, days of the week, etc. The display 105 for prompting creation is a sentence "The general X-ray reading report has not been created." and a frame line 105A surrounding the sentence is thickened, and the color of the frame line 105A is made different from the surrounding color. Note that in this case, if the user has already created the name of the document to be created next predicted by the prediction execution unit 64, the display 105 for prompting creation may not be performed.

[0103] Also, as a proposal made by the display screen generation unit 62, it may be made at a time after the creation timing predicted by the prediction execution unit 64. For example, when the document to be created has not been created when a predetermined time has elapsed from the predicted creation timing, a display 105 for prompting the creation of the document may be performed.

[0104] [Fifth Embodiment] In each of the above embodiments, as the content to be machine-learned from the operation history, machine learning is performed on the user-centered timing such as the order of the user's input operations, the creation frequency of the document, the creation timing when the document was created, and the creation status of the document. However, it is not limited to this, and the creation timing according to the medical treatment schedule of the patient in charge of the user may be machine-learned and predicted by the prediction execution unit 64. Note that the configurations of the medical treatment support system 10 and the medical treatment support device 11 are the same as those in the first embodiment above.

[0105] FIG. 17 shows an example of an operation history used in the present embodiment. This operation history arranges the medical schedules of the patients in charge of the user in chronological order, and is also what is called a so-called timeline. Further, the content of this timeline may be created by the medical support device 11 as a display screen and distributed to the client terminal 12, or the timeline may be edited by an input operation of the client terminal 12. Medical items are arranged in the upper row of the timeline, and the names of the created documents corresponding to the medical items are arranged below the medical items. The chronological order shown in the bottom row indicates the period when the patient undergoes diagnosis before surgery, the period when the patient is hospitalized for treatment and surgery, and the period when the patient receives follow-up after surgery, respectively. Note that FIG. 16 shows an example of a gastric cancer patient, and the occupation of the user of the client terminal 12 is a surgeon.

[0106] In the present embodiment, the learning unit 68 of the learning device 13 performs machine learning on the creation time corresponding to each medical item, with the name of the created document corresponding to each medical item as the operation target, for the medical items of the patients in charge of the user. That is, the learning unit 68 generates a learned model that outputs the creation times with a high frequency of creating the created document according to the medical schedule of the patient. The learned model generated by the learning unit 68 is transmitted to the medical support device 11, and constitutes the prediction execution unit 64 of the medical support device 11 in the same manner as in the first embodiment.

[0107] When the learned model is generated as described above, the medical support system 10 operates as follows. The prediction execution unit 64 outputs the created document corresponding to each medical item and the creation time with a high frequency of creating the created document in the medical schedule for the medical items of the patients in charge of the user extracted from various processing requests. For example, when a learned model is generated from an example of an operation history such as FIG. 16 described above, and the prediction execution unit 64 is configured from this learned model, paying attention to the medical schedule of the patient in charge of the user, for example, when there is an endoscopic examination in the medical item, the endoscopic consent form and the endoscopic report as the names of the created documents with a high possibility of being created are predicted, and the time before a predetermined time of the endoscopic examination or after a predetermined time of the endoscopic examination, etc. are predicted as the creation times with a high possibility of being created.

[0108] The display screen generation unit 62 makes a proposal to the client terminal 12 from the next operation candidate predicted by the prediction execution unit 64. In this case, the display screen generation unit 62 sets the creation timing (for example, after a predetermined time has elapsed since the endoscopy examination) with a high possibility of being created predicted by the prediction execution unit 64 as the creation timing at which the user should create a document, and at that creation timing, edits the display screen of the document to be created next (for example, an endoscopy report) to replace the currently displayed display screen or to superimpose it on the currently displayed display screen. Also, in this case, the document with a high possibility of being created next is set as the document to be created next. Then, the display screen generation unit 62 distributes the edited display screen to the client terminal 12.

[0109] The GUI control unit 41 of the client terminal 12 receives the distribution of the display screen edited as described above and displays it on the screen of the display unit 36. Note that if the user has already created the name of the document to be created next predicted by the prediction execution unit 64, it is not necessary to display the document.

[0110] Alternatively, when the prediction execution unit 64 predicts the creation timing, a display prompting creation may be performed in the same manner as in the fourth embodiment. Note that the creation timing here is not limited to after or before any medical treatment, and is not limited to a specific time such as hours and minutes, but also includes time zones such as morning and afternoon, dates, days of the week, etc. The display 105 prompting creation is a sentence "The general X-ray interpretation report has not been created yet." and the frame line 105A surrounding the sentence is thickened and the color of the frame line 105A is made different from the surrounding color. Note that in this case, if the user has already created the name of the document to be created next predicted by the prediction execution unit 64, it is not necessary to perform the display 105 prompting creation. By the above operations, it is possible to avoid the risk of leakage of personal information and improve the prediction accuracy of the learned model and the prediction execution unit 64, as in the first embodiment.

[0111] In the above-described fourth and fifth embodiments, machine learning is performed on the creation time of the document created by the user, and the prediction execution unit 64 makes a prediction and then displays the created document or displays a prompt for creation at the predicted creation time. However, the present invention is not limited to this. For example, from the user's operation history, after performing machine learning on the type of examination or treatment (including surgery or treatment) ordered by the user and the order time when the user ordered the examination or treatment, the prediction execution unit 64 makes a prediction in the same manner as in the above-described embodiments and the like. Then, based on the prediction of the prediction execution unit 64, the display screen generation unit 62 may display the examination or treatment to be ordered or display a prompt for the user to order the examination or treatment (for example, display a sentence such as "An MRI examination order has not been issued." on the display screen) at the order time when the user should place an order. Here, the order time referred to herein is not limited to after any medical treatment or before medical treatment, etc., and is not limited to a specific time such as hours and minutes, but also includes time zones such as morning and afternoon, dates, days of the week, etc. In this way, when predicting the order time when the user should place an order, in the operation history, regardless of the medical item such as after examination or after medical treatment, the time or time zone when the user placed an order may be obtained as the order time. Thereby, for example, since a pathological examination or the like takes time to process, an order for the examination may be placed in the morning time zone, or since the surgical day is determined by the medical facility, the necessary examinations and document creations may be performed before that, and the accurate tendencies of the user can be learned.

[0112] In the above-described embodiments, when transmitting the operation history to the learning device 13, a user ID that does not include personal information is attached to the operation history. However, as shown in FIG. 18, if the operation history acquired by the operation history acquisition unit 63 includes personal information, and if personal information is also included for the patient in charge of the user, the personal information part (user name, patient ID, patient name, etc.) may be deleted before transmitting it to the learning device 13. Thereby, the risk of personal information leakage can be more reliably avoided.

[0113] Also, as shown in FIG. 19, it is preferable to store the deleted personal information part in the server group 18 installed in the same medical facility as the medical support device 11 and transmit the operation history with the personal information part deleted to the learning device 13. Note that it is preferable to attach a user ID as user identification information corresponding to the operation history to the part of the personal information deleted from the operation history. Thereby, when using the learned model generated by the learning device 13 in the prediction execution unit 64, it is also possible to read out the personal information stored in the server group 18 and restore the part related to the personal information.

[0114] In each of the above embodiments, the hardware structure of the processing unit that executes various processes such as the GUI control unit 41, the request issuing unit 42, the request receiving unit 61, the display screen generation unit 62, the operation history acquisition unit 63, the prediction execution unit 64, the acquisition unit 65, the registration unit 66, the storage unit 67, the learning unit 68, and the control unit 69 is various processors as shown below. The various processors include a CPU (Central Processing Unit), which is a general-purpose processor that executes software (program) and functions as various processing units, a programmable logic device (PLD) such as an FPGA (Field Programmable Gate Array), which is a processor whose circuit configuration can be changed after manufacturing, and a dedicated electric circuit (Graphical Processing Unit: GPU), which is a processor having a circuit configuration designed specifically to execute various processes.

[0115] One processing unit may be composed of one of these various processors, or may be composed of a combination of two or more processors of the same or different types (for example, a plurality of FPGAs, a combination of a CPU and an FPGA, or a combination of a GPU and a CPU). Further, a plurality of processing units may be composed of one processor. As an example of configuring a plurality of processing units with one processor, first, as represented by a computer such as a client or a server, one processor is configured by a combination of one or more CPUs and software, and this processor functions as a plurality of processing units. Second, as represented by a System On Chip (SoC), there is a form in which a processor that realizes the functions of an entire system including a plurality of processing units with one IC (Integrated Circuit) chip is used. As described above, various processing units are configured using one or more of the above various processors as a hardware structure.

[0116] Furthermore, the hardware structure of these various processors is more specifically an electrical circuit (circuitry) in a form that combines circuit elements such as semiconductor elements. Another aspect of the present invention is a medical support device that includes a prediction execution unit that acquires an operation history when operating a terminal device from a plurality of terminal devices installed in a plurality of medical facilities, and uses a learned model generated by an external server installed outside the medical facility to learn the acquired operation history to predict the next operation candidate when inputting an operation to the terminal device, and proposes to the terminal device based on the next operation candidate predicted by the prediction execution unit.

[0117] The present invention is not limited to the above embodiments, and it goes without saying that various configurations can be adopted without departing from the gist of the present invention. Furthermore, the present invention extends to a storage medium that stores a program in addition to the program.

Explanation of Reference Numerals

[0118] From the above description, the medical support device described in claim 1 below can be understood. [Claim 1] A medical support device comprising a processor, wherein the processor, acquires an operation history when a user operates the terminal device from the terminal device installed in the medical facility, predicts the next operation candidate when the terminal device is input-operated using a learned model generated by having an external server installed outside the medical facility learn the acquired operation history, and makes a proposal to the terminal device from the predicted next operation candidate, Medical support device.

[0119] 10 Medical support system 11 Medical support device 12 Client terminal 13 Learning device 14, 16 Network 17 Medical information system 18 Server group 21 Electronic medical record server 21A Medical record database 22 Image server 22A Image database 23 Report server 23A Report database 31, 51 CPU (Central Processing Unit) 32, 52 Memory 33, 53 Storage 34, 54 Communication unit 35, 55 Connection circuit 36 Display unit 37 Operation unit 39, 59 Operating program 41 GUI (Graphical User Interface) control unit 42 Request issuing unit 61 Request reception unit 62 Display screen generation unit 63 Operation history acquisition unit 64 Prediction execution unit 65 Acquisition unit 66 Registration unit 67 Memory unit 68 Learning unit 69 Control unit 71 Initial screen 72 Schedule display column 73 Mail display column 74 List display column 78, 79 Scroll bar 81 Clinical flow screen 101 Layout display screen 102 Endoscopic image 102A, 105A Frame line 103 Operation content 105 Display A1, A2, B1 Doctors G1, G2, G19 Groups N1 Technician

Claims

1. Comprising a processor, The processor Obtains an operation history including operation information of an operation performed on the terminal device and user information of the user who performed the operation, Outputs the next operation candidate for the terminal device by using a learned model generated by learning based on the operation history including the user information, The learned model is received from an external server, Outputs the document to be created next as the next operation candidate, A medical support device that, depending on whether the next document to be created has been created, when outputting the next document to be created, if the next document to be created has not been created, performs a display prompting creation, and if the next document to be created has been created, does not perform the display prompting creation.

2. The user information includes at least one user attribute of the user's occupation, the user's medical department, or the disease name of the patient in charge of the user, The processor Outputs the next operation candidate for the terminal device by using a learned learned model generated by learning based on a plurality of the operation histories having the same user attribute. The medical support device according to claim 1.

3. The user information includes a user ID that identifies the individual who is the user, The processor Outputs the next operation candidate for the terminal device by using a learned model generated from the operation history having the same user ID or a learned model generated from the operation history of users having the same attributes. The medical support device according to claim 2.

4. The processor Outputs the next operation candidate for each user information by using the learned model. The medical support device according to claim 1.

5. The operation information includes at least one of the name of the operation target operated by the user, the type of the operation target, the frequency of operating the operation target, or the order of operating the operation target. The medical support device according to claim 1.

6. The processor Outputs the operation target to be operated next as the next operation candidate. The medical support device according to claim 5.

7. The processor Outputs the next operation candidate according to whether the next operation candidate has been operated. The medical support device according to claim 6.

8. The operation history includes at least one of the name of the document to be created, the type of the document to be created, the frequency of creating the document to be created, or the order of creating the document to be created. The medical support device according to claim 1.

9. The processor is configured to The medical support device according to claim 8, which proposes to display a screen for editing a document to be created next as the next operation candidate. **Claim 10** The operation history includes at least the creation time of creating the document to be created. The processor is configured to Using a learned model generated by learning based on the operation history including the user information, output the next document to be created for the terminal device and the creation time of the next document to be created. The medical support device according to claim 1. **Claim 11** The operation history includes medical items and medical schedules. The processor is configured to Using the learned model, output the document to be created and the creation time according to the medical schedule. The medical support device according to claim 10. **Claim 12** The operation history includes the type of examination or treatment ordered by the user and the order time when the user ordered the examination or treatment. The processor is configured to Using the learned model, output at least one of the types of the next examination or treatment and at least one of the order times of the next examination or treatment. The medical support device according to claim 1. **Claim 13** The medical support device according to any one of claims 1 to 12, The terminal device, A medical support system comprising an external server that receives the operation history, generates the learned model, and transmits the learned model to the medical support device. **Claim 14** Obtaining an operation history including operation information of an operation performed on a terminal device and user information of the user who performed the operation; Outputting a next operation candidate for the terminal device using a learned model generated by learning based on the operation history including the user information. The learned model is received from an external server. The learned model is received from an external server. Output a document to be created next as the next operation candidate. A method of operating a medical support device, in which, depending on whether the next document to be created has been created, when the next document to be created is output, if the next document to be created has not been created, a display prompting creation is performed, and if the next document to be created has been created, the display prompting creation is not performed. **Claim 15** A function of obtaining an operation history including operation information of an operation performed on a terminal device and user information of the user who performed the operation. Causing a computer to implement a function of outputting a next operation candidate for the terminal device by using a learned model generated by learning based on the operation history including the user information; The learned model is received from an external server; Outputting a document to be created next as the next operation candidate; An operation program for a medical support device that, when outputting the next document to be created, performs a display prompting creation if the next document to be created has not been created, and does not perform the display prompting creation if the next document to be created has been created, according to whether the next document to be created has been created.

Images