Report creation assistance device and report creation assistance system

Abstract

To automatically provide a person to have a consultation with according to the details of a medical report, and reflect a result of consultation on the medical report more simply without misunderstanding.SOLUTION: A report creation auxiliary device comprises an analysis unit, an evaluation unit, a determination unit, a specification unit, and a notification unit. The analysis unit analyzes a report and creates an analysis result. The evaluation unit evaluates the report based on the analysis result, and creates an evaluation result. The determination unit determines whether an opinion on the report needs to be sought based on the evaluation result. When the determination unit determines that an opinion on the report needs to be sought, the specification unit specifies a person to have a consultation with. The notification unit performs notification to the person.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The embodiments disclosed in this specification and the drawings relate to a report creation assistance device and a report creation assistance system. [Background technology]

[0002] Currently, radiology reports are often created in an image reading room by multiple radiologists. To complete the report, radiologists consult with other radiologists in the room and undergo double checks. This often requires verbal communication while mutually referencing the display screen of the report being created. Therefore, remote work, where verbal communication is difficult, may not be promoted when creating radiology reports.

[0003] Chat is an alternative to verbal communication. However, there are some problems with using chat to consult about radiology work. For example, some of the problems include "not knowing who to consult because you don't know the other person's skills," "not knowing whether to ask for help because you don't know the other person's situation," and "misunderstandings can occur because the person you are consulting doesn't refer to the display screen of the radiology report."

[0004] As a result, if communication is limited to chat, junior radiologists may be unable to (or choose not to) consult with others, or may complete their radiology reports with a misunderstanding. This could result in a decline in the quality of the radiology report. Therefore, it would be desirable to be able to automatically provide people who can be consulted based on the content of the radiology report. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2006-277219 Summary of the Invention [Problem to be solved by the invention]

[0006] One of the problems to be solved by the embodiments disclosed in this specification and the drawings is to automatically provide a person who can be consulted based on the contents of a medical report, and to reflect the consultation results in the medical report more concisely and without misunderstanding. However, the problems to be solved by the embodiments disclosed in this specification and the drawings are not limited to the above problem. Problems corresponding to the effects of each configuration shown in the embodiments described below can also be positioned as other problems. [Means for solving the problem]

[0007] A report creation assistance device according to an embodiment includes an analysis unit, an evaluation unit, a determination unit, an identification unit, and a notification unit. The analysis unit analyzes a report and generates an analysis result. The evaluation unit evaluates the report based on the analysis result and generates an evaluation result. The determination unit determines whether or not it is necessary to request an opinion on the report based on the evaluation result. If it is determined that it is necessary to request an opinion on the report, the identification unit identifies a party that can be consulted. The notification unit notifies the party. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of a medical information system according to the first embodiment. [Figure 2] FIG. 2 is a diagram illustrating a work status table according to the first embodiment. [Figure 3] FIG. 3 is a diagram illustrating a doctor information table according to the first embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of the configuration of the report creation assist device according to the first embodiment. [Figure 5] FIG. 5 is a diagram for explaining missing elements when the report contents are vectorized. [Figure 6]FIG. 6 is a sequence diagram illustrating the remote consultation pre-processing executed by the medical information system according to the first embodiment. [Figure 7] FIG. 7 is a flowchart illustrating the consultation proposal determination process executed by the report creation assist device according to the first embodiment. [Figure 8] FIG. 8 is a flowchart for explaining the destination specifying process executed by the report creation assisting device according to the first embodiment. [Figure 9] FIG. 9 is a diagram illustrating an example of the configuration of a report creation assisting device according to the second embodiment. [Figure 10] FIG. 10 is a sequence diagram illustrating a user confirmation process executed by the medical information system according to the second embodiment. [Figure 11] FIG. 11 is a diagram illustrating an example of the configuration of a report creation assisting device according to the third embodiment. [Figure 12] FIG. 12 is a sequence diagram illustrating a remote consultation process executed by a medical information system according to the third embodiment. [Figure 13] FIG. 13 is a diagram illustrating a support UI according to the third embodiment. [Figure 14] FIG. 14 is a flowchart illustrating a comment analysis process executed by the report creation assist device according to the third embodiment. [Figure 15] FIG. 15 is a diagram illustrating a support UI according to the third embodiment. [Figure 16] FIG. 16 is a diagram illustrating a support UI according to the third embodiment. [Figure 17] FIG. 17 is a diagram illustrating a support UI according to the third embodiment. [Figure 18] FIG. 18 is a diagram illustrating a support UI according to the third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, embodiments of a report creation assist device and a report creation assist system will be described in detail with reference to the drawings.

[0010] (First embodiment) Fig. 1 is a diagram showing an example of the configuration of a medical information system according to a first embodiment. For example, as shown in Fig. 1, a medical information system 100 according to the first embodiment includes a report creation assistant device 110, a hospital information system (HIS) 120, a terminal device 130, a radiology information system (RIS) 140, a medical image diagnostic device 150, and a medical image management system (Picture Archiving and Communication Systems (PACS)) 160. These devices and systems are connected to each other so as to be able to communicate with each other via a network such as a local area network (LAN). The medical information system 100 may also include a data warehouse (DWH) or the like.

[0011] The HIS 120 includes, for example, an electronic medical record system that manages information related to electronic medical records. The electronic medical record system includes an electronic medical record DB (Data Base) 121 that stores information related to electronic medical records. The information related to electronic medical records includes, for example, patient information and medical treatment information. The HIS 120 may also include a report creation assistant device 110 and a report creation device described below. The electronic medical record system may also incorporate various functions of the report creation assistant device 110.

[0012] Patient information is information specific to a patient, and includes, for example, a patient ID, patient name, gender, date of birth, and age. Medical information is information that medical professionals learn about a patient's physical condition, medical condition, and treatment during the course of medical care. Medical information includes, for example, image information, examination history information, electrocardiogram information, vital sign information, medication history information, report information, nursing record information, and medical chart information.

[0013] The image information is, for example, information indicating the location of a medical image acquired by photographing a patient, etc. The image information includes, for example, information indicating the location of a medical image file generated by a medical image diagnostic device as a result of an examination.

[0014] The test history information is information that represents the history of test results obtained as a result of specimen tests, bacteriological tests, and the like being performed on a patient, for example.

[0015] The electrocardiogram information is, for example, information relating to an electrocardiogram waveform measured from a patient.

[0016] Vital sign information is, for example, basic information related to the patient's life, including, for example, pulse rate, respiratory rate, oxygen concentration, body temperature, blood pressure, and level of consciousness. Medication history information is, for example, information indicating the history of the amount of medication administered to the patient.

[0017] The report information is, for example, information summarizing the patient's condition and disease after a radiologist interprets medical images such as X-ray images, CT images, MRI images, and ultrasound images in response to an examination request from a medical doctor in a clinical department. The report information includes, for example, radiology report information representing an interpretation report created by the radiologist with reference to a medical image file stored in the PACS 160. Since the report information is generally stored in the PACS 160, the electronic medical record system can display the report information by reading it from the PACS 160.

[0018] The nursing record information is, for example, information entered into an electronic medical record by a nurse, etc. The nursing record information includes nursing records at the time of hospitalization, etc.

[0019] The medical record information is, for example, information entered into an electronic medical record by a doctor or the like. The medical record information includes, for example, consultation records, medical records during hospitalization, the patient's medical history, and medication prescription history. The medical records during hospitalization are recorded, for example, in association with the date and time of entry. The medical record data in this embodiment is assumed to be, for example, the contents of the medical records during hospitalization, but may also include other information. The other information includes, for example, the patient's medical history and medication prescription history, as well as medical images or medical video images associated with the medical records. This also applies to each of the following embodiments.

[0020] Furthermore, the HIS 120 includes a doctor DB 122 that stores information about doctors. The information about doctors includes, for example, information about their working status and profiles (which may be translated into knowledge and careers). The doctor DB 122 may also include doctors from other hospitals or virtual doctors created by AI. Specific examples of information about doctors are described below using FIGS. 2 and 3.

[0021] FIG. 2 is a diagram illustrating a work status table according to the first embodiment. For example, as shown in FIG. 2, the work status table 1221 associates IDs assigned to doctors with work statuses. Work statuses are assigned, for example, "absent," "away from desk," and "present." The work status "absent" indicates, for example, that the doctor is not at work at the hospital. The work status "away from desk" indicates, for example, that the doctor is at work at the hospital but is not present because he or she is consulting a doctor or undergoing surgery. The work status "present" indicates, for example, that the doctor is logged in to a terminal device.

[0022] The work status is updated in real time or at predetermined time intervals based on information from, for example, an attendance management system included in the hospital information system and a monitoring system that monitors the operation of terminal devices. Note that the work status is not limited to the above assignments and may be further subdivided. Specifically, for example, a radiologist may be assigned the work status "emergency," "screening," or "follow-up observation" instead of "present." Furthermore, for a clinician, the work status "present" may be assigned the work status "consulting" or "outside of consultation." By subdividing the work status in this way, for example, even if a doctor is present, the status can be displayed as "absent" or "away from desk" depending on how busy he or she is.

[0023] FIG. 3 is a diagram illustrating a doctor information table according to the first embodiment. For example, as shown in FIG. 3, the doctor information table 1222 associates an ID assigned to a doctor with the doctor's name, the medical department to which the doctor belongs, knowledge, and career. The knowledge includes, for example, information on the specialty field. The specialty field information is, for example, a word-based list of the body parts and case names in which the doctor specializes. The career includes, for example, document information such as the contents of reports entered by the doctor and papers written by the doctor. The doctor information table 1222 may further associate information on patients in the doctor's care, costs incurred when the doctor works, the name of the hospital to which the doctor belongs, and the languages ​​the doctor speaks.

[0024] The work status table 1221 and the doctor information table 1222 may be integrated into one table. The doctor information table 1222 may be divided into multiple tables. That is, the tables included in the doctor DB 122 only need to associate the items included in the work status table 1221 and the doctor information table 1222 with each other using the doctor's ID.

[0025] The terminal device 130 corresponds to, for example, a PC (Personal Computer), a tablet PC, a PDA (Personal Digital Assistant), and a smartphone. The terminal device 130 is placed in each medical department in a medical facility. For example, if an image interpretation doctor uses the terminal device 130 to create an image interpretation report, or if a clinician uses the terminal device 130 to create a medical examination record, etc.

[0026] For example, application software (imaging report software) for creating an interpretation report is installed in the terminal device 130 used to create an interpretation report. Alternatively, the terminal device 130 may create an interpretation report in the form of a web application or the like that does not require the installation of interpretation report software. The terminal device 130 used to create an interpretation report may be called a "imaging report creation device."

[0027] 1 illustrates an example of the connection relationship between one terminal device 130 and other devices and systems in the medical information system 100. However, in the following description, it is assumed that multiple terminal devices 130 are connected to the medical information system 100.

[0028] The RIS 140 is a system that manages examination reservation information related to radiological examination work. For example, the RIS 140 adds various setting information to examination order information input by a medical doctor in an order system included in the HIS 120, accumulates the information, and manages the accumulated information as examination reservation information. The RIS 140 transmits an examination order to the medical image diagnostic device 150 in accordance with the examination reservation information. Furthermore, the RIS 140 transmits examination implementation information generated by the medical image diagnostic device 150 as a result of the examination to the electronic medical record system included in the HIS 120.

[0029] The medical image diagnostic device 150 is a device that performs an examination by taking an image of a patient. The medical image diagnostic device 150 includes, for example, an X-ray computed tomography device, an X-ray diagnostic device, a magnetic resonance imaging device, a nuclear medicine diagnostic device, and an ultrasound diagnostic device. The medical image diagnostic device 150 performs an examination based on examination reservation information transmitted from, for example, the RIS 140. The medical image diagnostic device 150 generates examination implementation information and transmits it to the RIS 140.

[0030] Furthermore, the medical image diagnostic device 150 generates medical image data by performing an examination. The medical image data includes, for example, X-ray CT image data, X-ray image data, MRI image data, nuclear medicine image data, and ultrasound image data. The medical image diagnostic device 150 generates a medical image file by converting the generated medical image data into a format that complies with, for example, the DICOM (Digital Imaging and Communication in Medicine) standard. The medical image file is, for example, a file in a format that complies with the DICOM standard. The medical image diagnostic device 150 transmits the generated medical image file to the PACS 160.

[0031] The PACS 160 is a system that manages various medical image files. The PACS 160 stores, for example, medical image files transmitted from the medical image diagnostic device 150. The PACS 160 may also store report information attached to the medical image files or report information for examinations related to multiple medical image files.

[0032] The DWH, not shown in FIG. 1, is a database system that collectively stores information generated by related medical and nursing care institutions, so-called clinical big data. The DWH is realized, for example, by a general server device. The DWH includes, for example, a processing circuit, a memory, and a communication interface. The processing circuit, the memory, and the communication interface are connected to each other so that they can communicate with each other, for example, via a bus.

[0033] FIG. 4 is a diagram showing an example of the configuration of a report creation assistance device according to the first embodiment. For example, as shown in FIG. 4, the report creation assistance device 110 includes a processing circuit 111, a memory 112, and a communication interface 113. The report creation assistance device 110 prompts the report creator to consult with others about the contents of the report and automatically matches the creator with a consultant who has the necessary knowledge. Below, the memory 112 and the communication interface 113 will be described first.

[0034] The memory 112 corresponds to, for example, a hard disk drive (HDD), a solid state drive (SSD), an integrated circuit storage device, etc. The memory 112 may also be a drive that reads and writes various information from and to a portable storage medium such as a CD-ROM drive, a DVD drive, or a flash memory.

[0035] The memory 112 stores, for example, a report vector data group 112a. The report vector data group 112a is, for example, past report information stored in the PACS 160 that has been subjected to a text classification process using machine learning in advance. Therefore, the report vector data group 112a is assumed to be vectorized reports that have been classified (clustered) into an arbitrary number of clusters. The report vector data group 112a may be created by the processing circuitry 111 or may be acquired from a server device (not shown). The memory 112 may also read and store a work status table 1221 and a doctor information table 1222 from the doctor DB 122 of the HIS 120. The same applies to each of the following embodiments.

[0036] The text classification process described above includes document vectorization and classification. Examples of vectorization include count-based methods (such as Bag of Words (BoW) and Term Frequency-Inverse Document Frequency (TF-IDF)) and inference-based methods (such as word2vec, doc2vec, fastText, and BERT). Examples of classification include unsupervised learning methods (such as Latent Semantic Indexing (LSI), Probabilistic LSI (PLSI), and Latent Dirichelet Allocation (LDA)) and supervised learning methods (such as SVM, RF, NN, LR, and Naive Bayes). When inference-based vectorization is performed, classification is performed using supervised learning.

[0037] The communication interface 113 performs data communication between the HIS 120 and the terminal device 130 connected via a network. Any standard may be used for communication with the HIS 120 and the terminal device 130, and examples of such standards include standards conforming to HL7 (Health Level Seven) and DICOM (Digital Imaging and Communications in Medicine).

[0038] The processing circuit 111 includes a processor (not shown), memory such as a ROM (Read-Only Memory) and RAM, and controls the report creation assist device 110. The processing circuit 111 includes a reception function 111a, an analysis function 111b, an evaluation function 111c, a determination function 111d, a specification function 111e, and a notification function 111f. The various functions performed by the reception function 111a, the analysis function 111b, the evaluation function 111c, the determination function 111d, the specification function 111e, and the notification function 111f are stored in the memory or memory 112 in the form of computer-executable programs. The processing circuit 111 is a processor that realizes the functions corresponding to these various functions by reading and executing the programs corresponding to each program from the memory. In other words, the processing circuit 111, after reading each program, has the multiple functions shown in the processing circuit 111 of FIG. 4.

[0039] In the above description, the various functions described above are realized by a single processing circuit 111, but the various functions may be realized by multiple independent processors executing programs. In other words, the various functions described above may be configured as programs and one processing circuit may execute each program, or specific functions may be implemented in dedicated, independent program execution circuits. Note that instead of storing programs in memory 112 or the like, the programs may be configured to be directly embedded in the processor circuit. In this case, the processor realizes the functions by reading and executing the programs embedded in the circuit.

[0040] The processor in the processing circuit 111 executes various functions by, for example, reading and executing programs stored in the memory 112. Furthermore, the reception function 111a, analysis function 111b, evaluation function 111c, determination function 111d, identification function 111e, and notification function 111f of the processing circuit 111 are examples of a reception unit, an analysis unit, an evaluation unit, a determination unit, an identification unit, and a notification unit, respectively.

[0041] The reception function 111a is a function that receives various notifications from terminal devices and databases. The various notifications include, for example, update information for reports and instructions to start predetermined processing. The processing circuit 111 receives instructions from operators (e.g., report creators and doctors) via the terminal device 130 using the reception function 111a. The processing circuit 111 also receives reports from the terminal device 130. The processing circuit 111 also receives information on multiple doctors who can be consulted (consultable doctor information) from the doctor DB 122. The consultable doctor information is information extracted from the work status table 1221 about doctors who can be consulted.

[0042] The analysis function 111b is a function that analyzes a report. The report is analyzed, for example, by performing vectorization processing on the report. The processing circuitry 111 analyzes the report using the analysis function 111b and generates an analysis result. Specifically, the processing circuitry 111 vectorizes the report and outputs (generates) the vectorized report as the analysis result. The processing circuitry 111 may generate image analysis information by performing image analysis on a medical image included in the report. The image analysis information includes, for example, information on structures (e.g., bones and organs) included in the medical image. The processing performed by the analysis function 111b may also be called a "report analysis process."

[0043] The evaluation function 111c is a function that evaluates a report based on the analysis results. For example, the report is evaluated by calculating the similarity between the report to be evaluated and a reference report. The reference report is, for example, a representative of any cluster in the report vector data group 112a. The processing circuit 111 evaluates the report based on the analysis results using the evaluation function 111c, and generates an evaluation result. Specifically, the processing circuit 111 identifies a cluster in the report vector data group for the vectorized report. Then, the processing circuit 111 calculates the similarity between the vectorized report and the representative of the identified cluster, and outputs (generates) the calculated similarity as an evaluation result. The processing executed by the evaluation function 111c may be called a "report evaluation process."

[0044] If image analysis information is generated when identifying the clusters, the processing circuitry 111 may further identify the clusters based on the image analysis information. In this case, clusters unrelated to the image analysis information are excluded, thereby reducing the dimension related to vectorization. This allows for faster determination processing, which will be described later. Furthermore, in terms of reducing the dimension related to vectorization, the processing circuitry 111 may identify clusters based on, in addition to the image analysis information, DICOM tags, examination purpose, or the like.

[0045] The determination function 111d is a function that determines whether or not it is necessary to seek an opinion on the report. The determination method is, for example, by comparing the similarity with a threshold. The processing circuit 111 determines whether or not it is necessary to seek an opinion on the report based on the evaluation result using the determination function 111d. Specifically, the processing circuit 111 determines whether or not the similarity is equal to or greater than a threshold. If the processing circuit 111 determines that the similarity is equal to or greater than the threshold, it outputs a determination result that consultation is unnecessary, and if the processing circuit 111 determines that the similarity is less than the threshold, it outputs a determination result that consultation is necessary. Note that the determination function 111d may also perform determination processing related to other functions. Furthermore, the processing executed by the determination function 111d may be called a "consultation determination processing."

[0046] The identification function 111e is a function for identifying a consultable party. The identification method is performed, for example, by extracting (searching) a doctor related to the report content from among available doctors as a consultable party. When the identification function 111e determines that an opinion on the report needs to be sought, the processing circuitry 111 identifies a consultable party. Specifically, the processing circuitry 111 extracts doctors related to the report content from the consultable doctor information based on the cluster to which the vectorized report belongs or any element of the cluster to which the vectorized report may belong. At this time, the processing circuitry 111 may prioritize the extracted doctors. The priority is determined based on at least one of, for example, the language of the report, the patient's case, the purpose of the examination, and the examination environment (type of medical image diagnostic device). Furthermore, the priority may be determined based on the doctor's cost. The consultable doctor information may be vectorized using any of the above elements. The processing performed by the identification function 111e may be referred to as a "party identification process."

[0047] The arbitrary element is, for example, an element that is missing when the report content is vectorized. Specifically, the arbitrary element is an element that is weakly related to the elements of the report content among multiple elements of the cluster with high contributions, i.e., an element that is missing from the report content. This will be explained in detail with reference to FIG. 5.

[0048] FIG. 5 is a diagram illustrating missing elements when report contents are vectorized. FIG. 5 illustrates a sample Sc corresponding to the report contents and a cluster CL to which the sample Sc may belong. For ease of explanation, two vectors, V1 and V2, are assumed to be elements that contribute highly to the cluster CL. Furthermore, the sample Sc is close to the cluster CL, and is therefore likely to belong to the cluster CL. The cluster CL is, for example, a collection of samples of past report contents related to a certain finding.

[0049] To identify the missing elements in sample Sc, the processing circuit 111 calculates a displacement vector d from sample Sc to sample Sr representing cluster CL. If displacement vector d contains only the components of vector V2, then sample Sc is missing elements of vector V2. Therefore, the processing circuit 111 can identify an appropriate consultant by extracting consultants associated with the elements of vector V2.

[0050] The notification function 111f is a function that notifies the terminal device of a consultation partner of a consultation acceptance request. The processing circuit 111 notifies the doctors extracted in relation to the report content by the notification function 111f in order of priority. Specifically, the processing circuit 111 notifies the doctor with the highest priority, and if a consultation acceptance notification is not received after a predetermined time has passed, it notifies the doctor with the next highest priority. Thereafter, the processing circuit 111 similarly notifies the doctors in order of priority until a consultation acceptance notification is received. Note that the notification function 111f may include a function that notifies suggestions and information from the report creation assistant device 220 to the terminal device 210 of the report creator.

[0051] Next, the operation of the medical information system according to the first embodiment configured as above will be described with reference to the sequence diagram of Fig. 6. In the following, the creation of an interpretation report will be described as an example.

[0052] FIG. 6 is a sequence diagram illustrating remote consultation pre-processing executed by the medical information system according to the first embodiment. The sequence diagram in FIG. 6 illustrates processing related to the terminal device 210 of the report creator, the report creation assistant device 220, the doctor DB 230, and the destination terminal device 240. The terminal device 210 and the destination terminal device 240 correspond to the terminal device 130 in FIG. 1, the report creation assistant device 220 corresponds to the report creation assistant device 110 in FIGS. 1 and 4, and the doctor DB 230 corresponds to the doctor DB 122 in FIG. 1. The report creation assistant device 220 has a configuration substantially similar to that of the report creation assistant device 110 in FIG. 4. The report creation assistant device 220 and two or more terminal devices (e.g., the terminal device 210 and the destination terminal device 240) may be collectively referred to as a "report creation assistant system." For example, in the report creation assistant system, the terminal device 210 and the destination terminal device 240 communicate with each other via the report creation assistant device 220. The same applies to the subsequent figures.

[0053] Specifically, the sequence diagram in FIG. 6 shows the pre-stage processing of the remote consultation processing described below. The pre-stage processing includes, for example, a consultation proposal determination process and a partner identification process. The consultation proposal determination process analyzes and evaluates the created report and determines whether or not it is necessary to request an opinion on the report. The partner identification process identifies a person who has agreed to be consulted (consultation agreement) from among candidates who can be consulted.

[0054] (Step ST110) The terminal device 210 accepts a report update operation. A report update operation is, for example, an operation in which the report creator clicks the confirm button on the image interpretation report creation screen, or an operation in which temporary saving is executed.

[0055] (Step ST120) After the report update operation is performed, the terminal device 210 notifies the report contents to the report creation assistant device 220. Specifically, the terminal device 210 outputs the radiology report at the time the report update operation is performed to the report creation assistant device 220. The report creation assistant device 220 accepts the radiology report from the terminal device 210.

[0056] (Step ST130) Upon receiving the radiology report, the report creation assistant device 220 executes a consultable doctor search. Specifically, the report creation assistant device 220 requests consultable doctor information from the doctor DB 230 (step ST131). The request for consultable doctor information (consultable doctor information request) includes, for example, an instruction to extract doctors whose duty status is "present." Upon receiving the request for consultable doctor information, the doctor DB 230 notifies the consultable doctor information to the report creation assistant device 220 (step ST132). As a result, the report creation assistant device 220 receives the consultable doctor information. Note that the processing of step ST130 is not limited to being executed upon receiving the radiology report, and may be performed at any timing. Furthermore, the processing of step ST130 may be repeatedly performed at any time interval. This allows the information on consultable doctors to be constantly updated.

[0057] (Step ST140) After the report update operation is performed, the report creation assistant device 220 executes a report analysis process. The report creation assistant device 220 performs vectorization processing on the radiology report and generates a vectorized radiology report as an analysis result.

[0058] (Step ST150) After the report analysis process is executed, the report creation assistant device 220 executes a report evaluation process, which evaluates the radiology report based on the analysis result and generates an evaluation result.

[0059] (Step ST160) After the report evaluation process is executed, the report creation assistant device 220 executes a consultation determination process. Based on the evaluation result, the report creation assistant device 220 determines whether or not it is necessary to request an opinion on the radiology report. In other words, based on the evaluation result, the report creation assistant device 220 determines whether or not a consultation is necessary. In the sequence diagram of FIG. 6, it is assumed that it is determined in step ST160 that a consultation is necessary.

[0060] (Step ST170) After determining that consultation is necessary, the report creation assistant device 220 notifies the terminal device 210 of a consultation proposal. The terminal device 210 accepts the consultation proposal notification.

[0061] (Step ST180) After receiving the consultation proposal notification, the terminal device 210 displays on the display screen a message indicating that the consultation proposal notification has been received. In response to this message, the report creator performs an operation to accept the consultation proposal. By this operation, the terminal device 210 receives an instruction from the report creator to accept the consultation proposal.

[0062] (Step ST190) After receiving an instruction to accept the consultation proposal from the report creator, the terminal device 210 makes a consultation request to the report creation assistant device 220. The report creation assistant device 220 accepts the consultation request as a response to the consultation proposal notification.

[0063] (Step ST200) After accepting the consultation request, the report creation assistant device 220 executes a destination identification process. Specifically, the report creation assistant device 220 requests the destination terminal device 240 to accept the consultation (step ST201). The destination terminal device 240, which has accepted the consultation acceptance request, displays a message indicating that the consultation acceptance request has been accepted at an arbitrary position on the display screen. For example, the destination terminal device 240 displays a pop-up notification having a UI (User Interface) that allows the user to select whether or not to accept the consultation acceptance request. In response to this message, the destination doctor performs an operation to accept the consultation acceptance request. By this operation, the destination terminal device 240 receives an instruction to accept the consultation acceptance request from the destination doctor (step ST202). After receiving the consultation acceptance request from the destination doctor, the destination terminal device 240 notifies the report creation assistant device 220 of the consultation acceptance. The report creation assistant device 220 accepts the consultation acceptance notification as a response to the consultation acceptance request (step ST203).

[0064] (Step ST210) After receiving the consultation acceptance notification, the report creation assistant device 220 notifies the information of the consultant to the terminal device 210. The terminal device 210 receives the consultant information notification. The consultant information includes, for example, the name and medical department.

[0065] (Step ST220) After receiving the consultation partner information notification, the terminal device 210, the report creation assistant device 220, and the destination terminal device 240 start the remote consultation process. The remote consultation process is performed, for example, on a UI that displays a report display screen and a chat screen side by side on a single screen. The remote consultation process will be described later.

[0066] The above has described the remote consultation pre-processing executed by the medical information system according to the first embodiment. Next, the processing executed mainly by the report creation auxiliary device 220 in the remote consultation pre-processing will be described with reference to FIGS. 7 and 8.

[0067] Fig. 7 is a flowchart for explaining the consultation proposal determination process executed by the report creation assistant device according to the first embodiment. The consultation proposal determination process roughly corresponds to the processes from step ST140 to step ST170 in the sequence diagram of Fig. 6. The flowchart of Fig. 7 is started, for example, when a report content notification is received from the terminal device 210, by the processing circuit 111 of the report creation assistant device 220 executing a program related to consultation proposal determination (consultation proposal determination program).

[0068] (Step ST310) When the consultation suggestion determination program is executed, the processing circuitry 111 calls the analysis function 111b. The analysis function 111b causes the processing circuitry 111 to analyze the report and generate an analysis result. Specifically, the processing circuitry 111 performs vectorization processing on the radiology report and generates the vectorized radiology report as the analysis result.

[0069] (Step ST320) After generating the analysis result, the processing circuitry 111 calls the evaluation function 111c. The evaluation function 111c causes the processing circuitry 111 to evaluate the report based on the analysis result and generate an evaluation result. Specifically, the processing circuitry 111 identifies a cluster in the report vector data group 112a for the vectorized radiology report. Then, the processing circuitry 111 calculates the similarity between the vectorized radiology report and a representative of the identified cluster, and generates the calculated similarity as an evaluation result.

[0070] (Step ST330) After generating the evaluation result, the processing circuit 111 calls the determination function 111d. The determination function 111d causes the processing circuit 111 to determine whether the evaluation result satisfies a predetermined condition. Specifically, the processing circuit 111 determines whether the similarity, which is the evaluation result, is equal to or greater than a threshold. If the similarity is equal to or greater than the threshold, i.e., if the evaluation result satisfies the predetermined condition, the processing ends. If the similarity is less than the threshold, i.e., if the evaluation result does not satisfy the predetermined condition, the processing proceeds to step ST340.

[0071] (Step ST340) After determining that the evaluation result does not satisfy the predetermined condition, the processing circuit 111 calls the notification function 111f. The notification function 111f causes the processing circuit 111 to notify the terminal of the report creator of the consultation proposal. After step ST340, the processing of the flowchart in FIG. 7 ends.

[0072] Fig. 8 is a flowchart for explaining the destination identification process executed by the report creation assist device according to the first embodiment. The destination identification process roughly corresponds to the process of step ST200 in the sequence diagram of Fig. 6. The flowchart of Fig. 8 is started, for example, when a consultation request is received from the terminal device 210, by the processing circuit 111 of the report creation assist device 220 executing a program related to destination identification (destination identification program).

[0073] (Step ST410) When the destination identification program is executed, the processing circuitry 111 calls the identification function 111e. The identification function 111e causes the processing circuitry 111 to search for a consultant related to the report content. Specifically, the processing circuitry 111 selects an arbitrary element representing a cluster to which the vectorized radiology report belongs or a cluster to which the vectorized report may belong, and narrows down the consultants related to the selected element from the consultable doctor information.

[0074] (Step ST420) After searching for a consultant, the processing circuitry 111 selects a consultant in descending order of priority. Specifically, the processing circuitry 111 determines the priority of the narrowed down list of consultants based on an arbitrary condition. The arbitrary condition may be, for example, at least one of the language written in the radiology report, the patient's case related to the radiology report, the examination items related to the radiology report, and the examination environment related to the radiology report. The arbitrary condition may also include the cost of the doctor. For example, if a consultation is desired at low cost, the processing circuitry 111 sorts the narrowed down list of consultants in ascending order of cost.

[0075] When the transition is made from step ST450, the processing circuit 111 selects a consultation partner that is one level lower in priority than the consultation partner selected immediately before.

[0076] (Step ST430) After selecting the consultation partner, the processing circuit 111 notifies the consultation partner's terminal device (destination terminal device) of a consultation acceptance request.

[0077] (Step ST440) After notifying the destination terminal device of the consultation acceptance request, the processing circuit 111 determines whether or not a consultation acceptance notice has been received from the destination terminal device. If a consultation acceptance notice has been received from the destination terminal device, the processing proceeds to step ST460. If a consultation acceptance notice has not been received from the destination terminal device, the processing proceeds to step ST450.

[0078] (Step ST450) If a consultation acceptance notification has not been received from the destination terminal device, the processing circuit 111 determines whether a predetermined time has elapsed. The predetermined time corresponds to the time from when a pop-up notification is displayed on the destination terminal device until an answer as to whether or not to accept the consultation is selected, and may be set arbitrarily. If the predetermined time has elapsed and a consultation acceptance notification has not been received, the process returns to step ST420. If the predetermined time has not elapsed, the process returns to step ST440 and waits for receipt of a consultation acceptance notification. The process of step ST450 may be included in the process of step ST440.

[0079] (Step ST460) After receiving the consultation acceptance notification from the destination terminal device, the processing circuit 111 notifies the consultation partner information to the terminal device of the report creator. After step ST460, the processing of the flowchart in FIG.

[0080] As described above, the report creation assistance device of the first embodiment analyzes a report, generates analysis results, evaluates the report based on the analysis results, generates evaluation results, and determines based on the evaluation results whether or not it is necessary to seek opinions on the report.If it is determined that it is necessary to seek opinions on the report, it identifies a party that can be consulted and notifies the party.

[0081] Therefore, the report creation assist device according to the first embodiment can improve the quality of medical reports by automatically providing a person who can be consulted depending on the contents of the medical report.

[0082] In the first embodiment, the report creator is not conscious of the suggestion as to whether or not to execute the remote consultation process, but this is not limited to this. For example, the report creation assistance device may receive an instruction to request the execution of the remote consultation process. In this case, the report creator issues an instruction to the report creation assistance device to request the execution of the remote consultation process in order to execute the remote consultation process independently.

[0083] (Second embodiment) In the first embodiment, a determination as to whether or not a consultation proposal is necessary is explained based on the report contents, whereas in the second embodiment, a determination as to how often a consultation proposal is made (proposal frequency) is explained based on information about the report creator.

[0084] 9 is a diagram showing an example of the configuration of a report creation assistant device according to the second embodiment. For example, as shown in FIG. 9, a report creation assistant device 110A includes a processing circuit 111A, a memory 112, and a communication interface 113.

[0085] The processing circuit 111A further has a frequency determination function 111g. The frequency determination function 111g of the processing circuit 111A is an example of frequency determination processing. Note that other functions are substantially the same as those of the processing circuit 111 in FIG. 4, and therefore description thereof will be omitted.

[0086] The frequency determination function 111g is a function that determines the frequency at which a consultation proposal notification is sent from the report creation assistant device to the terminal device of the report creator. When the processing circuit 111A receives the ID (user ID) of the report creator using the frequency determination function 111g, it requests user information for the received user ID from the doctor DB. Then, the processing circuit 111A determines the frequency at which to send a consultation proposal notification based on the user information. Note that the processing circuit 111A may also decide not to send a consultation proposal notification.

[0087] Fig. 10 is a sequence diagram for explaining the user confirmation process executed by the medical information system according to the second embodiment. The sequence diagram of Fig. 10 shows the process related to the terminal device 210 of the report creator, the report creation assistant device 220A, and the doctor DB 230. The report creation assistant device 220A corresponds to the report creation assistant device 110A in Fig. 9 and has substantially the same configuration as the report creation assistant device 110A. In addition, the destination terminal device 240 is omitted from the sequence diagram of Fig. 10.

[0088] Specifically, the sequence diagram of FIG. 10 determines whether or not to perform the consultation suggestion determination process in the sequence diagram of FIG. 6, or the frequency of sending the consultation suggestion notification in step ST170 in the sequence diagram of FIG. 6, depending on the career history of the report creator, etc.

[0089] (Step ST510) The terminal device 210 accepts a log-in operation by the report creator. The log-in operation is, for example, an operation in which the report creator logs in to the radiology report software.

[0090] (Step ST520) After the login operation is performed, the terminal device 210 notifies the report creation assistant device 220 A of the user ID of the report creator. The report creation assistant device 220 A accepts the user ID from the terminal device 210 .

[0091] (Step ST530) Upon receiving the user ID, the report creation assistant device 220 executes a report creator search. Specifically, the report creation assistant device 220A requests the doctor DB 230 for user information on the report creator (step ST531). The user information includes, for example, career information. Upon receiving the request for user information, the doctor DB 230 notifies the report creation assistant device 220A of the user information (step ST532). As a result, the report creation assistant device 220A receives the user information.

[0092] (Step ST540) After receiving the user information, the report creation assist device 220A executes a proposal frequency determination process. The proposal frequency determination process determines whether to execute the remote consultation pre-processing of FIG. 6 or determines the frequency of issuing the consultation proposal notification of step ST170, depending on the user information. Both of these determination processes may be executed, or whether to execute the process of determining the frequency of issuing the consultation proposal notification may be determined depending on the result of whether to execute the remote consultation pre-processing.

[0093] For example, when determining whether to perform remote consultation pre-processing, the report creation assistant device 220A determines whether to perform remote consultation pre-processing based on career information included in the user information. Specifically, for example, if the career is a predetermined number of years or more, the report creation assistant device 220A determines not to perform remote consultation pre-processing. This allows the report creation assistant device 220A to exclude users who do not require remote consultation, such as experienced radiologists.

[0094] Further, for example, when determining the frequency of issuing a consultation proposal notification in step ST170, the report creation assistant device 220A determines the frequency of issuing a consultation proposal notification in step ST170 based on the carrier included in the user information. The consultation proposal notification in step ST170 is issued in accordance with the determination process in step ST330 of the flowchart in FIG. 7. Therefore, for example, the report creation assistant device 220A determines the frequency of issuing a consultation proposal notification by setting a predetermined condition (here, a threshold value related to similarity) for the determination process in accordance with the carrier. In this way, the report creation assistant device 220A can increase or decrease the frequency of issuing a consultation proposal notification by, for example, setting a threshold value related to similarity in accordance with the carrier.

[0095] As described above, the report creation assist device according to the second embodiment can determine the frequency of consultation suggestions depending on the career of the report creator.

[0096] (Third embodiment) In the first and second embodiments, the processing before the remote consultation processing has been described, whereas in the third embodiment, the remote consultation processing will be described.

[0097] 11 is a diagram showing an example of the configuration of a report creation assistant device according to the third embodiment. For example, as shown in FIG. 11, a report creation assistant device 110B includes a processing circuit 111B, a memory 112, and a communication interface 113.

[0098] The processing circuit 111B further has a comment analysis function 111h and a comment reflection function 111i. The comment analysis function 111h and the comment reflection function 111i of the processing circuit 111B are examples of a comment analysis unit and a comment reflection unit, respectively. Note that other functions are substantially the same as those of the processing circuit 111A in FIG. 9, and therefore description thereof will be omitted.

[0099] The comment analysis function 111h is a function that analyzes the content of comments and classifies the comments. The processing circuit 111B performs vectorization processing on the comment content using the comment analysis function 111h and classifies the comment content according to a past comment vector data group. It is assumed that the comment vector data group is stored in the memory 112.

[0100] The indication reflection function 111i is a function that creates indication reflection information based on the analyzed comment content. The processing circuit 111B creates the indication reflection information based on the analyzed comment content using the indication reflection function 111i. Specifically, the processing circuit 111B generates a sentence corresponding to the comment content using, for example, a question-and-answer technique for document search or a technique for a dialogue system, and creates the generated sentence as indication reflection information.

[0101] The processing circuitry 111B may generate the indication reflection information based on information other than the comment content. For example, the processing circuitry 111B may generate the indication reflection information based on the patient's case, the examination site, the purpose of the examination, the contents of the radiology report, etc.

[0102] Fig. 12 is a sequence diagram for explaining the remote consultation process executed by the medical information system according to the third embodiment. The sequence diagram of Fig. 12 shows the process related to the terminal device 210 of the report creator, the report creation assistant device 220B, and the destination terminal device 240. The report creation assistant device 220B corresponds to the report creation assistant device 110B in Fig. 11 and has substantially the same configuration as the report creation assistant device 110B. The doctor DB 230 is omitted from the sequence diagram of Fig. 12.

[0103] Specifically, the sequence diagram of Fig. 12 shows the remote consultation process in step ST220 in the sequence diagram of Fig. 6. Therefore, in Fig. 12, the report creator and the person to be consulted have already been matched.

[0104] (Step ST610) When the remote consultation process starts, the report creation assistant device 220B instructs the terminal device 210 and the destination terminal device 240 to display a support UI. The support UI includes, for example, at least a display screen for the radiology report (radiology report display screen) and a chat screen for consultation. The terminal device 210 and the destination terminal device 240 accept the instruction to display the support UI. Note that the support UI may be different between the terminal device 210 and the destination terminal device 240.

[0105] (Step ST620) In response to the support UI display instruction, the terminal device 210 displays the support UI. The support UI of the terminal device 210 includes, for example, at least a display screen for the radiology report, a UI for reflecting comments, and a chat screen for consultation.

[0106] 13 is a diagram illustrating a support UI according to the third embodiment. The support UI 300 in FIG.

[0107] The image interpretation report display screen 310 includes an area for displaying patient information, an area for displaying findings and diagnoses, and an area for displaying medical images. The image interpretation report display screen 310 may directly display the image interpretation report creation screen of the image interpretation report software. In other words, the support UI of the terminal device 210 may allow editing of the image interpretation report on the image interpretation report display screen 310.

[0108] The viewer 320 displays medical images stored in the PACS 160. In the viewer 320, for example, medical images relating to the image interpretation report displayed on the image interpretation report display screen 310 are displayed.

[0109] The comment reflection UI 330 is a UI that allows the user to select text to be added to the radiology report. Specifically, one or more items are displayed in the comment reflection UI 330, and the report creator can reflect the contents of the selected item in the radiology report by selecting one of the items. The items include, for example, text to be added to the report. The one or more items displayed in the comment reflection UI 330 are created by the report creation assist device 220B analyzing comments on the chat screen 340. Hereinafter, information about the one or more items displayed in the comment reflection UI 330 will be referred to as "comment reflection information."

[0110] The real-time exchange between the report creator and the person seeking advice is displayed in the form of comments on the chat screen 340. Comments entered on the chat screen 340 are analyzed by the report creation assist device 220B, and an icon corresponding to the comment content on the chat screen 340 is added.

[0111] (Step ST630) In response to the support UI display instruction, the destination terminal device 240 displays the support UI. The support UI of the destination terminal device 240 includes, for example, at least a display screen for the radiology report, a comment reflection UI, and a chat screen for consultation. The support UI of the destination terminal device 240 is substantially the same as the support UI 300 of FIG. 13, and therefore a description thereof will be omitted. Note that the support UI of the destination terminal device 240 may be configured such that only the radiology report is displayed on the radiology report display screen 310 (i.e., it cannot be edited), and the items displayed in the comment reflection UI 330 cannot be selected.

[0112] (Step ST640) After the support UI is displayed, the destination terminal device 240 accepts a comment input operation. The comment input operation is, for example, an operation in which a consultation partner inputs a comment on the chat screen.

[0113] (Step ST650) After the comment input operation is performed, the destination terminal device 240 notifies the report creation assistant device 220B and the terminal device 210 of the input comment content. The report creation assistant device 220B and the terminal device 210 receive the comment content notification. As a result, the comment on the chat screen is updated on the terminal device 210. Note that the comment on the chat screen may be updated after the comment analysis process described below.

[0114] (Step ST660) Upon receiving the comment content notification, the report creation assist device 220B executes comment analysis processing. The comment analysis processing analyzes the comment content, classifies the comments, and generates comment classification information. The comment analysis processing may also generate indication reflection information based on the classified comments. In the sequence diagram of FIG. 12, it is assumed that the comment classification information and indication reflection information are generated in step ST660.

[0115] (Step ST670) After the comment analysis process is executed, the report creation assist device 220B notifies the terminal device 210 and the destination terminal device 240 of the comment classification information. The terminal device 210 and the destination terminal device 240 each receive the comment classification information. After receiving the comment classification information, an icon according to the classification result is added to the comment on the chat screen of each terminal device.

[0116] (Step ST680) After the comment analysis process is executed, the report creation assistant device 220B notifies the terminal device 210 and the destination terminal device 240 of the pointing out reflection information. The terminal device 210 and the destination terminal device 240 each receive the pointing out reflection information. After receiving the pointing out reflection information, one or more items are displayed in the pointing out reflection UI of each terminal device. Note that step ST680 may be executed simultaneously with step ST670.

[0117] (Step ST690) After the comment classification information and the indication reflection information are notified, the terminal device 210 updates the screen display and displays the indication reflection UI. The screen display update is, for example, adding an icon to the comment on the chat screen.

[0118] (Step ST700) After the comment classification information and the indication reflection information are notified, the destination terminal device 240 updates the screen display in a manner similar to that of the terminal device 210.

[0119] (Step ST710) After the indication reflection UI is displayed, the terminal device 210 accepts an indication reflection selection operation. The indication reflection selection operation is, for example, an operation in which the report creator selects one of one or more items displayed in the indication reflection UI. By this operation, the selected item is added to the radiology report display screen of the terminal device 210, and the radiology report is updated.

[0120] (Step ST720) After the indication reflection selection operation is performed, the terminal device 210 notifies the selected content to the report creation assistant device 220B and the destination terminal device 240. The report creation assistant device 220B and the destination terminal device 240 each receive the selected content notification. After receiving the selected content notification, the item selected by the report creator is added to the radiology report display screen of the destination terminal device 240.

[0121] The above has described the remote consultation process executed by the medical information system according to the third embodiment. Next, the process executed mainly by the report generation auxiliary device 220B in the remote consultation process will be described with reference to FIG.

[0122] Fig. 14 is a flowchart for explaining comment analysis processing executed by the report creation assist device according to the third embodiment. The comment analysis processing roughly corresponds to the processing from step ST660 to step ST680 in the sequence diagram of Fig. 12. The flowchart of Fig. 14 is started, for example, when a comment content notification is received from the destination terminal device 240, by the processing circuit 111B of the report creation assist device 220B executing a program related to comment analysis (comment analysis program).

[0123] (Step ST810) When the comment analysis program is executed, the processing circuit 111B calls the comment analysis function 111h. The comment analysis function 111h causes the processing circuit 111B to analyze the comment content and classify the comment. Specifically, the processing circuit 111B performs vectorization processing on the comment content and classifies the comment content according to a group of past comment vector data.

[0124] Comment content can be classified into, for example, confirmation comments, pointing out comments, and normal comments. Confirmation comments are, for example, comments that provide advice regarding the radiology report, and are comments that do not need to be added to the radiology report. Pointing out comments are, for example, comments that point out omissions in the information, and are comments that need to be added to the contents of the radiology report. Normal comments are, for example, comments other than confirmation comments and pointing out comments. Hereinafter, information indicating the classification of these comment contents will be referred to as "comment classification information."

[0125] (Step ST820) After analyzing the comment content, the processing circuit 111B determines whether or not to assign an icon to the classified comment content. This determination is made, for example, according to the classification of the comment content included in the comment classification information. If the comment content is a normal comment, the processing circuit 111B determines not to assign an icon, and if the comment content is other than a normal comment, the processing circuit 111B determines to assign an icon. If it is determined to assign an icon, the processing proceeds to step ST830. If it is determined not to assign an icon, the processing ends.

[0126] (Step ST830) After it is determined that an icon should be assigned, the processing circuit 111B determines whether or not to create pointing out reflection information according to the classified comment content. The determination here is made, for example, according to the classification of the comment content. If the comment content is a confirmation comment, the processing circuit 111B determines not to create pointing out reflection information, and if the comment content is a pointing out comment, the processing circuit 111B determines to create pointing out reflection information. If it is determined to create pointing out reflection information, the processing proceeds to step ST840. If it is determined not to create pointing out reflection information, the processing proceeds to step ST850.

[0127] (Step ST840) After determining that the comment reflection information should be created, the processing circuit 111B calls the comment reflection function 111i. The processing circuit 111B creates the comment reflection information based on the analyzed comment content using the comment reflection function 111i. Then, the processing circuit 111B notifies the terminal device of the comment classification information and the comment reflection information.

[0128] (Step ST850) After it is determined that the indication reflection information is not to be created, the processing circuit 111B notifies the terminal device of the comment classification information. After step ST850, the processing of the flowchart in FIG.

[0129] Next, the display of the support UI related to the flowchart of Fig. 14 will be explained using Fig. 15 to Fig. 18. The support UI explained below is assumed to be displayed on the terminal device 130 of the report creator. The support UI displayed on the destination terminal device 240 is substantially the same as the support UI displayed on the terminal device 130, except that the orientation of the comments on the chat screen is different.

[0130] 15 shows the support UI 300 displayed on the terminal device 130 immediately after the destination terminal device 240 inputs a comment 341, "Given the patient's age, I think a physiological test is also necessary for ***.", to the chat screen 340. The comment 341 is in response to the character string 311, "There is also the possibility of ***," written in the findings / diagnosis section of the radiology report display screen 310. At this time, the report creation assistant device 220B analyzes the comment 341 and classifies it as a comment that has been pointed out. Furthermore, the report creation assistant device 220B generates comment reflection information based on the analyzed comment 341. The report creation assistant device 220B notifies the terminal device 130 of the comment classification information and the comment reflection information.

[0131] 16 shows the support UI 300 after the comment classification information and the indication reflection information have been notified. After receiving the comment classification information and the indication reflection information, the terminal device 130 assigns an icon 342 relating to the indication comment to the comment 341. Furthermore, the terminal device 130 displays the indication reflection information in the indication reflection UI 330. Here, for example, it is assumed that the item 331 "Physiological test A is recommended" and the item 332 "Physiological test B is recommended" are displayed as the indication reflection information. Note that the report creation assist device 220B may include an instruction to apply markup, such as coloring, to the comment classification information in order to highlight the description of the comment 341 related to the indication reflection information.

[0132] FIG. 17 shows the support UI 300 after the report creator selects item 331 and updates the report description. After selecting item 331, the terminal device 130 automatically adds a character string 312, "Physiological test A is recommended.", to the findings and diagnosis on the radiology report display screen 310. At this time, the terminal device 130 may apply markup similar to that described above to the character string 312. After adding the character string 312 to the radiology report display screen 310, the report creator updates the report description by clicking the Confirm button. This causes the terminal device 130 to notify the report creation assistant device 220B of the updated report content. The report creation assistant device 220B then instructs each terminal device to display a comment 343, "Report description has been updated," on the chat screen 340, indicating that the report description has been updated during the remote consultation process.

[0133] 18 shows the support UI 300 after the report description has been updated. Chat screen 340 in the support UI in FIG. 18 displays comment 344 "It's been updated" by the report creator and comment 345 "There's no problem" by the person being consulted after the report description has been updated. Note that comment 346 "It's quicker to check from the back with MIP" immediately before comment 341, which has been classified as a suggestion comment, has been classified as a confirmation comment by the report creation support device 220B, and an icon 347 related to the confirmation comment has been added.

[0134] As described above, the report creation assistant device 220B creates the indication reflection information based on the chat comment, but this is not limited to this. The report creation assistant device 220B may create similar indication reflection information even if the chat comment does not contain a specific description (in the above example, "Physiological examination is also necessary"). For example, the report creation assistant device 220B may create indication reflection information that infers the content of the comment based on the content of the radiology report. At this time, the report creation assistant device 220B may create indication reflection information further based on the patient's case, the examination site, and the purpose of the examination.

[0135] As described above, the report creation assist device according to the third embodiment displays a support UI, allowing the terminal device of the report creator and the terminal device of the other party to communicate with each other, thereby enabling smooth communication between the report creator and the other party.

[0136] This report creation support device can also classify comments entered by the other party's terminal, add an icon to the chat screen of the support UI according to the content of the comment, and generate text to add to the report. This allows the report creator to point out and correct any omissions in the report content, and also allows the consultation results to be reflected in the medical report more concisely and without misunderstanding, thereby improving the quality of the medical report.

[0137] (Other embodiments) In the above embodiments, the creation of radiology reports has been described as a specific example, but this is not limiting. For example, the report creation assistance device can be applied to any document written by a medical professional, regardless of whether the medical professional is a doctor, such as a medical examination record, a medical care record, or a nursing record. For example, if the report creation assistance device is used to create a nursing record, the HIS 120 includes a nurse DB that stores information about nurses. Note that radiology reports, medical examination records, medical care records, and nursing records may also be called "medical reports" because they are reports used for medical purposes.

[0138] According to at least one of the embodiments described above, it is possible to automatically provide a person who can be consulted depending on the contents of a medical report.

[0139] Although several embodiments have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, modifications, and combinations of embodiments can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, as well as within the scope of the invention and its equivalents as defined in the claims. [Explanation of symbols]

[0140] 100 Medical Information Systems 110, 110A, 110B, 220, 220A, 220B Report creation aid 111, 111A, 111B processing circuit 111a reception function 111b Analysis function 111c evaluation function 111d Judgment function 111e Specific functions 111f Notification function 111g Frequency determination function 111h Comment analysis function 111i Pointing out function 112 memory 112a Report Vector Data Group 113 Communication Interface 120 Hospital Information Systems (HIS) 121 Electronic Medical Record DB 122 Doctors DB 1221 Work Status Table 1222 Doctor Information Table 130,210 Terminal Equipment 140 Radiology Information Systems (RIS) 150 Medical imaging diagnostic equipment 160 Medical Image Management Systems (Picture Archiving and Communication Systems: PACS) 240 Remote Terminal Device 310 Image interpretation report display screen 320 Viewer 340 Chat Screen

Claims

1. an analysis unit that analyzes the report and generates an analysis result; an evaluation unit that evaluates the report based on the analysis result and generates an evaluation result; a determination unit that determines whether or not it is necessary to request an opinion on the report based on the evaluation result and a threshold value according to the career of the report creator; an identification unit that identifies a party available for consultation when it is determined that an opinion on the report needs to be sought; a notification unit that notifies the other party; A report creation assistant device comprising:

2. the analysis unit performs vectorization processing on the report and generates a vectorized report as the analysis result. The report creation assisting device according to claim 1 .

3. the evaluation unit generates a similarity between the vectorized report and a reference report as the evaluation result. The report creation assisting device according to claim 2 .

4. The evaluation unit identifies a cluster in a report vector data group obtained by performing text classification processing on past reports for the vectorized report, and generates the similarity between the vectorized report and the reference report that is a representative of the cluster as the evaluation result. The report creation assisting device according to claim 3 .

5. A reception desk that accepts information on multiple doctors available for consultation Further comprising: the identifying unit selects an arbitrary element representative of the cluster, and extracts a plurality of contact points related to the report from the plurality of doctors available for consultation based on the arbitrary element; The report creation assisting device according to claim 4.

6. the specifying unit specifies the destination from the plurality of destinations based on a priority order; The report creation assisting device according to claim 5 .

7. The priority is determined based on at least one of a language described in the report, a patient case related to the report, a test item related to the report, and a test environment related to the report. The report creation assisting device according to claim 6.

8. The prioritization is further determined based on physician costs. The report creation assisting device according to claim 7.

9. a terminal device for creating the report; a destination terminal device of the destination; The report creation assisting device according to any one of claims 1 to 8, Equipped with The terminal device and the destination terminal device communicate with each other via the report creation assistant device.

10. the terminal device and the destination terminal device display a support UI (User Interface) including a display screen of the report and a chat screen; The report creation assistance system according to claim 9.

11. the report creation assistant device classifies the comments on the chat screen input by the destination terminal device; The report creation assistance system according to claim 10.

12. the report creation assistant device assigns an icon to the classified comment; The report creation assistance system according to claim 11.

13. the report creation assistant device generates one or more items to be added to the report based on the comments input by the destination terminal device; The support UI further includes a prompt display UI for displaying the one or more items; The report creation assistance system according to any one of claims 10 to 12.

14. the terminal device updates the report when any one of the one or more items displayed on the indication display UI is selected; The report creation assistance system according to claim 13.

Images