Diagnostic support device, diagnostic support method, and program

Abstract

A diagnostic support device is provided that creates a draft of a medical certificate and transmits the draft to an appropriate department. [Solution] In the diagnostic support device, a prompt creation means acquires information at the time of diagnosis and creates a prompt including the information at the time of diagnosis. An estimation means inputs the prompt into a diagnostic support AI and estimates a case candidate. A draft creation means creates a draft of a medical certificate based on the case candidate. A selection means selects a destination to send the draft of the medical certificate based on the case candidate. The diagnostic support device can support doctors' decision-making in the medical field.

Description

[Technical Field]

[0001] The present disclosure relates to techniques for assisting diagnosis. [Background technology]

[0002] Systems that use AI (Artificial Intelligence) to support diagnosis are known. For example, Patent Document 1 discloses a method for generating diagnostic information from patient interview information using a generation AI. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 7450310 Summary of the Invention [Problem to be solved by the invention]

[0004] The diagnostic information in Patent Document 1 is used as an aid to doctors' diagnoses. However, with the method in Patent Document 1, the diagnostic information is not always used effectively, for example, when the diagnostic information includes information outside the doctor's area of ​​expertise.

[0005] One object of the present disclosure is to provide a diagnosis support device that creates a draft of a medical report and sends the draft of the medical report to an appropriate department. [Means for solving the problem]

[0006] According to one aspect of the present disclosure, a diagnosis support device includes: a prompt generating means for acquiring diagnostic information and generating a prompt including the diagnostic information; an estimation means for inputting the prompt into a diagnostic support AI and estimating a case candidate; a draft preparation means for preparing a draft of a medical certificate based on the candidate case; a selection means for selecting a destination to which the draft of the medical certificate is to be sent based on the candidate case; Equipped with.

[0007] In another aspect of the present disclosure, a diagnostic assistance method includes: A computer-implemented diagnostic support method, comprising: Obtain diagnostic information, create a prompt that includes the diagnostic information, The prompt is input to the diagnostic support AI, and a case candidate is estimated. Prepare a draft medical certificate based on the candidate case; A destination for sending the draft medical certificate is selected based on the candidate case.

[0008] In yet another aspect of the disclosure, a program includes: Obtain diagnostic information, create a prompt that includes the diagnostic information, The prompt is input to the diagnostic support AI, and a case candidate is estimated. Prepare a draft medical certificate based on the candidate case; The computer is caused to execute a process of selecting a destination to which the draft of the medical certificate is to be sent based on the candidate case. [Effects of the Invention]

[0009] According to the present disclosure, it is possible to provide a diagnostic support device that creates a draft of a medical report and transmits the draft to an appropriate department. [Brief explanation of the drawings]

[0010] [Figure 1] 1 shows the overall configuration of a diagnosis support system to which a diagnosis support device according to the present disclosure is applied. [Figure 2] FIG. 2 is a block diagram illustrating a hardware configuration of a diagnosis support device according to the present disclosure. [Figure 3] 1 is a block diagram showing a functional configuration of a diagnosis support device according to the present disclosure. [Figure 4] This is an example of a draft medical certificate. [Figure 5] 10 is a flowchart of processing by the diagnosis support device according to the present disclosure. [Figure 6]FIG. 10 is a block diagram showing a functional configuration of another diagnosis support device according to the present disclosure. [Figure 7] 10 is a flowchart of processing by another diagnosis support device according to the present disclosure. [Figure 8] FIG. 10 is a block diagram showing a functional configuration of another diagnosis support device according to the present disclosure. [Figure 9] 10 is a flowchart of processing by another diagnosis support device according to the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, preferred embodiments of the present disclosure will be described with reference to the drawings.

[0012] First Embodiment A medical institution may include multiple medical departments or divisions (hereinafter referred to as "medical divisions"). During a patient's examination, a case may be considered minor for the medical division where the patient was examined, but not minor for other medical divisions, and information about the case may be stored in the medical institution's Hospital Information System (HIS). However, currently, it takes time to associate such information.

[0013] Therefore, as will be described in detail later, the diagnosis support device of this embodiment uses a generation AI to infer cases from information obtained when examining a patient. The diagnosis support device then prepares a draft medical report for the inferred case and sends the draft medical report to the appropriate medical department. In this way, the diagnosis support device of this embodiment can quickly infer cases across medical departments and prepare a draft medical report.

[0014] [Overall configuration] 1 shows the overall configuration of a diagnosis support system to which a diagnosis support device according to the present disclosure is applied. The diagnosis support system 1 includes a terminal device 5, a diagnosis support device 10, and terminal devices 20 in multiple medical departments. Note that when distinguishing between individual medical departments, the terminal devices 20 in the medical departments are referred to with a subscript, and when not distinguishing between them, they are simply referred to as "terminal devices 20."

[0015] The terminal device 5 is operated by a doctor or the like and transmits information obtained when a patient is examined to the diagnosis support device 10. The information obtained when a patient is examined includes, for example, conversations between the doctor and the patient and medical images taken during the examination. The information obtained when a patient is examined is hereinafter also referred to as "diagnosis information." The terminal device 5 is, for example, configured as a personal computer and communicates with the diagnosis support device 10 via a network such as the Internet.

[0016] The diagnostic support device 10 estimates the patient's condition from the information at the time of diagnosis and prepares a draft of the medical certificate. The diagnostic support device 10 then transmits the draft of the medical certificate to the appropriate medical department. The diagnostic support device 10 is configured, for example, by a server device, and communicates with the terminal device 5 and the terminal device 20 via a network such as the Internet.

[0017] The terminal device 20 in the medical department is operated by a medical department staff member or the like, and is used to view the draft of the medical certificate received from the diagnosis support device 10. The terminal device 20 is configured, for example, by a personal computer or the like, and communicates with the diagnosis support device 10 via a network such as the Internet.

[0018] [Hardware configuration] 2 is a block diagram showing the hardware configuration of a diagnosis support device 10 according to the first embodiment. As shown in the figure, the diagnosis support device 10 includes an interface (I / F) 11, a processor 12, a memory 13, a recording medium 14, and a database (DB) 15.

[0019] The I / F 11 communicates with the terminal device 5 and the terminal device 20 via a network such as the Internet.

[0020] The processor 12 is a computer such as a CPU (Central Processing Unit) and controls the entire diagnostic support device 10 by executing a prepared program. The processor 12 may be a GPU (Graphics Processing Unit), a DSP (Digital Signal Processor), an MPU (Micro Processing Unit), an FPU (Floating Point number Processing Unit), a PPU (Physics Processing Unit), a TPU (Tensor Processing Unit), a quantum processor, a microcontroller, or a combination thereof. The processor 12 executes a draft creation process, which will be described later.

[0021] The memory 13 is configured by a ROM (Read Only Memory), a RAM (Random Access Memory), etc. The memory 13 is also used as a working memory while the processor 12 is executing various processes.

[0022] The recording medium 14 is a non-volatile, non-transitory recording medium such as a disk-shaped recording medium or a semiconductor memory, and is configured to be detachable from the diagnosis support device 10. The recording medium 14 records various programs to be executed by the processor 12. When the diagnosis support device 10 executes various processes, the programs recorded on the recording medium 14 are loaded into the memory 13 and executed by the processor 12.

[0023] The DB 15 stores, for example, a table describing rules that associate cases with medical departments, which will be described later.

[0024] In addition to the above, the diagnosis support device 10 may also include a display device such as a liquid crystal display and input devices such as a keyboard and a mouse. These display devices and input devices are used by, for example, an administrator of the diagnosis support device 10 to perform necessary management.

[0025] [Function Configuration] 3 is a block diagram showing the functional configuration of the diagnosis support device 10 according to the first embodiment. The diagnosis support device 10 functionally includes a diagnostic information input unit 101, a case candidate estimation unit 102, a draft creation unit 103, and a draft sending unit 104.

[0026] Diagnosis information is input to the diagnosis support device 10 from the terminal device 5 via the I / F 11. The diagnosis information includes audio data of conversations between a doctor and a patient, medical image data captured during a medical examination, etc. The diagnosis information is input to a diagnosis information input unit 101.

[0027] The diagnostic information input unit 101 creates a prompt to be input to the generation AI and outputs it to the case candidate estimation unit 102. For example, the diagnostic information input unit 101 attaches diagnostic information to the following prompt and outputs it to the case candidate estimation unit 102. Note that the diagnostic information input unit 101 may attach the voice data directly to the prompt, or may convert the voice data into text data using an existing voice recognition model and insert the converted text data into the prompt. (Example prompt) The attached data is information at the time of diagnosis. Please output possible cases from the attached data. Also, please tell us the basis for selecting those cases.

[0028] The case candidate estimation unit 102 estimates case candidates (hereinafter also referred to as "case candidates") using a generation AI prepared in advance. The generation AI used by the case candidate estimation unit 102 is hereinafter referred to as "diagnostic support AI." The diagnostic support AI takes a prompt as input and outputs a case and the basis for selecting the case. In addition to the case and the basis for selecting the case, the diagnostic support AI may also output the likelihood of the case (hereinafter also referred to as "case accuracy"). The case accuracy may be expressed in words such as most likely or second most likely, or may be expressed numerically. The case candidate estimation unit 102 outputs the output of the diagnostic support AI to the draft creation unit 103 as a case candidate.

[0029] The diagnostic assistance AI is created by additionally training a general-purpose generative AI such as ChatGPT by OpenAI or Gemini by Google. The additional training includes fine tuning and transfer learning. For example, the case candidate estimation unit 102 performs additional training on the general-purpose generative AI using a dataset consisting of pairs of information at the time of diagnosis and answers to the information at the time of diagnosis. This creates a diagnostic assistance AI specialized for the medical field. The answers to the information at the time of diagnosis are case information (hereinafter also referred to as "case information"), and case information stored in the HIS can be used.

[0030] The case candidate estimation unit 102 may also use Retrieval Augmented Generation (RAG) instead of diagnostic support AI. RAG is a technology that combines general-purpose generation AI with external information sources to generate highly reliable answers. The case candidate estimation unit 102 may use databases owned by medical institutions as external information sources.

[0031] The draft creation unit 103 creates a draft of the medical report based on the output of the diagnostic support AI. Then, the draft creation unit 103 outputs the draft of the medical report to the draft sending unit 104. Specifically, the draft creation unit 103 creates a draft of the medical report by arranging the text output by the diagnostic support AI. For example, if the output of the diagnostic support AI is in bullet points, the draft creation unit 103 creates a draft of the medical report by converting the bullet points into text. Furthermore, if the diagnostic support AI outputs the certainty of a case as a score (numerical value), the draft creation unit 103 does not use the numerical value as it is, but replaces it with words such as "highly likely" or "lowly likely" to create a draft of the medical report.

[0032] When the diagnostic support AI outputs multiple cases, the draft creation unit 103 can create a draft of the medical report for each case. At this time, the draft creation unit 103 may set a priority for the draft of the medical report according to the accuracy of the case. The draft creation unit 103 sets a higher priority for the draft of the medical report as the accuracy of the case increases.

[0033] An example of a draft medical certificate is shown in Figure 4. In the example of Figure 4, cases are listed in descending order of probability, along with the reasons for selecting the cases.

[0034] Returning to FIG. 3 , the draft sending unit 104 refers to the cases included in the draft medical report and selects the medical department to which the draft medical report is to be sent. For example, the draft sending unit 104 can select the medical department from the draft medical report based on a rule. The rule is, for example, a rule that associates cases with medical departments. The draft sending unit 104 may also select the medical department using a machine learning model. This machine learning model is, for example, a trained machine learning model that is trained to input the draft medical report and output the medical department.

[0035] The draft sending unit 104 sends the draft of the medical report to the terminal device 20 of the selected medical department. At this time, the draft sending unit 104 may notify the doctor who examined the patient (hereinafter also referred to as the "examining doctor") of the draft of the medical report and the medical department to which the draft will be sent, and obtain permission from the examining doctor to send the draft of the medical report.

[0036] If there are multiple drafts of the medical report, the draft sending unit 104 may send all of the drafts of the medical report, or may send drafts of the medical report that satisfy predetermined conditions, such as a draft of the medical report specified by the examining physician, a draft of the medical report with a higher priority than a predetermined rank, or a draft of the medical report with a certainty of the case equal to or higher than a predetermined threshold.

[0037] Furthermore, when there are multiple drafts of medical certificates and the cases belong to different medical fields, the draft sending unit 104 may separate the drafts of medical certificates by medical field and send them to the corresponding medical departments. Note that instead of the draft sending unit 104 performing the above classification, the draft creating unit 103 may classify the cases and create a draft of the medical certificate for each medical field.

[0038] In the above configuration, the diagnostic information input unit 101 is an example of a prompt creation means, the case candidate estimation unit 102 is an example of an estimation means, the draft creation unit 103 is an example of a draft creation means, and the draft sending unit 104 is an example of a selection means.

[0039] [Draft creation process] Next, a description will be given of a draft creation process for creating a draft of the above-mentioned medical certificate. Fig. 5 is a flowchart of the draft creation process by the diagnosis support device 10. This process is realized by the processor 12 shown in Fig. 2 executing a program prepared in advance and operating as each element shown in Fig. 3.

[0040] First, diagnostic information is input to the diagnosis support device 10 via the I / F 11. The diagnostic information is input to the diagnostic information input unit 101. The diagnostic information input unit 101 creates a prompt including the diagnostic information (step S101). The diagnostic information input unit 101 outputs the prompt to the case candidate estimation unit 102.

[0041] Next, the case candidate estimation unit 102 inputs a prompt to the diagnostic support AI to estimate a case candidate (step S102). The case candidate estimation unit 102 outputs the case candidate to the draft creation unit 103. Next, the draft creation unit 103 creates a draft of the medical report based on the case candidate (step S103). Specifically, the draft creation unit 103 creates a draft of the medical report by adjusting the format of the text output by the diagnostic support AI. The draft creation unit 103 outputs the draft of the medical report to the draft sending unit 104.

[0042] Next, the draft sending unit 104 selects a medical department to which the draft of the medical certificate is to be sent, and sends the draft of the medical certificate to the selected medical department (step S104), after which the process ends.

[0043] [Variations] Next, a modification of the first embodiment will be described.

[0044] The diagnosis support device 10 may transmit a draft of a medical report to a medical department taking into consideration the urgency of the case. For example, the draft sending unit 104 of the diagnosis support device 10 may send a draft of a medical report for a case with a high urgency to the medical department in preference to other drafts of medical reports. Furthermore, the draft sending unit 104 may send an alert notification to a case with a high urgency so that the case is given priority over other patients. Examples of high urgency cases include cases involving the brain and the heart, which are life-threatening cases. Furthermore, the urgency is expressed, for example, in three levels, "high," "medium," and "low," or in two levels, "high" and "low."

[0045] The urgency of a case can be estimated using a diagnostic support AI. For example, the case candidate estimation unit 102 can estimate the urgency of a case in addition to the case and the reason for selecting the case by inputting a prompt to the diagnostic support AI that includes an instruction to output the urgency of the case.

[0046] The urgency of the case may also be estimated by the draft sending unit 104. The draft sending unit 104 can estimate the urgency and medical department of the case based on a rule base. The rule is, for example, a rule that associates the case, the urgency, and the medical department. The draft sending unit 104 may also estimate the urgency and medical department of the case using a machine learning model. This machine learning model is, for example, a trained machine learning model that is trained to input a draft of a medical certificate and output the urgency and medical department of the case.

[0047] [Usage example] Next, a specific example of use of the diagnosis support system according to the first embodiment will be described. In this example, a draft of a medical certificate is used as a draft of a referral letter.

[0048] When a patient speaks to a doctor during a medical examination at an emergency hospital or a local doctor, such as "I visited Southeast Asia a week ago and have had a stomachache since then," a terminal device 5 installed at the emergency hospital or the local doctor transmits the conversation to a diagnosis support device 10 as information at the time of diagnosis.

[0049] The diagnosis support device 10 prepares a draft of a referral letter for a tropical region-related illness based on the information at the time of diagnosis. Then, the diagnosis support device 10 selects a medical department (a medical department that is familiar with tropical region-related illnesses) to which the patient should be referred from the draft of the referral letter and transmits the draft of the referral letter. At this time, the diagnosis support device 10 may present the draft of the referral letter and the referral destination to a doctor at an emergency hospital or a local doctor and obtain permission to transmit the letter. In this way, use of the diagnosis support system can reduce the effort required to prepare a referral letter and also enable a smooth examination at the medical department to which the patient should be referred.

[0050] Second Embodiment Next, a second embodiment will be described. In the second embodiment, it is assumed that there is a group of medical institutions connected via a network, and that there are multiple medical departments that can handle a certain case. The diagnosis support device according to the second embodiment transmits a draft of a medical certificate taking into consideration the resources of each medical department. In this way, the diagnosis support device according to the second embodiment can level resources among medical departments.

[0051] The diagnosis support device according to the second embodiment has the same system configuration and hardware configuration as the diagnosis support device 10 according to the first embodiment, and therefore a description thereof will be omitted.

[0052] [Function Configuration] 6 is a block diagram showing the functional configuration of a diagnostic support device 10a according to the second embodiment. Functionally, the diagnostic support device 10a includes a diagnostic information input unit 101a, a case candidate estimation unit 102a, a draft creation unit 103a, a draft sending unit 104a, and a department information acquisition unit 105a. Note that the diagnostic information input unit 101a, the case candidate estimation unit 102a, and the draft creation unit 103a have the same configurations and operate in the same manner as the diagnostic information input unit 101, the case candidate estimation unit 102, and the draft creation unit 103 of the diagnostic support device 10 according to the first embodiment, and therefore, description thereof will be omitted.

[0053] Information about the medical department (hereinafter also referred to as "medical department information") is input to the diagnostic support device 10a from the terminal device 20 of each medical department via the I / F 11. The medical department information is input to the department information acquisition unit 105a. The medical department information includes, for example, fixed information such as the location of the medical department and the medical equipment (examination equipment) owned by the medical department, and variable information such as the doctor's workload, the doctor's schedule, and the operating status of the medical equipment. The variable information is periodically transmitted from the terminal device 20 of each medical department to the diagnostic support device 10a. The department information acquisition unit 105a outputs the medical department information to the draft sending unit 104a.

[0054] The draft sending unit 104a selects a medical department to which the draft of the medical certificate is to be sent, based on the draft of the medical certificate and the medical department information.

[0055] Specifically, the draft sending unit 104a first selects multiple medical departments as destination candidates based on the medical report draft. The draft sending unit 104a can select multiple medical departments by using a rule base or a machine learning model. The rule is, for example, a rule that associates cases with medical departments in a one-to-many manner. The machine learning model is, for example, a trained machine learning model that receives the medical report draft as input and is trained to output multiple medical departments.

[0056] Next, the draft sending unit 104a refers to the medical department information and selects an optimal medical department from among multiple medical departments. The draft sending unit 104a may select the medical department closest to the patient's current location as the optimal medical department. Alternatively, the draft sending unit 104a may select the medical department where the doctor's workload is not strained and the doctor has available time in his / her schedule as the optimal medical department. Alternatively, the draft sending unit 104a may select the optimal medical department by taking into consideration not only the doctor's schedule but also the availability of medical equipment required for the examination and treatment of the case and the reservation status of the medical equipment.

[0057] The draft sending unit 104a sends the draft of the medical certificate to the terminal device 20 of the selected medical department.

[0058] In the above configuration, the department information acquiring unit 105a is an example of a medical department information acquiring means.

[0059] [Draft creation process] Next, a description will be given of a draft creation process for creating a draft of the above-mentioned medical certificate. Fig. 7 is a flowchart of the draft creation process by the diagnosis support device 10a. This process is realized by the processor 12 shown in Fig. 2 executing a program prepared in advance and operating as each element shown in Fig. 6. Note that steps S201 to S203 are similar to the processes of steps S101 to S103 of the first embodiment shown in Fig. 4, and therefore a description thereof will be omitted.

[0060] Medical department information is input to the diagnosis support device 10a from the terminal device 20 of each medical department via the I / F 11. The medical department information is input to the department information acquisition unit 105a (step S204). The department information acquisition unit 105a outputs the medical department information to the draft sending unit 104a.

[0061] Next, the draft sending unit 104a selects a medical department to which the draft of the medical certificate will be sent based on the draft of the medical certificate and the medical department information.The draft sending unit 104a then sends the draft of the medical certificate to the selected medical department (step S205).Then, the processing ends.

[0062] [Variations] Next, a description will be given of modifications of the second embodiment. The following modifications can be applied to the second embodiment in appropriate combinations.

[0063] (Variation 1) The diagnosis support device 10a may transmit a draft of a medical report to a medical department taking into consideration the urgency of the case. For example, the draft sending unit 104a of the diagnosis support device 10a may transmit a draft of a medical report for a case with a high urgency to the medical department in preference to other drafts of medical reports. Furthermore, the draft sending unit 104a may send an alert notification to a case with a high urgency so that the case is treated with priority over other patients. High urgency cases include cases involving the brain and the heart, which are life-threatening cases. Furthermore, the urgency is expressed, for example, in three levels, "high," "medium," and "low," or in two levels, "high" and "low."

[0064] The urgency of a case can be estimated using a diagnostic support AI. For example, the case candidate estimation unit 102a can estimate the urgency of a case in addition to the case and the reason for selecting the case by inputting a prompt to the diagnostic support AI that includes an instruction to output the urgency of the case.

[0065] The urgency of the case may also be estimated by the draft sending unit 104a. The draft sending unit 104a can estimate the urgency and medical department of the case based on a rule base. The rule is, for example, a rule that associates the case, the urgency, and multiple medical departments. The draft sending unit 104a may also estimate the urgency and medical department of the case using a machine learning model. This machine learning model is, for example, a trained machine learning model that is trained to input a draft of a medical certificate and output the urgency and multiple medical departments of the case.

[0066] (Variation 2) The diagnosis support device 10a may make the draft of the medical report available to each medical department instead of sending the draft to the medical department. For example, the draft sending unit 104a of the diagnosis support device 10a makes the draft of the medical report available to the network of a group of medical institutions. In this case, the draft sending unit 104a may impose access restrictions so that only the medical departments selected by the draft sending unit 104a can access the draft of the medical report.

[0067] A person in charge of each medical department can access the diagnosis support device 10a and view the draft medical report. If the medical department is able to handle the case, it may notify the diagnosis support device 10a of this. Upon receiving the notification, the diagnosis support device 10a updates the published information. For example, the diagnosis support device 10a may add the name of the corresponding medical department to the draft medical report, or may update the access restrictions so that only the corresponding medical department can view the draft medical report.

[0068] Third Embodiment 8 is a block diagram showing the functional configuration of a diagnostic support device according to the third embodiment. The diagnostic support device 300 includes a prompt creating unit 301, an estimating unit 302, a draft creating unit 303, and a selecting unit 304.

[0069] 9 is a flowchart of processing by the diagnosis support device of the third embodiment. The prompt creation means 301 acquires information at the time of diagnosis and creates a prompt including the information at the time of diagnosis (step S301). The estimation means 302 inputs the prompt into the diagnosis support AI and estimates a case candidate (step S302). The draft creation means 303 creates a draft of a medical report based on the case candidate (step S303). The selection means 304 selects a destination to send the draft of the medical report based on the case candidate (step S304).

[0070] The diagnosis support device 300 of the third embodiment makes it possible to create a draft of a medical certificate and send the draft to an appropriate department. This allows the diagnosis support device 300 to support doctors in making decisions in the medical field.

[0071] A part or all of the above-described embodiments can be described as, but not limited to, the following supplementary notes.

[0072] (Appendix 1) a prompt generating means for acquiring diagnostic information and generating a prompt including the diagnostic information; an estimation means for inputting the prompt into a diagnostic support AI and estimating a case candidate; a draft preparation means for preparing a draft of a medical certificate based on the candidate case; a selection means for selecting a destination to which the draft of the medical certificate is to be sent based on the candidate case; A diagnostic support device comprising:

[0073] (Appendix 2) The diagnostic support device described in Appendix 1, wherein the diagnostic support AI is created by additionally training a general-purpose generation AI using pairs of information at the time of diagnosis and information about the case as learning data.

[0074] (Appendix 3) the estimation means estimates the case candidate, a reason for selecting the case candidate, and a probability indicating a possibility that the case candidate is the case candidate; 2. The diagnostic support device according to claim 1, wherein the draft creation means creates a draft of the diagnosis report including the case candidates and the grounds in descending order of the degree of certainty.

[0075] (Appendix 4) The diagnostic support device according to claim 3, wherein the selection means transmits to the destination a draft of the diagnosis report for the case candidate whose certainty is equal to or greater than a predetermined threshold.

[0076] (Appendix 5) the estimation means estimates the urgency of the candidate case; 2. The diagnostic support device according to claim 1, wherein the selection means transmits a draft of the medical report and an alert notification to the destination when the urgency is high.

[0077] (Appendix 6) a medical department information acquiring means for acquiring medical department information from a plurality of medical departments; 2. The diagnostic support device according to claim 1, wherein the selection means selects a medical department to which the draft of the medical certificate is to be sent based on the case candidate and the medical department information.

[0078] (Appendix 7) the medical department information includes physician workload and physician schedule; 7. The diagnostic support device according to claim 6, wherein the selection means selects the medical department to which the draft medical certificate is to be sent based on the doctor's workload and the doctor's schedule.

[0079] (Appendix 8) The medical department information includes a reservation status of testing equipment owned by the medical department; The diagnostic support device according to claim 7, wherein the selection means selects the medical department to which the draft of the medical certificate is to be sent based on the availability of the testing equipment and reservation status.

[0080] (Appendix 9) A computer-implemented diagnostic support method, comprising: Obtain diagnostic information, create a prompt that includes the diagnostic information, The prompt is input to the diagnostic support AI, and a case candidate is estimated. Prepare a draft medical certificate based on the candidate case; A diagnostic support method for selecting a destination to which the draft medical certificate is to be sent based on the candidate case.

[0081] (Appendix 10) Obtain diagnostic information, create a prompt that includes the diagnostic information, The prompt is input to the diagnostic support AI, and a case candidate is estimated. Prepare a draft medical certificate based on the candidate case; A program that causes a computer to execute a process of selecting a destination to which the draft medical certificate will be sent based on the candidate case.

[0082] Although the present disclosure has been described above with reference to the embodiments and examples, the present disclosure is not limited to the above-described embodiments and examples. Various modifications that can be understood by a person skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. [Explanation of symbols]

[0083] 10, 10a Diagnostic support device 101, 101a Diagnostic information input section 102, 102a Case candidate estimation section 103, 103a Draft Creation Section 104, 104a Draft Sending Section 105a Department Information Acquisition Department

Claims

1. a prompt generating means for acquiring diagnostic information and generating a prompt including the diagnostic information; an estimation means for inputting the prompt into a diagnostic support AI and estimating a case candidate; a draft preparation means for preparing a draft of a medical certificate based on the candidate case; a selection means for selecting a destination to which the draft of the medical certificate is to be sent based on the candidate case; A diagnostic support device comprising:

2. The diagnostic support device according to claim 1 , wherein the diagnostic support AI is created by additionally training a general-purpose generation AI using pairs of information at the time of diagnosis and information about the case as learning data.

3. the estimation means estimates the case candidate, a reason for selecting the case candidate, and a probability indicating a possibility that the case candidate is the case candidate; The diagnostic support device according to claim 1 , wherein the draft creating unit creates a draft of the medical report including the case candidates and the grounds in descending order of the degree of certainty.

4. The diagnostic support device according to claim 3 , wherein the selection means transmits to the destination a draft of the diagnosis report for the case candidate whose certainty is equal to or greater than a predetermined threshold value.

5. the estimation means estimates the urgency of the candidate case; The diagnostic support device according to claim 1 , wherein the selection means transmits the draft of the medical report and an alert notification to the destination when the urgency is high.

6. a medical department information acquiring means for acquiring medical department information from a plurality of medical departments; The diagnosis support device according to claim 1 , wherein the selection means selects a medical department to which the draft of the medical certificate is to be sent based on the case candidate and the medical department information.

7. the medical department information includes physician workload and physician schedule; 7. The diagnostic support device according to claim 6, wherein the selection means selects the medical department to which the draft medical certificate is to be sent based on the doctor's workload and the doctor's schedule.

8. The medical department information includes a reservation status of testing equipment owned by the medical department; 8. The diagnostic support device according to claim 7, wherein the selection means selects the medical department to which the draft of the medical certificate is to be sent based on the availability of the testing equipment and reservation status.

9. A computer-implemented diagnostic support method, comprising: Obtain diagnostic information, create a prompt that includes the diagnostic information, The prompt is input to a diagnostic support AI to estimate a case candidate; Prepare a draft medical certificate based on the candidate case; A diagnostic support method for selecting a destination to which the draft medical certificate is to be sent based on the candidate case.

10. Obtain diagnostic information, create a prompt that includes the diagnostic information, The prompt is input to a diagnostic support AI to estimate a case candidate; Prepare a draft medical certificate based on the candidate case; A program that causes a computer to execute a process of selecting a destination to which the draft medical certificate will be sent based on the candidate case.

Images