Medical support systems, medical support methods, medical support programs

Abstract

There is a need for a medical support system that can provide appropriate medical care tailored to each patient. [Solution] A medical support system comprising: an audio information acquisition unit; a patient information acquisition unit; a medical determination unit; a calculation processing unit that generates support information to be presented to a physician; and an information output unit that outputs the support information. The calculation processing unit performs: an information conversion process that converts audio information acquired by the audio information acquisition unit into processing information that can be recognized by the medical determination unit; an input process that inputs the processing information and patient information to the medical determination unit as medical necessary information; an extraction process that extracts the medical procedures performed; and a support information generation process that extracts unperformed medical procedures based on the difference between the medical procedure list output by the medical determination unit and the medical procedures extracted by the extraction process, and generates support information to include the unperformed medical procedures.

Description

Technical Field

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

Background Art

[0002] For example, Japanese Unexamined Patent Application Publication No. 2023-26640 (Patent Document 1) discloses a technique for creating an electronic medical record based on voice data of a doctor and a patient in medical treatment, thereby reducing the burden on the doctor for creating the medical record.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] On the other hand, the medical treatment content provided to patients is likely to be personalized, and depending on the capabilities of medical staff, even if the above-mentioned electronic medical record is utilized, appropriate medical treatment corresponding to the patient may not be provided.

[0005] Therefore, it is desired to provide a medical support system capable of providing appropriate medical treatment corresponding to the patient.

Means for Solving the Problems

[0006] The medical support system according to the present disclosure is a medical support system that supports medical treatment by a doctor, a voice information acquisition unit that acquires the voice during the medical treatment as voice information, a patient information acquisition unit that acquires patient information, which is information of a patient receiving the medical treatment, Using a pre-trained model,A medical treatment determination unit outputs a list of medical treatments that are potentially effective for the patient, based on the medical treatment information which is the information necessary to perform the aforementioned medical treatment. Using the aforementioned trained model, A calculation processing unit that generates support information to be presented to the physician, It comprises an information output unit that outputs the aforementioned support display information, The aforementioned arithmetic processing unit is Information conversion processing that converts the voice information acquired by the voice information acquisition unit into processing information that can be recognized by the medical diagnosis determination unit, Input processing involves inputting the processing information and the patient information into the medical treatment determination unit as the medical treatment necessary information. An extraction process to extract the medical procedures performed by the physician to the patient during the consultation, based on the aforementioned audio information, A process to generate a list of medical procedures performed, which is a list of the medical procedures performed extracted by the extraction process, The medical treatment list output by the medical treatment determination unit And, referring to the aforementioned list of medical procedures performed, the medical procedures not included in the aforementioned list of medical procedures performed are, Unperformed medical procedures are medical procedures that may not have been performed by the physician during the aforementioned medical treatment. as The process involves extracting and generating support information that includes the aforementioned unperformed medical procedures, and then executing a support information generation process. death, In the support information generation process, a medical evaluation process is performed to generate medical evaluation information that evaluates the physician's response to the medical treatment and its medical appropriateness, based on the medical information required for the treatment and the medical treatment that has not yet been performed. .

[0007] In this configuration, processing information converted from voice information and patient information are used as necessary medical information, and a list of medical procedures is output based on this necessary medical information. Then, support information is generated to include unperformed medical procedures extracted based on the difference between the list of medical procedures and the medical procedures performed, which are extracted based on the voice information. Therefore, by checking the outputted support information, physicians can reduce the possibility of completing a consultation without performing medical procedures that should have been performed for the patient. Thus, the quality of medical care provided by physicians can be improved. Thus, this configuration allows for the provision of appropriate medical care tailored to each patient.

[0008] Further features and advantages of the medical support system will become clear from the following description of exemplary and non-limiting embodiments described with reference to the drawings.

Brief Description of the Drawings

[0009] [Figure 1] Schematic diagram of the medical support system [Figure 2] System configuration diagram of the medical support system [Figure 3] System configuration diagram of the medical support system in another embodiment [Figure 4] Figure showing a display screen on which support prompt information is displayed [Figure 5] Figure showing a display screen on which support prompt information is displayed [Figure 6] Figure showing a display screen on which support prompt information is displayed [Figure 7] Figure showing a display screen on which support prompt information and a chat bar are displayed [Figure 8] Figure showing a display screen on which support prompt information and a chat bar are displayed [Figure 9] Figure showing a display screen on which pre-prompt information is displayed [Figure 10] Control flow diagram [Figure 11] Control flow diagram [Figure 12] Control flow diagram

Modes for Carrying Out the Invention

[0010] Hereinafter, embodiments of the medical support system will be described based on the drawings. In the medical support system, the processing executed by each functional unit is executed by a computer using a medical support program. Further, the processing executed by each functional unit in the medical support system can be replaced with "steps" respectively, and the content according to the present disclosure also includes a medical support method including various steps executed in the medical support system.

[0011] As shown in FIG. 1, the medical support system is a system that supports the medical treatment by doctor 9. As shown in FIGS. 2 and 3, the medical support system includes an audio information acquisition unit 10, a patient information acquisition unit 20, a medical treatment determination unit 30, an information output unit 40, and an arithmetic processing unit 100 that generates support presentation information for presentation to doctor 9. In the present embodiment, the medical support system includes a user terminal 1 and a management server 2. Here, the "doctor" can include those who perform medical treatment on non-humans such as veterinarians. Of course, the "doctor" also includes dentists. In this example, the patient 8 who receives medical treatment from doctor 9 is regarded as a patient, but it is not limited to this. The "patient" can also include pet animals and livestock that are the objects of medical treatment in animal hospitals and the like. In addition, the "medical treatment by doctor" includes medical treatment in hospitals, home visits (including house calls), telemedicine, and the like. In the following, the notations of the symbols indicating the doctor and the patient are omitted.

[0012] The management server 2 that plays the core role of the medical support system is communicably connected to the user terminal 1 via a network. The management server 2 may be a physical server or a cloud server. In the example of FIG. 2, the medical support system is provided as an on-premises type service. The network is constructed by wire or wirelessly, or by using both of them. Also, a closed area network can be used as the network, or an open line equipped with a firewall can also be used. Examples of the user terminal 1 include a smartphone, a tablet, and a personal computer. Also, the user terminal 1 can be a wearable terminal such as smart glasses or a smartwatch. The medical support system may include a plurality of user terminals 1. In that case, the management server 2 is communicably connected to each of the plurality of user terminals 1. In this example, the doctor performs medical treatment on the patient while using the user terminal 1.

[0013] The user terminal 1 comprises a voice input unit 60, an information input unit 70, and a display unit 80. The voice input unit 60 is configured to accept voice input from the user. The voice input unit 60 is also configured to accept voice input of conversations from multiple users. The voice input unit 60 is equipped with a microphone for receiving such voice input. The information input unit 70 is configured to accept text input from a keyboard, touch panel, etc. The information input unit 70 is also configured to accept imaging information captured by an imaging device (in this case, a camera). The information input unit 70 may also accept image information input by a scanner.

[0014] The display unit 80 displays visible information such as text and images. The display unit 80 is configured to display information input by the voice input unit 60 and the information input unit 70. In addition, the display unit 80 can display information output from the information output unit 40 (management server 2). The display unit 80 is equipped with a display (for example, a touch panel) for displaying this information.

[0015] The patient information acquisition unit 20, the information output unit 40, and the voice information acquisition unit 10 are located on the management server 2. In this example, the medical diagnosis unit 30 and the calculation processing unit 100 are also located on the management server 2.

[0016] The voice information acquisition unit 10 acquires audio information from the audio during the medical examination. In this embodiment, the voice information acquired is obtained from the audio input unit 60 during the medical examination. Here, the audio during the medical examination includes the voice spoken by the doctor performing the examination and the voice of the patient. In addition, the audio during the medical examination may also include voices other than those of the doctor or patient. The voice information acquisition unit 10 is configured to transmit the acquired voice information to the arithmetic processing unit 100. As an example, the voice information acquisition unit 10 performs a process to chunk the audio during the examination at regular intervals. In this way, the voice information acquisition unit 10 acquires the audio during the examination as voice information. The voice information acquisition unit 10 may acquire voice information when the doctor inputs information to start the medical examination into the audio input unit 60 or the information input unit 70. Similarly, the voice information acquisition unit 10 may terminate the acquisition of voice information when the doctor inputs information to complete the medical examination into the audio input unit 60 or the information input unit 70.

[0017] The patient information acquisition unit 20 acquires patient information, which is information about patients receiving medical treatment. In this embodiment, patient information includes medical record information about the patient. Medical record information includes basic information of the patient (in this case, the patient) (name, age, gender, etc.), medical history, medical condition, test results, treatment details, prescriptions, etc. In this example, the management server 2 is equipped with a storage unit 50. In this example, such patient information is stored in the storage unit 50 (database) as an electronic medical record. The patient information acquisition unit 20 inputs (transmits) the acquired patient information to the calculation processing unit 100.

[0018] The medical treatment determination unit 30 outputs a list of medical procedures that may be effective for the patient based on the medical treatment necessary information, which is the information required to perform medical treatment. In this embodiment, the medical treatment necessary information includes the patient information described above. The medical treatment necessary information also includes processing information converted from the voice information acquired by the patient information acquisition unit 20. In this example, the medical treatment determination unit 30 outputs a list of medical procedures based on the medical treatment necessary information, i.e., the patient information and the processing information converted from the voice information. The medical treatment determination unit 30 acquires the medical treatment necessary information from the calculation processing unit 100.

[0019] The information output unit 40 outputs support information. In this embodiment, the information output unit 40 outputs the support information generated by the arithmetic processing unit 100 to the display unit 80 of the user terminal 1. This allows the support information to be displayed on the display unit 80. The information output unit 40 may also output at least a portion of the support information as sound, light, or the like, for example, from the user terminal 1.

[0020] In this example, the management server 2 is equipped with a trained model 3, which is a combination of trained parameters and a computation program based on collected information. The medical judgment unit 30 and the computation processing unit 100 then perform processing using the trained model 3. In this embodiment, as shown in Figure 2, the medical judgment unit 30 and the computation processing unit 100 are constructed as the trained model 3. The trained model 3 includes a speech recognition model and a large-scale language model. The trained model 3 is described as a deep learning model, but is not limited to this. The trained model 3 may be a machine learning model other than a deep learning model. Here, by having the management server 2 equipped with the trained model 3, it is possible to provide services in an on-premise environment. The trained model 3 is a deep learning model that has learned medical information. The medical support system can also use the trained model 3 on the data server 5. It should be noted that, not limited to such a trained model 3, the medical judgment unit 30 and the computation processing unit 100 may have a database that they can use, and the medical judgment unit 30 and the computation processing unit 100 may perform processing by referring to the database.

[0021] As shown in Figure 3, the medical support system may also include a data server 5 that can communicate with the management server 2, in addition to the management server 2. In this case, the medical support system may include multiple management servers 2. In this case, it is preferable that the data server 5 can communicate with multiple management servers 2. In the example in Figure 3, the medical determination unit 30 and the calculation processing unit 100 are located on the data server 5. The data server 5 may be a physical server or a cloud server. Also, in the example in Figure 3, the medical support system is provided as an on-premise service, but it is not limited to this. For example, the data server 5 can be an external cloud server. In this case, information (data) is sent and received to and from the data server 5 in an off-premise environment. In this case, the information provided from the management server 2 to the data server 5 is anonymized. Depending on the scale of the medical institution, etc., it is preferable to appropriately select the configuration shown in Figure 2 or Figure 3 for the medical support system.

[0022] In clinical practice, medical care tends to rely heavily on the physician's observational skills, interviewing abilities, and clinical experience (hereinafter, these abilities may be collectively referred to as "clinical competence"). Consequently, some physicians may overlook important changes in a patient's condition or make mistakes in treatment plans. Furthermore, insufficient medical information entered into electronic medical records may result in patients not receiving adequate care at subsequent appointments. Additionally, if a patient is treated by a different physician, insufficient information sharing or significant differences in clinical competence between the two physicians can also lead to patients not receiving adequate care. Therefore, in this embodiment of the medical support system, support information is generated to ensure that a certain level of adequate medical care is consistently provided, regardless of the physician's clinical competence. The configuration for generating this support information will be described in detail below.

[0023] As shown in Figure 10, in this embodiment, the arithmetic processing unit 100 performs information conversion processing (S01), input processing (S02), first extraction processing (S03), second extraction processing (S04), third extraction processing (S05), and support information generation processing (S06).

[0024] The information conversion process is the process of converting the voice information acquired by the voice information acquisition unit 10 into processing information that can be recognized by the medical diagnosis determination unit 30. In this embodiment, the processing information is character information obtained by converting the voice information into text. Here, the voice acquired as voice information during the examination is converted into character information by the information conversion process. The converted character information can be stored in, for example, the storage unit 50. Note that the processing information is not limited to the above-mentioned character information obtained by converting voice information into text, as long as it is information that can be recognized by a computer. In addition to the above-mentioned character information, the processing information may also include information related to the emotions and physical condition of the doctor and patient, such as the tone and volume of their voice. Specifically, the arithmetic processing unit 100 may generate estimated information in the information conversion process, such as estimating the emotions and mood of the doctor and patient from the tone and volume of their voice, or estimating the patient's physical condition, and include it in the processing information. The processing information may also include the content of statements made by persons other than the doctor and patient.

[0025] In this embodiment, the processing information includes medical information related to the patient's medical treatment extracted from the voice information. In this example, the processing information includes multiple types of medical information. Specifically, the processing information includes multiple types of medical information such as condition information regarding the symptoms and medical state reported by the patient, prescription information regarding the prescribed treatment, examination information regarding tests, and date and time information regarding the date and time of the medical consultation. These multiple types of medical information include suggestion information proposed by the doctor or the patient, agreement information indicating agreements made between the doctor and the patient, and consent information for which the doctor or the patient seeks consent. The arithmetic processing unit 100 may generate the processing information by associating each of these multiple types of medical information with the estimated information described above. In that case, the arithmetic processing unit 100 may generate the entire text information including the medical information as the processing information in the information conversion process, or it may generate the processing information excluding text information other than the medical information (text information converted from the voice information).

[0026] The input process involves inputting processing information and patient information as necessary medical information into the medical determination unit 30. The calculation processing unit 100 inputs the processing information converted from the voice information and the patient information obtained from the patient information acquisition unit 20 as necessary medical information into the medical determination unit 30. Here, the processing by the medical determination unit 30 will be explained. Based on the patient information obtained from the patient information acquisition unit 20 and the processing information, the medical determination unit 30 outputs a list of medical procedures that may be effective for the patient (effective medical procedures). Here, effective medical procedures include all medical procedures performed by a physician. In addition, the medical procedures performed may include procedures such as taking temperature, measuring blood pressure, administering eye drops, trimming nails, psychological counseling, and tracheal suctioning (referred to here as "peripheral medical procedures"). In addition, the medical procedures performed may include decisions made by the physician from a medical standpoint (for example, decisions on medication intervals and consultation intervals). On the other hand, actions unrelated to medical care (for example, simply walking or inputting some kind of information into user terminal 1) are not included in the medical treatment activities performed.

[0027] The medical treatment determination unit 30 refers to patient information corresponding to the patient receiving treatment and lists one or more effective medical procedures. Similarly, the medical treatment determination unit 30 refers to processing information and lists one or more effective medical procedures. If there are many effective medical procedures, the medical treatment determination unit 30 can determine the importance of each of the multiple effective medical procedures and prioritize including the effective medical procedures with the highest importance in the medical procedure list. On the other hand, the medical treatment determination unit 30 can include all effective medical procedures in the medical procedure list regardless of the number of effective medical procedures. The medical treatment determination unit 30 inputs the output medical procedure list to the calculation processing unit 100.

[0028] For example, the medical treatment determination unit 30 can compare patient information and processing information and prioritize including common effective medical treatments in the medical treatment list. Furthermore, the medical treatment determination unit 30 can also prioritize including in the medical treatment list multiple effective medical treatments based on processing information that are not included in the effective medical treatments based on patient information.

[0029] For example, in a conversation between a doctor and a patient during a medical examination, if the patient says, "For the past few days, I've been having trouble urinating. I haven't been able to urinate at all since yesterday. Something feels wrong," this statement is converted from audio information into processing information (in this case, text-based medical information). Based on this processing information, the medical assessment unit 30 lists several effective medical procedures, such as "palpation of the bladder and abdominal observation," "abdominal ultrasound or simple ultrasound examination," "perform urinary catheterization," "re-evaluate blood tests," and "investigate the cause of urinary retention, examine and explain the presence or absence of signs of infection." The medical assessment unit 30 may include all of these effective medical procedures in the medical procedure list, or it may include only some of them. The medical assessment unit 30 may also generate a medical procedure list using estimated information based on the above statement.

[0030] For example, the medical treatment determination unit 30 can refer to patient information and, as described above, modify the content of multiple effective medical treatments based on processing information, add other effective medical treatments, or delete parts of effective medical treatments. In this way, the medical treatment determination unit 30 can modify and change effective medical treatments based on processing information according to the patient's age, medical condition, medical history, test results, etc., included in the patient information.

[0031] Furthermore, the medical treatment determination unit 30 can also include in the medical treatment list effective medical treatments based on the contents of the patient information, even if they are not included in the processing information. For example, if a predetermined period of time has elapsed since a predetermined medical treatment (e.g., a blood test) was previously performed, the medical treatment determination unit 30 can include a suggestion for that predetermined medical treatment in the medical treatment list. In this way, the medical treatment determination unit 30 can generate a medical treatment list that corresponds to the specific contents of the processing information and the patient information. The following section returns to the explanation of the processing performed by the calculation processing unit 100.

[0032] The first extraction process is a process of extracting the medical actions performed by the doctor on the patient during the consultation, based on the voice information. The first extraction process (S03) corresponds to the "extraction process". In this embodiment, the arithmetic processing unit 100 extracts the medical actions performed from the processing information converted from the voice information as described above. The arithmetic processing unit 100 extracts and lists all the medical actions performed by the doctor during the consultation. In other words, the arithmetic processing unit 100 generates a list of medical actions performed. Medical actions performed include all medical procedures performed by the doctor. Medical actions performed may also include the medical-related procedures described above. Medical actions performed may also include decisions made by the doctor from a medical standpoint (for example, decisions on the interval between taking medication or the interval between consultations). On the other hand, actions unrelated to medical care (for example, simply walking or inputting information into the user terminal 1) are not included in medical actions performed. The calculation processing unit 100 can determine whether the actions of the physician during the consultation constitute a medical procedure, based on the physician's statements during the consultation as well as the patient's statements. The calculation processing unit 100 may also refer to patient information as needed to make the above determination.

[0033] For example, if the physician during the examination says, "Let's listen to the heart sounds. Yes, everything is fine," the arithmetic processing unit 100 will determine from the processing information based on that statement that "auscultation was performed." In other words, the arithmetic processing unit 100 extracts "auscultation" as an performed medical procedure from the processing information. Also, if the physician during the examination says, "Let's keep an eye on it," the arithmetic processing unit 100 will determine from the processing information based on that statement that "a suggestion for follow-up observation was made." In other words, the arithmetic processing unit 100 extracts "a suggestion for follow-up observation" as an performed medical procedure from the processing information.

[0034] The calculation processing unit 100 extracts unperformed medical procedures, which are medical procedures that may not have been performed by the physician during treatment, based on the difference between the list of medical procedures output by the medical treatment determination unit 30 and the medical procedures performed extracted by the extraction process. In this embodiment, the calculation processing unit 100 refers to the list of medical procedures and the list of medical procedures performed and performs a second extraction process to extract medical procedures not included in the list of medical procedures performed as unperformed medical procedures (S04). That is, the calculation processing unit 100 extracts valid medical procedures not included in the medical procedures performed as unperformed medical procedures. The calculation processing unit 100 then lists the extracted unperformed medical procedures.

[0035] For example, if the calculation processing unit 100 finds that "palpation of the bladder and abdominal observation" is included in the list of medical procedures, but not in the list of performed medical procedures, it will extract "palpation of the bladder and abdominal observation" as an unperformed medical procedure.

[0036] In this example, the calculation processing unit 100 performs a third extraction process (S05) to extract agreement information from the processing information that indicates the agreements made between the patient and the doctor as described above. The calculation processing unit 100 extracts one or more agreements from the statements made by the doctor and the patient during the medical consultation in the processing information. Here, the agreements include decisions and changes to treatment plans, decisions and changes to prescribed medications, decisions and changes to consultation dates and times, decisions and changes to consultation intervals, etc., which are included in the above medical-related information. The agreements also include requests from the patient regarding medical-related information. Furthermore, the agreements also include proposals from the doctor and the patient regarding deadlines for medical-related information (for example, matters to be considered by the next consultation, matters to be considered at the next consultation, matters to be decided after a predetermined period, etc.).

[0037] The processing unit 100 determines whether there are any agreements in the processing information based on statements made by at least one of the doctors and patients during the consultation. For example, if a patient says, "Is it possible to change the examination from once every three months to once every six months?", and the doctor responds, "Let's decide after we see the results of the next examination," the processing unit 100 determines that there is an agreement based on the processing information derived from those statements. The processing unit 100 then extracts the agreement, "At the next consultation, we will consider whether or not to change the examination from once every three months to once every six months," from the processing information. The processing unit 100 then lists the extracted agreements. The processing unit 100 can also determine whether there are any agreements in the processing information based on statements made by persons other than the doctor and patient during the consultation (for example, the patient's attendant).

[0038] The arithmetic processing unit 100 generates support information to include unperformed medical procedures. In this embodiment, the arithmetic processing unit 100 performs a support information generation process to generate support information to include one or more unperformed medical procedures (S06). In addition, in this embodiment, if the voice information includes agreements made between the patient and the doctor, the arithmetic processing unit 100 generates support information to include these agreements. The generated support information is output to the display unit 80 of the user terminal 1 by the information output unit 40 as described above (Figures 4 to 8). The arithmetic processing unit 100 can generate support information when, for example, an instruction to generate support information is input from the voice input unit 60 or the information input unit 70 of the user terminal 1. Input from the information input unit 70 may include a window displayed on the display unit 80 that allows the user to select whether or not to generate support information, and the doctor can input an instruction to generate support information by operating this window. Furthermore, the arithmetic processing unit 100 is also capable of regenerating support information based on input from the voice input unit 60 and the information input unit 70.

[0039] In this example, as shown in Figure 11, the arithmetic processing unit 100 is configured to perform importance evaluation processing (S11), commitment evaluation processing (S12), and medical evaluation processing (S13) in the support information generation process.

[0040] When the calculation processing unit 100 extracts multiple unperformed medical procedures, it generates support information so that the unperformed medical procedures with the highest importance are output preferentially. In the importance evaluation process (S11), support information is generated so that the unperformed medical procedures with the highest importance are output preferentially to the information output unit 40. As a result, one or more unperformed medical procedures with high importance are preferentially displayed on the display unit 80. In the importance evaluation process, the calculation processing unit 100 lists each of the extracted multiple unperformed medical procedures in order of the highest risk to the patient by comparing them with the medical information necessary for treatment (processing information, patient information). The calculation processing unit 100 can also perform the above listing by considering the medical knowledge learned by the trained model 3 for these unperformed medical procedures. Furthermore, the calculation processing unit 100 generates support information for each of these unperformed medical procedures, linking them with the relevant medical information necessary for treatment and the above medical knowledge, and creating text to encourage advice and warnings to physicians.

[0041] Figure 4 shows an example of unperformed medical information displayed on the display unit 80. In the illustrated example, two unperformed medical procedures deemed to pose a high risk to the patient are listed as problems. Solutions are also listed for each unperformed medical procedure. In this way, the calculation processing unit 100 can generate unperformed medical information that prioritizes displaying multiple unperformed medical procedures deemed to pose a high risk to the patient, and displays these unperformed medical procedures as problems and solutions on the display unit 80. This allows physicians to identify high-priority unperformed medical procedures and incorporate them into treatment as needed. For example, a physician can provide unperformed medical procedures to a patient during treatment.

[0042] Furthermore, the calculation processing unit 100 can also include unperformed medical procedures that it has determined pose a relatively low risk to the patient in the unperformed medical information. In the example shown in Figure 5, several unperformed medical procedures that were determined to pose a relatively low risk to the patient are displayed on the display unit 80 as problems and supplementary information. This allows physicians to review even unperformed medical procedures of relatively low importance and reflect them in their treatment as needed.

[0043] Furthermore, even if the calculation processing unit 100 extracts only one unperformed medical procedure, it can perform importance evaluation processing and generate unperformed medical information for that procedure as support display information. In addition, the calculation processing unit 100 can appropriately change the display content to be shown on the display unit 80 when generating the unperformed medical information.

[0044] In the commitment evaluation process (S12), commitment information is generated as support information to encourage advice and warnings to the physician, based on the commitments extracted in the third extraction process. In the commitment evaluation process, the calculation processing unit 100 documents the commitments in relation to patient information and medical knowledge as needed. Figure 6 shows an example of commitment information displayed on the display unit 80. In the illustrated example, the commitment information is shown as "reminder regarding prescription changes" and "reminder regarding consultation date". As shown in the illustrated example, the calculation processing unit 100 can also include matters that were not agreed upon between the physician and the patient (in this case, "reminder regarding prescription changes") in the commitment information. This allows the physician to check the commitment information and reflect it in treatment as needed. Furthermore, it reduces the possibility that promises agreed upon between the physician and the patient will not be fulfilled, making it easier to increase trust in the physician. The calculation processing unit 100 can appropriately change the display content shown on the display unit 80 when generating the commitment information.

[0045] In the medical evaluation process (S13), medical evaluation information, which is an evaluation of the physician's medical treatment, is generated as support display information based on the medical information required for treatment (processing information, patient information) and the information on unperformed treatments. In the medical evaluation process, the calculation processing unit 100 associates the medical information required for treatment, the information on unperformed treatments, and medical knowledge as necessary to create a documented evaluation (summary) of the physician's medical treatment. Figure 7 shows an example of medical evaluation information displayed on the display unit 80. In the illustrated example, the medical evaluation information is shown as "evaluation of physician's response, evaluation of medical appropriateness." Specifically, it displays an evaluation of the medical actions that the physician should have performed, but is not limited to this. For example, the medical evaluation information may include information that evaluates the physician's communication with the patient based on estimated information included in the processing information (e.g., the emotions of the physician and patient estimated from the tone of voice, volume, intonation, etc., the patient's physical condition, etc.). This allows the physician to confirm their own objective evaluation of their medical treatment. Furthermore, the calculation processing unit 100 can appropriately change the display content to be displayed on the display unit 80 when generating medical evaluation information.

[0046] In this embodiment, the arithmetic processing unit 100 can update the generated support information in real time during medical treatment. The information output unit 40 is configured to output the updated support information in real time. In this example, the arithmetic processing unit 100 can update the support information in real time during medical treatment by generating update information based on the information input by the voice input unit 60 and information input unit 70 of the user terminal 1. In the examples of Figures 7 and 8, the user terminal 1 is equipped with a chat interface 86 as the information input unit 70. Therefore, the user (here, a doctor) can input text information via the chat interface 86. The doctor may input new information related to the medical treatment or questions using the voice input unit 60 or information input unit 70. In the example of Figure 8, questions regarding the support information displayed on the display unit 80 have been input. The arithmetic processing unit 100 generates answer information (update information) for the input questions, thereby updating the support information. Furthermore, the support information may be updated when the physician inputs into the voice input unit 60 or the information input unit 70 that they have performed an unperformed medical procedure.

[0047] In this example, the calculation processing unit 100 updates patient information (in this case, the electronic medical record) based on necessary medical information (processing information, patient information) and unperformed medical information. In other words, the calculation processing unit 100 updates patient information based on patient information acquired in advance, voice information during medical treatment, and unperformed medical information. Furthermore, the calculation processing unit 100 can also update patient information based on generated support information. The calculation processing unit 100 updates patient information when medical treatment for the patient is completed. The updated patient information is acquired by the patient information acquisition unit 20 and stored in the storage unit 50. Note that patient information may also be updated directly by the physician during or after medical treatment. That is, the physician can update patient information by inputting information into the voice input unit 60 or information input unit 70 during or after medical treatment. Naturally, the physician can refer to patient information using the user terminal 1 regardless of whether or not they are in the middle of a medical consultation.

[0048] As shown in Figures 9 and 12, in this embodiment, the calculation processing unit 100 generates advance presentation information to be presented to the doctor performing the next consultation, based on the support presentation information and the patient information newly acquired by the patient information acquisition unit 20, after the completion of the consultation but before the next consultation. The information output unit 40 is further configured to output the advance presentation information. In this example, the calculation processing unit 100 performs an advance information generation process to generate the advance presentation information a predetermined period before the patient's next scheduled consultation date (S21). The advance information generation process may also be performed after a predetermined period has elapsed after the completion of the patient's consultation. Thus, the timing of the advance information generation process performed by the calculation processing unit 100 can be changed as appropriate.

[0049] In this example, the calculation processing unit 100 generates pre-presentation information based on updated patient information through a pre-information generation process. The pre-presentation information includes advice and warnings for the doctor regarding the next consultation. The pre-presentation information also generates text for providing advice and warnings to the doctor, linking the evaluation of test results and physical findings that the doctor should check at the next consultation with the patient's primary diagnosis, symptoms, chief complaint, etc. The pre-presentation information generated by the calculation processing unit 100 is output by the information output unit 40 and displayed on the display unit 80. In the example in Figure 9, "Pre-consultation advice," "Evaluation of blood tests," and "Evaluation of vital signs" are displayed as pre-presentation information, but this is not limited to these. For example, advice on how to communicate appropriately with the patient may be displayed as pre-presentation information. Thus, the content displayed as pre-presentation information can be changed as appropriate. By checking the pre-presentation information in advance, the doctor can minimize the omission of medical procedures that should be provided to the patient during the consultation.

[0050] As described above, physicians can review the support information presented during consultations, making it easier to incorporate its content into their practice immediately. Furthermore, by reviewing the support information, physicians can gain an objective evaluation of their own practice, making it easier to improve the quality of care they provide to patients. Additionally, physicians can review the pre-consultation information before the consultation, enabling them to provide appropriate care tailored to each patient, thus improving the quality of care. Moreover, it is easier to avoid personalization of the care provided to patients, allowing physicians to consistently provide care above a certain standard, regardless of their individual abilities.

[0051] [Other Embodiments] (1) In the above embodiment, the arithmetic processing unit 100 was described as being able to generate support information when an instruction to generate support information is input from the voice input unit 60 or information input unit 70 of the user terminal 1, but it is not limited to this. The arithmetic processing unit 100 can generate support information when it determines that the processing information converted from the voice information contains predetermined words as spoken by the doctor. Examples of predetermined words include "That's enough," "We're finished now," "Take care," and "Goodbye." The arithmetic processing unit 100 may also generate support information in real time during the consultation, or it may generate support information after a certain period of time has elapsed since the start of the consultation (for example, after 5 minutes). In this way, the timing of generating support information can be changed as appropriate by the arithmetic processing unit 100.

[0052] (2) In the above embodiment, the arithmetic processing unit 100 was described as having a configuration that allows it to update the generated support information in real time during medical treatment, but it is not limited to this. The arithmetic processing unit 100 may also be configured not to perform any updating process on the generated support information. On the other hand, the arithmetic processing unit 100 may also be configured to automatically update the support information after a certain period of time has elapsed since the start of medical treatment.

[0053] (3) In the above embodiment, the arithmetic processing unit 100 was described as having a configuration in which it generates advance presentation information to be presented to the doctor performing the next medical examination in advance, based on the support presentation information and the patient information newly acquired by the patient information acquisition unit 20, after the completion of the medical examination but before the next medical examination. However, it is not limited to this configuration. The arithmetic processing unit 100 may be configured not to generate advance presentation information. In that case, it is preferable that the support presentation information generated in the previous medical examination is displayed on the display unit 80 when the medical examination is performed.

[0054] (4) In the above embodiment, the arithmetic processing unit 100 was described as having a configuration in which, when multiple unperformed medical procedures are extracted, support information is generated in such a way that the unperformed medical procedures of high importance are output preferentially. However, the arithmetic processing unit 100 may generate support information in such a way that, when multiple unperformed medical procedures are extracted, all unperformed medical procedures are output regardless of whether they are of high importance or not.

[0055] (5) The configurations disclosed in each of the embodiments described above can be applied in combination with configurations disclosed in other embodiments, as long as no inconsistencies arise. With regard to other configurations, the embodiments disclosed herein are merely illustrative in all respects. Therefore, various modifications can be made as appropriate without departing from the spirit of this disclosure.

[0056] [Summary of the above embodiments] The following is a summary of the medical support system described above.

[0057] The medical support system related to this disclosure is a medical support system that assists physicians in providing medical care, The audio information acquisition unit acquires audio information from the audio recorded during the medical examination, A patient information acquisition unit acquires patient information, which is information about the patient receiving the aforementioned medical treatment. A medical treatment determination unit outputs a list of medical treatments that are potentially effective for the patient, based on the medical treatment information which is the information necessary to perform the aforementioned medical treatment. A calculation processing unit that generates support information to be presented to the physician, It comprises an information output unit that outputs the aforementioned support display information, The aforementioned arithmetic processing unit is Information conversion processing that converts the voice information acquired by the voice information acquisition unit into processing information that can be recognized by the medical diagnosis determination unit, Input processing involves inputting the processing information and the patient information into the medical treatment determination unit as the medical treatment necessary information. An extraction process to extract the medical procedures performed by the physician to the patient during the consultation, based on the aforementioned audio information, Based on the difference between the medical procedure list output by the medical procedure determination unit and the medical procedures performed extracted by the extraction process, the system extracts unperformed medical procedures, which are medical procedures that the physician may not have performed during the medical treatment, and performs a support information generation process to generate support information that includes the unperformed medical procedures.

[0058] In this configuration, processing information converted from voice information and patient information are used as necessary medical information, and a list of medical procedures is output based on this necessary medical information. Then, support information is generated to include unperformed medical procedures extracted based on the difference between the list of medical procedures and the medical procedures performed, which are extracted based on the voice information. Therefore, by checking the outputted support information, physicians can reduce the possibility of completing a consultation without performing medical procedures that should have been performed for the patient. Thus, the quality of medical care provided by physicians can be improved. Thus, this configuration allows for the provision of appropriate medical care tailored to each patient.

[0059] Here, the patient information includes medical record information relating to the patient. Preferably, the processing information includes medical information related to the patient's medical treatment, extracted from the audio information.

[0060] According to this configuration, a list of medical procedures is generated based on the patient's medical record information and medical-related information extracted from conversations during the consultation. Therefore, the medical procedures necessary for the patient can be accurately included in the list of medical procedures.

[0061] Furthermore, if the arithmetic processing unit includes agreements made between the patient and the physician in the voice information, it is preferable that the support presentation information be generated to include such agreements.

[0062] This configuration allows for the inclusion of verbal agreements between the patient and the doctor in the support information provided. Therefore, the doctor can easily confirm the agreements made during the consultation.

[0063] Furthermore, the calculation processing unit generates advance presentation information to be presented in advance to the physician performing the next consultation, based on the support presentation information and the patient information newly acquired by the patient information acquisition unit, after the completion of the consultation and before the next consultation. The information output unit may further output the pre-presented information.

[0064] This configuration allows physicians to conduct medical treatment by reviewing pre-examination information generated based on support information created during the previous consultation and newly acquired patient information. Therefore, physicians can more easily select appropriate medical procedures tailored to the patient during the consultation. This effectively improves the quality of medical care.

[0065] Furthermore, the calculation processing unit can update the generated support information in real time during the medical examination. Preferably, the information output unit is configured to output the updated support information in real time.

[0066] With this configuration, support information is updated in real time during medical consultations. Therefore, physicians can check the latest support information, which effectively improves the quality of medical care.

[0067] Furthermore, when the calculation processing unit extracts multiple unperformed medical procedures, it is preferable that it generates the support information so that the unperformed medical procedures of higher importance are output preferentially.

[0068] This configuration allows for the prioritization of outputting only the most important unperformed medical procedures among multiple unperformed procedures. This makes it easy for physicians to identify the most important unperformed procedures, even when a large number are identified. Therefore, it facilitates improvements in the quality of medical care.

[0069] The technology relating to this disclosure includes a method comprising each step performed in the above system, and a program for causing a computer to perform each process performed in the above system, and the above effects can also be achieved by these methods and programs. [Industrial applicability]

[0070] The technology disclosed herein can be used in medical support systems, medical support methods, and medical support programs. [Explanation of Symbols]

[0071] 10: Voice Information Acquisition Unit 20: Patient Information Acquisition Department 30: Medical evaluation department 40: Information output unit 100: Arithmetic Processing Unit

Claims

1. A medical support system that assists physicians in providing medical care, The audio information acquisition unit acquires audio information from the audio recorded during the medical examination, A patient information acquisition unit acquires patient information, which is information about the patient receiving the aforementioned medical treatment. A medical treatment determination unit that uses a trained model to output a list of medical treatments that are potentially effective for the patient based on medical treatment information, which is the information necessary to perform the medical treatment, and A computation processing unit that generates support information to be presented to the physician using the aforementioned trained model, It comprises an information output unit that outputs the aforementioned support display information, The aforementioned arithmetic processing unit is Information conversion processing that converts the voice information acquired by the voice information acquisition unit into processing information that can be recognized by the medical diagnosis determination unit, Input processing involves inputting the processing information and the patient information into the medical treatment determination unit as the medical treatment necessary information. An extraction process to extract the medical procedures performed by the physician to the patient during the consultation, based on the aforementioned audio information, A process to generate a list of medical procedures performed, which is a list of the medical procedures performed extracted by the extraction process, The medical treatment determination unit outputs the medical treatment procedure list and the medical treatment procedure list that was performed, and the medical treatment procedure list that is not included in the medical treatment procedure list is extracted as unperformed medical treatment procedures, which are medical treatment procedures that the physician may not have performed during the treatment, and the support information generation process is executed to generate the support information to include the unperformed medical treatment procedures. A medical support system that, in the support information generation process, executes a medical evaluation process to generate medical evaluation information that evaluates the physician's response to the medical treatment and its medical appropriateness based on the medical information required for the medical treatment and the medical treatment not performed.

2. The aforementioned patient information includes medical record information relating to the patient. The medical support system according to claim 1, wherein the processing information includes medical information related to the patient's medical treatment extracted from the voice information.

3. The medical support system according to claim 1, wherein the calculation processing unit generates the support presentation information to include the agreements made between the patient and the doctor when the voice information includes such agreements.

4. The calculation processing unit, after the completion of the medical treatment but before the next medical treatment, uses the support presentation information and the patient information newly acquired by the patient information acquisition unit as a basis. , generate advance information to be presented to the physician who will perform the next consultation, The medical support system according to claim 1, wherein the information output unit further outputs the pre-presented information.

5. The calculation processing unit is capable of updating the generated support information in real time during the medical treatment. The medical support system according to claim 1, wherein the information output unit is configured to output the updated support information in real time.

6. The medical support system according to claim 1, wherein the calculation processing unit generates the support presentation information such that, when a plurality of unperformed medical procedures are extracted, the unperformed medical procedures of high importance are output preferentially.

7. The medical support system according to claim 1, wherein the calculation processing unit generates the support information in the support information generation process such that the unperformed medical procedures are clearly indicated as problems in the medical treatment and solutions corresponding to those problems.

8. A medical support method comprising each step performed in the medical support system according to any one of claims 1 to 7.

9. A medical support program for causing a computer to perform each of the processes performed in the medical support system according to any one of claims 1 to 7.

Images