Medical information processing device, medical information processing method, and medical information processing program
The medical information processing system infers patient thoughts and behaviors to create timely consultation opportunities, addressing the challenge of unrecognized medication side effects, enhancing patient care outside the hospital setting.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- CANON KK
- Filing Date
- 2025-01-07
- Publication Date
- 2026-07-17
AI Technical Summary
Patients often fail to recognize symptoms associated with medication side effects or their urgency, making it difficult for them to consult a medical professional until their next appointment, leading to potential health issues.
A medical information processing system that utilizes a trained model to infer patient thoughts and behaviors based on medical and behavioral information, comparing these inferences with actual patient expressions to create timely consultation opportunities through a healthcare professional's terminal.
Enables patients to receive timely consultations without needing to visit the hospital, improving communication quality during follow-up appointments by understanding patient wishes in advance.
Smart Images
Figure 2026119561000001_ABST
Abstract
Description
Technical Field
[0001] The embodiments disclosed in this specification and the drawings relate to a medical information processing device, a medical information processing method, and a medical information processing program.
Background Art
[0002] After examining a patient, a doctor comprehensively judges the patient's condition, treatment schedule, etc. and determines the next appointment date. Even if the prescribed medicine is ineffective or the patient dislikes the symptoms associated with side effects, the patient has to endure until the next appointment date or inquire at the hospital. However, it may be difficult for the patient to consult the hospital due to reasons such as not being able to recognize that the symptoms are due to side effects or not knowing the urgency of the symptoms.
Prior Art Documents
Patent Documents
[0003]
Patent Document 1
Patent Document 2
Patent Document 3
Summary of the Invention
Problems to be Solved by the Invention
[0004] One of the problems to be solved by the embodiments disclosed in this specification and the drawings is to provide the patient with a well-timed opportunity to consult a medical professional. However, the problems to be solved by the embodiments disclosed in this specification and the drawings are not limited to the above problems. The problems corresponding to the effects of each configuration shown in the embodiments described later can also be regarded as other problems.
Means for Solving the Problems
[0005] The medical information processing device according to this embodiment includes an inference information acquisition unit that acquires inference information, and an inference result information acquisition unit that acquires inference result information showing the result inferred by a trained model based on the inference information. The inference information includes the patient's medical information and the patient's behavioral information or the patient's thoughts information. [Brief explanation of the drawing]
[0006] [Figure 1] A diagram showing an example of the schematic configuration of a medical information processing system according to the embodiment. [Figure 2] A block diagram showing an example of the configuration of a medical information processing device according to the first embodiment. [Figure 3] A schematic diagram of a trained model according to the first embodiment. [Figure 4] A flowchart illustrating the medical information method according to the first embodiment. [Figure 5] A diagram showing an example of the display screen of a physician's terminal according to the first embodiment. [Figure 6] A diagram showing an example of the display screen of a physician's terminal according to the first embodiment. [Figure 7] A block diagram showing an example of the configuration of a medical information processing device according to the second embodiment. [Figure 8] A schematic diagram of a trained model according to the second embodiment. [Figure 9] A flowchart illustrating the medical information method according to the second embodiment. [Modes for carrying out the invention]
[0007] Embodiments of a medical information processing system, a medical information processing device, and a medical information processing method will be described below with reference to the drawings.
[0008] <Medical Information Processing System 1> Figure 1 shows an example of a schematic configuration of the medical information processing system 1 common to each embodiment. In the medical information processing system 1, the hospital (Hospital A) is connected to one or more cloud services CS via API integration through a network N such as the Internet. Although only one cloud service CS is shown in Figure 1, there are multiple cloud services CS depending on the type of service. For example, multiple cloud services such as a cloud service that provides SNS (Social Networking Service) and a cloud service that provides medication record services are connected to the hospital's medical information processing device 10 via API integration.
[0009] As shown in Figure 1, a medical information processing device 10, a physician terminal 20, and a medical information storage device 30 are connected to the in-hospital network of Hospital A.
[0010] As will be described in more detail later, the medical information processing device 10 acquires inference information (such as patient medical information and behavioral information) from the medical information storage device 30 and / or the cloud service CS, acquires inference result information from a trained model located inside or outside the medical information processing device 10, compares the inference result information with the actual information, and if the similarity between the two is high, displays the common points (matching parts) on the doctor's terminal 20.
[0011] The physician terminal 20 is a terminal device used by medical professionals such as physicians, and is, for example, a desktop or laptop computer installed in an examination room. The physician terminal 20 is communicatively connected to the medical information processing device 10 and the medical information storage device 30. The physician terminal 20 retrieves patient information (patient's name, age, gender, date of visit, examination information, medical history, etc.) from the electronic medical record stored in the medical information storage device 30 and displays it on the display. The physician terminal 20 also displays information received from the medical information processing device 10. The physician terminal 20 may also be a portable information terminal such as a smartphone or tablet carried by the medical professional.
[0012] The medical information storage device 30 is a device that stores medical information such as a patient's electronic medical record information and test results, and is communicably connected to the medical information processing device 10 and the doctor terminal 20.
[0013] The cloud service CS provides cloud services such as SNS, medication records, diet records, or blood glucose records to users (such as patients). The patient terminal PD is connected to the cloud service CS and transmits information input by the user, sensor information obtained by the sensors of the patient terminal PD, etc. to the cloud service CS. The patient terminal PD is a portable information terminal such as a smartphone or a tablet. In the case of the cloud service CS that provides SNS, articles posted by the patient are stored in the storage of the cloud service CS. Also, in the case of the cloud service CS that provides a medication record service, the medication record input by the patient to the patient terminal PD is stored in the storage of the cloud service CS.
[0014] Note that the patient terminal PD may be a wearable device such as a smartwatch, or a home medical device (such as a sphygmomanometer) having a communication function. Also, one patient may use a plurality of patient terminals PD such as a smartphone and a smartwatch.
[0015] Note that, as shown in FIG. 1, hospitals B and C other than hospital A may also be connected to the network N. Thereby, hospital A (such as the medical information processing device 10) can acquire the electronic medical record information, test records, etc. existing in hospitals B and C.
[0016] <Medical Information Processing Device According to the First Embodiment> While referring to the functional block diagram of FIG. 2, the medical information processing device 10 according to the first embodiment will be described in detail.
[0017] The medical information processing device 10 includes a communication interface 11, a memory circuit 12, and a processing circuit 13. Hereinafter, the details of each component will be described.
[0018] The communication interface 11 communicates with other components of the medical information processing system 1 (doctor's terminal 20, medical information storage device 30, cloud service CS, etc.) via the communication network according to various communication protocols.
[0019] The memory circuit 12 is connected to the processing circuit 13 and stores various types of information used by the processing circuit 13. The memory circuit 12 can be implemented using, for example, a semiconductor memory element such as RAM (Random Access Memory) or flash memory, a hard disk, or an optical disc. The memory circuit 12 stores various programs necessary for the processing circuit 13 to perform its various functions, as well as various types of data processed by these programs. In this specification, the various types of data dealt with are typically digital data.
[0020] The memory circuit 12 stores trained model data 12a, an adverse event information database 12b, and a medication record database 12c.
[0021] The trained model data 12a includes parameters and programs for constructing and executing a pre-trained model (inference model) generated in advance by machine learning such as deep learning. In this embodiment, the trained model data 12a is data for constructing a pre-trained model that takes a patient's medical information and behavioral information as input and outputs the patient's thoughts. For example, the medical information is information on side effects of medications the patient takes, and the behavioral information is the patient's medication record.
[0022] Patient information refers to information that patients have expressed since their most recent medical visit, such as by posting on social media. However, patient information is not limited to social media posts; it may also include information written by patients on blogs, online forums, review sites, comment sections, group chats, or information written in emails or chat messages sent by patients.
[0023] Furthermore, the trained model data 12a is not limited to being provided in the medical information processing device 10, but may also be provided in a cloud service CS that provides services such as generation AI.
[0024] The adverse drug reaction information database 12b is a database that stores information on drug side effects. For example, for drugs for high blood pressure, symptoms such as dizziness, headache, lightheadedness, and palpitations are stored as adverse drug reaction information, and for drugs for diabetes, symptoms such as loss of appetite and vomiting are stored as adverse drug reaction information. It should be noted that the adverse drug reaction information database 12b is not limited to being installed in the medical information processing device 10, but may also be installed in the medical information storage device 30 or on a server of a public institution (such as the Pharmaceuticals and Medical Devices Agency).
[0025] The medication record database 12c is a database that stores patients' medication records. The medication record database 12c is constructed and updated, for example, by the medical information processing device 10 collecting medication records from a cloud service CS that provides medication record services. The medication record database 12c may store medication records for multiple patients.
[0026] The processing circuit 13 is an arithmetic circuit that performs various calculations and controls the operation of the medical information processing device 10. The processing circuit 13 has an information acquisition function 13a, an information output function 13b, an inference function 13c, a comparison function 13d, and a consultation opportunity creation function 13e.
[0027] The information acquisition function 13a is an example of the inference information acquisition unit, inference result information acquisition unit, thought information acquisition unit, and action information acquisition unit in the claims. The information output function 13b is an example of the information output unit in the claims. The inference function 13c is an example of the inference unit in the claims. The comparison function 13d is an example of the comparison unit in the claims. The consultation opportunity creation function 13e is an example of the consultation opportunity creation unit in the claims.
[0028] In this embodiment, each processing function performed by the information acquisition function 13a, information output function 13b, inference function 13c, comparison function 13d, and consultation opportunity creation function 13e is stored in the memory circuit 12 in the form of a program that can be executed by a computer. The processing circuit 13 is composed of a processor and realizes the function corresponding to each program by reading and executing the program from the memory circuit 12. In other words, the processing circuit 13 in the state in which each program has been read will have the functions shown in the processing circuit 13 of Figure 2.
[0029] In Figure 2, the information acquisition function 13a, information output function 13b, inference function 13c, comparison function 13d, and consultation opportunity creation function 13e are shown to be implemented by a single processing circuit 13, but the embodiments are not limited to this. For example, the processing circuit 13 may be configured as a combination of multiple independent processors, and each processing function may be implemented by each processor executing each program. Each processing function of the processing circuit 13 may be implemented by appropriately distributing or integrating it across one or more processing circuits.
[0030] The information acquisition function 13a acquires desired information from the memory circuit 12, the medical information storage device 30, and the cloud service CS. For example, the information acquisition function 13a acquires patient prescription information (drug name, dosage, usage, etc.) from the electronic medical record in the medical information storage device 30. The information acquisition function 13a also acquires inference information (in this embodiment, patient medical information and behavioral information) necessary for performing inference using a trained model. The information acquisition function 13a also acquires inference result information (in this embodiment, inference thought information) output by the trained model. The information acquisition function 13a acquires patient thought information (actual thoughts) from the cloud service CS, such as an SNS.
[0031] The information output function 13b outputs various types of information to the physician's terminal 20, etc. For example, the information output function 13b outputs information for displaying electronic medical record information acquired from the medical information storage device 30 on the physician's terminal 20.
[0032] The inference function 13c performs inference based on the inference information and outputs the resulting inference result information. The inference function 13c performs inference using a trained model constructed based on the trained model data 12a. That is, the inference function 13c inputs the inference information into the trained model and obtains the inference result information output from the trained model. As shown in Figure 3, in this embodiment, the inference information is the patient's medical information and behavioral information. The inference result information is the inference result information inferred based on the medical information and behavioral information. Note that the trained model shown in Figure 3 is a schematic diagram, and the number of neurons in the input layer, hidden layer, and output layer, as well as the number of layers in the hidden layer, differ from the actual model.
[0033] Furthermore, the medical information processing device 10 does not necessarily have to have an inference function 13c, and may utilize an external inference function such as a generative AI. In this case, the medical information processing device 10 accesses a cloud service CS connected to the network N that provides services such as generative AI. In addition, the inference function 13c is not limited to directly inputting the patient's medical information and behavioral information into the trained model, but may also input input data created based on the patient's medical information and behavioral information into the trained model.
[0034] The comparison function 13d performs a comparison between the inference result information obtained by the inference function 13c and the actual information. In this embodiment, the comparison function 13d compares the thought information (information actually expressed by the patient) with the inferred thought information. For example, the comparison function 13d calculates the similarity between the thought information and the inferred thought information. If the similarity is greater than or equal to a predetermined value, the comparison function 13d determines that the thought information and the inferred thought information are similar.
[0035] The similarity score is calculated, for example, by extracting vectors for each word and calculating the cosine similarity. Methods such as SentenceBERT and BERTScore may be used.
[0036] The consultation opportunity creation function 13e creates opportunities for consultation between patients and medical professionals such as doctors, based on the comparison results of the comparison function 13d. For example, the consultation opportunity creation function 13e displays common points (matching parts) between the patient's thought information and the inferred thought information on the doctor's terminal 20. The consultation opportunity creation function 13e may also display means for providing consultation opportunities between patients and medical professionals on the doctor's terminal 20. As such means, the doctor's terminal 20 may display a button (see Figure 5) for sending an email to the patient terminal PD with a consultation invitation. In addition, an example message to be included in the email may be displayed on the doctor's terminal 20.
[0037] <Medical information processing method according to the first embodiment> Referring to the flowchart in Figure 4, an example of a medical information processing method according to the first embodiment will be described. The process shown in Figure 4 is executed by the medical information processing device 10 at predetermined intervals from the most recent consultation date (last visit date) to the next consultation date (next follow-up consultation date). Note that this process may also be performed even if the next follow-up consultation date is not scheduled. Furthermore, if there are multiple patients, this process is executed for each patient.
[0038] Step S11: The information acquisition function 13a acquires the patient's medical information (in this case, side effect information). Specifically, the information acquisition function 13a first acquires prescription information (drug name, dosage, usage, etc.) from the target patient's electronic medical record stored in the medical information storage device 30. Then, the information acquisition function 13a searches the side effect information database 12b using the drug name as a key and acquires side effect information for the prescribed drug. For example, if a drug for high blood pressure is prescribed, dizziness, headache, lightheadedness, palpitations, etc., will be acquired as side effect information. If the patient is using a medication record app, the information acquisition function 13a may acquire the patient's prescription information from the cloud service CS that provides the medication record service.
[0039] Step S12: The information acquisition function 13a acquires patient behavior information (in this case, medication records). More specifically, the information acquisition function 13a acquires medication record information from the medication record database 12c. For example, it acquires information that the patient takes their medication on the days determined by the medication's dosage (Monday, Wednesday, Friday, Sunday). Alternatively, the information acquisition function 13a may directly acquire the patient's medication record information via API from the cloud service CS that provides the medication record service.
[0040] Step S13: Obtain inferred thought information based on the medical information obtained in Step S11 and the behavioral information obtained in Step S12. In this example, the inference function 13c performs inference based on side effect information and medication records, and outputs the thought information obtained through inference (inferred thought information). The information acquisition function 13a acquires the inferred thought information. For example, inference result information such as "feeling dizzy," "having a headache," and "feeling lightheaded" is acquired.
[0041] In step S13, if the medical information processing device 10 is not equipped with an inference function 13c, the information output function 13b outputs medical information and behavioral information as inference information to a cloud service CS that provides services such as generation AI. Subsequently, the information acquisition function 13a acquires inference result information (inference information) from the cloud service CS.
[0042] Step S14: The information acquisition function 13a acquires the patient's thoughts (actual thoughts). Specifically, the information acquisition function 13a acquires the target patient's posted information from a cloud service CS such as SNS. Specifically, it acquires information posted by the patient since the most recent visit date. For example, it acquires posted information such as, "I have a headache and feel dizzy." If the thoughts information collected and stored from the cloud service CS is stored in the memory circuit 12 as a thoughts information database (not shown), the information acquisition function 13a may acquire the thoughts information from that database.
[0043] Step S15: The comparison function 13d compares the inferred thought information obtained in step S13 with the thought information obtained in step S14. The comparison function 13d calculates the similarity between the inferred thought information and the thought information using methods such as SentenceBERT and BERTScore.
[0044] Step S16: The comparison function 13d determines whether the inferred thought information and the thought information are similar. Specifically, if the similarity calculated in step S15 is greater than or equal to a predetermined value, the comparison function 13d determines that the inferred thought information and the thought information are similar. If the determination in this step is Yes, the process proceeds to step S17; otherwise, the process ends.
[0045] Step S17: The consultation opportunity creation function 13e creates opportunities for consultation between patients and healthcare professionals. For example, the consultation opportunity creation function 13e sends a consultation notification to the patient's terminal PD.
[0046] Furthermore, the decision of a healthcare professional may be involved in sending the consultation notice. For example, the consultation opportunity creation function 13e causes the physician terminal 20 to display a means for sending the consultation notice to the patient's terminal PD. As such a means, a button B for sending the consultation notice to the patient may be displayed, as shown in the display screen I1 of Figure 5. For example, if the physician considers the content of the patient's wishes and the medical schedule and determines that it is urgent, they press button B. This sends the consultation notice to the patient's terminal PD. The consultation notice may include a message suggesting that the patient move up the consultation date, a link to the appointment booking site, etc.
[0047] As shown in Figure 5, screen I1 displays basic information such as the patient's name (John Doe), age (34 years old), and gender (male), as well as the date of the last visit and the date of the next follow-up appointment. Screen I1 is the screen with the Summary Report tab selected. It is also possible to select other tabs (Consultation Information, Questionnaire Results, Medical History, Impact on Daily Life, Assistant) and display the information for each.
[0048] Figure 6 shows screen I2 with the Assistant tab selected. On screen I2, "Recommended Example Sentences" are displayed as a function to support questioning patients during consultations. These example sentences may be generated by a generation AI. The consultation opportunity creation function 13e may also include example sentences for questioning support in the consultation guidance when creating the consultation guidance. The example sentences to be included in the consultation guidance may be selected by the healthcare professional. For example, the healthcare professional can select sentences to include in the consultation guidance by clicking the checkbox CB for the example sentences.
[0049] Furthermore, the consultation opportunity creation function 13e may display commonalities between the patient's thoughts and the inferred thoughts on the physician's terminal 20. On screen I1, the commonalities between the patient's thoughts and the inferred thoughts are shown as "feeling dizzy and having a headache." By displaying such information on the physician's terminal 20, healthcare professionals can better understand the patient's condition and support the patient even outside the hospital.
[0050] As explained above, according to the first embodiment, the patient's thoughts are inferred based on the patient's medical information and behavioral information, and the thoughts actually expressed by the patient are compared with the thoughts obtained through inference. If the two are similar, an opportunity for consultation between the patient and a healthcare professional is created. This allows patients to be provided with timely opportunities for consultation with healthcare professionals without having to access the hospital. Since consultation opportunities are created when the inferred thoughts and actual thoughts are similar, situations where consultation opportunities are created when the patient does not need consultation are avoided as much as possible, and consultation opportunities can be provided to patients with high accuracy.
[0051] According to the first embodiment, hospitals (healthcare workers) can support patients even outside the hospital. Furthermore, even if a notification of an appointment is not sent to the patient, healthcare workers such as the attending physician can improve the quality of communication with the patient during scheduled follow-up appointments by understanding the patient's wishes in advance.
[0052] In the above explanation, medical information was side effect information and behavioral information was medication records, but this embodiment is not limited to this. For example, medical information may be side effect information and behavioral information may be meal record information and / or blood glucose level information. This allows for inference of the patient's feelings, such as not wanting to eat. Meal record information is, for example, meal record information (such as daily calorie intake) entered into the patient terminal PD. Blood glucose level information is, for example, blood glucose levels obtained from the patient terminal PD, such as a wearable device. As another example, medical information may be body composition information such as limb skeletal muscle mass, and behavioral information may be calendar information and / or exercise level information. This allows for inference of the patient's feelings, such as body pain or itchiness due to muscle soreness. Calendar information is, for example, information showing the patient's schedule entered into the calendar app of the patient terminal PD. Exercise level information is, for example, information showing the amount of exercise, such as steps taken, calories burned, and exercise time, obtained from the patient terminal PD.
[0053] Furthermore, in the description of the first embodiment above, a consultation opportunity was created when the actual thought information and the inferred thought information were similar. However, the consultation opportunity creation function 13e may create a consultation opportunity between a healthcare professional and a patient based on the inferred result information. For example, if the inferred result information indicates that the patient is unwell, the consultation opportunity creation function 13e sends a notice to the patient terminal PD to request a medical consultation. Alternatively, the consultation opportunity creation function 13e may cause the physician terminal 20 to display a means for sending a notice to the patient terminal PD to request a medical consultation.
[0054] <Medical information processing device according to the second embodiment> Next, the medical information processing device 10A according to the second embodiment will be described in detail with reference to the functional block diagram in Figure 7. In Figure 7, components having the same function are denoted by the same reference numerals as in Figure 2. Since the medical information processing device 10A according to the second embodiment has the same configuration as the medical information processing device 10 described in the first embodiment, only the differences will be described.
[0055] The memory circuit 12 stores not only the side effect information database 12b described in the first embodiment, but also the trained model data 12d and the memory information database 12e.
[0056] The trained model data 12d includes parameters and programs for constructing a pre-trained model (inference model) generated in advance by machine learning such as deep learning. The trained model data 12d is data for constructing a trained model that takes a patient's medical information and thought information as input and outputs the patient's behavioral information. Note that the trained model data 12d is not limited to being provided on the medical information processing device 10A, but may also be provided on an external inference server or the like that provides services such as generation AI.
[0057] The thoughts information database 12e is a database that stores patient thoughts information. The thoughts information database 12e is constructed and updated, for example, by the medical information processing device 10 collecting patient posting information from a cloud service CS that provides social networking services. The thoughts information database 12e may store thoughts information for multiple patients.
[0058] In the second embodiment, the inference function 13c inputs inference information into a trained model composed of trained model data 12d and obtains inference result information output from the trained model. In this embodiment, as shown in Figure 8, the inference information is the patient's medical information and thought information. The inference result information is inferred behavior information inferred based on the medical information and thought information. Note that the trained model shown in Figure 8 is a schematic diagram, and the number of neurons in the input layer, hidden layer, and output layer, as well as the number of layers in the hidden layer, differ from the actual model.
[0059] <Medical information processing method according to the second embodiment> Referring to the flowchart in Figure 9, an example of a medical information processing method according to the second embodiment will be described. The process shown in Figure 9 is performed by the medical information processing device 10 at predetermined intervals from the patient's last visit date to their next visit date. This process may also be performed even if no next follow-up visit is scheduled. Furthermore, if there are multiple patients, this process is performed for each patient.
[0060] Step S21: The information acquisition function 13a acquires the patient's medical information (in this case, side effect information). The details of this step are the same as those of step S11 described above, so the explanation is omitted.
[0061] Step S22: Information acquisition function 13a acquires patient information. Specifically, information acquisition function 13a acquires the patient's posted information from cloud services such as SNS. In particular, information posted by the patient since the most recent medical consultation date is acquired.
[0062] Step S23: Obtain inferred behavioral information inferred based on the medical information obtained in Step S21 and the thought information obtained in Step S22. Specifically, the inference function 13c performs inference based on the medical information and thought information and outputs the patient's behavioral information (inferred behavioral information) obtained through inference. The information acquisition function 13a acquires the inferred behavioral information. In this case, the inferred behavioral information is the inference result of the medication record.
[0063] In step S23, if the medical information processing device 10 is not equipped with an inference function 13c, the information output function 13b outputs the medical information and thought information as inference information to the cloud service CS which provides services such as generation AI. Subsequently, the information acquisition function 13a acquires the inference result information (inference behavior information) from the cloud service CS.
[0064] Step S24: Information acquisition function 13a acquires patient behavior information (actual behavior). Specifically, information acquisition function 13a acquires medication record information of the target patient via API from the cloud service CS, which provides medication record services.
[0065] Step S25: The comparison function 13d compares the inferred behavior information obtained in step S23 with the behavior information obtained in step S24. For example, the comparison function 13d calculates the similarity between the inferred behavior information and the behavior information.
[0066] Step S26: The comparison function 13d determines whether the inferred behavior information and the behavior information are similar. Specifically, if the similarity calculated in step S25 is greater than or equal to a predetermined value, the comparison function 13d determines that the inferred behavior information and the behavior information are similar. If the determination in this step is Yes, the process proceeds to step S27; otherwise, the process ends.
[0067] Step S27: The consultation opportunity creation function 13e creates opportunities for consultation between patients and healthcare professionals. For example, the consultation opportunity creation function 13e causes the physician terminal 20 to display a means for sending a consultation notification to the patient's terminal PD. The consultation opportunity creation function 13e may also display commonalities between behavioral information and inferred behavioral information on the physician terminal 20. The consultation opportunity creation function 13e may also display dates and days of the week when the patient has not taken their medication despite it being the day they should have taken it. This allows physicians to send emails or other messages to the patient urging them to take their medication properly.
[0068] Furthermore, the consultation opportunity creation function 13e may display on the physician terminal 20 a means (for example, button B mentioned above) for sending a consultation invitation to the patient terminal PD. In addition, the consultation opportunity creation function 13e may display on the physician terminal 20 recommended example sentences, such as "Are you taking your medication properly?", to support the physician in asking questions to the patient. Healthcare professionals may select a sentence they deem appropriate from these example sentences and include it in the consultation invitation sent to the patient.
[0069] As explained above, according to the second embodiment, the patient's behavioral information is inferred based on the patient's medical information and belief information, and the patient's actual behavioral information is compared with the behavioral information obtained through inference. If the two are similar, an opportunity for consultation between the patient and a healthcare professional is created. This makes it possible to provide patients with timely opportunities for consultation with healthcare professionals without them having to access the hospital. Since the consultation opportunity is created when the inferred behavioral information and actual behavioral information are similar, it is possible to avoid creating a consultation opportunity when the patient does not need it as much as possible, and to provide consultation opportunities with high accuracy.
[0070] According to the second embodiment, hospitals can support patients even outside the hospital. Furthermore, even if a notification of an upcoming appointment is not sent to the patient, the quality of communication with the patient during scheduled follow-up appointments can be improved by having the attending physician or other healthcare professionals understand the patient's wishes in advance.
[0071] In the description of the second embodiment above, a consultation opportunity was created when the actual behavioral information and the inferred behavioral information were similar. However, the consultation opportunity creation function 13e may create a consultation opportunity between a healthcare professional and a patient based on the inferred result information. For example, if the inferred result information indicates that the patient is not taking medication according to the instructions or is not eating properly, the consultation opportunity creation function 13e sends a notice to the patient terminal PD. Alternatively, the consultation opportunity creation function 13e may cause the physician terminal 20 to display a means for sending a notice to the patient terminal PD.
[0072] In the above explanation, the term "processor" refers to circuits such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an Application Specific Integrated Circuit (ASIC), or a programmable logic device (e.g., a Simple Programmable Logic Device (SPLD), a Complex Programmable Logic Device (CPLD), and a Field Programmable Gate Array (FPGA)). The processor functions by reading and executing a program stored in the memory circuit 12. Alternatively, instead of storing the program in the memory circuit 12, the processor may be configured to directly incorporate the program into its circuitry. In this case, the processor functions by reading and executing the program incorporated into the circuitry.
[0073] Furthermore, the medical information processing method described in Figures 4 and 9 can be implemented by executing a pre-prepared medical information processing program on a computer such as a personal computer or workstation. This medical information processing program can be distributed via a network such as the Internet. Alternatively, this medical information processing program can be recorded on a computer-readable, non-transient recording medium such as a hard disk, flexible disk (FD), CD-ROM, MO, or DVD, and executed by reading it from the recording medium by a computer.
[0074] Although several embodiments have been described above, these embodiments are presented only as examples and are not intended to limit the scope of the invention. The novel apparatus and methods described herein can be implemented in a variety of other forms. Furthermore, various omissions, substitutions, and modifications can be made to the embodiments of the apparatus and methods described herein, without departing from the spirit of the invention. The appended claims and equivalents are intended to include such embodiments and modifications that are included in the scope and spirit of the invention. [Explanation of Symbols]
[0075] 1. Medical Information Processing System 10,10A Medical Information Processing Device 11 Communication Interface 12 Memory circuit 12a, 12d Pre-trained model data 12b Adverse Reaction Information Database 12c Medication Record Database 12e Thoughts Information Database 13 Processing Circuit 13a Information acquisition function 13b Information output function 13c Inference Function 13d Comparison Function 13e Function to create opportunities for consultation 20 Doctor's terminal 30 Medical information storage device B button CS Cloud Service I1,I2 display screen N Network PD patient terminal
Claims
1. An inference information acquisition unit that acquires inference information including the patient's medical information and the patient's behavioral information or the patient's thoughts information, An inference result information acquisition unit acquires inference result information that shows the results inferred by a trained model based on the aforementioned inference information. A medical information processing device equipped with [a specific feature].
2. The inference information includes the medical information and the behavioral information, The medical information processing apparatus according to claim 1, wherein the inference result information is inferred information inferred based on the medical information and the behavioral information.
3. A thought information acquisition unit that acquires thought information of the aforementioned patient, A comparison unit that compares the aforementioned thought information with the aforementioned inferred thought information, The medical information processing apparatus according to claim 2, further comprising the above.
4. The medical information processing apparatus according to claim 3, further comprising a consultation opportunity creation unit that, when the comparison unit determines that the thought information and the inferred thought information are similar, causes the medical professional's terminal to display means for sending a consultation guidance to the patient's terminal.
5. The medical information processing apparatus according to claim 3, further comprising a consultation opportunity creation unit that sends a consultation guidance to the patient's terminal when the comparison unit determines that the thought information and the inferred thought information are similar.
6. The medical information processing device according to claim 3, further comprising a consultation opportunity creation unit that, when the comparison unit determines that the thought information and the inferred thought information are similar, displays the commonalities between the thought information and the inferred thought information on the terminal of a medical professional.
7. The aforementioned inference information includes the medical information and the aforementioned thought information, The medical information processing device according to claim 1, wherein the inference result information is inferred behavior information inferred based on the medical information and the thought information.
8. A behavioral information acquisition unit that acquires the behavioral information of the aforementioned patient, A comparison unit that compares the aforementioned behavioral information with the aforementioned inferred behavioral information, The medical information processing device according to claim 7, further comprising the above.
9. The medical information processing apparatus according to claim 8, further comprising a consultation opportunity creation unit that, when the comparison unit determines that the behavioral information and the inferred behavioral information are similar, causes the medical professional's terminal to display means for sending a consultation guidance to the patient's terminal.
10. The medical information processing apparatus according to claim 8, further comprising a consultation opportunity creation unit that sends a consultation guidance to the patient's terminal when the comparison unit determines that the behavioral information and the inferred behavioral information are similar.
11. The medical information processing device according to claim 8, further comprising a consultation opportunity creation unit that, when the comparison unit determines that the behavioral information and the inferred behavioral information are similar, displays the commonalities between the behavioral information and the inferred behavioral information on the terminal of a medical professional.
12. The medical information processing apparatus according to claim 1, wherein the aforementioned information is information posted by the patient on social media.
13. The medical information processing device according to claim 1, wherein the behavioral information includes at least one of the patient's medication record, meal record, calendar information, and exercise information.
14. The medical information processing device according to claim 1, further comprising a consultation opportunity creation unit that creates a consultation opportunity between a medical professional and the patient based on the aforementioned inference result information.
15. The medical information processing device according to claim 14, wherein the consultation opportunity creation unit causes the terminal of a medical professional to display means for sending a consultation guidance to the terminal of the patient.
16. The medical information processing device according to claim 14, wherein the consultation opportunity creation unit transmits a consultation guidance to the patient's terminal.
17. Obtain inference information including the patient's medical information and the patient's behavioral information or the patient's thoughts information. Based on the aforementioned inference information, obtain inference result information showing the results inferred by the trained model. Medical information processing method.
18. On the computer, Obtain inference information including the patient's medical information and the patient's behavioral information or the patient's thoughts information. A medical information processing program that performs the task of obtaining inference result information, which shows the results inferred by a trained model based on the aforementioned inference information.