Information processing system, information processing device, information association method, learning method, and information association program

Abstract

This information processing system comprises: an interface device that acquires medical treatment information related to medical treatment for a user and medical interview information related to symptoms of the user before and after the medical treatment; a memory in which the medical treatment information and the medical interview information are recorded; and a processor. The processor determines the treatment effect of the medical treatment for the user on the basis of the medical interview information acquired by the interface device, and records in the memory the medical treatment information, the medical interview information, and the treatment effect in association with each other.

Description

Information processing system, information processing device, information association method, learning method, and information association program

[0001] The present invention relates to an information processing system, an information processing device, an information association method, a learning method, and an information association program for supporting treatment in a medical institution.

[0002] In recent years, medical institutions have begun to digitize various types of information, such as medical records, making it easier to manage and use patient medical records. For example, patients who fall ill may visit the hospital multiple times as needed to receive treatment. Hospitals record the surgeries, examinations, and other treatments that patients undergo in their medical records. Electronic medical records make it easy to search and share patient information, and by referring to the medical records at the time of diagnosis, doctors can understand the patient's progress and provide more appropriate treatment to the patient.

[0003] As a method for confirming whether a doctor's treatment was appropriate, Japanese Patent Application Publication No. 2022-180692 (hereinafter referred to as Patent Document 1) proposes a technology for confirming the effectiveness of medication prescribed to a patient.

[0004] Japanese Patent Application Laid-Open No. 2022-180692

[0005] However, doctors cannot always examine patients until their symptoms are completely cured. For example, even if a patient is not completely cured, they may stop visiting the hospital or change hospitals if their symptoms improve slightly or they are short on time. As a result, the information recorded at the hospital alone may not determine why the patient stopped visiting the hospital, and it may remain unclear whether the hospital's treatment was effective for the patient's symptoms. Furthermore, patients may self-prescribe medication before visiting the hospital, making it difficult to determine the optimal treatment, etc., by comprehensively considering the patient's condition from the past to the present. However, it is difficult to require the patient to input all of this information. In consideration of these issues, the present invention aims to provide an information processing system, information processing device, information association method, learning method, and information association program that can easily obtain knowledge about effective medical procedures tailored to the patient's condition.

[0006] An information processing system according to one aspect of the present invention includes an interface device that acquires medical treatment information regarding a medical treatment performed on a user and medical interview information regarding the user's symptoms before and after the medical treatment, a memory that records the medical treatment information and the medical interview information, and a processor, wherein the processor determines the treatment effectiveness of the medical treatment performed on the user based on the medical interview information acquired by the interface device, and records the medical treatment information, the medical interview information, and the treatment effectiveness in association with each other in the memory.

[0007] An information processing device according to one aspect of the present invention comprises an interface device having an information acquisition unit that acquires medical treatment information regarding medical treatment for a user, an information presentation unit that presents information to the user, and a processor, wherein when the information acquisition unit acquires the medical treatment information, the processor creates a questionnaire based on the medical treatment information to acquire information regarding the user's symptoms before and after the medical treatment for the user, and causes the information presentation unit to present the questionnaire.

[0008] An information association method according to one aspect of the present invention is an information association method for associating information on subjective symptoms and treatment effects, and includes a treatment result acquisition step for acquiring treatment results for a patient in accordance with medical procedure information, a treatment effect confirmation step for confirming the treatment effects based on changes in the subjective symptoms after the treatment corresponding to the treatment, and a step for associating the treatment results with changes in the patient's subjective symptoms before and after the treatment.

[0009] A learning method according to one aspect of the present invention involves training a neural network using training data obtained by annotating a data set consisting of two pieces of data: initial symptoms before medical intervention, the time from the onset of the initial symptoms to the medical intervention, and at least one of the medical interventions, and annotating the time until recovery is realized, and using this training data to construct an inference model.

[0010] An information association program according to one aspect of the present invention is an information association program for associating information on subjective symptoms and treatment effects, and causes a computer to execute a treatment result acquisition step of acquiring treatment results for a patient in accordance with medical procedure information, a treatment effect confirmation step of confirming the treatment effects based on changes in the subjective symptoms after the treatment corresponding to the treatment, and a step of associating the treatment results with changes in the patient's subjective symptoms before and after the treatment.

[0011] The present invention has the effect of making it possible to obtain knowledge about effective medical treatment suited to the patient's condition.

[0012] FIG. 1 is a block diagram showing an information processing system according to an embodiment of the present invention. FIG. 2 is an explanatory diagram showing an example of a message presented to a user by a control unit 11. FIG. 3 is an explanatory diagram for explaining an example of knowledge information in an external knowledge DB 19. FIG. 4 is a flowchart for explaining the operation of an embodiment. FIG. 5 is an explanatory diagram showing an example of a display by an information presenting unit 17 using knowledge information from an external knowledge DB 19. FIG. 6 is an explanatory diagram showing an example of an inference model created by a learning unit 19c.

[0013] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[0014] (Embodiment) Figure 1 is a block diagram showing an information processing system according to one embodiment of the present invention. This embodiment acquires information about medical procedures, such as medication, surgery, and treatment, from past to present, for a patient (hereinafter referred to as medical procedure information), as well as information about the patient's symptoms, health, and so on (hereinafter referred to as medical interview information), and determines the effectiveness of the medical procedure based on this information, making it possible to present the determination result to a doctor or other such person as a treatment result. This allows a doctor or other such person to gain knowledge about effective medical procedures according to the patient's condition. Furthermore, this embodiment makes it possible to obtain knowledge about effective medical procedures for a case by utilizing a knowledge database based on the processing results.

[0015] 1 , the mobile terminal 10 can be operated by, for example, the patient himself / herself. Examples of the mobile terminal include various information processing terminals such as a smartphone, a tablet, and a personal computer (hereinafter referred to as a PC). The mobile terminal 10 includes a control unit 11, a sensor unit 12, an information input unit 13, a treatment result determination unit 14, a database linkage unit 15, an information processing unit 16, an information presentation unit 17, and an information recording unit 18. Each unit of the mobile terminal 10, such as the control unit 11, may be configured by a processor using a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an FPGA (Field Programmable Gate Array), or the like, may operate according to a program stored in a memory (not shown) to control each unit, or may implement some or all of its functions using hardware electronic circuits.

[0016] The control unit 11 receives user operations on input operation units such as a keyboard or a touchpad (not shown) provided on the mobile terminal 10, and controls each unit of the mobile terminal 10 based on the user operations. The control unit 11 also controls each unit of the mobile terminal 10 in accordance with an application running on the mobile terminal 10.

[0017] The sensor unit 12 and the information input unit 13 are interface devices that acquire various information and constitute an information acquisition unit. The sensor unit 12 can be configured with one or more sensors, such as a camera or a microphone. The sensor unit 12 is controlled by the control unit 11 to acquire medical treatment information related to the medical treatment provided to the user. For example, if the sensor unit 12 is configured with a camera, the sensor unit 12 captures an image of a medical receipt, such as a medical statement, or a medical chart, thereby acquiring the medical treatment information printed on the medical receipt. The medical receipt includes receipts issued by the medical institution (hospital) where the medical treatment was performed, as well as receipts issued when purchasing over-the-counter drugs. The control unit 11 recognizes character strings and the like written on the medical receipt or the like using known image analysis of the captured image, acquires the user's own medical treatment information based on the recognition result, and records the acquired information in the information recording unit 18.

[0018] Medical treatment information is information about medical treatments performed on the user. For example, medical treatment information includes information about medical institutions visited by the user, information about treatments performed at medical institutions (treatment information), and information about prescriptions (prescription information). Prescription information includes not only prescription drugs but also information about over-the-counter drugs, i.e., medication information about the user's own medication, including information about the type of medication and the status of administration. Although supplements are different from pharmaceuticals, in this embodiment, the medication information may also include supplement intake as needed. Even if medical treatment information is not clearly stated on a medical receipt or the like, if the medical treatment information can be inferred based on the information stated on the medical receipt or the like, the control unit 11 may record the inferred medical treatment information in the information recording unit 18.

[0019] As described above, medical treatment information is generated in various situations, so it is sufficient to be able to acquire and record it as needed. For example, medical treatment information may be recorded in the information recording unit 18 of the mobile terminal 10, or the medical treatment information recording function may be provided in a recording unit of an in-hospital system (not shown), a recording unit of a pharmacy system, or the external knowledge database 19 (shown). In this case, an input unit may be provided in the in-hospital system, and the medical treatment information input by the input unit may be sent to the mobile terminal 10.

[0020] The information input unit 13 may be configured with an input operation unit such as a keyboard or a touchpad, or may be configured with various interfaces for receiving data from outside. The information input unit 13 acquires medical interview information through user input operations or reception of external data, and outputs the information to the control unit 11. For example, the information input unit 13 may acquire the medical interview information through text input or voice input by the user. The control unit 11 provides the medical interview information to the information recording unit 18 for recording. Medical procedure information may also be obtained in cooperation with a system of a hospital or pharmacy.

[0021] The medical interview information includes information such as the user's medical history, symptoms, lifestyle, and medical history. In this embodiment, the information includes not only information obtained as a result of a medical interview by a doctor or the like, but also information about the characteristics and progression of symptoms (subjective symptoms) experienced by the user from the past to the present, as assessed by the user. Furthermore, in this embodiment, the medical interview information includes information about the user's behavioral history, such as sleep and exercise, and biometric information, such as the user's body temperature, pulse, and blood pressure, although these are not considered subjective symptoms. Furthermore, the medical interview information may include information about the user's lifestyle and preferences (hereinafter, referred to as lifestyle information). For example, lifestyle information about the user's illness, such as drinking and smoking, may be acquired as medical interview information. Because subjective symptoms may sometimes be unconscious, the medical interview content may be selected rather than simply written down, to raise patient awareness. For such a medical interview, the medical interview items may be inferred using an inference engine or the like, or the contents of the medical interview items may be recorded in a database and selectively read and displayed.

[0022] While the example in which medical treatment information is acquired by the sensor unit 12 and medical interview information is acquired by the information input unit 13 has been described, the sensor unit 12 may be configured to acquire not only medical treatment information but also medical interview information, and the information input unit 13 may be configured to acquire not only medical interview information but also medical treatment information. For example, the sensor unit 12 may acquire biometric information (medical interview information) such as pulse rate. For example, if the user owns a biometric information acquisition device capable of acquiring biometric information, such as a smartwatch, medical interview information such as symptoms may be acquired from this biometric information acquisition device. Furthermore, medical treatment information may be acquired by inputting information into the information input unit 13.

[0023] The information recording unit 18 is configured as a recording device equipped with a recording medium (not shown) and has multiple areas for recording medical procedure information, medical interview information, and lifestyle information (hereinafter, these pieces of information are referred to as medical information). In the example of FIG. 18 , the control unit 11 records medical procedure information in the medical procedure information recording area 18a, medical interview information in the medical interview information recording area 18b, and lifestyle information included in the medical interview information in the lifestyle information recording area 18c. The recording media may be provided in an in-hospital system, a pharmacy system, or an external knowledge database 19 separate from the mobile terminal 10. When recording in such locations, a recording control unit corresponding to each recording medium may be provided to link each recording medium with the information recording unit 18 of the terminal.

[0024] In the example of Figure 1, it can be seen that on November 3rd, a patient visited the hospital, received treatment A, was prescribed prescription 1, and experienced subjective symptoms of lower back pain and fatigue. It can also be seen that on November 1st, two days before this visit, the patient experienced no subjective symptoms, but on November 2nd, the day before the visit, the patient experienced subjective symptoms of lower back pain and fatigue. Furthermore, if it is also possible to know that the user has a drinking and smoking habit, this information may be used to improve the accuracy of diagnosis, predict complications, provide advice during treatment, and provide precautions when taking prescribed medications. It can also be seen that on November 4th, the day after the visit, the patient still felt fatigued, but the subjective symptoms of lower back pain disappeared, and on November 5th, the day after the visit, both the lower back pain and fatigue disappeared, and the patient recovered.

[0025] In this manner, in this embodiment, not only information after a hospital visit (after a medical procedure) but also information before a hospital visit (before a medical procedure) is acquired and recorded. For example, medical procedure information regarding over-the-counter medications taken by the user before the hospital visit is also recorded, as well as medical interview information including the user's subjective symptoms, behavioral history, biological information, and lifestyle information from before the hospital visit to the present. The control unit 11 also records profile information, including the user's name, age, sex, nationality, height, weight, and other information, in the information recording unit 18. This makes it possible to organize knowledge information according to the patient's profile.

[0026] The control unit 11 may be configured to display a message for acquiring medical interview information on a display device (not shown). For example, when the sensor unit 12 reads a medical receipt or the like, the control unit 11 determines the contents of the medical receipt or the like, creates a medical interview form for acquiring medical interview information corresponding to the determined medical procedure information, and displays a message on the display screen of the display device to prompt the user to enter information into the medical interview form. The medical interview form is a table for acquiring medical interview information from the past to the present of the user, and shows medical interview items for obtaining predetermined medical interview information.

[0027] 2 is an explanatory diagram showing an example of a message for inputting information into a medical questionnaire presented to a user by the control unit 11. In the example of Fig. 2, the control unit 11 may display a message such as "You visited the hospital on November 3rd. Please write down as many subjective symptoms as you can remember" as a "confirmation of symptoms up to the time of treatment" message, and may allow the information input unit 13 to accept an input operation of the user's medical questionnaire information from the past to the present.

[0028] The external knowledge DB 19 stores knowledge information including information about what symptoms are likely to appear in what kind of disease (the name of the disease may be unknown), and what kind of examinations, tests, treatments, and prescriptions are performed. This knowledge information can also be loaded into the information recording unit 18 of the mobile terminal 10. By using this knowledge information, it is possible to search for and infer what symptoms were present from information about examinations, tests, and prescriptions. By using this inference, the control unit 11 can present candidate symptoms to the user and allow the user to select a symptom. In other words, by using the knowledge information, the control unit 11 does not need to ask questions about content that is unrelated to the user.

[0029] Furthermore, after reading a receipt or the like, the control unit 11 may display a message such as "You visited the hospital on November 3rd. Please enter your current symptoms" as a "confirmation of the effectiveness of treatment after treatment," for example, every day, and accept the user's input operation of medical interview information via the information input unit 13.

[0030] In this case, since the treatment or prescription was likely initiated due to a concern about the symptoms, the subjective symptoms alleviated will be the same as those before treatment, and the same information as before treatment can be used for the medical interview information. Furthermore, by referencing the results of the user's past medical interviews, it is possible to search for the subjective symptoms the user experienced, and by asking questions based on the search results, such as "Did your lower back pain improve before the prescription?", the medical interview can be configured to allow the user to answer with a "yes" or "no." Furthermore, by asking questions such as "How much has the pain improved from the condition before treatment?", it is possible to quantify the results. This analog information can be used to determine how long it will take to recover, allowing for more accurate information on the recovery period, and data can be acquired to present that information.

[0031] In this way, the user himself records in the information recording unit 18 information from the past, including before visiting the hospital (before medical treatment), to the present, including after visiting the hospital (after medical treatment).

[0032] Here, we have shown an example of conducting a post-medical procedure interview after a pre-medical procedure interview, but even if it is not possible to input or record the interview when the medical procedure occurs, it is also possible to conduct an interview after visiting the hospital (after the medical procedure) or when taking medicine and seeing its effect, and based on the results, recall the time before the medical procedure and obtain interview information. In this case, knowledge information can also be searched to narrow down the interview.

[0033] It should be noted that the user's own behavioral history and biological information can be automatically acquired by the sensor unit 12 capable of acquiring such information, and the information can be recorded in the information recording unit 18. For example, information that can be automatically acquired by the sensor unit 12 includes the number of times the user has gone to the toilet, weight, blood oxygen concentration, blood pressure, number of exercises, exercise frequency, number of times the user has gone out, number of meals, number of coughs, voice quality, etc. This makes it possible to acquire a certain amount of medical interview information even if the user does not input information themselves.

[0034] The treatment result determination unit 14 reads the medical treatment information and medical interview information recorded in the information recording unit 18 and determines the treatment effectiveness of the medical treatment based on the read information. For example, the treatment result determination unit 14 may determine whether the treatment was effective based on whether subjective symptoms of recovery were observed within a number of days based on the treatment information and prescription information from the date of treatment. The treatment result determination unit 14 outputs the medical treatment information, medical interview information, and information on the treatment effectiveness determination results (hereinafter, this information is referred to as terminal medical information) read from the information recording unit 18 to the DB linkage unit 15 and the information processing unit 16. The terminal medical information may also include user profile information.

[0035] The processing result determination unit 14 performs learning, such as deep learning, using a large amount of training data. Deep learning is a multilayered version of the machine learning process using neural networks. A typical example is a forward propagation neural network, which sends information from front to back and makes a judgment. In its simplest form, this requires three layers: an input layer consisting of m1 neurons, a middle layer consisting of m2 neurons given by parameters, and an output layer consisting of m3 neurons corresponding to the number of classes to be discriminated. The neurons in the input layer and middle layer, and those in the middle layer and output layer, are connected by connection weights, and bias values ​​are added to the middle layer and output layer, making it easy to form logic gates. While three layers are sufficient for simple discrimination, increasing the number of middle layers makes it possible to learn how to combine multiple features during the machine learning process. In recent years, networks with 9 to 152 layers have become practical due to the learning time, judgment accuracy, and energy consumption. Various well-known networks may be used as the network N1 used in machine learning. For example, R-CNN (Regions with CNN features) or FCN (Fully Convolutional Networks) using CNN (Convolution Neural Network) may be used. This involves a process called "convolution" that compresses image features, operates with minimal processing, and is strong in pattern recognition. Furthermore, "recurrent neural networks" (fully connected recurrent neural networks) that can handle more complex information and flow information bidirectionally may be used to handle information analysis in which the meaning changes depending on the order or sequence of information.

[0036] The DB linking unit 15 organizes the terminal medical information from the treatment result determination unit 14 into information necessary for an external knowledge database (DB) 19, and then outputs the information to the external knowledge DB 19. The external knowledge DB 19 is a database that stores knowledge information that is the result of previously associating medical procedure information, interview information, and treatment effect determination results.

[0037] The external knowledge DB 19 digitizes and records the knowledge of doctors and other academic experts, as well as the contents of various papers and medical books. In other words, treatments for various illnesses are organized and recorded as knowledge information, and this knowledge information is configured to enable searching and inferring the symptoms from information on examinations, tests, and prescriptions.

[0038] Furthermore, the external knowledge DB 19 stores knowledge information including information on the general time from onset of symptoms to medical treatment, such as how long it will take to receive medical treatment after the onset of symptoms or the appearance of subjective symptoms, and how long it will take to recover (this will vary from person to person, but an average trend will suffice), as well as information on the general time from medical treatment to recovery (this will vary from person to person, but an average trend will suffice). Using this knowledge information, doctors can inform patients of a guideline for their next examination, and patients can concentrate on their treatment by estimating how long it will take for them to recover.

[0039] As described above, some or all of the knowledge information may be recorded in the information recording unit 18 of the mobile terminal 10. The knowledge information recorded in the external knowledge DB 19 is extremely comprehensive and general, encompassing information such as gender, age, race, and medical history, and is also an accumulation of past information. Therefore, for more individual situations and diseases and symptoms that have not occurred before, further accumulation of information is required. Recent advances in IT technology and mobile terminal technology have made it relatively easy to collect individual cases, and the information can also be converted into big data.

[0040] In the present invention, treatment results corresponding to the medical information of each patient are acquired from the mobile terminal 10 or the like, and the effectiveness of the treatment is confirmed by checking the patient's subjective symptoms up until the treatment and by checking changes in the subjective symptoms after the treatment, thereby associating the treatment results with changes in the patient's subjective symptoms before and after the treatment. That is, in this embodiment, the same type of information as that given by the knowledge information is acquired, and the external knowledge DB 19 can be constantly updated by importing or referring to knowledge information from the external knowledge DB 19.

[0041] Furthermore, by organizing data as in the information recording unit 18, it becomes easy to acquire large amounts of data that serve as training data for an inference model. For example, it is possible to construct an inference model that obtains the output of medical interview questions by inputting medical receipts, and the inference results of the constructed inference model can be used instead of searching the external knowledge DB 19.

[0042] To convert the data into big data, a case record unit for recording individual cases may be provided in the external knowledge DB 19. Of course, individual cases may also be collected and recorded in a record unit provided in a hospital's in-house system or in the system of other dedicated services. This application assumes collaboration with such other systems, and big data can be collected and recorded by communicating in some form with personal mobile devices 10 owned by individuals. If the collected big data is made available under the necessary management and contractual conditions, it can be used to build a database according to standardized special rules or as training data for inference models for specific applications.

[0043] Furthermore, in addition to recording raw data, the external knowledge DB 19 may also record data obtained by performing statistical processing on the raw data so that personal information and the like cannot be identified.

[0044] The external knowledge DB 19 also includes a control unit 19a, a teacher data conversion unit 19b, a learning unit 19c, and an inference unit 19d. The control unit 19d manages communications and records to prevent privacy issues and properly organizes the database. The teacher data conversion unit 19b converts data recorded in the external knowledge DB 19 into teacher data. The learning unit 19c constructs an inference model by learning using the teacher data created by the teacher data conversion unit 19d. The inference unit 19d uses the inference model constructed by the learning unit 19c to output a specific inference result in response to a specific data input. The control unit 19a performs various controls, including data exchange between the teacher data conversion unit 19b, the learning unit 19c, and the inference unit 19d. The control unit 19a also controls returning inference results to the mobile terminal 10 upon request from the mobile terminal 10, and searching for specific data upon request from the mobile terminal 10 and returning the results to the mobile terminal 10.

[0045] 6 is an explanatory diagram showing an example of an inference model created by the learning unit 19c. The example in FIG. 6 shows the time required for recovery for, for example, three examples of the same case, with time on the horizontal axis and the degree of deterioration of the disease on the vertical axis. Time t0 indicates the timing when initial symptoms appeared before medical intervention, time t1 indicates the timing when medical intervention began in each example, and time t2 indicates the timing when the user realized that they had recovered in each example.

[0046] The bottom row of Figure 6 is an example where medical intervention was performed relatively early after the initial symptoms appeared, the middle row of Figure 6 is an example where medical intervention was performed relatively late after the initial symptoms appeared, and the top row of Figure 6 is an example where medical intervention was performed after the initial symptoms appeared, but between the bottom row and the middle row of Figure 6. In the example of Figure 6, it is thought that there is a relationship between the timing of medical intervention after the initial symptoms appeared and the time required for recovery.

[0047] The training data generation unit 19b generates training data by annotating the time (t2-t1) obtained by the medical interview until the patient realizes that they have recovered from three types of data sets: information on early symptoms before medical intervention obtained by interview, information on the time (t1-t0) from the time the early symptoms appeared until medical intervention, and information on the details of the medical intervention. When generating training data, age, gender, race, etc. may be added to the above data sets.

[0048] The learning unit 19c learns using the training data created by the training data generation unit 19b and constructs an inference model. According to this inference model, the initial symptoms and the time from the onset of the initial symptoms to the start of medical treatment are input, and the time required for recovery is obtained as an inference result. This inference result can be displayed on the mobile terminal 10.

[0049] Furthermore, specific information about medical procedures can be acquired through interviews and included in the training data. Therefore, the training data during learning also includes information about the time from the onset of symptoms to the medical procedure, such as purchasing and taking over-the-counter medication at a pharmacy or going to a clinic and taking prescribed medication. Therefore, the inference model can output inference results for recovery times for cases where medical procedures are performed immediately after the onset of symptoms, when medical procedures are performed two or three days later, and when medical procedures are performed more than a week later. By transmitting and displaying this information on the portable information terminal 10, the user can obtain extremely useful information about recovery.

[0050] If the external knowledge DB 19 exists on a network line such as the Internet, the DB linking unit 15 uses a communication function (not shown) of the mobile terminal 10 to transmit terminal medical information to the external knowledge DB 19. The DB linking unit 15 also exchanges information with the external knowledge DB 19 and outputs knowledge information about a specific case to the information processing unit 16. As described above, the mobile terminal 10 may be given the same functions as the external knowledge DB 19 by sending a portion of the knowledge information in the external knowledge DB 19 to the mobile terminal 10 and recording the knowledge information in the information recording unit 18 of the mobile terminal 10.

[0051] FIG. 3 is an explanatory diagram for explaining an example of knowledge information in the external knowledge DB 19. As shown in FIG.

[0052] The external knowledge DB 19 has a plurality of recording areas R1, R2, ... (hereinafter, these areas are collectively referred to as recording area R) for recording knowledge information (related information) for each symptom. In the example of Fig. 1, for example, knowledge information related to abdominal pain is recorded in recording area R1, and knowledge information related to fever is recorded in recording area R2. Each recording area R has a plurality of disease-specific recording areas RD for recording knowledge information for each disease.

[0053] The disease recording area RD records knowledge information for each candidate disease. A candidate disease is a candidate disease that presents with that symptom, and multiple candidate diseases are possible for one symptom. For example, even if a patient has a fever, there are many possible disease candidates, such as COVID-19 or influenza. Each disease recording area RD records knowledge information such as the disease's representative symptoms, causes, test results, in-hospital treatment, and prescription medications.

[0054] Furthermore, in this embodiment, the external knowledge DB 19 receives terminal medical information from the DB linking unit 15. In the example of FIG. 1 , the terminal medical information is recorded as profile information in the disease-specific record area RD for candidate disease 1. In this profile information, the terminal medical information from the DB linking unit 15 is recorded as information associated with existing knowledge information. The external knowledge DB 19 may, for example, determine which disease-specific record area RD to associate and record the terminal medical information by comparing subjective symptoms included in the terminal medical information from the DB linking unit 15 with representative symptoms in each disease-specific record area RD in the external knowledge DB 19. The external knowledge DB 19 may also determine which disease-specific record area RD to associate and record the terminal medical information by comparing medical procedure information included in the terminal medical information with in-hospital treatments and prescription drugs in each disease-specific record area RD in the external knowledge DB 19. In the example of Fig. 1, the associated information includes the subjective symptom of abdominal pain, throbbing pain, prescription 1 being prescribed, and recovery in 2 days. As with the treatment result determination unit 14, an inference model using deep learning may be used as the association method.

[0055] The information processing unit 16 performs necessary processing for presentation of the terminal medical information from the treatment result determination unit 14 or the knowledge information from the DB linkage unit 15, and then outputs the information to the information presentation unit 17. The information presentation unit 17 may be configured, for example, with a display device or speaker (not shown) provided in the mobile terminal 10, and may present the terminal medical information from the information processing unit 16 by displaying it on the display screen of the display device or outputting it as audio. The information presentation unit 17 may also be controlled by the control unit 11 to present a questionnaire created by the control unit 11 to the user. The mobile terminal 10 may also have a communication function (not shown) that can connect to a medical institution, and the information presentation unit 17 may use this communication function to transmit the terminal medical information from the information processing unit 16 to the medical institution (or doctor). The presentation by the information presentation unit 17 allows the user or the medical institution (or doctor) to determine the extent to which the user's or the medical institution's treatment has improved the user's symptoms.

[0056] Furthermore, the information presenting unit 17 can present knowledge information from the external knowledge DB 19 to the user, a medical institution, or the like. This makes it easy for the user, the medical institution, or the like to determine what treatment is effective for the current symptoms. For example, when the treatment result determining unit 14 does not obtain a treatment effect determination result, or in response to a user's operation, the control unit 11 may control the DB linking unit 15, the information processing unit 16, and the information presenting unit 17 to present knowledge information read from the external knowledge DB 19 to the user, a medical institution (doctor), or the like. For example, in response to a medical treatment for the user or an input operation of the user's symptoms, the control unit 11 may search the external knowledge DB 19 for knowledge information similar to the medical treatment for the user or the user's symptoms from the external knowledge DB 19, extract the knowledge information from the search result, and present it.

[0057] Next, the operation of the embodiment configured as above will be described with reference to Figures 4 and 5. Figure 4 is a flowchart for explaining the operation of the embodiment. Figure 5 is an explanatory diagram showing an example of display using knowledge information from the external knowledge DB 19 by the information presentation unit 17.

[0058] 4, the control unit 11 determines whether an input operation has been performed to wake up the mobile terminal 10 from the sleep mode. The control unit 11 is in a standby state for a wake-up request in S1, and when a wake-up request is generated (YES in S1), the control unit 11 displays a predetermined top screen display on the display screen of a display device (not shown) (S2). The control unit 11 determines whether a request has been made to launch an application (medical app) for creating and using terminal medical information in S3.

[0059] When a request to start the medical application is made in S3 (YES in S3), the control unit 11 determines whether or not there has been an input operation of medical receipt information, etc. If there has been an input operation of medical receipt information, etc. (YES in S4), the control unit 11 acquires medical procedure information in S5 (S5), and acquires medical inquiry information by displaying a message for medical inquiry based on the medical receipt information, etc., based on a search or inference from a knowledge database (S6).

[0060] The medical interview information is obtained by a user's input referring to a display (or voice transmission) as shown in Figure 2, but the questions can also be simplified by narrowing down the list of medical receipt information. User input methods include button operation, text input, character recognition, voice recognition, and other methods such as a Yes or No selection input using a touch input or a vibration sensor. In other words, the processor of the mobile terminal 10 controls these inputs and outputs so that when the information acquisition unit acquires medical procedure information related to a medical procedure for the user, the processor determines the medical procedure information, creates medical interview information for acquiring information about the user's symptoms before and after the medical procedure, and presents the information to the information presentation unit.

[0061] In addition, the processor selects candidate symptoms and creates the medical interview information when acquiring information about the user's symptoms before the medical procedure according to the acquired medical procedure information, and selects candidate symptoms and creates the medical interview information when acquiring information about the user's symptoms after the medical procedure according to the input information.

[0062] When creating the medical interview information, information on the initial symptoms before the medical procedure, as determined from the previous interview, and the time from when the symptoms appeared until the medical procedure was performed should also be obtained. Furthermore, the time from the date and time of the medical procedure (examination or medication) until the patient realized they had recovered should also be obtained in the interview after the medical procedure. In this case, the medical interview information may be obtained by having the user input the information via the information input unit 13 in the form of voice, text, or answers to questions. If there is no user input, the information may be calculated using information obtained from biometric information or the like. After the medical procedure, the information may be obtained by repeatedly checking the patient's physical condition every day.

[0063] The mobile terminal 10 has a recording unit (database recording unit) that records knowledge information in the DB linking unit 15 or the information recording unit 18. When medical procedure information related to a medical procedure is acquired, the mobile terminal 10 links with the records in the external knowledge DB 19 or the information recording unit 18 to determine example symptoms for a medical interview that are candidates before the medical procedure. Linking refers to performing search or inference. The external knowledge DB 19 linked by the DB linking unit 15 records knowledge information that is information on the relationship between the symptoms of each disease, medical procedures, and the effects of the medical procedures. Information entered by a doctor or pharmacist may be used instead of the medical procedure information.

[0064] The control unit 11 records the acquired medical treatment information and medical interview information in the information recording unit 18 (S7) and returns the process to S2. If activation of the medical app is not requested (NO in S3), the control unit 11 executes another function of the mobile terminal 10 in S18 and returns the process to S2.

[0065] If the control unit 11 determines in S4 that no input operation for prescription information or the like has been performed, it proceeds to S10 to determine whether the information presentation mode has been instructed. If the information presentation mode has not been instructed (NO in S10), the control unit 11 returns to S2. If the information presentation mode has been instructed (YES in S10), the control unit 11 proceeds to S11 to determine the type of information to be presented. If terminal medical information has been instructed to be presented, the control unit 11 controls the information processing unit 16 and the information presentation unit 17 to display the medical procedure information and interview information read from the information recording unit 18 and the terminal medical information including the treatment effect assessment result from the treatment result assessment unit 14 on the display screen of the display device or transmit it to the medical institution or doctor (S12). In this way, the user or the medical institution (or doctor) can determine the effectiveness of the user's medical procedures from the past to the present by referring to the presented terminal medical information.

[0066] Furthermore, if knowledge information is specified as the information to be presented, the control unit 11 displays a symptom input message for identifying the knowledge information to be searched for and acquires the symptom input by the user (S13). The control unit 11 controls the DB linkage unit 15 to send the symptom input result to the external knowledge DB 19 and acquire knowledge information corresponding to the symptom. The external knowledge DB 19 searches for people (disease candidates) corresponding to the input symptoms and transmits the knowledge information of the search result to the DB linkage unit 15. Note that if the external knowledge DB 19 does not obtain any relevant search results, it transmits information to that effect to the DB linkage unit 15.

[0067] When a search result is obtained from the external knowledge DB 19, i.e., when a person with a similar case (disease candidate) corresponding to the symptom input by the user can be searched for in the external knowledge DB 19 (YES in S14), the control unit 11 controls the information processing unit 16 and the information presentation unit 17 to display knowledge information indicating the progress of the searched person (S15), and returns the process to S13. When a search result corresponding to the symptom input by the user cannot be obtained (NO in S14), the control unit 11 proceeds to the process of S16 and presents a message that no information was found.

[0068] There are several possible methods for displaying the knowledge information in S15. First, a member who manages the external knowledge DB 19 selects examples of typical patients for a specific disease, and transmits the selected information to the mobile terminal 10, where it can be displayed as shown in Fig. 5, which will be described later.

[0069] Furthermore, in situations where a large amount of patient information (dataset) has been collected, one method is to create a graph with the time taken to heal on the horizontal axis and the corresponding number of cases on the vertical axis, and obtain information on the average healing time, or information on the healing time that maximizes the number of cases by referring to the peak on the vertical axis, and send that information to the mobile terminal 10 for display. Alternatively, examples of healing that occurred faster than the average healing time and examples of healing that took longer than the average healing time, such as characteristic patient behavior corresponding to a time that is 2 sigma or 3 sigma away from the average healing time using statistical calculations, can be searched for and sent to the mobile terminal 10, where it can be displayed. This allows the user to obtain reference information on what to do to prevent the condition from worsening and how to heal more quickly.

[0070] In addition, the output of an inference model learned in the external knowledge DB 19 may be acquired and displayed. In addition, if the mobile terminal 10 has a system configuration in which an inference model is built in, a display based on the output of the inference model may be performed.

[0071] As mentioned above, this inference model is trained using training data obtained by annotating the time it takes for a patient to recover from a medical procedure (i.e., whether symptoms before the procedure are alleviated) based on subjective symptoms, medical prescription information, and pharmacy purchase history information. Therefore, this inference model can infer the outcome of various medical procedures. This inference allows users to select and implement the appropriate medical procedure. In other words, when experiencing critical symptoms, users can determine whether to rest at home or go to the hospital based on the experiences of other users. Since recovery time varies significantly among individuals, such as physical strength, individual profile information may also be input during inference model training to enable more accurate inference. For example, even if a similar symptom, such as "fever," is measured using medical equipment, the inference described above will vary depending on the disease, such as "cold," "COVID," or "influenza." However, differences in medical procedures can be determined by inputting information from medical prescriptions and receipts. Therefore, when "fever" is input as a symptom, information specific to each disease may be output. It is also possible to determine which of these diseases the patient may have based on information on the patient's behavior, etc. For example, if the patient's residential area is entered in the profile, the name of the disease with the highest probability can be displayed based on information on medical receipts incurred in that area at that time.

[0072] FIG. 5 shows an example of the knowledge information displayed in S15.

[0073] The example in Fig. 5 shows an example of a knowledge information display Dk displayed on the display screen 17a of a display device (not shown) provided in the mobile terminal 10. The knowledge information display Dk in Fig. 5 shows that search results for three disease candidates shown in patterns A to C were obtained from the external knowledge DB 19. Pattern A indicates that a person who visited a hospital, took medication, took leave, and recuperated recovered from the illness. Pattern B indicates that a person who only took medication without visiting a hospital and continued working normally took longer to recover. Pattern C indicates that a person who neither visited a hospital nor took medication experienced a worsening of the illness.

[0074] The inference model of the external knowledge DB19 is trained using training data obtained by annotating how long it takes for a patient to recover after a medical procedure (whether the symptoms experienced before the medical procedure are alleviated) for medical procedures that assume symptoms, prescription information, and pharmacy purchase history information, so it is possible to infer what kind of medical procedure will result in the patient receiving the symptoms they have entered.

[0075] By referring to the knowledge information display Dk, the user can refer to the medical procedures and their effectiveness results for people with symptoms similar to the user's own when deciding which medical procedures to perform. Note that knowledge information can also be obtained by medical institutions or doctors, and by referring to the knowledge information, medical institutions or doctors can receive effective support for medical procedures for patients.

[0076] In this manner, in this embodiment, medical procedure information such as past procedures is acquired, and medical interview information corresponding to the acquired medical procedure information is acquired, and the results of the treatment effectiveness are determined based on the medical procedure information and the medical interview information, and terminal medical information including the medical procedure information, the medical interview information, and the results of the treatment effectiveness assessment can be associated and presented. This makes it possible to judge the effectiveness of medical procedures based on the presented terminal medical information, and to receive support for future medical procedures.

[0077] The present invention is not limited to the above-described embodiments, and the components can be modified and embodied in practice without departing from the spirit of the invention. Furthermore, various inventions can be formed by appropriately combining multiple components disclosed in the above-described embodiments. For example, some of the components shown in the embodiments may be omitted. Furthermore, components from different embodiments may be appropriately combined.

[0078] Furthermore, among the technologies described herein, many of the controls and functions, mainly those described in the flowcharts, can be set by a program, and the above-described controls and functions can be realized by a computer reading and executing the program. The program can be recorded or stored, in whole or in part, as a computer program product on a portable medium such as a flexible disk, CD-ROM, or nonvolatile memory, or on a storage medium such as a hard disk or volatile memory, and can be distributed or provided at the time of product shipment, via a portable medium, or via a communication line. A user can easily realize the information processing system, information processing device, information association method, learning method, and information association program of the present embodiment by downloading the program via a communication network and installing it on a computer, or by installing it on a computer from a recording medium.

Claims

1. An information processing system comprising: an interface device that acquires medical treatment information regarding a medical treatment performed on a user and medical interview information regarding the user's symptoms before and after the medical treatment; a memory that records the medical treatment information and the medical interview information; and a processor, wherein the processor determines the treatment effectiveness of the medical treatment performed on the user based on the medical interview information acquired by the interface device, and records the medical treatment information, the medical interview information, and the treatment effectiveness in association with each other in the memory.

2. The information processing system of claim 1, wherein the processor determines the treatment effectiveness of the medical procedure based on a comparison result between a first symptom, which is the user's symptom before the medical procedure, and a second symptom, which is the user's symptom after the medical procedure.

3. The information processing system of claim 1, wherein the processor queries a database in which knowledge information, which is the result of associating medical treatment information regarding medical treatment, medical interview information regarding symptoms, and information on the evaluation results of the treatment effectiveness of the medical treatment, is accumulated to extract candidate symptoms, prompts the user via the interface device to select from the extracted candidate symptoms, and uses the symptom selected by the user as the medical interview information.

4. The information processing system according to claim 1, wherein the processor acquires the medical treatment information based on the user's medical prescription or medical record.

5. The information processing system according to claim 1, wherein the processor acquires the medical treatment information from a hospital that performed the medical treatment.

6. The information processing system of claim 1, wherein the processor, when the interface device is configured as a device that accepts text input or voice input by the user, acquires the medical interview information based on text input or voice input by the user, and when the interface device is configured as a biometric information acquisition device owned by the user, acquires the medical interview information based on the output of the biometric information acquisition device.

7. The information processing system according to claim 1, wherein the processor acquires lifestyle information of the user in addition to the symptoms as the medical interview information.

8. The information processing system of claim 1, wherein the processor queries a database in which knowledge information resulting from associating medical treatment information related to medical treatment, medical interview information related to symptoms, and information on the evaluation results of the treatment effectiveness of the medical treatment is accumulated, and extracts and presents knowledge information similar to the medical treatment for the user or the user's symptoms from the database.

9. The information processing system according to claim 8, wherein the processor, if no judgment result of the treatment effect is obtained, extracts knowledge information from the database that is similar to medical treatment for the user or symptoms of the user and presents it.

10. An information processing apparatus comprising: an interface device having an information acquisition unit that acquires medical treatment information regarding medical treatment for a user; an information presentation unit that presents information to the user; and a processor, wherein when the information acquisition unit acquires the medical treatment information, the processor creates a questionnaire based on the medical treatment information to acquire information regarding the user's symptoms before and after the medical treatment for the user, and causes the information presentation unit to present the questionnaire.

11. The information processing device according to claim 10, wherein the processor, when acquiring information about the user's symptoms before the medical procedure according to the acquired medical procedure information, selects candidate symptoms and determines the medical interview items for acquiring the medical interview information.

12. The information processing device according to claim 11, wherein the processor, when acquiring information about the user's symptoms after the medical procedure, selects candidate symptoms and determines the medical interview items for acquiring the medical interview information according to the acquired information about the user's symptoms before the medical procedure.

13. The information processing device according to claim 10, wherein the processor sets questionnaire items for obtaining, for information regarding the user's symptoms before the medical procedure, the time from when the symptoms appeared until the medical procedure, and / or the time from the date and time when the medical procedure was performed until the user realized that they had recovered.

14. An information processing device as described in claim 10, further comprising a database linkage unit that links with a database in which knowledge information resulting from associating medical treatment information regarding medical treatment, medical interview information regarding symptoms, and information on the evaluation results of the treatment effectiveness of the medical treatment is accumulated, and when the processor obtains medical treatment information regarding medical treatment for the user, it links with the database to determine candidate symptoms before the medical treatment.

15. The information processing device according to claim 14, wherein the database stores information on the relationship between symptoms of each disease, medical treatment, and the effects of said medical treatment.

16. An information association method for associating information on subjective symptoms and treatment effects, comprising: a treatment result acquisition step for acquiring treatment results for a patient in accordance with medical procedure information; a treatment effect confirmation step for confirming the treatment effects based on changes in the subjective symptoms after the treatment corresponding to the treatment; and a step for associating the treatment results with changes in the patient's subjective symptoms before and after the treatment.

17. The information association method according to claim 16, wherein the patient's symptom confirmation step searches for candidate symptoms related to medical treatment using a database in which information on the relationship between symptoms, medical treatment, and the effects of said medical treatment is recorded.

18. The information association method according to claim 16, wherein one of the step of confirming subjective symptoms and the step of confirming treatment effects is confirmed first, and then the other is confirmed based on the result.

19. A learning method in which a dataset consisting of two pieces of data, namely, initial symptoms before medical treatment, the time from the time when said initial symptoms appeared until medical treatment and at least one of said medical treatments, is used to annotate the time until recovery is realized and turn it into training data, and then a neural network is trained using the training data to construct an inference model.

20. An information association program for associating information on subjective symptoms and treatment effects, the information association program causing a computer to execute the following steps: a treatment result acquisition step for acquiring treatment results for a patient in accordance with medical procedure information; a treatment effect confirmation step for confirming the treatment effects based on changes in the patient's subjective symptoms after the treatment corresponding to the treatment; and a step for associating the treatment results with changes in the patient's subjective symptoms before and after the treatment.

Images