阅读说明：本技术 支援系统、支援方法以及支援程序 (Support system, support method, and support program ) 是由 坂口雄纪 野村治 木下康 关根佑辅 于 2018-07-31 设计创作，主要内容包括：本发明提供一种有助于医疗费的削减的支援系统、支援方法以及支援程序。支援系统(100)具备：数据获取部(111),其获取与预定了在医疗机关就诊的就诊预定者有关的就诊预定者数据(D1)、与就诊预定者向医疗机关来访的来访履历有关的来访数据(D2)以及与就诊预定者过于在医疗机关就诊的诊察内容有关的就诊数据(D3)；学习部(112),其使用就诊预定者数据、来访数据以及就诊数据进行机械学习；以及提示部(113),其基于机械学习的结果,提示是否需要对就诊预定者进行诊察。(The invention provides a support system, a support method and a support program which are helpful for reducing medical fees. The support system (100) is provided with: a data acquisition unit (111) for acquiring preschool patient data (D1) relating to a preschool patient who is scheduled to visit the medical institution, visit data (D2) relating to a history of visits the preschool patient visits the medical institution, and visit data (D3) relating to the content of the visits the preschool patient visits too much in the medical institution; a learning unit (112) that performs machine learning using the scheduled visit data, the visit data, and the visit data; and a presentation unit (113) that presents whether or not the examination of the person scheduled to see a doctor is necessary, based on the result of the machine learning.)

1. A support system that supports a medical examination by a healthcare practitioner, the support system comprising:

a data acquisition unit that acquires pre-visit person data relating to a pre-visit person who is scheduled to visit the medical institution, visit data relating to a history of visits made by the pre-visit person to the medical institution, and visit data relating to the content of previous visits made by the pre-visit person to the medical institution;

a learning unit that performs machine learning using the scheduled visit data, the visit data, and the visit data; and

and a presentation unit that presents whether or not the scheduled patient needs to be examined based on a result of the machine learning.

2. The support system according to claim 1, wherein,

the presentation unit presents another examination behavior that replaces the examination by the healthcare worker when presenting that the examination is not necessary.

3. The support system according to claim 2, wherein,

the presentation unit presents, as the another examination action, communication with the person scheduled to visit based on a dialogue-type device.

4. The support system according to any one of claims 1 to 3, wherein,

the visit data includes prescription data relating to medical supplies prescribed to the scheduled visit,

the learning unit performs mechanical learning of prescription conditions of recommended medical supplies based on the preschool visit data, the visit data, and the prescription data,

the presentation unit presents the prescription condition based on a result of the machine learning.

5. The support system according to any one of claims 1 to 4,

the presentation unit presents presentation contents and presentation results together.

6. A support method for supporting a medical examination by a healthcare practitioner, the support method comprising:

a data acquisition step of acquiring president data on a president who has scheduled a visit in a medical institution, visit data on a history of visits of the president to the medical institution, and visit data on a content of previous visits of the president in the medical institution;

a learning step of performing machine learning using the scheduled visit data, the visit data, and the visit data; and

and a prompting step of prompting whether the examination of the scheduled patient needs to be performed or not based on the result of the mechanical learning.

7. A support program for supporting a medical examination by a healthcare practitioner, the support program executing:

a data acquisition step of acquiring president data on a president who has scheduled a visit in a medical institution, visit data on a history of visits of the president to the medical institution, and visit data on a content of previous visits of the president in the medical institution;

a learning step of performing machine learning using the scheduled visit data, the visit data, and the visit data; and

and a prompting step of prompting whether the examination of the scheduled patient needs to be performed or not based on the result of the mechanical learning.

Technical Field

The present invention relates to a support system, a support method, and a support program for supporting a medical practitioner's examination.

Background

In recent years, our country has come to an ultra-advanced society, and there is a fear that the number of medical practitioners is insufficient, the medical quality is degraded, and the like. Therefore, for the purpose of improving medical efficiency and medical quality, a plurality of medical institutions cooperate to treat patients and advance regional medical cooperation.

For example, patent document 1 listed below describes a regional medical cooperation system that supports the introduction of patients between medical institutions.

Disclosure of Invention

One of the major problems in the aging society is the rise of medical fees, which is social insurance fees. The medical cost is increased due to various reasons such as a high medical administration field and medical upgrade, and particularly, the problem of excessive medical treatment of elderly people is emphasized.

For example, when an elderly person experiences physical discomfort, it is difficult to determine the health status of the elderly person, and the elderly person actively visits medical institutions such as a hospital. In addition, even if a self-diagnosis is actually not necessary among the elderly, many elderly frequently visit the medical institutions of the community to eliminate psychological loneliness and the like. As a result, medical costs increase, the burden on medical practitioners such as doctors increases, and further, there arises a problem that medical supplies are excessively opened.

On the other hand, even if the medical practitioner thinks that prescription is unnecessary, there is a possibility that the medical practitioner will overdue medical supplies such as medicines to the elderly who come to the hospital.

As described above, when a healthy elderly person receives "a doctor and a doctor" and "a doctor should make a diagnosis" and "a doctor's unnecessary diagnosis" or the like, the "medical supplies are opened too much" due to these actions, resulting in a problem that medical costs are increased.

The present invention has been made in view of the above circumstances, and an object thereof is to provide a support system, a support method, and a support program that contribute to reduction of medical fees.

The support system according to the present invention for achieving the above object supports examinations by medical practitioners, and includes: a data acquisition unit that acquires pre-visit person data relating to a pre-visit person who is scheduled to visit the medical institution, visit data relating to a history of visits made by the pre-visit person to the medical institution, and visit data relating to the content of previous visits made by the pre-visit person to the medical institution; a learning unit that performs machine learning using the scheduled visit data, the visit data, and the visit data; and a presentation unit that presents whether or not the scheduled patient needs to be examined based on a result of the machine learning.

The support method of the present invention for achieving the above object supports examinations by medical practitioners, and includes: a data acquisition step of acquiring president data on a president who has scheduled a visit in a medical institution, visit data on a history of visits of the president to the medical institution, and visit data on a content of previous visits of the president in the medical institution; a learning step of performing machine learning using the scheduled visit data, the visit data, and the visit data; and a prompting step of prompting whether the examination of the person scheduled for medical examination is required or not based on the result of the mechanical learning.

The support program according to the present invention for supporting a medical examination by a healthcare practitioner, which achieves the above object, executes the steps of: a data acquisition step of acquiring president data on a president who has scheduled a visit in a medical institution, visit data on a history of visits of the president to the medical institution, and visit data on a content of previous visits of the president in the medical institution; a learning step of performing machine learning using the scheduled visit data, the visit data, and the visit data; and a prompting step of prompting whether the examination of the person scheduled for medical examination is required or not based on the result of the mechanical learning.

Effects of the invention

The invention prompts whether the medical practitioner needs to examine the person scheduled for seeing a doctor or not based on the result of the mechanical learning. The medical practitioner can avoid examining the scheduled patient who lacks the necessity of the treatment by referring to the presented contents. As a result, it is possible to prevent an increase in the business load of medical practitioners and an excessive opening of medical supplies due to visits from elderly people to medical institutions, and it is possible to effectively reduce medical costs.

Drawings

Fig. 1 is a diagram showing an outline of the support system of the present embodiment.

Fig. 2 is a diagram showing a state in which the support system of the present embodiment is connected to a medical institution terminal and a scheduled visit person terminal via a network.

Fig. 3A is a block diagram showing a hardware configuration of the support system according to the present embodiment.

Fig. 3B is a block diagram showing a functional configuration of the support system according to the present embodiment.

Fig. 4A is a diagram showing the scheduled visit data, the visit data, and the visit data of the support system according to the present embodiment.

Fig. 4B is a diagram showing prescription data of the support system according to the present embodiment.

Fig. 4C is a diagram showing region data of the support system according to the present embodiment.

Fig. 4D is a diagram showing climate data of the support system of the present embodiment.

Fig. 4E is a diagram showing medical institution data of the support system of the present embodiment.

Fig. 5 is a flowchart showing the support method according to the present embodiment.

Fig. 6 is a diagram showing the contents of a prompt and the basis of the prompt displayed on the display of the medical institution terminal.

Detailed Description

Embodiments of the present invention will be described below with reference to the drawings. In the description of the drawings, the same elements are denoted by the same reference numerals, and redundant description is omitted. In addition, the dimensional ratio of the drawings is exaggerated for convenience of explanation and is different from the actual ratio.

Fig. 1 and 2 are diagrams for explaining the overall configuration of the support system 100 according to the present embodiment. Fig. 3A and 3B are diagrams for explaining each part of the support system 100. Fig. 4A to 4E are diagrams for explaining data processed by the support system 100.

As shown in fig. 1, the support system 100 is a system that presents whether or not a person scheduled to visit a patient who desires to make a visit needs to be examined, and presents prescription conditions of medical supplies (for example, whether or not a prescription of a medicine is needed, the type of a medicine, the amount of a medicine, the dosage form of a medicine, and the like), using the data of scheduled patients D1, the visit data D2, the visit data D3, the other data D4 (the regional data D41, the climate data D42, the medical institution data D43), and the like. The "medical institution" is not particularly limited, and refers to, for example, facilities for providing medical care to a person scheduled to visit a doctor or a nurse, and includes, for example, a hospital and a clinic. The "specific (fixed) region" is not particularly limited, and may be, for example, a region divided into city units, prefecture units, country units, and the like.

As shown in fig. 2, the support system 100 is connected to the medical institution terminals 200 of the respective medical institutions and the patient terminals 300 owned by the respective preschoolers via a network, and is configured as a server for receiving and transmitting data between the medical institution terminals 200 and the patient terminals 300. When a person scheduled to visit, such as an elderly person, arrives at a medical institution or before the visit, the person terminal 300 is operated to receive a presentation of a treatment policy from the support system 100. The healthcare practitioner (doctor, nurse, or the like) can confirm the medical guideline by using the medical institution terminal 200. The network can employ a wireless communication system based on a communication function such as Wifi (registered trademark) or Bluetooth (registered trademark), or other contactless wireless communication or wired communication.

In the present embodiment, the support system 100 is configured by an interactive device that can communicate with a person based on an interactive session. As the interactive device, a robot with an interactive function mounted with, for example, an AI can be used. The interactive device can be equipped with, for example, a display capable of displaying still images and video, a speaker capable of outputting audio and music, a camera function capable of capturing still images and video, and the like. The design of the interactive robot is not particularly limited, and examples thereof include a human type and an animal type.

The support system 100 will be described in detail below.

The configuration of hardware of the support system 100 will be described.

The support system 100 is not particularly limited, and may be configured by a mainframe, a computer cluster, or the like. As shown in fig. 3A, the support system 100 includes a CPU (Central processing unit) 110, a storage unit 120, an input/output I/F130, and a communication unit 140. The CPU110, the storage unit 120, the input/output I/F130, and the communication unit 140 are connected to a bus 150, and receive and transmit data and the like to and from each other via the bus 150.

The CPU110 executes control of each unit, various arithmetic processes, and the like in accordance with various programs stored in the storage unit 120.

The storage unit 120 is configured by a ROM (Read Only Memory) for storing various programs and various data, a RAM (random Access Memory) for temporarily storing programs and data as a work area, a hard disk for storing various programs including an operating system and various data, and the like.

The input/output I/F130 is an interface for connecting an input device such as a keyboard, a mouse, a scanner, and a microphone and an output device such as a display, a speaker, and a printer.

The communication unit 140 is an interface for communicating with the medical institution terminal 200, the doctor terminal 300, and the like.

Next, the main functions of the support system 100 will be described.

The storage unit 120 stores various data such as the preschool data D1, the visiting data D2, the visiting data D3, and the other data D4. The storage unit 120 stores a support program for providing the support method according to the present embodiment.

As shown in fig. 3B, the CPU110 functions as a data acquisition unit 111, a learning unit 112, and a presentation unit 113 by executing the support program stored in the storage unit 120.

The data acquisition unit 111 will be described.

The data acquiring unit 111 acquires the scheduled visit data D1, the visiting data D2, the visit data D3, and the other data D4.

As shown in fig. 4A, the scheduled visit data D1 includes, for example, an identification ID of the scheduled visit (e.g., data that can be acquired by MyNumber or the like), a scheduled visit name, an address, and an age. The visiting data D2 includes, for example, a past medical history (record of visiting to a medical institution). The visit data D3 includes, for example, the results of the previous visit to the medical institution and the results of the previous visits to the medical institution. The visit data D3 can include data acquired at the time of the visit (visit) when the scheduled person has experience in home medical care and home care.

The preschool data D1 can also include, for example, data relating to genetic information of the preschool. The genetic information includes not only the genetic information of the person scheduled for medical consultation but also the genetic information of relatives. The genetic information can be constituted by, for example, a DNA examination result or the like. The genetic information can be used to determine whether or not a disease is greatly affected by a genetic element when determining a disease of a person scheduled to see a doctor, for example.

The data D1 of the scheduled visit persons, the visiting data D2, and the data D3 of the visited persons are stored in the storage unit 120 in a state associated with each scheduled visit person. The data D1, D2, and D3 can be stored and managed by, for example, a known electronic medical record.

As shown in fig. 4B, the visit data D3 can include medical institution-side prescription data (prescription data) D31 and pharmacy-side prescription data (prescription data) D32. The medical institution-side prescription data D31 includes various data related to the prescription when, for example, a medical supply (e.g., a medicine or the like) is turned off by a medical institution by a past medical president. The medical institution-side prescription data D31 includes data relating to, for example, the date and time of prescription, the type of medical product, the amount of prescription, the dosage form, and the like. Further, the pharmacy-side prescription data D32 includes data on medical supplies actually prescribed to the person scheduled for the visit at the pharmacy based on the prescription provided at the medical institution. The pharmacy-side prescription data D32 includes data (prescription records and the like described in the drug record book) relating to date and time of prescription, type of medical product, prescription amount, dosage form, and the like, as in the case of the medical institution-side prescription data D31, for example. The medical product according to the present embodiment includes a so-called digital medical product having a digital function (a function of detecting biological information of a biological organ and acquiring the information after taking a medicine, for example). The medical service system can be used for sharing information and the like related to a medical treatment reservation acquired from a digital medical supply by a medical institution, a medical treatment reservation person and a medical practitioner, or for monitoring the medication state of the medical treatment reservation person.

The data acquiring unit 111 can acquire, for example, the scheduled-visit data D1, the visiting data D2, and the visiting data D3 from the medical institution terminal 200 of each medical institution and the scheduled-visit person terminal 300 of each scheduled-visit person.

The other data D4 to be acquired by the data acquiring unit 111 may include region data D41 shown in fig. 4C, climate data D42 shown in fig. 4D, and medical institution data D43 shown in fig. 4E.

As shown in fig. 4C, the region data D41 includes information on a specific region name, a population in the specific region, a main family structure in the specific region (for example, an average value of the number of people in the family in the specific region), an age group in the specific region (for example, an average value of the age group in the specific region), and whether or not the scheduled person has a medical record and a prescription record in the specific region. The region data D41 may include data of diseases and the like that are prevalent in a specific region, for example. In addition, the region data D41 can include, for example, data relating to traffic information in a specific region. The data related to the traffic information includes, for example, the distance from the home of the person scheduled for medical treatment to the position of the medical institution, and the type of available transportation means (e.g., public transport, electric train).

As shown in fig. 4D, the climate data D42 includes data on the climate (weather) related to the surrounding environment of each medical institution. The climate data D42 includes the weather, temperature, humidity, sunshine time of the surrounding environment.

The data acquiring unit 111 can acquire, for example, the region data D41 and the climate data D42 from the internet.

As shown in fig. 4E, the medical institution data D43 includes data on the name (medical institution name) of each medical institution, the address, the medical subject, the number of devices (including a bed, an ambulance, a medical instrument, a business instrument, and the like) held, the layout, the clinical route, the guideline, the doctor, and the like. These data are stored in the storage unit 120 in a state associated with each medical institution. The layout data can be composed of a layout diagram of a medical institution indicating the position and distance of each device, examination room, nurse station, general ward, ICU (Intensive Care Unit), HCU (High Care Unit), and the like. The data of the clinical route can be constituted by, for example, a schedule in which schedules from the admission to the discharge positions of a plurality of the scheduled patients are collected. Examples of the data of the guideline include data related to an educational guideline such as research and repair, and data related to a medical guideline such as intensive care. Note that, although not shown, the data of a doctor and the like includes data of a doctor name, medical subjects, medical experience, surgical experience, a service schedule, and the like. These data are stored in the storage unit 120 in a state associated with each doctor.

The medical institution data D43 may include data related to a miscellaneous condition of the medical institution, for example. The data related to the miscellaneous condition includes, for example, a miscellaneous condition (a miscellaneous condition related to an external situation, a miscellaneous condition related to admission, and the like) of a medical institution within a certain range from the own home of the person scheduled for medical treatment. For example, the support system 100 can provide the scheduled patient visit information (such as a schedule and a transfer guide) of the optimal transportation means or recommend a doctor having an excellent treatment result for a specific disease and present the medical institution in which such doctor works, based on the data on the traffic information and the data on the congestion condition, when the scheduled patient visit the predetermined medical institution. The support system 100 can automatically perform a medical institution presentation by the transportation means and a medical appointment in accordance with the arrival time to the medical institution.

In addition, the other data D4 can include reuse data relating to, for example, medical instruments and medical supplies. The reuse data includes information on whether the medical instrument can be reused by, for example, cleaning or sterilization. The medical device is, for example, a single-use medical device, but may be a medical device (a component of the medical device) other than the single-use medical device. In addition, the reuse data can include information relating to, for example, remaining medical supplies. The remaining medical supplies include information on whether or not a medicine (e.g., a liquid medicine) stored in a predetermined amount in a container such as a bottle can be used for a plurality of persons scheduled to visit. For example, a drug stored in a specific container can be administered to a person scheduled for a medical visit, and a drug stored in the same container can be administered to another person scheduled for a medical visit, the drug is regarded as reusable.

The reuse data can be acquired in real time from, for example, a Hospital Information System (Hospital Information System) of a medical institution having medical instruments and medical supplies to be reused.

The data acquisition unit 111 can acquire, for example, medical data as other information that helps support healthcare practitioners. The medical data is, for example, data related to medical knowledge, and includes disease data related to a disease (disease name, symptom, necessity of receiving treatment, and the like), treatment data related to treatment (treatment method, period required for treatment, necessary equipment, medicine, wholesale price thereof, and the like), data related to a medical insurance system, and the like. The data acquiring unit 111 can acquire medical data from the internet, or from electronic data of a medical professional book acquired by a scanner or the like, for example.

Next, the learning unit 112 will be described.

The learning unit 112 performs machine learning using the preschool data D1, the visiting data D2, the visiting data D3, and the other data D4. In the present specification, "machine learning" means that an algorithm is used to analyze input data, and useful rules, criteria for determination, and the like are extracted from the analysis result to expand the algorithm.

The support system 100 of the present embodiment provides both a presentation of whether or not a medical practitioner needs to make a medical examination and a presentation of the prescription of a medical product. The support system 100 performs the machine learning based on the above-described data so that the contents of the presentation do not become inappropriate contents. The support system 100 predicts the current and future dynamics of the scheduled visit person from the past dynamics of the scheduled visit person (visit frequency to the medical institution, contents of the visit, results of the visit, prescription of medical supplies, use status of medical supplies, etc.) by the mechanical learning by the learning unit 112, and presents an appropriate corresponding method to the healthcare practitioner based on the prediction result. The learning unit 112 can learn the prescription of an appropriate medical product based on, for example, the medical institution prescription data D31 and/or the pharmacy prescription data D32 of a plurality of persons.

Specifically, when a request for a medical visit is made from a person scheduled for a medical visit who visits a medical institution or a person scheduled for a medical visit before visiting the medical institution, the presentation unit 113 presents the medical practitioner whether or not to visit the medical practitioner based on the result of the mechanical learning by the learning unit 112. The presentation unit 113 also presents the prescription of the medical product to the person scheduled to visit the doctor by the healthcare worker. The prescription case mentioned here includes, for example, determination of whether or not a prescription of a medical product is required, and determination of the type, amount, usage, formulation, and the like of a medicine. As an example of the presentation by the presentation unit 113, it is also possible to present, for example, a plurality of persons who share an extra medical supply in one family (for example, a couple, a parent, a child, or the like) based on the medical institution-side prescription data D31 and/or the pharmacy-side prescription data D32, or present another person who uses a medical supply of a person who does not need to be taken due to a certain relationship in a predetermined population, and present another person who opens the same medical supply to purchase medical supplies in bulk, thereby reducing the purchase cost.

When presenting whether or not the medical practitioner needs to see a doctor and the prescription of the medical supplies, the presentation unit 113 presents the basis for presenting the presentation together with the presentation contents. For example, in the present embodiment, as will be described later, when it is determined that the examination by the healthcare practitioner is not necessary, the basis is presented based on each data. When there are a plurality of responses, a plurality of responses can be presented. The medical practitioner can adopt each presentation content in a manner of being able to recognize whether or not the medical practitioner needs to make a diagnosis and to give a prescription of the medical supplies and the basis thereof. In addition, the presentation method according to (evidence) may use a graph or a table to show, for example, a relationship between data, or specifically, figures to be used to derive things and phenomena according to factors, contribution rates, and the like.

In the present embodiment, the presentation unit 113 performs presentation when receiving a presentation request from a medical practitioner or a person scheduled to visit a doctor. However, the timing at which the presentation unit 113 performs presentation is not particularly limited. For example, the presentation unit 113 may automatically acquire data on an irregular or regular basis, and automatically present an appropriate course of action for the scheduled visit to the medical institution or medical practitioner when the scheduled visit to the medical institution is predicted even if the scheduled visit is not requested by the medical practitioner or scheduled visit person. For example, the presentation unit 113 may acquire data on the movement of the scheduled visit person irregularly or periodically and present a future prediction such as a treatment policy to the scheduled visit person who is predicted to visit the medical institution.

Fig. 5 and 6 are diagrams for explaining the support method according to the present embodiment. The supporting method of the present embodiment will be described below with reference to fig. 5 and 6.

Referring to fig. 5, the support method generally includes: a data acquisition step (S1) of acquiring the scheduled visit data D1, the visiting data D2, the visit data D3, and the other data D4; a learning step (S2) of performing machine learning by using the preschool data D1, the visiting data D2, the visiting data D3, and the other data D4; and a presentation step (S3) for presenting whether or not the medical practitioner needs to see a doctor and the medical product prescription condition based on the result of the machine learning. Hereinafter, each step will be described.

In addition, algorithms for mechanical learning are generally classified into supervised learning, unsupervised learning, reinforcement learning, and the like. In the supervised learning algorithm, a data set of an input and a result is given to the learning unit 112 to perform machine learning. In the unsupervised learning algorithm, the learning unit 112 is given only a large amount of input data to perform machine learning. In the reinforcement learning algorithm, correction is applied based on the environment changed by the solution output by the algorithm and on the degree to which the output solution is correct reward. The algorithm of the mechanical learning of the learning unit 112 may be any one of supervised learning, unsupervised learning, and reinforcement learning, and in the present embodiment, a case where the learning unit 112 performs the mechanical learning based on the algorithm of the supervised learning will be described as an example.

First, the data acquisition step (S1) will be explained.

In the data acquisition step (S1), the data acquisition unit 111 acquires the scheduled visit data D1, the visit data D2, the visit data D3, and the other data D4, and stores them in the storage unit 120. The timing at which the data acquiring unit 111 acquires the scheduled visit data D1, the visiting data D2, the visiting data D3, and the other data D4 is not particularly limited, and may be acquired at a predetermined time interval or at a timing at which these data change. The data acquiring unit 111 acquires the scheduled visit data D1, the visiting data D2, the visiting data D3, and the other data D4 over a predetermined period, and stores them in the storage unit 120. Therefore, the input data and the resolved data set for the supervised learning are stored in a large amount in the storage section 120.

For example, in the present embodiment, when a person scheduled to visit the medical institution visits a registration certificate, insurance card, or electronic medical record, common data of regional medical care, and data of persons scheduled to visit inside and outside a predetermined area (person scheduled to visit data D1, visit data D2, and visit data D3) are acquired and confirmed. In addition, at this time, the response to the person scheduled to visit is performed by a plurality of or a single dialogue-type device, and the testimony relating to the visit is listened from the person scheduled to visit. The results of auscultation are used together with the respective data in the learning step described later.

The method of acquiring information from the person scheduled for a visit is not limited to acquiring language information by auscultation as described above. For example, the support system 100 may acquire biological information. Examples of a method for acquiring biological information include a method for acquiring the degree of progression of arteriosclerosis by acquiring body temperature and oxygen saturation using infrared rays, and measuring a pulse wave of a peripheral blood vessel. In addition, the support system 100 can acquire information on the reaction (the degree of red tide of the face, the motor function, etc.) of the person scheduled for a visit during auscultation via the dialogue-type device. The support system 100 may also include an algorithm for determining the reliability of speech movement of the person scheduled to visit based on the auscultation and the information obtained by the above methods, and for confirming the validity of the information obtained from the person scheduled to visit.

The acquisition of the information from the person scheduled to visit may be performed only by the interactive device provided in the support system 100, but may be performed manually (by a healthcare professional, etc.), or may be performed by both the interactive device and the human. For example, when a person communicates with the person scheduled to visit through the interactive device and inputs the acquired information, it is possible to acquire the information from the person scheduled to visit more accurately and smoothly.

Next, the learning step (S2) will be described.

In the learning step (S2), the learning section 112 applies an algorithm of supervised learning to a large number of data groups stored in the storage section 120. The algorithm for supervised learning is not particularly limited, and examples thereof include known algorithms such as a least squares method, a linear regression, an autoregressive method, and a neural network.

The learning unit 112 predicts the current and future trends of the visit of the scheduled patient to the medical institution based on the acquired data. Further, referring to the auscultation result and the prediction result, a prompt indicating whether or not the medical practitioner is required to make a diagnosis and a prompt indicating the prescription dynamics of the medical supplies are executed.

The learning unit 112 can perform machine learning of information that contributes to determination of reuse of a medical instrument, for example, based on information such as whether the medical instrument can be reused in surgery, medical care, and the like, and whether reuse of the medical instrument is possible in a case where the medical instrument can be reused, which method (cleaning/sterilization method) can be used to realize reuse, and which constituent member of the medical instrument can be reused. The learning unit 113 can perform machine learning of information that contributes to determination of reuse of medical supplies for use in surgery, diagnosis, and the like, based on information such as whether or not the medical supplies can be reused, and by which method (storage method of medical supplies, providing method to a person scheduled to see a doctor) the medical supplies can be reused when the medical supplies can be reused. The presentation unit 113 can provide information on reuse of medical instruments and medical supplies to the medical institution by presenting the learning result of the machine learning. The medical institution acquires or shares the above-described learning result concerning reuse among a specific medical institution or a plurality of medical institutions, thereby effectively reducing medical costs.

Next, the presentation step (S3) will be described.

The presentation unit 113 can display presentation contents and presentation results on the display 210 of the medical institution terminal 200, as shown in fig. 6, for example. The presentation content and the presentation information can be displayed on, for example, a display 310 (see fig. 1) of the doctor terminal 300 owned by the person scheduled to visit, a display provided in the interactive device, or the like.

Referring to fig. 6, an example of the presentation contents and the presentation basis will be described.

For example, when it is determined that the doctor's examination is not necessary as the content of the presentation, the main cause of the determination result is displayed as the presentation. In addition, as the content of the presentation, the necessity of a medical examination and the result of determination of the prescription of the medicine may be displayed together.

As shown in fig. 6, the prompt contents contain, for example, the second opinion. The second opinion includes, for example, both a judgment as to whether a medical practitioner is required to make a medical examination and a judgment as to the prescription status of a medical product. If it is determined by the second opinion that a new prescription for medical supplies is required (for example, a medicine having a different prescription from the previous prescription is prescribed), the recommendation of the new prescription is presented, and if the same prescription as the previous prescription is prescribed, the remaining amount of the medical supplies is predicted based on the respective prescription data D31 and D32 (see fig. 4B), and the recommendation of the prescription for which only the shortage is made is presented.

As shown in fig. 6, the prompt content contains, for example, a notification. The notification means that, when it is determined that the result of auscultation by the person scheduled to visit, the previous examinations, and the prescription of medical supplies are not appropriately performed, it is proposed to notify the person scheduled to visit, the medical institution, the relative of the person scheduled to visit, and the like. The presentation unit 113 presents, for example, a notification to a public institution or the like when a determination result is obtained that the person scheduled for medical service intends to make a new visit or to repeat prescription of medical supplies.

In addition, as shown in fig. 6, the prompt content includes, for example, the utilization of a dialogue-type device. If it is determined that the person scheduled for medical examination has not visited the medical institution for the purpose of medical examination, the person scheduled for medical examination can be satisfied even if the person is not under the examination of the medical practitioner by executing an interactive device (communication). Therefore, the scheduled patient can be smoothly returned to home.

In addition, when presenting that the doctor is not required to make a study, for example, the presentation unit 113 may present a method other than a session by an interactive facility as another study behavior instead of the study by the doctor. The presentation unit 113 can present, for example, a conversation with a volunteer staff, a conversation with another person scheduled to visit a doctor, a petting of an animal, and the like.

When the support system 100 presents the support to the specific scheduled visit person, the data acquisition unit 111 may acquire the scheduled visit data D1, the visiting data D2, the visiting data D3, and other data again. Then, the learning unit 112 may perform the machine learning again using the newly acquired data to update the learning model. The support system 100 can predict, for example, future dynamics of the same scheduled visit person or different scheduled visit persons based on the updated learning model, and accumulate the results as new data to be applied at the next proposal.

As described above, the support system 100 according to the present embodiment includes: a data acquisition unit 111 for acquiring president data D1 relating to a president who is scheduled to visit the medical institution, visit data D2 relating to a history of visits made by the president to the medical institution, and visit data D3 relating to the content of previous visits made by the president in the medical institution; a learning unit 112 for performing machine learning using the preschool data D1, the visiting data D2, and the visiting data D3; and a presentation unit 113 that presents whether or not a diagnosis of the person scheduled to see a doctor is necessary based on the result of the machine learning.

As described above, the support system 100 presents whether or not the medical practitioner needs to make a diagnosis of the person scheduled to visit based on the result of the machine learning. The medical practitioner can avoid examining the scheduled patient who lacks the necessity of the treatment by referring to the presented contents. As a result, it is possible to prevent an increase in the workload of the healthcare worker and an excessive prescription of medical supplies due to the visit of the elderly to the medical institution, and it is possible to effectively reduce medical costs.

When presenting that no examination is necessary, the presentation unit 113 presents other examination behaviors instead of the examination by the healthcare worker. Therefore, even when the person scheduled for medical examination does not receive the examination of the medical practitioner, the person can obtain a high satisfaction feeling of visiting the medical institution.

The presentation unit 113 presents the interaction with the person scheduled to visit by the interactive device as another examination action. Therefore, the satisfaction of the person scheduled to visit can be further improved while suppressing an increase in the business load of the healthcare worker.

The visit data D3 includes prescription data D31 and D32 relating to medical supplies prescribed to the person scheduled to visit. The learning unit 112 learns the recommended prescription of the medical product based on the preschool data D1, the visiting data D2, the visiting data D3, and the prescription data D31 and D32. Then, the presentation unit 113 presents the prescription of the medical product based on the result of the machine learning. Therefore, the support system 100 can more appropriately determine whether or not a prescription of a medical product is required, and can provide an appropriate prescription amount and an appropriate type of medical product when the medical product is prescribed.

The presentation unit 113 presents presentation contents and presentation results together. Therefore, the medical practitioner, the person scheduled for a medical visit, and the like can adopt the presentation contents with approval.

Further, the support method of the present embodiment includes: a data acquisition step (S1) for acquiring preschool patient data D1 relating to a preschool patient who is scheduled to visit the medical institution, visit data D2 relating to the history of visits made by the preschool patient to the medical institution, and visit data D3 relating to the content of previous visits made by the preschool patient in the medical institution; a learning step (S2) of performing machine learning using the prescheduled visit data D1, the visiting data D2 and the visiting data D3; and a presentation step (S3) for presenting whether or not the examination of the person scheduled to see a doctor is necessary based on the result of the machine learning. Therefore, the medical practitioner can avoid performing a diagnosis on the scheduled patient who lacks the necessity of a diagnosis by referring to the presented contents. As a result, it is possible to prevent an increase in the business load of healthcare practitioners and an excessive prescription of medical supplies due to the visit of elderly persons to medical institutions, and it is possible to effectively reduce medical costs.

The support program according to the present embodiment executes the steps of: acquiring president data D1 relating to a president who is scheduled to visit the medical institution, visit data D2 relating to the history of visits made by the president to the medical institution, and visit data D3 relating to the content of previous visits made by the president in the medical institution; a learning step (S2) of performing machine learning using the prescheduled visit data D1, the visiting data D2 and the visiting data D3; and a presentation step (S3) for presenting whether or not the examination of the person scheduled to see a doctor is necessary based on the result of the machine learning. Therefore, the medical practitioner can avoid performing a diagnosis on the scheduled patient who lacks the necessity of a diagnosis by referring to the presented contents. As a result, it is possible to prevent an increase in the business load of healthcare practitioners and an excessive prescription of medical supplies due to the visit of elderly persons to medical institutions, and it is possible to effectively reduce medical costs.

The support system, the support method, and the support program according to the present invention have been described above by way of embodiments, but the present invention is not limited to the respective configurations described in the specification, and can be modified as appropriate based on the description of the scope of the patent claims.

For example, the support system, the support method, and the support program according to the above embodiments may share each acquired data and presentation content among a plurality of medical institutions, or may use only a single medical institution.

The data used for the machine learning in the support system of the present invention is not particularly limited as long as at least the scheduled visit data, the visit data, and the visit data are used. The content of the presentation may include whether or not the examination of the person scheduled to visit is required.

In addition, when the medical data includes prescription data, the prescription data may include data of at least one of the medical institution-side prescription data and the pharmacy-side prescription data.

In the support system according to the above-described embodiment, the learning unit performs the machine learning using an algorithm for supervised learning, but the algorithm used by the learning unit for the machine learning may be an algorithm for unsupervised learning or an algorithm for reinforcement learning. The learning unit may perform machine learning using a plurality of types of algorithms.

The means and method for performing various processes in the support system according to the above embodiment can be implemented by any of dedicated hardware circuits or programmed computers. The support program may be provided by a computer-readable recording medium such as a CD-ROM (Compact Disc Read Only Memory), or may be provided on-line via a network such as the internet. In this case, the program recorded on the computer-readable recording medium is usually transferred to a storage unit such as a hard disk and stored therein. The support program may be provided as a separate application.

The present application is proposed based on japanese patent application No. 2017-230847, which was proposed on 2017, 11, 30, the disclosure of which is incorporated herein by reference in its entirety.

Description of the reference numerals

100 support system (dialogue type equipment)

111 data acquisition part

112 learning part

113 presentation unit

D1 data of prescheduler

D2 visiting data

D3 data of medical treatment

D31 medical institution side prescription data (prescription data)

D32 pharmacy side prescription data (prescription data)

D4 other data

D41 geographical data

D42 climate data

D43 medical institution data.