JP7398849B1 - Programs, methods, information processing systems, and defecation sheets - Google Patents

Abstract

The present invention provides a system that allows continuous evaluation of disease activity through extremely simple operations by a patient. One form of the present disclosure is a program that is executed by an information processing system equipped with a server having a processor, the program causes the processor to execute an image of an image photographed by a terminal device used by a patient or a medical worker. a step of acquiring a stool image of a patient; a step of estimating a disease activity index related to a specific disease of the patient by image processing of the acquired stool image; and a step of estimating a disease activity index of a specific disease of the patient; A program that outputs steps and executes them. [Selection diagram] Figure 10

Description

The present disclosure relates to a program, a method, an information processing system, and a defecation sheet.

BACKGROUND ART Conventionally, systems are known that evaluate the current disease activity of a specific disease using information related to health conditions input by a user.
For example, the following non-patent literature describes how to accept daily input operations from patients regarding multiple input items related to their health status, and use the input information to perform scoring according to unique evaluation criteria. A system for estimating disease activity for ulcerative colitis is disclosed.

My IBD Care:Crohn’s & Colitis (https://play.google.com/store/apps/details?id=nhs.ibd.com.nhsibd&hl=ja&gl=US&pli=1)

However, in conventional systems, patients are required to perform input operations on multiple items every day regarding their daily health status. For this reason, patients may feel burdened by daily input operations, and may not be able to continue appropriate input operations, making it impossible to appropriately evaluate disease activity.

The present disclosure aims to provide a system that allows continuous evaluation of disease activity through extremely simple operations by the patient.

One aspect of the present disclosure is a program that is executed by an information processing system including a server having a processor, the program causing the processor to display a stool image of a patient taken by a terminal device used by the patient or a medical worker. a step of estimating a disease activity index related to a specific disease of a patient by image processing on the obtained stool image; and a step of outputting the estimated disease activity index of the patient to a terminal device. A program that executes .

According to the present disclosure, disease activity can be continuously evaluated through extremely simple operations by the patient.

FIG. 1 is a diagram illustrating an overview of a disease activity evaluation system of the present invention. 2 is a diagram showing the hardware configuration of the terminal device shown in FIG. 1. FIG. 3 is a diagram showing a functional configuration of the terminal device shown in FIG. 2. FIG. 2 is a diagram showing the hardware configuration of the evaluation server shown in FIG. 1. FIG. 5 is a diagram showing a functional configuration of the evaluation server shown in FIG. 4. FIG. It is a diagram showing an example of the structure of each database stored in the evaluation server. 2 is a diagram showing the hardware configuration of the medical information management server shown in FIG. 1. FIG. 7 is a diagram showing the functional configuration of the medical information management server shown in FIG. 6. FIG. It is a diagram showing an example of the structure of each database stored in the medical information management server. FIG. 1 is a diagram showing an overview of the present embodiment. FIG. 3 is a diagram illustrating a flow of disease activity evaluation processing according to the present invention. FIG. 3 is a diagram illustrating a flow of reservation processing according to the present invention. FIG. 3 is a diagram illustrating a flow of evaluation history aggregation processing according to the present invention. FIG. 2 is a diagram illustrating a flow of a process for generating an endoscope-equivalent image. 2 is a diagram showing an example of a first screen of the system 1. FIG. 3 is a diagram showing an example of a second screen of the system 1. FIG. 3 is a diagram showing an example of a third screen of the system 1. FIG. It is a figure explaining the flow of evaluation processing of disease activity of system 1 concerning the 1st modification. It is a figure explaining the flow of evaluation processing of disease activity of system 1 concerning the 2nd modification.

Hereinafter, one embodiment of the present invention will be described with reference to the drawings. In the following description, the same parts are given the same reference numerals. Their names and functions are also the same. Therefore, detailed description thereof will not be repeated.

<1. System 1 configuration>
The configuration of the system 1 will be explained below.

(1-1. Overall composition)
The information processing system 1 (hereinafter simply referred to as system 1) according to the present embodiment is a system that evaluates the disease activity of a specific disease in a user (patient or medical worker) using stool images taken by the user. be. In the following description, a fecal image refers to an image in which excrement including feces is the subject. In the following description, inflammatory bowel disease will be taken as an example of a specific disease.

Here, disease activity is an index quantitatively indicating the state of a disease. In general, disease activity is determined by scoring a plurality of evaluation items according to the condition, and comparing the total score with a preset standard to assign a grade of disease activity. For example, the following grades can be set as disease activity in inflammatory bowel disease.
・Lv1: The disease is very good. ・Lv2: There are mild lesions, but the disease is stable. ・Lv3: There are moderate lesions, and mild to moderate symptoms. ・Lv4: Severe lesions.・Lv5: Very severe lesions and very severe symptoms In the following explanation, the above grade will be referred to as the index of disease activity. Note that the evaluation criteria for disease activity can be set arbitrarily.

FIG. 1 is a diagram showing the overall configuration of a system 1. As shown in FIG.
As shown in FIG. 1, the system 1 includes a plurality of terminal devices 10 (in FIG. 1, a patient terminal 10A and a medical terminal 10B are shown. Hereinafter, they may be collectively referred to as "terminal devices 10"). , an evaluation server 20, and a medical institution server 30.

The terminal device 10 is an information processing device operated by each user who uses the system 1. The terminal device 10 is realized by a stationary PC (Personal Computer), a laptop PC, or a mobile terminal such as a smartphone or tablet compatible with the system 1. Terminal device 10 is connected to network 80 .

Here, the users of the system 1 include the following persons.
・Patients with a history of specific diseases ・Medical workers including doctors who provide medical care to patients ・System administrators who maintain and operate System 1

In the following description, the terminal device 10 will be distinguished as follows depending on the operator.
・Patient terminal 10A: Terminal device used by a patient ・Medical terminal 10B: Terminal device used by a medical worker ・Administrator terminal: Terminal device used by a system administrator (not shown)
Note that the configuration common to each of these terminals will be collectively described as the terminal device 10.

The evaluation server 20 is an information processing device that executes processing to evaluate the disease activity of the patient, mainly through image processing using stool images transmitted from the patient terminal 10A.
Evaluation server 20 may include a variety of computers, such as a personal computer, a server computer (eg, a web server, an application server, a database server, or a combination thereof). In this embodiment, the evaluation server 20 will be explained using a server computer as an example.

The medical institution server 30 is an information processing device that executes processing for managing information related to daily medical treatment of patients by medical personnel, mainly in medical institutions such as hospitals.
Medical institution server 30 may include a variety of computers, such as a personal computer, a server computer (eg, a web server, an application server, a database server, or a combination thereof). In this embodiment, the medical institution server 30 will be explained using a server computer as an example.

(1-2. Hardware configuration of terminal device 10)
FIG. 2 is a diagram showing the hardware configuration of the terminal device 10.
As shown in FIG. 2, the terminal device 10 includes a communication IF (Interface) 12, an input device 13, an output device 14, a memory 15, a storage section 16, and a processor 19.

The terminal device 10 is communicably connected to the evaluation server 20 and the medical institution server 30 via the network 80. The terminal device 10 has a wireless base station 81 that is compatible with communication standards such as 5G and LTE (Long Term Evolution), and a wireless LAN (Local Area Network) that is compatible with wireless LAN (Local Area Network) standards such as IEEE (Institute of Electrical and Electronics Engineers) 802.11. It is connected to the network 80 by communicating with a communication device such as a wireless LAN router 82.

The communication IF 12 is an interface for inputting and outputting signals so that the terminal device 10 communicates with an external device.
The input device 13 is an input device (for example, a touch panel, a touch pad, a pointing device such as a mouse, a keyboard, etc.) for receiving input operations from a user.

The output device 14 is an output device (display, speaker, etc.) for presenting information to the user.
The memory 15 is for temporarily storing programs and data processed by the programs, and is a volatile memory such as DRAM (Dynamic Random Access Memory).

The storage unit 16 is a storage device for storing programs and data, and is, for example, a flash memory or an HDD (Hard Disc Drive).
The programs include, for example, the following programs.
- Program of OS (Operating System) - Program of application (for example, web browser) that executes information processing The data includes, for example, the following data.
・Databases referenced in information processing ・Data obtained by executing information processing (that is, execution results of information processing)

The processor 19 is hardware for executing a set of instructions written in a program stored in the storage unit 16, and is composed of an arithmetic unit, registers, peripheral circuits, and the like.

(1-3. Functional configuration of terminal device 10)
FIG. 3 is a diagram showing the functional configuration of the terminal device 10.
As shown in FIG. 3, the terminal device 10 includes a plurality of antennas (an antenna 111, an antenna 112), a wireless communication section (a first wireless communication section 121, a second wireless communication section 122) corresponding to each antenna, and an operation It includes a reception section 130 (including a touch sensitive device 131 and a display 132), a position information sensor 150, a camera 160, a storage section 170, and a control section 180.

The terminal device 10 also has functions and configurations that are not particularly illustrated in FIG. 3 (for example, a battery for holding power, a power supply circuit that controls the supply of power from the battery to each circuit, etc.). As shown in FIG. 3, each block included in the terminal device 10 is electrically connected by a bus or the like.

The antenna 111 radiates a signal emitted by the terminal device 10 as a radio wave. Further, the antenna 111 receives radio waves from space and provides a received signal to the first wireless communication unit 121.
The antenna 112 radiates a signal emitted by the terminal device 10 as a radio wave. Further, the antenna 112 receives radio waves from space and provides a received signal to the second wireless communication unit 122.

The first wireless communication unit 121 performs modulation and demodulation processing for transmitting and receiving signals via the antenna 111 so that the terminal device 10 communicates with other wireless devices.
The second wireless communication unit 122 performs modulation and demodulation processing for transmitting and receiving signals via the antenna 112 so that the terminal device 10 communicates with other wireless devices.

The first wireless communication unit 121 and the second wireless communication unit 122 are communication modules including a tuner, an RSSI (Received Signal Strength Indicator) calculation circuit, a CRC (Cyclic Redundancy Check) calculation circuit, a high frequency circuit, and the like. The first wireless communication unit 121 and the second wireless communication unit 122 perform modulation/demodulation and frequency conversion of wireless signals transmitted and received by the terminal device 10 and provide received signals to the control unit 180.

The operation reception unit 130 has a mechanism for accepting user input operations. Specifically, the operation reception unit 130 is configured as a touch screen and includes a touch sensitive device 131 and a display 132.

Touch sensitive device 131 receives input operations from the user of terminal device 10 . The touch sensitive device 131 uses, for example, a capacitive touch panel to detect the position of the user's touch on the touch panel. The touch sensitive device 131 outputs a signal indicating the user's touch position detected by the touch panel to the control unit 180 as an input operation.

The display 132 displays data such as images, videos, and text under the control of the control unit 180. The display 132 is realized by, for example, an LCD (Liquid Crystal Display) or an organic EL (Electro-Luminescence) display.

The position information sensor 150 is a sensor that detects the position of the terminal device 10, and is, for example, a GPS (Global Positioning System) module. A GPS module is a receiving device used in a satellite positioning system. The satellite positioning system receives signals from at least three or four satellites, and detects the current position of the terminal device 10 equipped with a GPS module based on the received signals.
The camera 160 is a device that receives light with a light receiving element and outputs it as a photographed image.

The storage unit 170 is configured of, for example, a flash memory, and stores data and programs used by the terminal device 10. The storage unit 170 stores at least user information 171, a detection model 172, and various programs (not shown).

User information 171 is information regarding the user who uses the terminal device 10. The user information 171 includes account information (user ID, password for identifying the user), etc. that are requested to be input when the user logs into the system 1 using the terminal device 10. In addition, the user information 171 may include various types of information regarding user attributes that are registered when the user installs a predetermined application on the terminal device 10 in order to use the system 1.

The user ID may be a customer ID issued to the user.
The user ID is a service that provides various functions such as email, video calls, calendars, storage, document creation, spreadsheet creation, and news distribution in the form of SaaS (Software as a Service) (including for corporations). It may be a user account of

The detection model 172 is obtained by causing a machine learning model to perform machine learning in accordance with a model learning program based on the learning data. For example, in this embodiment, the detection model 172 is trained to output the position and size of human excrement and the type of excrement in response to input image information. Types of excrement include solid stool, mushy stool, watery stool, blood, mucus, and contaminants.

At this time, the learning data is, for example, image information of human body excrement as input data, and for the input image information, information about the position and size of the excrement, and information about the type of the excrement. Let , and be the correct output data.

Specifically, the detection model 172 compares the feature quantities of images of various types of excrement (reference feature quantities) with the feature quantities of the excrement captured in the image to be evaluated, and calculates the similarity. Determine whether any type of excrement is present by evaluating. When evaluating the similarity, if the feature amount of the image to be evaluated is within a preset threshold value with respect to the reference feature amount, the presence of the excrement is detected. Note that reference feature amounts are set for various types of excrement.

The detection model 172 according to this embodiment is, for example, a composite function with parameters that is a composite of a plurality of functions. A parameterized composite function is defined by a combination of multiple tunable functions and parameters. The detection model 172 according to the present embodiment may be any parameterized composite function that satisfies the above requirements, but is assumed to be a multilayer neural network model (hereinafter referred to as a "multilayer network"). The detection model 172 using a multilayer network has an input layer, an output layer, and at least one intermediate layer or hidden layer provided between the input layer and the output layer. Detection model 172 is envisioned for use as a program module that is part of artificial intelligence software.

As the multilayer network according to this embodiment, for example, a deep neural network (DNN), which is a multilayer neural network that is a target of deep learning, can be used. As the DNN, for example, a convolutional neural network (CNN) that targets images may be used.

The control unit 180 controls the operation of the terminal device 10 by reading a program stored in the storage unit 170 and executing instructions included in the program. The control unit 180 is, for example, an application installed in the terminal device 10 in advance. The control section 180 functions as an input reception section 181, a transmission/reception section 182, a photographing section 183, a detection section 184, a data processing section 185, and a display control section 186 by operating according to a program.

The input receiving unit 181 performs a process of receiving a user's input operation on the input device 13 such as the touch sensitive device 131.

The transmitting/receiving unit 182 performs processing for the terminal device 10 to transmit and receive data with an external device such as the evaluation server 20 according to a communication protocol.

The photographing unit 183 performs processing for activating the camera 160 and photographing an image of the subject in response to an operation from the patient.

The detection unit 184 detects stool included in the image photographed by the photographing unit 183. At this time, the detection unit 184 inputs the image data to be evaluated into the detection model 172, and detects excrement such as feces included in the image based on the information output from the detection model 172. This process will be described later.

The data processing unit 185 performs calculations on the data input by the terminal device 10 according to a program, and outputs the calculation results to a memory or the like. For example, the data processing unit 185 controls storage processing of captured images in the storage unit 170. When feces is detected in the image, the data processing unit 185 transmits the image to the evaluation server 20 without storing the image in the storage unit 170 of the terminal device 10.

The display control unit 186 performs a process of displaying information on the display 132 to present the information to the user. The display control unit 186 has a function as a web browser, and accesses the information output by the evaluation server 20 to the logical line (TCP connection) with the terminal device 10, and displays the information on the display 132 of the terminal device 10. Perform display processing (rendering).

(1-4. Hardware configuration of evaluation server 20)
FIG. 4 is a diagram showing the hardware configuration of the evaluation server 20.
As shown in FIG. 4, the evaluation server 20 includes a communication IF 22, an input/output IF 23, a memory 25, a storage 26, and a processor 29.

The communication IF 22 is an interface for inputting and outputting signals so that the evaluation server 20 communicates with external devices.
The input/output IF 23 functions as an interface with an input device for accepting input operations from a user and an output device for presenting information to the user.

The memory 25 is for temporarily storing programs and data processed by the programs, and is a volatile memory such as DRAM (Dynamic Random Access Memory).

The storage 26 is a storage device for storing programs and data, and is, for example, a flash memory or an HDD (Hard Disc Drive).
The programs include, for example, the following programs.
- Program of OS (Operating System) - Program of application (for example, web browser) that executes information processing The data includes, for example, the following data.
・Databases referenced in information processing ・Data obtained by executing information processing (that is, execution results of information processing)

The processor 29 is hardware for executing a set of instructions written in a program stored in the storage 26, and is composed of an arithmetic unit, registers, peripheral circuits, and the like.

(1-5. Functional configuration of evaluation server 20)
FIG. 5 is a diagram showing the functional configuration of the evaluation server 20.
As shown in FIG. 5, the evaluation server 20 functions as a communication section 201, a storage section 202, and a control section 203.

The communication unit 201 performs processing for the evaluation server 20 to communicate with external devices.

The storage unit 202 stores data and programs used by the evaluation server 20. The storage unit 202 stores, for example, the following data.
・User database (user DB) 2021
・Image database (image DB) 2022
・Object database (object DB) 2023
・Tag database (tag DB) 2024
・Comment database (comment DB) 2025
・Evaluation history database (evaluation history DB) 2026
・Activity estimation model 2027
・Image synthesis model 2028

The user DB 2021 is a database that manages information regarding users who use the system 1 as patients. Details of the data structure of the user DB 2021 will be described later.

The image DB 2022 is a database that manages information regarding images acquired by the terminal device 10. Details of the data structure of the image DB 2022 will be described later.

The object DB 2023 is a database that manages information regarding objects that constitute excrement detected from images. Examples of objects constituting excrement include feces, blood, mucus, and contaminants. Details of the data structure of the object DB 2023 will be described later.

The tag DB 2024 is a database that manages information regarding tags selected by the patient when photographing stool. In the system 1, a plurality of items for history of health conditions and excretory behavior are set as tags, and the user can select tags for the corresponding items when photographing stool. Details of the data structure of the tag DB 2024 will be described later.

The comment DB 2025 is a database that manages information regarding comments arbitrarily input by the user when photographing a stool. Details of the data structure of the comment DB 2025 will be described later.

The evaluation history DB 2026 is a database that manages the history of the evaluation results of disease activity by the system 1. Details of the data structure of the evaluation history DB 2026 will be described later.

The activity estimation model 2027 is obtained by causing a machine learning model to perform machine learning according to a model learning program based on the learning data. For example, in the present embodiment, the activity estimation model 2027 outputs an estimated value of the disease activity index for inflammatory bowel disease in the person (patient) who excreted the stool, based on the input stool image. being learned. Furthermore, the activity estimation model 2027 compares the input stool image with the learned dataset and outputs a likelihood indicating the likelihood of the estimated value of the disease activity index.

At this time, the learning data is, for example, image information of excrement from a human body as input data, and based on the input image information, it is determined by a doctor about the person (patient) who excreted the stool at the same time. The value of the disease activity index for inflammatory bowel disease is used as the correct output data.

Furthermore, the activity estimation model 2027 may further output the estimated position of the lesion range in the disease with respect to the input stool image. In this case, the image information of the excrement is used as input data, and the information regarding the lesion range of the inflammatory bowel disease of the person (patient) who excreted the stool is used as the correct output data.

The activity estimation model 2027 according to this embodiment is, for example, a composite function with parameters that is a composite of a plurality of functions. A parameterized composite function is defined by a combination of multiple tunable functions and parameters. The activity estimation model 2027 according to the present embodiment may be any composite function with parameters that satisfies the above requirements, but is assumed to be a multilayer neural network model (hereinafter referred to as "multilayer network"). The activity estimation model 2027 using a multilayer network has an input layer, an output layer, and at least one intermediate layer or hidden layer provided between the input layer and the output layer. The activity estimation model 2027 is assumed to be used as a program module that is part of artificial intelligence software.

The image synthesis model 2028 is obtained by having a machine learning model perform machine learning in accordance with a model learning program based on the learning data. For example, in this embodiment, the image synthesis model 2028 learns to output an endoscope-equivalent image that is estimated as the state of the intestine of the person (patient) who excreted the stool, based on the input stool image. has been done.

At this time, the learning data is, for example, image information of human excrement as input data, and based on the input image information, an endoscopy is performed on the person (patient) who excreted the stool at the same time. The photographed endoscopic image of the intestine is used as the correct output data.

The image synthesis model 2028 according to this embodiment is, for example, a parameterized synthesis function in which a plurality of functions are synthesized. A parameterized composite function is defined by a combination of multiple tunable functions and parameters. The image synthesis model 2028 according to this embodiment may be any parameterized synthesis function that satisfies the above requirements, but is assumed to be a multilayer neural network model (hereinafter referred to as a "multilayer network"). The image synthesis model 2028 using a multilayer network has an input layer, an output layer, and at least one intermediate layer or hidden layer provided between the input layer and the output layer. Image synthesis model 2028 is envisioned for use as a program module that is part of artificial intelligence software.

The control unit 203 performs functions shown as various modules when the processor 29 of the evaluation server 20 performs processing according to a program. The control unit 203 functions as a transmission/reception control module 2031 , an acquisition module 2032 , an evaluation module 2033 , a determination module 2034 , a totaling module 2035 , an image generation module 2036 , and an output module 2037 .

The transmission/reception control module 2031 controls processing in which the evaluation server 20 transmits/receives signals with an external device according to a communication protocol.

The acquisition module 2032 acquires information regarding the stool image transmitted from the patient terminal 10A in response to the patient's operation. The acquisition module 2032 stores information regarding the acquired stool images in a predetermined storage area, and then records a new record in the database to be updated among the image DB 2022, object DB 2023, and comment DB 2025.

The evaluation module 2033 evaluates the patient's disease activity through image processing using the acquired stool images. Specifically, the evaluation module 2033 obtains the estimated value of the disease activity index output from the activity estimation model 2027 by inputting the stool image to the activity estimation model 2027. The evaluation module 2033 records the obtained estimated value of the disease activity index as a new record in the evaluation history DB 2026.
Note that the evaluation module 2033 may input object data related to excrement detected from the stool image to the activity estimation model 2027.

The determination module 2034 determines whether or not the patient requires examination by a doctor, based on the evaluation result of disease activity performed by the evaluation module 2033, using preset determination conditions regarding the necessity of an appointment. judge. If the determination module 2034 determines that the patient requires medical examination by a doctor, the determination module 2034 notifies the patient and the attending physician to that effect.

The aggregation module 2035 aggregates the evaluation history according to a preset frequency. The aggregation module 2035 refers to the evaluation history DB 2026 and aggregates, for example, the following information as the evaluation history of disease activity for the patient to be aggregated.
・Percentage of days on which stool images were acquired during the period covered by the calculation ・Changes in evaluation results during the period covered by the calculation ・Contents of comments from patients or doctors during the period covered by the calculation

The image generation module 2036 performs image processing using the acquired stool image to generate an endoscope-equivalent image of the patient when excreting stool related to the stool image. An endoscope-equivalent image is information that represents the state of the patient's intestine as an image equivalent to an endoscopic image, which is assumed to be taken when actually taken using an endoscope. .
The image generation module 2036 acquires an endoscope-equivalent image output from the image synthesis model 2028 by inputting the stool image to the image synthesis model 2028. After storing the endoscope-equivalent image in a predetermined storage area, the image generation module 2036 creates address information (for example, URL information) indicating the storage area.

The output module 2037 outputs various types of information obtained from the executed processing in response to operations by a patient or a medical worker. The output module 2037 outputs, for example, the following information.
・Information on the evaluation results of disease activity for the patient terminal 10A and the medical terminal 10B ・Information on the aggregated results of the evaluation history for the patient terminal 10A and the medical terminal 10B ・Consultation reservation for the patient terminal 10A Information that the proposal is being made/Information that the generated endoscope-equivalent image has been created for the medical terminal 10B

(1-6. Data structure of each database managed by evaluation server 20)
Next, an example of the data structure of each database stored in the storage unit 202 of the evaluation server 20 will be described using FIG. 6.

(1-6-1. User DB2021)
FIG. 6A is a diagram showing an example of the data structure of the user DB 2021.
As shown in FIG. 6A, the user DB 2021 stores information regarding users who use the system 1 as patients. A new record is recorded in the user DB 2021 by a user registration operation when a patient starts using the system 1.

The user DB 2021 includes the item "user ID", the item "name", the item "date of birth", the item "gender", and the item "patient ID".

The item "user ID" stores user identification information that can identify a user who uses the system 1 as a patient.

The item "Name" stores information regarding the name of the user corresponding to the user ID.

The item "date of birth" stores information regarding the date of birth of the user corresponding to the user ID.

The item “gender” stores information regarding the gender of the user corresponding to the user ID.

The item "Patient ID" stores identification information as a patient managed by the medical institution server 30 of the user corresponding to the user ID. Note that in place of the patient ID managed only by a specific medical institution, the insured number of the health insurance card, the My Number, etc. may be used.
Note that the structure of the user DB 2021 shown in FIG. 6A is merely an example, and the user DB 2021 may include columns in which other data items are stored.

(1-6-2. Image DB2022)
FIG. 6B is a diagram showing an example of the data structure of the image DB 2022.
As shown in FIG. 6B, the image DB 2022 stores information regarding captured stool images. A new record is recorded in the image DB 2022 when a stool image is transmitted to the evaluation server 20 and acquired by the acquisition module 2032.

The image DB 2022 includes the item "Image ID", the item "User ID", the item "Photographing date and time", the item "Photographing terminal", the item "Photographing location", the item "Tag ID", and the item "Storage area". ”.

The item "Image ID" stores identification information of a stool image that can identify the stool image acquired by the evaluation server 20.

The item "user ID" stores the user ID of the patient who excreted the stool associated with the stool image corresponding to the image ID.

The item "photographing date and time" stores information regarding the date and time when the flight image corresponding to the image ID was photographed.

The item "photographing terminal" stores identification information of the terminal device 10 that can identify the terminal that photographed the stool image corresponding to the image ID.

The item “photographing location” stores information regarding the place where the stool image corresponding to the image ID was photographed. Information regarding the photographing location can be obtained by linking the position information acquired by the position information sensor 150 in the terminal device 10 that took the photograph with the photographed image.

The item "tag ID" stores identification information of a tag selected by the user as a corresponding item at the time of photographing the flight image corresponding to the image ID.

The item "save area" stores address information (for example, URL information) of a storage area where the stool image corresponding to the image ID is saved. The system 1 may utilize an external storage area for storing image data.
Note that the structure of the image DB 2022 shown in FIG. 6B is merely an example, and the image DB 2022 may include columns in which other data items are stored.

(1-6-3. Object DB2023)
FIG. 6C is a diagram showing an example of the data structure of the object DB 2023.
As shown in FIG. 6C, the object DB 2023 stores information regarding objects that constitute excrement included in the stool image. A new record is recorded in the object DB 2023 when information regarding excrement detected by the detection unit 184 of the patient terminal 10A is transmitted to the evaluation server 20 together with a stool image and acquired by the acquisition module 2032.

The object DB 2023 includes the item "object ID", the item "image ID", the item "object type", and the item "position and range of object".

The item "object ID" stores identification information of an object that can identify an object constituting excrement detected to be included in the stool image.

The item "image ID" stores the image ID of the stool image in which the object corresponding to the object ID is detected.

The item "object type" stores the type of excrement to which the object corresponding to the object ID corresponds. Examples of the types of excrement include the following:
・Solid stool, muddy stool, watery stool, blood, mucus, contaminants

The item "position and range of object" stores information that can specify the position and range of the object in the image in which the object corresponding to the object ID is detected.
Note that the structure of the object DB 2023 shown in FIG. 6C is merely an example, and the object DB 2023 may include columns in which other data items are stored.

(1-6-4. Tag DB2024)
FIG. 6D is a diagram showing an example of the data structure of the tag DB 2024.
As shown in FIG. 6D, the tag DB 2024 stores information regarding tags input by the patient when photographing stool. A plurality of pieces of information regarding tags are prepared in advance and are already stored when the system 1 is started to be used.

The tag DB 2024 includes the item "tag ID", the item "tag name", and the item "tag description".

The item "tag ID" stores tag identification information that allows identification of the tag.

The item "tag name" stores the name of the tag corresponding to the tag ID.

The item "tag description" stores a description of the tag corresponding to the tag ID. Here, specific examples of tag names and descriptions will be given.
・Abdominal pain tag: Select this tag if you feel abdominal pain on the day of excretion or the day before ・Bleeding tag: Select this tag if you experience bleeding during excretion ・Anorexia: Select this tag if you have a loss of appetite on the day of excretion or the day before Tag/fever: A tag to be selected when there is a fever on the day of excretion or the day before. Note that the structure of the tag DB 2024 shown in FIG. 6D is just an example, and the tag DB 2024 includes columns in which other data items are stored. Good too.

(1-6-5.Comment DB2025)
FIG. 6E is a diagram showing an example of the data structure of the comment DB 2025.
As shown in FIG. 6E, the comment DB 2025 stores information regarding comments input by the patient when photographing the stool. A new record is recorded in the object DB 2023 when the stool image is transmitted to the evaluation server 20 and acquired by the acquisition module 2032.

The comment DB 2025 includes the item "comment ID", the item "image ID", and the item "patient comment".

The item "comment ID" stores comment identification information that allows the comment to be identified.

The item "image" stores the image ID of the stool image in which the comment corresponding to the comment ID has been input.

The item "patient comment" stores the content of the comment corresponding to the comment ID.
Note that the structure of the comment DB 2025 shown in FIG. 6E is merely an example, and the comment DB 2025 may include columns in which other data items are stored.

(1-6-6. Evaluation history DB2026)
FIG. 6F is a diagram showing an example of the data structure of the evaluation history DB 2026.
As shown in FIG. 6F, the evaluation history DB 2026 stores the history of evaluation results regarding disease activity by the system 1. A new code is recorded in the evaluation history DB 2026 when the evaluation module 2033 evaluates disease activity.

The evaluation history DB 2026 includes the item "evaluation history ID", the item "image ID", the item "evaluation date and time", the item "evaluation result", the item "likelihood", the item "attending physician's comment", and the item " including whether medical examination is necessary.

The item “evaluation history ID” stores evaluation history identification information that allows identification of evaluation results obtained through processing of the system 1.

The item "image ID" stores the image ID of the stool image that is the subject of evaluation corresponding to the evaluation history ID.

The item "evaluation date and time" stores information regarding the date and time when the evaluation corresponding to the evaluation history ID was performed. As the evaluation date and time, the date and time when the flight to be evaluated was photographed may be managed.

The item “evaluation result” stores information regarding the evaluation result of the disease activity index in the evaluation history corresponding to the evaluation history ID.

The item “likelihood” stores the likelihood of the evaluation result of disease activity in the evaluation history corresponding to the evaluation history ID. The likelihood of an evaluation result is a value that quantitatively indicates the probability of the evaluation result, and is a value that suggests the credibility of the evaluation result.

The item "attending physician comment" stores comments input by the attending physician regarding the evaluation history corresponding to the evaluation history ID and the stool image that was the subject of the evaluation.

The item "Necessity of medical examination" stores the determination result as to whether or not a medical examination by a doctor at a medical institution is necessary for the patient who is the subject of the evaluation by the system 1, based on the evaluation history corresponding to the evaluation history ID. be done.
Note that the structure of the evaluation history DB 2026 shown in FIG. 6F is merely an example, and the evaluation history DB 2026 may include columns in which other data items are stored.

(1-7. Hardware configuration of medical institution server 30)
FIG. 7 is a diagram showing the hardware configuration of the medical institution server 30.
As shown in FIG. 7, the medical institution server 30 includes a communication IF 32, an input/output IF 33, a memory 35, a storage 36, and a processor 39.

The communication IF 32 is an interface for inputting and outputting signals so that the medical institution server 30 communicates with an external device.
The input/output IF 33 functions as an interface with an input device for accepting input operations from a user and an output device for presenting information to the user.

The memory 35 is for temporarily storing programs and data processed by the programs, and is a volatile memory such as DRAM (Dynamic Random Access Memory).

The storage 36 is a storage device for storing programs and data, and is, for example, a flash memory or an HDD (Hard Disc Drive).
The programs include, for example, the following programs.
- Program of OS (Operating System) - Program of application (for example, web browser) that executes information processing The data includes, for example, the following data.
・Databases referenced in information processing ・Data obtained by executing information processing (that is, execution results of information processing)

The processor 39 is hardware for executing a set of instructions written in a program stored in the storage 36, and is composed of an arithmetic unit, registers, peripheral circuits, and the like.

(1-8. Functional configuration of medical institution server 30)
FIG. 8 is a diagram showing the functional configuration of the medical institution server 30.
As shown in FIG. 5, the medical institution server 30 functions as a communication section 301, a storage section 302, and a control section 303.

The communication unit 301 performs processing for the medical institution server 30 to communicate with an external device.

The storage unit 302 stores data and programs used by the medical institution server 30. The storage unit 302 stores, for example, the following data.
・Patient database (patient DB3021) 3021
・Doctor database (physician DB3023) 3022
・Medical history database (medical history DB3022) 3023
・Medical treatment time frame database (medical treatment time frame DB3024) 3024
・Reservation database (reservation DB3025) 3025

The patient DB 3021 is a database that manages information regarding patients receiving medical treatment at a medical institution. Details of the data structure of the patient DB 3021 will be described later.

The doctor DB 3023 is a database that manages information regarding doctors belonging to a medical institution. Details of the data structure of the doctor DB 3023 will be described later.

The medical history DB 3022 is a database that manages information regarding the history of medical treatments performed at medical institutions. Details of the data structure of the medical history DB 3022 will be described later.

The medical treatment time frame DB 3024 is a database that manages information regarding medical treatment time frames scheduled at medical institutions. Details of the data structure of the medical treatment time frame DB 3024 will be described later.

The reservation DB 3025 is a database that manages information regarding medical treatment reservations at medical institutions. Details of the data structure of the reservation DB 3025 will be described later.

The control unit 303 performs functions shown as various modules when the processor 39 of the medical institution server 30 performs processing according to a program. The control unit 303 functions as a transmission/reception control module 3031 , an input reception module 3032 , an inquiry module 3033 , a reservation module 3034 , and an output module 3035 .

The transmission/reception control module 2031 controls processing in which the medical institution server 30 transmits/receives signals with an external device according to a communication protocol.

The input reception module 3032 performs a process of accepting input information in response to an operation from the terminal device 10. The input reception module 3032 receives, for example, a medical history input during daily medical treatment by a doctor at a medical institution, and records it as a new record in the medical history DB.

The inquiry module 3033 refers to the medical treatment time slot DB 3024 and the reservation DB 3025 for the desired reservation date included in the reservation request, and inquires about available reservation slots.

The reservation module 3034 receives a reservation request for a reservation available slot in response to an operation from the terminal device 10, records the reservation request as a new record in the reservation DB 3025, and makes a reservation for a medical examination.

The output module 3035 outputs various types of information obtained from the executed processing in response to operations by the patient or medical personnel. The output module 3035 outputs, for example, the following information.
・Information regarding available reservation slots for the patient terminal 10A ・Reservation completion notification for the patient terminal 10A ・Past medical history of patients to be treated for the medical terminal 10B

(1-9. Data structure of each database managed by the medical institution server 30)
Next, an example of the data structure of each database stored in the storage unit 302 of the medical institution server 30 will be described using FIG. 9.

(1-9-1. Patient DB3021)
FIG. 9A is a diagram showing an example of the data structure of the patient DB 3021.
As shown in FIG. 9A, the patient DB 3021 stores information regarding patients who use medical institutions. A new record is recorded in the patient DB 3021 when a patient is registered at a medical institution for the first time.

The patient DB 3021 includes the following items: "Patient ID", "Name", "Date of Birth", "Gender", "Address", "Telephone Number", and "Insured Person Information". ” and the item “attending physician.”

The item "Patient ID" stores patient identification information that can identify the patient at a medical institution.

The item "Name" stores information regarding the name of the patient corresponding to the patient ID.

The item "date of birth" stores information regarding the date of birth of the patient corresponding to the patient ID.

The item “gender” stores information regarding the gender of the patient corresponding to the patient ID.

The item "address" stores information regarding the address of the patient corresponding to the patient ID.

The item "telephone number" stores information regarding the patient's telephone number corresponding to the patient ID.

The item "Insured Number" stores the insured number in the health insurance card of the patient corresponding to the patient ID.

The item "attending physician" stores identification information of the attending physician in charge of the patient corresponding to the patient ID.
Note that the structure of the patient DB 3021 shown in FIG. 9A is merely an example, and the patient DB 3021 may include columns in which other data items are stored.

(1-9-2.Doctor DB3023)
FIG. 9B is a diagram showing an example of the data structure of the doctor DB 3023.
As shown in FIG. 9B, the doctor DB 3023 stores information regarding doctors who belong to medical institutions. A new record is recorded in the doctor DB 3023 when a doctor is assigned to a medical institution.

The doctor DB 3023 includes the item "Doctor ID", the item "Name", the item "Clinical department", the item "Qualification", the item "Position", and the item "Contact information".

The item "Doctor ID" stores identification information of a doctor that can identify the doctor at the medical institution.

The item "Name" stores information regarding the name of the doctor corresponding to the doctor ID.

The item “Clinical Department” stores information regarding the medical department to which the doctor corresponding to the doctor ID belongs.

The item "Qualification" stores information regarding qualifications such as specialist qualification held by the doctor corresponding to the doctor ID.

The item "position" stores information regarding the position of the doctor corresponding to the doctor ID in the medical department.

The item "contact information" stores information such as a telephone number or email address as the contact information of the doctor corresponding to the doctor ID.
Note that the structure of the doctor DB 3023 shown in FIG. 9B is merely an example, and the doctor DB 3023 may include columns in which other data items are stored.

(1-9-3. Medical history DB3022)
FIG. 9C is a diagram showing an example of the data structure of the medical history DB 3022.
As shown in FIG. 9C, the medical history DB 3022 stores information regarding the history of medical treatment at medical institutions. In the medical history DB 3022, a new record is recorded when a medical professional including a doctor inputs the contents of medical treatment performed at a medical institution.

The medical history DB 3022 includes the item "medical history ID", the item "patient ID", the item "physician in charge", the item "medical date and time", the item "medical content", the item "medical content", and the item " "Test results", the item "Prescription contents", and the item "Treatment, surgery information".

The item "Medical history ID" stores identification information of a medical history that can identify the medical treatment performed at a medical institution.

The item “Patient ID” stores identification information of a patient who received medical treatment corresponding to the medical history ID.

The item "Physician in Charge" stores identification information of the doctor in charge of the medical treatment corresponding to the medical history ID.

The item "Date and time of medical treatment" stores information regarding the date and time when the medical treatment corresponding to the medical history ID was performed.

The item "medical treatment content" stores the contents of the medical treatment corresponding to the medical history ID. The contents of medical treatment include, for example, the following:
・Information regarding subjective symptoms complained of by the patient to the doctor ・Medical findings made by the doctor during the relevant treatment (including index of disease activity at the time of treatment)

The item “test results” stores information regarding the results of tests performed in the medical treatment corresponding to the medical history ID.

The item "prescription content" stores information regarding medicines prescribed by a doctor to a patient in the medical treatment corresponding to the medical history ID.

The item "Treatment, surgery information" stores information regarding the process or surgery performed in the medical treatment corresponding to the medical history ID.
Note that the structure of the medical history DB 3022 shown in FIG. 9C is merely an example, and the medical history DB 3022 may include columns in which other data items are stored.

(1-9-4. Medical treatment time frame DB3024)
FIG. 9D is a diagram showing an example of the data structure of the medical treatment time frame DB 3024.
As shown in FIG. 9D, the medical treatment time frame DB 3024 stores information regarding medical treatment schedules for each clinical department at a medical institution. The medical treatment time frame DB 3024 is set based on the medical treatment time zone set at the medical institution, and a new record is recorded when the work shift of the doctor in charge in the target period is set.

The medical treatment time frame DB 3024 includes the item “medical treatment time frame ID”, the item “medical department”, the item “in charge doctor”, the item “medical treatment date and time”, and the item “medical treatment time”.

The item "medical treatment time frame ID" stores identification information of a medical treatment time frame that can identify a medical treatment time frame at a medical institution.

The item “Clinical Department” stores information regarding the medical department corresponding to the medical treatment time frame corresponding to the medical treatment time frame ID.

The item “Physician in Charge” stores identification information of a doctor in charge of the medical treatment corresponding to the medical treatment time slot ID.

The item "Date and time of medical treatment" stores information regarding the date and time of the medical treatment corresponding to the medical treatment time frame ID.

The item “medical treatment time slot” stores information regarding the medical treatment time zone corresponding to the medical treatment time slot ID.
Note that the structure of the medical treatment time frame DB 3024 shown in FIG. 9D is merely an example, and the medical treatment time frame DB 3024 may include columns in which other data items are stored.

(1-9-5. Reservation DB3025)
FIG. 9E is a diagram showing an example of the data structure of the reservation DB 3025.
As shown in FIG. 9E, the reservation DB 3025 stores information regarding medical treatment reservations at medical institutions. In the reservation DB 3025, a new record is recorded when a patient makes a reservation for medical treatment.

The reservation DB 3025 includes the item “reservation ID”, the item “patient ID”, the item “attending doctor”, and the item “reservation time frame”.

The item "Reservation ID" stores reservation identification information that can identify a medical treatment reservation at a medical institution.

The item "Patient ID" stores identification information of a patient receiving treatment in a medical treatment reservation corresponding to the reservation ID.

The item "Physician in Charge" stores identification information of the doctor in charge of the medical treatment in the medical treatment reservation corresponding to the reservation ID.

The item "reservation time frame" stores identification information of the medical treatment time frame to which the medical treatment reservation corresponding to the reservation ID corresponds.
Note that the structure of the reservation DB 3025 shown in FIG. 9E is merely an example, and the reservation DB 3025 may include columns in which other data items are stored.

<2. Overview of embodiment>
Hereinafter, an overview of an embodiment of the present invention will be described with reference to FIG. FIG. 10 is a diagram showing an overview of an embodiment of the present invention.

As shown in FIG. 10, in the system 1, a photograph of stool by a patient is performed. Stool photography is performed, for example, at the patient's home on a regular basis (for example, every day). A terminal device 10 (patient terminal 10A) such as a smartphone that the patient uses in daily life is used to photograph the stool. It is preferable to photograph the first stool of the day.

Next, the acquired stool image is transmitted to the evaluation server 20, and the evaluation server 20 determines the patient's disease activity. The evaluation server 20 notifies the patient terminal 10A of the evaluation results. Users can understand their own disease activity based on the state of their stool that day.

The evaluation server 20 also notifies the medical terminal 10B used by the attending physician of the evaluation results. The attending physician will confirm the notified patient's disease activity as necessary. At this time, the evaluation server 20 may issue an alert to the attending physician regarding a patient whose disease activity is worsening.

The evaluation server 20 also performs API cooperation with the medical institution server 30 to perform cooperation regarding medical reservations. The evaluation server 20 notifies the patient that medical treatment by a doctor is required when the predetermined criteria set in advance are met. Here, for example, the following patterns are assumed as criteria for determining whether medical treatment is required.
・If the disease activity on the day exceeds the threshold index (determined based on the evaluation results on the day)
・If there is a noticeable trend of deterioration over a certain period of time

In this way, System 1 evaluates the disease activity of patients with a history of specific diseases by using the daily and continuous act of excretion, and Support the treatment of the disease by notifying the patient.
The processing of the system 1 will be described in detail below.

<3. System 1 operation>
Various processes performed by the system 1 will be described below.

(3-1. Disease activity evaluation process)
First, disease activity evaluation processing, which is the main function of the system 1, will be explained.
FIG. 11 is a flowchart showing the disease activity evaluation process performed by the system 1.

As shown in FIG. 11, in the disease activity evaluation process, first, the patient operates the patient terminal 10A to take a photograph of the stool he or she has excreted (step S101).
Specifically, the photographing unit 183 of the patient terminal 10A activates the camera 160 in response to an operation from the user, and photographs the stool serving as the subject. A screen example at this time will be described later as a first screen example P1. The patient can also select a tag and enter a comment if necessary.
The data processing section 185 inputs image data photographed by the photographing section 183 to the detecting section 184 .

After step S101, the patient terminal 10A detects stool from the captured image (step S102).
Specifically, the detection unit 184 of the patient terminal 10A inputs the image input from the imaging unit 183 to the detection model 172 in step S101, thereby detecting the excrement contained in the image output from the detection model 172. Get object detection results. The detection unit 184 inputs the obtained object detection result to the data processing unit 185.

After step S102, the patient terminal 10A transmits the stool image to the evaluation server 20 (step S103).
Specifically, the data processing unit 185 of the patient terminal 10A transmits the following information to the evaluation server 20.
- Stool image data - Information regarding the detected object inputted from the detection unit 184 in step S102 - Tag data and comment data inputted at the time of photographing the stool image In other words, the data processing unit 185 detects that an object related to excrement has been detected. As a trigger, the corresponding image (stool image) is transmitted to the evaluation server 20 via the transmitter/receiver 182. Therefore, if an object related to excrement is not detected from the captured image, the data processing unit 185 does not transmit the corresponding image to the evaluation server 20.

Furthermore, after step S103, the patient terminal 10A deletes the stool image (step S104).
Specifically, the data processing unit 185 of the patient terminal 10A deletes the data of the image in which the object related to excrement was detected in step S102 without storing it in the storage area of the patient terminal 10A. Furthermore, the data processing unit 185 deletes the data without storing the stool image in an external storage area where captured images are stored, such as a cloud server that the patient uses on a daily basis.

Further, after step S103, the evaluation server 20 acquires a stool image (step S201).
Specifically, the transmission/reception control module 2031 of the evaluation server 20 receives various input data transmitted from the terminal device 10 and inputs it to the acquisition module 2032. The acquisition module 2032 acquires input data, stores it in each storage area, and records a new record in the image DB 2022.

Here, the type of data newly recorded in step S201 and the storage area will be explained.
・Stool image data: Image storage area (external server)
・Attribute information regarding flight images: Image DB2022
・Information regarding detected objects: Object DB2023
・Tag data input from the patient: Image DB2022
・Comments entered by patients: Comment DB2025

After step S201, the evaluation server 20 estimates a disease activity index for a specific disease in the patient (step S202).
Specifically, the evaluation module 2033 of the evaluation server 20 inputs the newly recorded image data in step S201 to the activity estimation model 2027, and uses the estimated value of the disease activity index output from the activity estimation model 2027. , along with its likelihood. The evaluation module 2033 uses the obtained estimated value of disease activity and records it in a new record in the evaluation result DB.

Further, in step S202, the determination module 2034 refers to a new record in the evaluation result DB and determines whether the patient in question requires medical examination by a doctor. The determination module 2034 updates the item “necessity of treatment” in the evaluation result DB based on the determination result.

After step S202, the evaluation server 20 outputs the evaluation result to the patient terminal 10A (step S203).
Specifically, the output module 2037 of the evaluation server 20 outputs the following information to the patient terminal 10A.
・Evaluation results of disease activity (estimated value)
・Address information regarding the necessity of medical treatment by a doctor and medical reservations

That is, if the determination module 2034 determines in step S202 that medical treatment by a doctor is necessary, the output module 2037 performs API cooperation with the medical institution server 30 in step S203. The output module 2037 acquires address information (for example, URL information) regarding a site related to medical examination reservations provided by the medical institution server 30, and outputs it to the patient terminal 10A.

After step S203, the patient terminal 10A presents the evaluation results to the patient (step S105).
Specifically, the display control unit 186 of the patient terminal 10A displays the information received by the transmitting/receiving unit 182 and output from the evaluation server 20 on the display 132 of the patient terminal 10A. As a result, the evaluation results of disease activity using stool images are presented to the patient. The evaluation results include address information regarding medical reservations. A screen example at this time will be described later as a third screen example P3.

Further, after step S203, the evaluation server 20 transmits the evaluation result to the medical terminal 10B used by the attending physician (step S204).
Specifically, the output module 2037 of the evaluation server 20 queries the patient DB 3021 and doctor DB 3023 to the medical institution server 30, and sends the following information to the doctor registered as the patient's attending physician. .
・Image data of the flight to be evaluated ・Attribute data related to the flight image (items included in the image DB 2022)
・Tag data and comment data entered by the patient ・Evaluation results of disease activity (estimated value)

At this time, the doctor can input comments regarding the patient prior to the examination at the actual medical institution. Comments input by the doctor by operating the medical terminal 10B are stored in the item "attending physician comment" of the evaluation history DB 2026.

Note that the process in step S204 is carried out by the attending physician only when the disease activity index estimated in step S202 indicates that the patient's condition is of concern (in other words, when it can be determined that the patient's condition is worsening). You may notify the terminal to that effect. Examples of cases in which a doctor is notified of the evaluation results of disease activity include any of the following.
・When the disease activity index is worse than a predetermined threshold value ・When the disease activity index is worse than the previous evaluation result ・When the disease activity index tends to worsen during the evaluation period More than one attending physician If you are in charge of several patients, there is a concern that if you notify the attending physician of all evaluation results without any problems, the amount of notifications will be enormous. can be sent.
With the above steps, the disease activity evaluation process by the system 1 is completed.

(3-2. Reservation processing)
Next, reservation processing by the system 1 will be explained.
FIG. 12 is a flowchart showing reservation processing by the system 1.

As shown in FIG. 12, in the reservation process, first, the patient terminal 10A makes a reservation request (step S111).
Specifically, the input receiving unit 181 of the patient terminal 10A receives an operation in which the patient clicks on the address information related to the medical reservation transmitted from the evaluation server 20. The patient terminal 10A then transmits a request regarding a medical appointment to the medical institution server 30 by accessing a site regarding the medical appointment provided by the medical institution server 30. The reservation request may include a patient ID at the patient's medical institution.

After step S111, the medical institution server 30 receives the reservation request (step S311).
Specifically, the input reception module 3032 of the medical institution server 30 receives a reservation request transmitted from the patient terminal 10A, and inputs the patient ID included in the reservation request to the inquiry module 3033. The inquiry module 3033 refers to the patient DB 3021 using the input patient ID, specifies the attending physician, refers to the medical treatment time slot DB 3024 and the reservation DB 3025, and extracts available reservation slots.

After step S311, the medical institution server 30 outputs the available reservation slots to the patient terminal 10A. (Step S312)
Specifically, the output module 3035 of the medical institution server 30 outputs the available reservation slot extracted in step S311 to the patient terminal 10A.

After step S312, the patient terminal 10A presents available reservation slots to the patient (step S112).
Specifically, the display control unit 186 of the patient terminal 10A displays on the display 132 the information regarding the available reservation slots received by the transmitting/receiving unit 182. As a result, available reservation slots are presented to the patient. The patient considers the time slot he or she would like to receive treatment from among the available appointment slots presented.

After step S112, the patient terminal 10A issues an instruction to select a reservation slot (step S113).
Specifically, the input receiving unit 181 of the patient terminal 10A receives the designation of the desired time slot for medical treatment input by the patient, and transmits it to the evaluation server 20 via the transmitting/receiving unit 182.

After step S113, the medical institution server 30 performs reservation setting (step S313).
Specifically, the reservation module 3034 of the medical institution server 30 records the desired time slot for medical treatment received from the patient terminal 10A in step S113 as a new record in the reservation DB 3025. As a result, a medical treatment reservation has been made.

After step S313, the medical institution server 30 transmits a reservation completion notification to the patient terminal 10A (step S314).
Specifically, the output module 3035 of the medical institution server 30 creates information regarding reservation completion and transmits it to the patient terminal 10A.

After step S314, the patient terminal 10A receives the reservation completion notification (step S114).
Specifically, the display control unit 186 of the patient terminal 10A displays information regarding the reservation completion notification received by the transmitting/receiving unit 182 on the display 132. As a result, the patient is informed that the medical appointment has been completed.
With the above, the reservation processing by the system 1 is completed.

(3-3. Evaluation history aggregation processing)
Next, evaluation history aggregation processing by the system 1 will be described.
FIG. 13 is a flowchart showing evaluation history aggregation processing performed by the system 1.

As shown in FIG. 13, in the evaluation history aggregation process, the evaluation server 20 first refers to the evaluation history DB 2026 and aggregates the disease activity evaluation history (step S221).
Specifically, the aggregation module 2035 of the evaluation server 20 aggregates the evaluation history in a predetermined evaluation target period for a patient to be evaluated. The aggregation module 2035 may perform aggregation processing every frequency of a preset period. Further, the aggregation module 2035 may aggregate the evaluation history in response to an operation by the patient or a medical professional including the attending physician.

After step S221, the evaluation server 20 outputs the total results (step S222).
Specifically, the output module 2037 of the evaluation server 20 creates a total report including the following information and outputs it to the patient terminal 10A.
・Evaluation frequency of disease activity during the evaluation period (frequency of acquisition of stool images)
・Time-series trends in disease activity index as evaluation results

After step S222, the patient terminal 10A presents the total results of the evaluation history to the patient (step S122).
Specifically, the display control unit 186 of the patient terminal 10A displays the total report received by the transmitting/receiving unit 182 on the display 132. As a result, a summary report of the evaluation results is presented to the patient. The aggregate report includes the following information:
- Notification to certify continuous acquisition of stool images when the frequency of acquisition of stool images exceeds a certain standard - Continuous acquisition of stool images when the frequency of acquisition of stool images falls below a certain standard notification prompting

In other words, in System 1, if a patient is able to take stool images every day and maintain a state where disease activity is continuously evaluated, this will be evaluated and the patient's attitude toward health management will be adjusted. Reward as appropriate.
On the other hand, in System 1, if a patient neglects to take daily stool images and continuous assessment of disease activity is not possible, this will be pointed out and the patient's attitude regarding health management will be inappropriate. request for improvement. With the above, the reservation processing by the system 1 is completed.

(3-4. Endoscope equivalent image generation process)
Next, the process of generating an endoscope-equivalent image by the system 1 will be described. This process is a secondary process, and is a process that is selectively executed based on a doctor's judgment, for example. Specifically, in step S204 shown in FIG. 11, the doctor who has been notified of the disease activity evaluation results sends an instruction to create an endoscopic image to the evaluation server 20 for the purpose of confirming the state of the patient's intestines. This process is executed using the input as a trigger.
FIG. 14 is a flowchart showing the process of generating an endoscope-equivalent image.

As shown in FIG. 14, in the endoscope-equivalent image generation process, the evaluation server 20 first generates an endoscope-equivalent image (step S241).
Specifically, the image generation module 2036 of the evaluation server 20 obtains the designated intestinal image output from the image synthesis model 2028 by inputting the stool image specified by the doctor into the image synthesis model 2028. After storing the endoscope-equivalent image in a predetermined storage area, the image generation module 2036 creates address information (for example, URL information) indicating the storage area.

After step S241, the output module 2037 outputs an endoscope-equivalent image to the medical terminal 10B (step S242).
Specifically, the output module 2037 of the evaluation server 20 outputs address information regarding the endoscope-equivalent image generated by the image generation module 2036 in step S241 to the medical terminal 10B. By accessing the output address information, the doctor can check the patient's endoscope-equivalent image. This allows doctors to estimate the state of the patient's intestines.
With the above steps, the process of generating an endoscope-equivalent image by the system 1 is completed.

<4. Screen example>
Examples of screens in the system 1 will be described below.

(4-1. First screen example P1)
FIG. 15 is a diagram showing a first screen example P1. The first screen example P1 is a screen displayed on the patient terminal 10A at step S101 shown in FIG. 11, and is a diagram showing a stool photographing operation.

As shown in FIG. 15, in the first screen example P1, stool is displayed as the subject to be photographed. After defecating, the patient activates the camera 160 from the patient terminal 10A and photographs the stool. A stool image is photographed by adjusting the toilet bowl so that it fits within the field of view and photographing by pressing the photographing button B1.

In addition, on the setting screen displayed by pressing the settings button B2 before shooting, settings regarding the storage process of the shot image can be selected. The following patterns are included in the process of saving captured images.
- Whether to save in the storage area of the terminal device 10 (default setting is not saved)
- Whether to save in the storage area of the cloud server used by the terminal device 10 (default setting is not saved)
For example, when a photograph is taken using a medical terminal 10B such as a tablet at a medical institution, the photographed stool image may be stored in the medical terminal 10B.

(4-2. Second screen example P2)
FIG. 16 is a diagram showing a second screen example P2. The second screen example P2 is a screen displayed on the patient terminal 10A at step S101 shown in FIG. 11, and is a diagram showing an operation for registering a stool image used for evaluation of disease activity. In other words, it is a diagram illustrating an operation for performing evaluation of disease activity.

As shown in FIG. 16, in the second screen example P2, along with the photographed flight image, information regarding the attributes of the image (photographing date, photographing time, user ID) is displayed in the attribute column C1.
Further, the second screen example P2 displays a tag input field C2 in which a tag corresponding to the user's physical condition or the like can be selected.

Further, the second screen example P2 displays a patient comment field C3 in which the patient can input the content he/she wishes to input in the form of text data.
When the patient presses the evaluation execution button B3, the process from step S102 shown in FIG. 11 is executed as evaluation of disease activity using this content.

(4-3. Third screen example P3)
FIG. 17 is a diagram showing a third screen example P3. The third screen example P3 is a screen displayed on the patient terminal 10A at step S105 shown in FIG. 11, and is a diagram showing a state in which the evaluation results of disease activity are presented to the patient.

As shown in FIG. 17, the third screen example P3 displays an evaluation result column C5 indicating the evaluation result of disease activity. The evaluation result column C5 also includes the following information.
・Evaluation date: Date of evaluation ・Patient name: Patient's name (can be omitted when displayed on patient terminal 10A)
・Disease activity (estimated value): The estimated value of the disease activity index that is the evaluation result. ・Likelihood: An index that indicates the likelihood of the estimated value. ・Necessity of medical treatment: Judgment result of whether medical treatment by a doctor is required. ・Evaluation details : Information about entered tags and comments

Here, in displaying the necessity of medical treatment, if the evaluation result of the disease activity index indicates that the patient is in a state of concern, the deadline for the next medical treatment may be notified. Examples of notification of the deadline for medical treatment include any of the following.
・When the disease activity index is worse than a predetermined threshold value ・When the disease activity index is worse than the previous evaluation result ・When the disease activity index tends to worsen during the evaluation period

In these cases, it is possible to effectively encourage a visit to a medical institution by not only notifying the patient that medical treatment is necessary, but also setting a deadline for the next medical treatment and encouraging the patient to receive treatment. Further, the number of days until the next medical treatment date may be adjusted according to the evaluation result of the disease activity index, that is, according to the estimated degree of the disease state. For example, if the estimated medical condition is particularly bad, a shorter deadline is set than if it is slightly worse.

Furthermore, a notification may be displayed to encourage the use of a highly effective drug that has been prescribed in advance, depending on the evaluation result of the disease activity index. In general, highly effective drugs include those that come with certain side effects and those that some patients are hesitant to use, such as suppositories. Such highly effective drugs are sometimes prescribed by a doctor during medical treatment and used as an as-needed drug with a limited administration period depending on changes in the patient's medical condition in daily life. In such cases, if the evaluation result display screen indicates that the disease state estimated from the evaluation results of the disease activity index is in a state of concern, the patient can be encouraged to use the drug with strong efficacy. Timely use can be made.

Further, if the estimated likelihood is lower than the threshold value, this may be displayed as an evaluation result, and the patient may be notified that the reliability of the estimation result is low.
In addition, if the estimated likelihood is high and the disease activity index is high (the condition is bad), this will be displayed as an evaluation result and a notification will be sent to strongly encourage treatment at a medical institution. good. Further, a deadline until the next scheduled medical treatment date may be set according to the likelihood and the disease activity index.

Further, in the third screen example P3, a medical appointment column C6 for making a medical appointment is displayed. The medical reservation column C6 displays address information exclusively for the patient who can access the reservation site. That is, by accessing the reservation site from the address information through API cooperation between the evaluation server 20 and the medical institution server 30, it is possible to receive inquiries about available reservation slots without having to input a patient ID.

Furthermore, an advice column C7 is displayed on the third screen example P3. When a patient clicks on the advice column C7, general matters to be kept in mind in daily life in the current state of disease activity are displayed as advice to the patient.

Further, in the third screen example P3, an attending physician comment field C8 is displayed. When the patient clicks on the attending physician's comment field C8, the content of the comment input by the patient's attending physician is displayed.

Note that the following information may be displayed on the evaluation result display screen shown in FIG. 17.
・Number of stools on the current day or in the past ・Number of bloody stools on the same day or in the past ・Number of days until the next medical appointment

<5. Summary>
As explained above, in the system 1, a disease activity index related to a specific disease of a patient is estimated by image processing on a stool image taken by the patient using the patient terminal 10A, and is output to the patient terminal 10A. .
Therefore, unlike conventional systems, patients do not have to input multiple evaluation items such as their health status and the presence or absence of subjective symptoms, and disease activity can be continuously evaluated through extremely simple operations by patients. It can be done in a specific manner.

In addition, in System 1, when estimating the disease activity index, the activity estimation model 2027 is created by machine learning using learning data with stool images as input data and disease activity index as correct output data. is used to output the estimated value of the disease activity index. Therefore, by performing machine learning using a large amount of learning data, accuracy in estimating the disease activity index can be ensured.

Furthermore, in the system 1, when estimating the disease activity index, the estimated position of the lesion range in the disease is output. Therefore, when the attending physician wants to check the evaluation results obtained by the system 1, information that contributes to treatment can be provided to the physician.

Moreover, in the system 1, when estimating the disease activity index, the likelihood of the estimated disease activity index is output. Therefore, by mentioning how likely the estimated value of disease activity, which is the evaluation result, is, it is possible to encourage judgment on how to handle the evaluation result.

Further, in the system 1, when feces is detected in an image taken by a patient, the image is not stored in the terminal device 10 but is sent to the evaluation server 20. Therefore, in general, stool images that the patient does not want to save in his or her terminal device 10 can be automatically deleted from the terminal device 10, and the convenience of the system 1 for the patient can be improved.

Further, in the system 1, when the estimated disease activity index exceeds a predetermined threshold set in advance, the patient is notified that medical treatment by the attending physician is required. Therefore, it becomes possible for doctors to provide timely medical treatment, and it is possible to support treatment by medical institutions.

In addition, the system 1 aggregates the frequency of acquisition of stool images by patients, and when the aggregated acquisition frequency falls below a certain standard, notifies the patient terminal 10A of a notification urging continuous acquisition of stool images. . Therefore, if a patient neglects to take stool images for continuous evaluation of disease activity, this can be pointed out to the patient.

Furthermore, in the system 1, when the aggregated frequency of acquisition exceeds a certain standard, a notification certifying continuous acquisition of stool images is sent to the terminal device 10. Therefore, it is possible to evaluate the posture of a patient who continuously takes fecal images to appropriately understand his or her health condition, and to motivate the patient to continue working on the posture.

In addition, the system 1 notifies the terminal device 10 used by the patient's attending physician of the evaluation results.
Therefore, it is possible to smoothly share with the attending physician that the patient's condition may have worsened, which can contribute to subsequent treatment at the medical institution.

Further, the system 1 may adopt a configuration in which the attending physician is notified only when a predetermined threshold value set in advance is exceeded. In this case, even if the attending physician is in charge of multiple patients, the evaluation results that should be notified to the attending physician will be automatically selected to prevent the attending physician from receiving a huge amount of notifications. be able to.

In addition, in System 1, through image processing using the acquired stool images, an endoscopic-equivalent image is generated that is estimated as the state of the patient's intestines when the stool associated with the stool image is excreted, and an image corresponding to an endoscope is generated. A notification is sent to the terminal used by the attending physician. As a result, when a doctor wants to confirm the state of the patient's intestine due to a change in the patient's condition, the doctor can be provided with an image showing the estimated state of the patient's intestine before performing an endoscopic examination. be able to.

<6. Modified example>
A modification of this embodiment will be described.
In the following description, descriptions of configurations similar to those of the above-described embodiments will be omitted.

(6-1) First modification: Evaluation method via estimation of specific protein content In the first modification of System 1, the method of estimating the disease activity index is different from the above embodiment. . That is, in this modification, the evaluation server 20 does not output the estimated value of the disease activity index in response to the input of the stool image, but estimates the content of a specific protein in the stool, and then outputs the estimated value of the disease activity index. A disease activity index is estimated from the estimated content. Such processing will be explained using FIG. 18.

FIG. 18 is a diagram illustrating a flow of a disease activity index estimation process according to the first modification. This figure explains only the process corresponding to step S202 shown in FIG. 11. The processing before and after that is the same as shown in FIG. 11, and the explanation thereof will be omitted.

As shown in FIG. 18, in the first modification, after step S201 shown in FIG. 11, the evaluation server 20 estimates the content of calprotin, which is a specific protein contained in stool (step S2021).
Specifically, the evaluation module 2033 acquires the content of calprotin output from the content estimation model by inputting an image to the content estimation model.

Here, the system 1 according to the first modification includes a content estimation model in place of the activity estimation model 2027 described above. The content estimation model was created by machine learning using training data, with stool images (or object data related to detected excrement) as input data and the content of calprotin in the stool as correct output data. There is. The evaluation module 2033 inputs the obtained calproctin content to the determination module 2034.

After step S2021, the evaluation server 20 determines disease activity based on the estimated content of calprotin (step S2022).
Specifically, the determination module 2034 uses the estimated value of the content of calprotin input from the evaluation module 2033 in step S2021 and compares it with the pre-stored determination criteria for the disease activity index. Determine the disease activity index. The criteria used in this case are set for each content range by associating the content range of calprotin with the disease activity index estimated in that range. In this way, an estimate of the disease activity index is determined.

In this way, in the system 1 according to the first modification, the acquired stool image is input to the content estimation model, thereby outputting an estimated value of the content of calprotin in the stool. Then, in the system 1 according to the first modification, the estimated value of the content is used to determine whether a disease occurs based on the criteria in which the preset content and the estimated value of the disease activity index are associated with each other. Determine an estimate of the activity index. Therefore, the estimated value of the disease activity index can be evaluated with high accuracy by estimating the content of a specific protein (calprotin) that is said to be closely related to the disease activity of inflammatory bowel disease. It is expected that evaluation will be carried out.

(6-2) Second modification: combination of rule-based determination In the second modification of System 1, the method of estimating the disease activity index is further different from the two patterns described above. That is, in this modification, first, when photographing a stool image, input of at least one of the following behavioral history and biological information is accepted from the patient.

1) Information regarding the history of excretory behavior (behavior history)
・Number of stools on the previous day or on the day, time of defecation, nature of stool, degree of bloody stool, image data of fecal incontinence, etc.

2) Information regarding the history of eating behavior including medication (behavioral history)
- Medication history (presence of use of antidiarrheals and analgesics)
・Meal details from the previous day or the day (including meal images), etc.

3) Biological information: height, weight, BMI
・Body temperature, pulse rate, degree of abdominal pain, etc.

In the system 1 according to the second modification, the disease activity index is estimated using the stool image and also by determination using the behavioral history and biological information input by the patient. Such processing will be explained using FIG. 19.

FIG. 19 is a diagram illustrating the flow of the disease activity index estimation process according to the second modification. This figure explains only the process corresponding to step S202 shown in FIG. 11. The processing before and after that is the same as shown in FIG. 11, and the explanation thereof will be omitted.

As shown in FIG. 10, in the third modification, after step S201 shown in FIG. 11, the evaluation server 20 estimates the disease activity index using the stool image (step S2023).
Specifically, the evaluation module 2033 of the evaluation server 20 inputs the stool image to the activity estimation model 2027, similar to step S202 shown in FIG. Obtain an estimate of the activity index.

After step S2023, the evaluation server 20 uses other inputs from the patient to determine the disease activity index on a rule basis (step S2024).
Specifically, the determination module 2034 of the evaluation server 20 compares the behavioral history and biological information input by the patient with preset evaluation criteria to determine the patient's disease activity index. The evaluation criteria used at this time are set for each numerical range by associating each numerical range of the behavior history and biological information with the disease activity index estimated in the range. In this way, a determined value of the disease activity index is obtained.

After step S2024, the evaluation server 20 performs a comprehensive evaluation by comparing the first disease activity index estimated from the stool image and the second disease activity index determined from the behavioral history and biological information. (Step S2025).
Specifically, if the first disease activity index and the second disease activity index have the same grade, the evaluation module 2033 adopts the grade.

On the other hand, if the first disease activity index and the second disease activity index are not of the same grade, the evaluation module 2033 adopts the worse value as the disease state. Note that the judgment method is not necessarily limited to adopting a bad value. For example, if the first disease activity index and the second disease activity index are not of the same grade, any value located between them (for example, the arithmetic mean value of both) may be adopted. Any method of comprehensive evaluation using these two indices can be set.

In this way, the system 1 according to the second modification acquires at least one of the behavioral history and biological information input from the patient, estimates the disease activity index using stool images, and estimates the behavioral history and biological information. Use the information to determine the disease activity index. Then, in the system 1 according to the second modification, comprehensive evaluation is performed using the two obtained disease activity indices.
Therefore, more reliable evaluation of disease activity can be performed using knowledge of various criteria accumulated in the past. In addition, by comparing the estimation results of the activity estimation model 2027 and the rule-based judgment results, it is possible to obtain information that can be used to retrain the activity estimation model 2027, and disease activity from stool images can be obtained. This can lead to improved accuracy in gender estimation processing.

Note that the above-described rule-based determination may use the stool image and input information input from the patient terminal 10A, as well as the medical treatment history at a past medical institution. For example, the determination rule shown below may be used.
1) Defecation frequency (user input information)
・Normal number of times: +0 points ・1 to 2 times/day more than the normal number of times: +1 point ・3 to 4 times/day more than the normal number of times: +2 points ・5 or more times/day more than the normal number of times: +3 points

2) Bloody stool (determined from user input information or stool image)
・No bloody stool: +0 points ・Slight blood (stripes) on less than half of the defecations: +1 point ・Clear blood contamination is seen during most of the defecations: +2 points ・Most of the stool is blood: +3 points

3) Mucosal findings (using the results of the most recent endoscopy)
・Normal or inactive findings: +0 points ・Mild symptoms (redness, decreased vascular transparency, mild fragility): +1 point ・Moderate symptoms (significant redness, loss of vascular fluoroscopy, fragility, erosion): +2 Points/Severe (spontaneous bleeding, ulcer): +3 points

4) General evaluation by the doctor (results of the most recent doctor's examination)
・Normal: +0 points ・Mild: +1 point ・Moderate: +2 points ・Severe: +3 points

5) Comprehensive Judgment The second disease activity index is determined based on the total score of 1) to 4) above.
・0 points: Lv1
・1-2 points: Lv2
・3 to 5 points: Lv3
・6-10 points: Lv4
・11-12 points: Lv5

(6-3) Third modified example: Re-learning the activity estimation model 2027 for each patient The third modified example of the system 1 is different from the above-described embodiments in that the activity estimation model 2027 is re-learned for each patient. It's different. That is, in this modification, the activity estimation model 2027 is customized for each patient.

Specifically, the administrator of the system 1 uses the patient's own stool image and the actual disease activity index at the time of excretion of stool corresponding to the stool image as learning data, and adds it to the activity estimation model 2027 for learning. Data can be retrained. Here, the actual disease activity index refers to a disease activity index obtained through medical treatment by a doctor.

In this case, the activity estimation model 2027 will be optimized uniquely for each patient, and for inflammatory bowel disease, a disease with large individual differences in symptoms, the activity estimation model 2027 will be optimized based on the stool condition and current disease activity. The activity estimation model 2027 can be made to learn the correspondence relationship with. This allows for more accurate evaluation of disease activity.

<7. Other variations>
Other modifications will be described below.

In the embodiment described above, a configuration is shown in which the feces discharged into the toilet are photographed, but this is not the only option. For example, the system 1 may perform imaging using a dedicated defecation sheet. The defecation sheet is a sheet that is used while floating in the sealed water in the toilet, and prevents feces from flowing into the sealed water during the time required for photographing. That is, it is preferable that the defecation sheet has a certain level of water resistance and also has hydrophilicity that allows it to be flushed away with excrement. Further, it is preferable that at least a portion of the surface of the sheet body constituting the defecation sheet is colored. The colored surface of the sheet body makes it easier to identify white mucus contained in excrement.

Further, color standards or size standards may be displayed on the sheet body of the defecation sheet. The color reference is a marker that serves as a color reference used in image processing of a stool image, and is used in image processing of a stool image. Color standards can be used, for example, to detect blood in stool, or to distinguish between mucus and watery stool.

In image processing using color standards, feature quantities such as hue, saturation, and brightness of objects related to excrement contained in a stool image are extracted based on color standards, and the type of the object is determined. Alternatively, as a method using a color histogram, the type of object included in the stool image may be determined by analyzing the color distribution of pixels in the stool image using a color standard.

As the color reference color, a color equivalent to the object related to excrement, a color equivalent to its complementary color, or a color whose brightness is contrasting may be adopted. Examples of detected objects and corresponding color reference candidates are shown below.
・When detecting blood (red) (red color or complementary green color)
・When detecting feces (brown) (brown color or complementary blue-purple color)
・When detecting mucus (white) (black with contrasting brightness)

The size standard is a display that serves as a size standard used in the image processing of the stool image, and is used in the image processing of the stool image. Size criteria can be used, for example, to detect the amount of stool and bleeding.

In image processing using size standards, a trained model (detection model 172, activity estimation model 2027, or content estimation model) that has been trained on images of excrement on which size standards have been photographed is used to process stool images. Accurately estimate the actual size and position of the object by comparing the size and position of the included object with the size and position of the size standard.

In other words, when performing image analysis using color standards or size standards written on such a defecation sheet, an image in which the color standards and size standards are taken together with excrement is used as training data for a trained model. Learning may be performed using input data.
In this way, by performing image processing on the stool image using the color standards and size standards written on the defecation sheet, it can be expected that the accuracy of estimation by the activity estimation model 2027 will be improved.

Further, in the system 1, the display mode of the evaluation results may be changed on the evaluation result display screen shown in FIG. 17 according to the estimated value of the disease activity index. In other words, if the estimated disease activity index is worse than the threshold value, worse than the previous evaluation result, or if there is a tendency for it to worsen during the evaluation period, the display should emphasize that fact. May be adopted. Examples of such display mode change patterns are listed below.
・Change the background color of the display screen ・Change the color of the characters on the display screen ・Change the size of the characters on the display screen ・Generate an alert sound along with the display screen ・Set the patient terminal 10A along with the display screen vibrate

Further, the system 1 may receive input of information regarding the degree of odor related to excrement from the patient. The odor information is determined subjectively by the patient, for example, in the following five steps, and is input to the patient terminal 10A.
・Normal stool odor ・Strong (or foreign) stool odor ・Extremely strong (or extremely foreign) stool odor

When estimating a disease activity index using stool odor, for example, a stool image and the degree of stool odor reflected in the stool image are used as input data, and the input image information and odor are used as input data. An estimation model obtained by machine learning using learning data with the value of the disease activity index as the correct output data can be used for information on the degree of disease activity index.
In addition, for the estimation results from the activity estimation model 2027 or the content estimation model described above, the overall disease activity index is calculated using a correction standard in which the degree of odor is associated with the assumed disease activity index. You may also make a subjective judgment.

Further, in the above embodiment, the specific disease is described using inflammatory bowel disease as an example, but the present invention is not limited to this embodiment. The disease for which the system 1 evaluates disease activity may be a disease other than inflammatory bowel disease.

Further, in the above embodiment, a configuration is shown in which the patient uses the patient terminal 10A to photograph the excrement excreted by the patient, but the present invention is not limited to this. For example, a medical worker at a medical institution, a staff member at a nursing care facility, or the like may acquire a stool image by photographing excrement excreted by a patient using the terminal device 10 used by each of them.

For example, disease activity can be evaluated by the system 1 for hospitalized patients. In this case, the nurse or the patient himself/herself who has been lent the medical terminal 10B by the nurse may use the medical terminal 10B to take a stool image.
In such evaluation of disease activity for hospitalized patients, the doctor on duty may be notified of the evaluation, depending on the evaluation result. Specifically, the medical terminal 10B used by the doctor on duty or the contact information of the doctor on duty is notified of the evaluation result of disease activity.

For example, in the following situations, the doctor on duty may be notified of the evaluation results of disease activity.
1) When the disease activity index is in a state of concern based on the evaluation results. - When the disease activity index is worse than a predetermined threshold. - When the disease activity index is worse than the previous evaluation result. - Disease activity When the index shows a tendency to worsen during the evaluation period 2) When the estimated likelihood is high and there is concern about the medical condition

In addition, in accordance with the evaluation results of disease activity by the system 1, that is, in cases where the patient's condition is of concern, the nurse sends a message to the medical terminal 10B to the effect that it is necessary to call the doctor on duty. may be notified. As a result, by checking the guidance displayed on the medical terminal 10B, the nurse can smoothly make a decision on whether or not to immediately seek a doctor's judgment based on the evaluation results of disease activity by the system 1. It can be carried out.

In addition, the system 1 can automatically record the number of defecation times and defecation time. In other words, when a patient acquires a stool image, by treating the time at which the stool image was taken as the time of defecation, the following information can be recorded over time by simply taking an image each time the patient defecates: .
・Number of flights per day / Number of flights during the day / Number of flights at night

Furthermore, based on the above-mentioned excrement detection results, the following information can be further recorded over time.
・Number of normal stools ・Number of diarrheal stools and their specific characteristics ・Number of bloody stools ・Number of bloody stools 100 times ・Number of bloody stools of 50% or more ・Number of bloody stools of less than 50% This can be used in medical treatment by referring to the patient's defecation history and determining the actual disease state.

Further, in the medical terminal 10B, medical history information recorded in the medical history DB 3022 shown in FIG. 9C may be accessible from the evaluation result screen shown in FIG. 17. That is, past medical results and the results of various tests such as blood sampling actually performed in the most recent medical treatment are displayed on the medical terminal 10B, as required by the medical worker who has confirmed the evaluation results of disease activity by the system 1. can be done.

Furthermore, the evaluation results of the disease activity index by the system 1 accumulated in the evaluation history DB 2026 shown in FIG. 6F may be accessible from the patient's electronic medical record displayed on the medical terminal 10B. In this case, the electronic medical record mainly displays medical history information recorded in the medical history DB 3022 shown in FIG. 9C. Then, the evaluation history of the disease activity index by the system 1 can be displayed on the medical terminal 10B according to the needs of the medical worker who has confirmed the contents of the electronic medical record.

Furthermore, in the system 1, an endoscope-equivalent image may be output to the patient terminal 10A. In general, treatment for inflammatory bowel disease tends to be long-term. In order to continue treatment, it is effective for patients to intuitively understand their current medical condition. Therefore, in addition to the estimated disease activity index, by presenting patients with endoscopic-equivalent images that allow them to visually understand the disease state, it is possible to check their own subjective symptoms related to the disease and the estimated actual state. This allows the patient to make adjustments, and motivates the patient to continue with the treatment.

In addition, System 1 combines endoscopic-equivalent images generated at each of multiple evaluation periods in chronological order, and generates a video in which the images change over time in accordance with display instructions. It is also possible to generate an endoscope-equivalent video that allows confirmation of changes in the state of the patient over time. This allows medical personnel or patients to understand changes in medical conditions due to treatment activities.

Furthermore, the system 1 may provide information that provides certain suggestions for medical treatment by a doctor. Specifically, using information on prescriptions and treatments recorded in medical history, for example, the evaluation results of disease activity estimated before and after the timing of a change in medicine or the timing of surgery are analyzed in chronological order. You can analyze it accordingly. This makes it possible to provide information that contributes to evaluating the non-inferiority of a newly prescribed drug and determining the timing of a change in treatment policy.

Although the embodiments of the present invention have been described in detail above, the scope of the present invention is not limited to the above embodiments. Moreover, various improvements and changes can be made to the embodiments described above without departing from the spirit of the present invention. Furthermore, the above embodiments and modifications can be combined, or some of them can be omitted.

<8. Additional notes>
The matters explained in the embodiment and the modified example will be added below.

(Additional note 1)
A program to be executed by an information processing system including a server having a processor, the program causing the processor to:
a step of acquiring a stool image of the patient taken by a terminal device used by the patient or a medical worker (step S201);
estimating a disease activity index related to a specific disease of the patient by image processing the acquired stool image (step S202);
A program that causes a program to execute a step (step S203) of outputting an estimated disease activity index to a terminal device.

(Additional note 2)
In the step of estimating the disease activity index (step S202), the processor:
Patients can be identified by inputting the acquired stool images to an activity estimation model created by machine learning using learning data with stool images as input data and disease activity index as correct output data. The program according to supplementary note 1, which causes the program to execute the step of outputting an estimated value of a disease activity index regarding the disease.

(Appendix 3)
In the step of estimating the disease activity index (step S202), the processor:
By inputting the acquired stool images into an activity estimation model created by machine learning using learning data with stool images as input data and the content of calproctin in the stool as correct output data. , outputting an estimated value of the content of calprotin in the stool;
outputting an estimated value of a disease activity index based on a criterion in which a preset content and an estimated value of a disease activity index are associated using the estimated value of the content; The program described in Supplementary Note 1, which causes the program to execute.

(Additional note 4)
In the step of estimating the disease activity index (step S202), further:
The program according to supplementary note 1, which outputs an estimated position of a lesion range in the disease.

(Appendix 5)
In the step of estimating the disease activity index (step S202), further:
The program according to supplementary note 1, which outputs the likelihood of the estimated disease activity index.

(Appendix 6)
In addition, the processor
Execute the step of relearning the activity estimation model used to estimate the disease activity index using the patient's stool image and the actual disease activity index at the time of excretion of stool corresponding to the stool image as learning data. The program described in Appendix 1.

(Appendix 7)
In addition, the processor
executing a step of acquiring at least one of behavioral history indicating a history of excretion behavior and eating behavior including medication inputted into the terminal device, and biological information;
In the step of estimating the disease activity index (step S202), the disease activity index is estimated based on preset evaluation criteria using the acquired behavioral history and biological information as well as the image processing of the stool image. The program described in Appendix 1.

(Appendix 8)
In the step of estimating the disease activity index (step S202), the program described in Appendix 1 performs image processing on the stool image using color standards or size standards written on the defecation sheet used by the patient when defecating. .

(Appendix 9)
In the step of acquiring a stool image (step S201),
a step of detecting excrement contained in an image taken by the terminal device (step S102);
The program according to supplementary note 1, which executes the step (step S103) of transmitting the image to a server without storing the image in the terminal device when excrement is detected in the image.

(Appendix 10)
In addition, the processor
The program according to supplementary note 1, which executes the step of notifying a patient that medical treatment by an attending physician is necessary when the estimated disease activity index exceeds a predetermined threshold set in advance.

(Appendix 11)
In addition, the processor
a step of totalizing the frequency of acquisition of stool images by patients (step S221);
The program according to supplementary note 1, which causes a terminal device to receive a notification prompting continuous acquisition of stool images when the aggregated acquisition frequency falls below a certain standard (step S222).

(Appendix 12)
In addition, the processor
a step of totalizing the frequency of acquisition of stool images by patients (step S221);
The program according to supplementary note 1, which causes the terminal device to perform the step (step S222) of notifying the terminal device of certifying continuous acquisition of stool images when the aggregated acquisition frequency exceeds a certain standard. .

(Appendix 13)
In addition, the processor
In Supplementary Note 1, when the estimated disease activity index exceeds a predetermined threshold set in advance, a step (step S204) of notifying the terminal used by the patient's attending physician to that effect is executed. Programs listed.

(Appendix 14)
In addition, the processor
A step of generating an endoscope-equivalent image that is estimated as the state of the patient's intestine when the stool associated with the stool image is excreted through image processing using the acquired stool image (step S241);
The program according to supplementary note 1, which executes the step of notifying a terminal used by a patient's attending physician that an endoscope-equivalent image has been created (step S242).

(Appendix 15)
A method performed by an information processing system including a server having a processor, the method comprising:
a step of acquiring a stool image of the patient taken by a terminal device used by the patient or a medical worker (step S201);
estimating a disease activity index related to a specific disease of the patient by image processing the acquired stool image (step S202);
A method comprising: outputting the estimated disease activity index to a terminal device (step S203).

(Appendix 16)
An information processing system including a server having a processor, the information processing system comprising:
means for acquiring a stool image of the patient taken by a terminal device used by the patient or a medical worker;
means for estimating a disease activity index related to a specific disease of a patient through image processing of the acquired stool image;
An information processing system comprising: means for outputting an estimated disease activity index to a terminal device.

(Appendix 17)
A defecation sheet used for photographing stool images used when executing the program according to any one of Supplementary Notes 1 to 14,
A defecation sheet is used to cover a water seal in a toilet where a patient defecates, and has a water-resistant sheet body to prevent feces from flowing into the water seal during the time required for imaging.

(Appendix 18)
The defecation sheet according to appendix 17, wherein at least a portion of the sheet body is colored, and at least one of a color standard and a size standard used in image processing of a stool image is displayed on the surface of the sheet body.

1 system 1
10 Terminal device 12 Communication IF
13 input device 14 output device 15 memory 16 storage unit 19 processor 20 evaluation server 20
20 Terminal device 22 Communication IF
23 input device 24 output device 25 memory 26 storage unit 29 processor 80 network

Claims (13)

A program that is executed by an information processing system including a server having a processor, the program causing the processor to:
acquiring a stool image of the patient taken by a terminal device used by the patient or a medical worker;
a step of acquiring at least one of behavioral history indicating the history of eating behavior including excretion behavior and medication input into the terminal device, biological information, and past medical treatment history at a medical institution;
The acquired stool image is used as input data, and the acquired stool image is applied to an activity estimation model created by machine learning using learning data with the disease activity index related to a specific disease as correct output data. inputting a first estimate of the disease activity index for the disease in the patient ;
obtaining a second estimated value of the disease activity index based on preset evaluation criteria using at least one of the acquired behavioral history, biological information, and medical history;
Comparing the first estimate and the second estimate to provide an overall evaluation of the disease activity index;
A program that causes the program to execute the step of outputting the disease activity index obtained by the comprehensive evaluation to the terminal device. The processor further includes:
The image information of the excrement is input data, and the correct output data is an endoscopic image of the intestine taken by endoscopy of the patient who excreted the excrement, based on the input image information of the excrement. By inputting the acquired stool image into an image synthesis model created by machine learning using learning data, the endoscope estimates the state of the intestine of the patient who excreted the stool. a step of outputting a corresponding image;
2. The program according to claim 1, wherein the program executes the step of notifying a terminal used by the patient's attending physician that the endoscope-equivalent image has been created. The activity estimation model is created by machine learning using learning data whose correct output data is information regarding the lesion range of inflammatory bowel disease of the person who excreted the stool included in the stool image,
In the step of outputting the first estimated value of the disease activity index , further:
The program according to claim 1 , wherein the activity estimation model outputs an estimated position of a lesion range in the disease. In the step of outputting the first estimated value of the disease activity index , further:
The program according to claim 1, which outputs the estimated likelihood of the disease activity index. The processor further includes:
Relearning the activity estimation model used for estimating the disease activity index using the patient's stool image and the actual disease activity index at the time of excretion of stool corresponding to the stool image as learning data. The program according to claim 1, which causes the program to execute the step of causing the computer to perform the steps. In the step of outputting the first estimated value of the disease activity index , image processing is performed on the stool image using color standards or size standards written on a defecation sheet used by the patient when defecating. The program described in Section 1. The terminal device is a patient terminal used by the patient in daily life,
In the step of acquiring the stool image,
Detecting excrement included in the image taken by the patient terminal ;
The program according to claim 1, which executes the step of transmitting the image to the server without storing the image in the patient terminal when the excrement is detected in the image. The processor further includes:
2. The method according to claim 1, further comprising the step of notifying the patient that medical treatment by an attending physician is required when the estimated disease activity index exceeds a predetermined threshold set in advance. program. The processor further includes:
aggregating the frequency of acquisition of the stool images by the patient;
2. The program according to claim 1, wherein the program causes the terminal device to perform the step of notifying the terminal device to urge continuous acquisition of the stool images when the aggregated acquisition frequency falls below a certain standard. The processor further includes:
aggregating the frequency of acquisition of the stool images by the patient;
2. The program according to claim 1, wherein the program causes the terminal device to perform the step of notifying the terminal device of certifying continuous acquisition of the stool images when the aggregated acquisition frequency exceeds a certain standard. . The processor further includes:
According to claim 1, when the estimated disease activity index exceeds a predetermined threshold set in advance, a terminal used by the patient's attending physician is caused to perform a step of notifying a terminal to that effect. program. A method executed by an information processing system including a server having a processor, the method comprising:
acquiring a stool image of the patient taken by a terminal device used by the patient or a medical worker;
a step of acquiring at least one of behavioral history indicating the history of eating behavior including excretion behavior and medication intake entered into the terminal device, biological information, and past medical treatment history at a medical institution;
The acquired stool image is used as input data, and the acquired stool image is applied to an activity estimation model created by machine learning using learning data with the disease activity index related to a specific disease as correct output data. inputting a first estimate of the disease activity index for the disease of the patient ;
obtaining a second estimated value of the disease activity index based on preset evaluation criteria using at least one of the acquired behavioral history, biological information, and medical history;
Comparing the first estimate and the second estimate to provide an overall evaluation of the disease activity index;
A method comprising: outputting the disease activity index obtained by the comprehensive evaluation to the terminal device. An information processing system comprising a server having a processor, the information processing system comprising:
means for acquiring a stool image of the patient taken by a terminal device used by the patient or a medical worker;
means for acquiring at least one of behavioral history indicating a history of eating behavior including excretion behavior and medication intake entered into the terminal device, biological information, and past medical treatment history at a medical institution;
The acquired stool image is used as input data, and the acquired stool image is applied to an activity estimation model created by machine learning using learning data with the disease activity index related to a specific disease as correct output data. Means for outputting a first estimate of the disease activity index regarding the disease in the patient upon input ;
means for acquiring a second estimated value of the disease activity index based on preset evaluation criteria using at least one of the acquired behavioral history, biological information, and medical history;
means for comparing the first estimated value and the second estimated value to perform a comprehensive evaluation of the disease activity index;
An information processing system comprising: means for outputting the disease activity index obtained by the comprehensive evaluation to the terminal device.