JP7193188B1 - Medical item extraction system - Google Patents

Abstract

A medical item extraction system configured to extract items related to medical care can improve convenience. According to one aspect of the present disclosure, a medical item extraction system configured to extract medical items includes a data acquisition unit, a text recognition unit, a treatment extraction unit, and a notification transmission unit. , provided. The data acquisition unit is configured to acquire image data related to medical information including test results for each user. The text recognizer is configured to recognize text contained in the image data. The action extractor is configured to extract unnecessary or redundant action items for individual users based on the recognized text. The notification transmission unit is configured to notify a preset contact when the action item is extracted. [Selection drawing] Fig. 2

Description

Article 30, Paragraph 2 of the Patent Law is applied Published by releasing application software (Published date: March 26, 2021) Posted on website (Published date: March 26, 2021)) (Address 1) https ://first-priority-inc. co. jp/buddydoctor/phrase/ (Address 2) https://first-priority-inc. co. jp/#service

The present disclosure relates to a medical item extraction system configured to extract medical items.

Patent Document 1 below discloses a medical item extraction system that checks for duplicate medical treatments such as medication and examinations as items related to medical care, and when it is determined that the same medical practice is performed at different medical institutions in the same month , a technique of notifying the insured person has been proposed.

JP 2014-157393 A

However, as a result of a detailed study by the inventor, the technique of Patent Document 1 above requires manual input of original data used to determine duplicate medical care. There were problems with convenience, such as the difficulty of obtaining cooperation from

One aspect of the present disclosure is to improve convenience in a medical item extraction system configured to extract medical items.

One aspect of the present disclosure is a medical item extraction system configured to extract medical items, comprising a data acquisition component, a text recognition component, a treatment extraction component, and a notification transmission component. The data acquisition unit is configured to acquire image data related to medical information including test results for each user. The text recognizer is configured to recognize text contained in the image data. The action extractor is configured to extract unnecessary or redundant action items for individual users based on the recognized text. The notification transmission unit is configured to notify a preset contact when the action item is extracted.

According to such a configuration, the text included in the image data can be recognized, and the treatment item can be extracted based on this text. Therefore, since there is no need to input text manually, convenience can be improved.

1 is a block diagram showing the configuration of a medical support system (medical item extraction system) according to a first embodiment; FIG. 4 is a flowchart showing processing in the first embodiment; FIG. 4 is an image diagram showing an example of display on a smartphone when medical information is captured. It is a table showing an example of cloud data. 4 is a table showing setting examples of determination thresholds; FIG. 11 is an image diagram showing a display example of duplicate notifications; It is a block diagram which shows the structure of the medical assistance system (insurance premium calculation system) of 2nd Embodiment. It is a flow chart which shows processing in a 2nd embodiment. FIG. 11 is an image diagram showing a display example on a smartphone when a living will is input;

Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.
[1. First Embodiment]
[1-1. composition]
A medical support system 1 shown in FIG. 1 is a system configured to provide medical support such as examinations and medical examinations. The medical support system 1 particularly has a function as a medical item extraction system for extracting redundant or useless items. The medical support system 1 is configured around an arithmetic device 10 . The medical support system 1 may include a user device 20 , a medical institution device 25 , a management device 30 and an examination device 35 .

The arithmetic device 10 is a device that implements the main processing in this embodiment. The computing device 10 is configured as, for example, a server device. The computing device 10 is configured to communicate with the user device 20, the medical institution device 25, the management device 30, and the examination device 35 via an Internet network, a mobile phone base station, or the like (not shown).

The computing device 10 includes a microcomputer having a CPU 11 and a semiconductor memory such as RAM or ROM (hereinafter referred to as memory 12). Each function of the arithmetic device 10 is implemented by the CPU 11 executing a program stored in a non-transitional substantive recording medium. In this example, the memory 12 corresponds to a non-transitional substantive recording medium storing programs. Also, by executing this program, a method corresponding to the program is executed. Note that the non-transitional substantive recording medium means a recording medium excluding electromagnetic waves. Further, the computing device 10 may have one microcomputer or may have a plurality of microcomputers.

The computing device 10 is configured to be able to record the cloud data 16 on a recording medium such as a hard disk or flash memory, and has a function as an overlap waste determining section 17 . The function of the duplication waste determination unit 17 is realized by the first server processing described later. The method of realizing the functions included in the arithmetic device 10 is not limited to software, and some or all of the functions may be realized using one or more pieces of hardware. For example, when the above functions are realized by an electronic circuit that is hardware, the electronic circuit may be realized by a digital circuit, an analog circuit, or a combination thereof.

The user device 20 is configured as, for example, a device such as a smartphone possessed by the user. A user is a person who uses an application for the medical support system 1 on the user device 20 . In addition, the user includes the patient himself/herself who undergoes examination, medical treatment, etc., and the family member of the patient himself/herself.

Hereinafter, the user device 20 is also called a smart phone. However, the user device 20 is not limited to a smart phone, and may be a personal computer, a tablet terminal, or the like. The user device 20 includes a CPU and a memory (not shown), and, like the computing device 10, executes a program stored in the memory to implement first smartphone processing, which will be described later.

The user device 20 downloads an application for executing the first smartphone process from an application server (not shown) that provides the application by the user's operation, and the application is installed, thereby enabling execution of the first smartphone process. Become. Although only one user device 20 is illustrated in FIG. 1 , the medical support system 1 may be provided with a plurality of user devices 20 . For example, each company may have user devices 20 for all employees. The same applies to the medical institution device 25, the management device 30, and the examination device 35.

The medical institution device 25 is a computer (for example, a personal computer) installed in medical institutions such as clinics and hospitals. The medical institution device 25 may be a smart phone, a tablet terminal, or the like. The medical institution device 25 includes a CPU and a memory (not shown), and implements various processes by executing programs stored in the memory, similar to the arithmetic device 10 . For example, the medical institution device 25 can transmit a receipt (medical fee statement) to the examination device 35 .

Also, the medical institution device 25 can receive medical records from the user device 20 and the computing device 10 . A medical record is a record of medical practices such as test results and consultation details. The medical institution device 25 can receive the examination result regarding the receipt from the examination device 35 .

The management device 30 is a computer (for example, a personal computer) installed in a company's health care office, health insurance association, or the like. The management device 30 may be a smart phone, a tablet terminal, or the like. The management device 30 includes a CPU and a memory (not shown), and, like the arithmetic device 10, executes a program stored in the memory to implement management processing, which will be described later. The management device 30 can receive from the computing device 10 medical records relating to employees of a company to be managed (for example, all employees of Company A).

The examination device 35 is configured as a computer (for example, a personal computer) owned by an examination agency that examines the medical bill requested from the medical institution device 25 . The examination device 35 may be a smart phone, a tablet terminal, or the like. The examination device 35 includes a CPU and a memory (not shown), and, like the arithmetic device 10, implements various processes by executing programs stored in the memory.

For example, the examination device 35 examines a medical bill and notifies the medical institution device 25 and the management device 30 of the examination result. In addition, the examination device 35 receives a reexamination request for the examination result of the receipt (for example, a reduction assessment) from the management device 30 , performs reexamination, and notifies the medical institution device 25 and the management device 30 of the reexamination result.

[1-2. process]
Next, the first smartphone process executed by the user device 20 (smartphone) will be described using the flowchart of FIG. 2(a).

First, in S110, the smartphone activates the camera and captures an image of the inspection result according to the user's instruction.
At this time, an application related to the medical support system 1 is activated on the smartphone, and an image for selecting various medical information is displayed as shown in FIG. 3(a). Here, it is configured such that an image related to examination results including text, an image such as an X-ray photograph or a CT photograph not including text, or the like can be selected. In addition, it may be configured to enable uploading of surgical photographs, chart information, and the like.

As an example, when an image of test results containing text (e.g., “Upload blood collection/measurement data” button) is selected on the smartphone, the camera starts up on the smartphone, and the user takes an image of the test result form. Then, on the smartphone, as shown in FIG. 3B, the captured image of the inspection result is synthesized and displayed on the application. A captured image is data obtained by capturing an inspection result (for example, a sheet or a document) in which inspection items and inspection results are described in text with a terminal such as a smartphone.

When the “upload” button is operated on this image, the captured image including the inspection result is transmitted to the computing device 10 . At this time, the smartphone generates image data by adding information such as an ID for specifying the user to the captured image, and transmits this image data.

In addition, as shown in FIG. 3(a), if there is a high possibility that the image captured by the smartphone camera will be unclear, as in the case where the CT examination is selected, as shown in FIG. 3(c) Alternatively, an image prompting the user to send the photograph of the inspection result to a predetermined destination by mail may be displayed.
Subsequently, the process of S120, which will be described later, is executed. However, after the process of S110, the first smartphone process may be temporarily ended.

Next, the first server processing executed by the computing device 10 will be described using the flowchart of FIG. 2(b). The first server process is, for example, a process that starts when image data is transmitted from a smartphone.

In the first server process, first, in S210, the computing device 10 acquires the image data transmitted from the smartphone in S110. The image data is classified for each user and recorded as cloud data 16 .

Subsequently, in S220, the computing device 10 recognizes text included in the image data. At this time, the arithmetic unit 10 recognizes the text included in the image data using OCR (Optical Character Recognition) using machine learning (AI), so-called AI-OCR. AI/OCR is a character recognition technique that uses machine learning to improve character recognition accuracy.

In AI/OCR here, images of inspection results (for example, paper and documents) in various formats can be used as teacher data. In particular, the AI/OCR of this embodiment extracts a character string from an image, and searches for the extracted character string (for example, the type and arrangement of candidate characters) from information on the connected computer or the Internet. Then, the extracted character string is matched with all information existing on the computer and the Internet to recognize the correct character string. In other words, in AI/OCR, not only the data in the computing device 10 but also all information existing on other computers and on the Internet can be used as teaching data. Therefore, high accuracy can be obtained with a small amount of training data.

According to such processing, the meaning of the character string can be correctly understood, and the inspection item and the inspection result can be easily associated with each other. As a result, the reading accuracy of inspection results can be improved.

When the computing device 10 recognizes the text in the image data, a list of inspection results is arranged for each item and recorded as the cloud data 16 as shown in FIG. 4, for example. The inspection result list is data in which the past inspection results for each user are arranged in a matrix for each inspection item. The inspection result list is devised so that each user can easily recognize the transition of inspection results for each inspection item.

Subsequently, at S230, the computing device 10 extracts unnecessary or redundant action items for individual users based on the recognized text. A treatment item represents an item about medical treatment. Treatment includes any one of examination, examination, and surgery. “Unnecessary or overlapping treatment items” are items extracted from the viewpoint of saving medical costs. For example, items that have undergone multiple examinations or consultations in close proximity are extracted as duplicate items, and if there is little variation in examination results, they are extracted as items that do not require examinations or consultations.

Specifically, at S240, the arithmetic device 10 compares the inspection results and calculates the rate of change. In this process, the computing device 10 extracts a plurality of past inspection results for the same user (specifically, one past inspection result and a past (for example, immediately preceding) inspection result). Then, a difference is obtained for each item of the plurality of test results, and the rate of change of the plurality of test results is calculated. The rate of change is the rate of increase or decrease in test results.

Subsequently, in S250, the computing device 10 compares the difference between the plurality of test results, here the rate of change, with the determination threshold, and extracts items that are less than the determination threshold. The extracted items are unnecessary or redundant action items for individual users. The computing device 10 may extract treatment items for each user. In particular, unnecessary or overlapping treatment items for the one user may be extracted based on a plurality of past test results for the one user obtained from the same medical institution.

Here, the determination threshold can be set using the threshold setting table shown in FIG. The threshold setting table exemplifies many determination threshold candidates for each item, and is configured such that the server administrator can select and set an arbitrary determination threshold from a plurality of threshold candidates. However, the threshold setting table here is used for extracting items for which the judgment threshold is “or more”, that is, items for which examination or consultation is necessary. However, using the threshold setting table, it is possible to extract the items "less than" the determination threshold as unnecessary or overlapping treatment items.

For example, in the item "red blood cell count", if the rate of change "10% or more" is selected, the determination threshold is set so as to extract "450 or less and 550 or more" corresponding to "10% or more". be. In this case, two determination thresholds are set. In addition, in order to extract items that require examination or consultation, it is sufficient to extract examination results in which the “erythrocyte count” is “greater than 450 and less than 550”.

Although the determination threshold can be set using the threshold setting table in this manner, which determination threshold is to be adopted is set, for example, by a majority decision by medical personnel. In the case of this example, a questionnaire was given to 50 doctors about the conditions under which medical treatment is unnecessary, and the numerical value obtained from the agreement of 3/4 or more of the results of this questionnaire was adopted as the judgment threshold.

For example, "10% or more" is adopted for "red blood cells", and "5% or more" is adopted for "total protein TP" and "hemoglobin". For "white blood cells" and the like, a determination threshold value is set for each medical institution, and "non-facility standard value" is adopted to indicate a case where the threshold is deviated from this value. In other words, depending on the item, the determination threshold can be set for each medical institution.

The computing device 10 compares the period from when the latest test result was obtained with the period set in advance for each test item, and no test has been performed for a long time since the latest test result was obtained. Items may be excluded from the action items so as not to be extracted as duplicate action items.

For example, treatment items that have not changed in a predetermined period of time, such as the past six months, are set not to be tested on the assumption that there is no other change in the medical condition. However, if the predetermined period of time is exceeded, even a treatment item that is determined to be unchanged may be set to be inspected.

Subsequently, in S260, the computing device 10 notifies a preset contact when the treatment item is extracted. This notification includes the fact that there are unnecessary or redundant action items, the content of the action items, and the like. Also, this notification is transmitted to the user device 20 and the management device 30, for example. After S260, the first server process ends.

As shown in FIG. 2(a), when the smartphone application receives the notification regarding the treatment item from the arithmetic device 10 in S120, the user device 20 displays the notification result. In the smartphone application, for example, as shown in FIG. A pop-up is an image that is superimposed on the image being displayed. Here, it is notified that there is a "notification". When the user selects the pop-up on the image, for example, as shown in FIG. 6(b), the fact that there are overlapping items and a specific name for identifying the overlapping items are displayed on the smartphone screen. Let A user who sees this result is more likely to take actions such as refraining from duplicate examinations and medical care. When the process of S120 ends, the smartphone process of FIG. 2(a) ends.

Next, management processing executed by the management device 30 will be described with reference to the flowchart of FIG. 2(c). The management process is a process that starts when a notification transmitted from the computing device 10 is received.

In the management process, first, in S<b>310 , the management device 30 receives a notification transmitted from the arithmetic device 10 . This notification is displayed as a message on the screen of the management device 30, and the administrator who operates the management device 30 can recognize the existence of unnecessary or redundant action items and the contents of the action items. After the process of S310, the management process of FIG. 2(c) ends.

[1-3. effect]
According to 1st Embodiment detailed above, there exist the following effects.
(1a) One aspect of the present disclosure is a medical support system 1 configured to extract medical items. The medical support system 1 includes an arithmetic device 10 and the like. The computing device 10 is configured to acquire image data relating to medical information including test results for each user. Computing device 10 is configured to recognize text contained in image data. The computing device 10 is configured to extract unnecessary or redundant action items for individual users based on the recognized text. The computing device 10 is configured to notify a preset contact when a treatment item is extracted.

According to such a configuration, the text included in the image data can be recognized, and the treatment item can be extracted based on this text. Therefore, since there is no need to input text manually, convenience can be improved.

(1b) In one aspect of the present disclosure, the computing device 10 is configured to extract unnecessary or redundant treatment items for individual users by comparing multiple test results for the same user in the past.

According to such a configuration, since a plurality of examination results in the past are compared, it is possible to extract unnecessary or overlapping treatment items more satisfactorily.

(1c) In one aspect of the present disclosure, the computing device 10 is configured to extract unnecessary or redundant treatment items for individual users according to the rate of change of a plurality of test results in the past.
According to such a configuration, unnecessary or overlapping treatment items are extracted using the rate of change of a plurality of test results in the past. It is possible to satisfactorily extract treatment items such as making them unnecessary.

(1d) In one aspect of the present disclosure, the computing device 10 is configured to extract unnecessary or redundant treatment items for individual users based on medical information obtained from the same medical institution.

According to such a configuration, treatment items are extracted based on medical information obtained from the same medical institution, so it is possible to more easily detect fraud such as duplicate examinations at the same medical institution.

(1e) In one aspect of the present disclosure, the computing device 10 is configured to recognize text included in image data using machine learning.
According to such a configuration, machine learning is used to recognize the meaning of the text, so it is possible to improve the recognition accuracy of medical information.

[2. Second Embodiment]
[2-1. Difference from First Embodiment]
Since the basic configuration of the second embodiment is the same as that of the first embodiment, differences will be described below. Note that the same reference numerals as in the first embodiment indicate the same configurations, and refer to the preceding description. The second embodiment differs from the first embodiment in that insurance premiums are calculated using living wills.

[2-2. composition]
A medical support system 2 of the second embodiment includes, as shown in FIG. 7, an arithmetic device 10, a user device 20, and a medical institution device 25, as in the first embodiment. The medical support system 2 includes a processing device 40 instead of the management device 30 and examination device 35 of the first embodiment.

The processing device 40 is, for example, a computer of an insurance company (for example, a personal computer), can communicate with the arithmetic device 10 and the user device 20, and has a function of calculating insurance premiums. The processing device 40 may be a smart phone, a tablet terminal, or the like.
The processing device 40 includes a microcomputer having a CPU 41 and a semiconductor memory 42 such as RAM or ROM, as shown in FIG. As with the arithmetic unit 10, various processes are realized by executing programs stored in the semiconductor memory 42. FIG.

In the medical support system 2, the company that owns the computing device 10 and the insurance company that owns the processing device 40 encourage the user of the user device 20 to enroll in insurance. At this time, the user is notified that the insurance premium can be discounted by the living will. A user interested in insurance downloads an application from an application server (not shown), and executes second smartphone processing on the application, which will be described later. Note that the application in the second embodiment and the application in the first embodiment may be the same application or different applications.

Also, a living will is a document for expressing the intention of an individual to receive medical care at the end of life. Also called an advance directive. In addition, the user device 20 is configured to be able to disclose the living will to the medical institution device 25 with family permission or the like in an emergency (when the user is in a serious condition). When the user's living will is registered in the computing device 10 , the medical institution can provide the user's living will to the medical institution device 25 and the processing device 40 .

[2-2. process]
Next, the second smartphone process executed by the user device 20 will be described using the flowchart of FIG. 8(a). The second smartphone process is, for example, a process that starts when an application related to Living Will starts.

In the second smartphone process, first, in S410, the user device 20 inputs a living will on the application by the user. For example, as shown in FIGS. 9(a) to 9(d), the user inputs a selection regarding the presence or absence of organ donation for each organ and a selection regarding life-prolonging treatment, and finally enters a signature to make living. Complete Will.

Subsequently, in S<b>420 , the user device 20 transmits the living will to the computing device 10 . After the process of S420, the second smartphone process of FIG. 8A ends.
Next, the second server processing executed by the computing device 10 will be described using the flowchart of FIG. 8(b). The second server process is, for example, a process started when a living will is transmitted from the user device 20 .

In the second server process, first, in S<b>510 , the computing device 10 receives a living will from the user device 20 .
Subsequently, at S<b>520 , the computing device 10 transmits to the processing device 40 will data including the living will and medical records relating to the user. A medical record is just enough to determine whether the user's health condition meets insurance requirements. After the process of S520, the second server process of FIG. 8B ends.

Next, the fee calculation process executed by the processing device 40 will be described with reference to the flowchart of FIG. 8(c). The fee calculation process is, for example, a process that is started when will data is received from the calculation device 10 .

In the charge calculation process, the processing device 40 first receives will data from the calculation device 10 in S510. Subsequently, in S520, the processing device 40 utilizes the will data to calculate the insurance premium for the user regarding the will data.

Here, the insurance premium is, for example, a fee for hospitalization insurance or nursing care insurance. For example, it is calculated in the following procedure.
<<1>> If a living will is entered, the first stage discount (for example, about 10%) is applied, regardless of the content of the statement of the living will or the state of health.

<<2>> If the health condition based on the medical record is excellent, the second stage discount (for example, about 15 to 20% in total) is applied.
<<3>> If the above <<2>> discount is not applied, and regular hospital visits can be confirmed, the second stage discount (for example, about 15 to 20% in total) is applied.

<<4>> If the above <<2>> or <<3>> discount does not apply, and the living will refuses at least one of life-prolonging treatment and gastrostomy, the second stage discount (for example, about 15 to 20% in total) is applied.

That is, the processing device 40 sets multiple levels of discounts for insurance premiums. Note that the discounts <<2>> to <<4>> may be applied in duplicate. At this time, the processing device 40 determines which discount is to be applied in duplicate according to the content of the input living will. For example, if all the conditions <<1>> to <<4>> above are satisfied, the third stage discount (total, for example, about 20 to 25%) may be applied. After the process of S520, the charge calculation process of FIG. 8(c) ends.

[2-3. effect]
According to the second embodiment described in detail above, the effect (1a) of the first embodiment described above is obtained, and the following effects are also obtained.

(2a) One aspect of the present disclosure includes an arithmetic device 10 and a processing device 40 . Computing device 10 is configured to obtain a living will entered by a user. The computing device 10 is configured to transmit will data representing data containing living wills to a preset processing device 40 . The processing device 40 is configured to calculate medical insurance premiums for the user based on the received will data.
According to such a configuration, the living will can be reflected in the insurance premium.

(2b) The processing device 40 determines whether or not a plurality of conditions are met, and sets the insurance premium so as to reduce the insurance premium in stages each time the plurality of conditions are met. For example, if there is a living will, a first-stage discount is applied, and if the living will is set to refuse prescribed medical care, a further discount is applied. That is, if a living will is entered, the first stage discount is applied, and depending on the content of the living will, a larger discount than the first stage discount is applied. At this time, in addition to the first stage discount, a further second stage discount may be implemented.
According to such a configuration, the insurance premium can be set in detail according to the user's selection.

[3. Other embodiments]
Although the embodiments of the present disclosure have been described above, the present disclosure is not limited to the above-described embodiments, and various modifications can be made.

(3a) In the above embodiment, the configuration of the first embodiment and the configuration of the second embodiment are explained as different medical support systems 1 and 2, but the configuration is not limited to this. For example, the configuration of the first embodiment and the configuration of the second embodiment may be combined to form one medical support system.

(3b) The computing device 10 and techniques thereof described in this disclosure are provided by configuring a processor and memory programmed to perform one or more functions embodied by a computer program. It may be implemented by a computer. Alternatively, the computing device 10 and techniques described in this disclosure may be implemented by a dedicated computer provided by configuring a processor with one or more dedicated hardware logic circuits. Alternatively, the computing device 10 and techniques described in this disclosure are a combination of a processor and memory programmed to perform one or more functions and a processor configured by one or more hardware logic circuits. may be implemented by one or more dedicated computers configured by Computer programs may also be stored as computer-executable instructions on a computer-readable non-transitional tangible storage medium. The method of realizing the function of each part included in the arithmetic device 10 does not necessarily include software, and all the functions may be realized using one or a plurality of pieces of hardware.

(3c) A plurality of functions possessed by one component in the above embodiment may be realized by a plurality of components, or a function possessed by one component may be realized by a plurality of components. . Also, a plurality of functions possessed by a plurality of components may be realized by a single component, or a function realized by a plurality of components may be realized by a single component. Also, part of the configuration of the above embodiment may be omitted. Moreover, at least part of the configuration of the above embodiment may be added or replaced with respect to the configuration of the other above embodiment.

(3d) In addition to the medical support system 1 described above, the computing device 10 and the processing device 40 that are components of the medical support system 1, a program for causing a computer to function as the computing device 10 or the processing device 40, and recording this program The present disclosure can also be realized in various forms such as a non-transitional substantive recording medium such as a semiconductor memory, a medical support method, and the like.

[4. Correspondence between the configuration of the embodiment and the configuration of the present disclosure]
Among the functions executed by the computing device 10, S210 corresponds to the data acquisition unit in the present disclosure, S220 corresponds to the text recognition unit in the present disclosure, S230 corresponds to the treatment extraction unit in the present disclosure, and S240 corresponds to the notification transmission unit in the present disclosure. Also, among the functions executed by the processing device 40, S620 corresponds to the insurance premium calculation unit in the present disclosure.

DESCRIPTION OF SYMBOLS 1, 2... Medical support system 10... Arithmetic apparatus 11... CPU, 12... Memory, 16... Cloud data, 17... Duplication waste determination part, 20... User apparatus, 25... Medical institution apparatus, 30... Management apparatus, 35 ... examination device, 40 ... processing device, 41 ... CPU, 42 ... semiconductor memory.

Claims (6)

A medical item extraction system configured to extract medical items,
a data acquisition unit configured to acquire image data including inspection results for each user;
a text recognition unit configured to recognize text contained in the image data;
a treatment extraction unit configured to extract unnecessary or redundant treatment items for individual users based on the recognized text;
a notification transmission unit configured to notify a preset contact when the action item is extracted;
with
The treatment extraction unit compares a plurality of test results included in the recognized text with test results within a past preset period, and the rate of change of the test results is less than a preset threshold. Extract items as said unnecessary or duplicate action items
A medical item extraction system configured to: The medical item extraction system according to claim 1,
A medical item extraction system, wherein the treatment extracting unit is configured to extract unnecessary or redundant treatment items for individual users by comparing a plurality of test results for the same user in the past. The medical item extraction system according to claim 1 or claim 2 ,
A medical item extraction system, wherein the treatment extraction unit extracts unnecessary or redundant treatment items for individual users based on medical information obtained from the same medical institution. The medical item extraction system according to any one of claims 1 to 3 ,
The medical item extraction system, wherein the text recognition unit recognizes text included in the image data using machine learning. The medical item extraction system according to any one of claims 1 to 4 ,
With the data acquisition unit as the first data acquisition unit,
a second data acquisition unit configured to acquire a living will entered by a user;
an insurance premium calculation unit configured to calculate an insurance premium for the user based on will data representing data including the living will;
further comprising
The insurance premium calculation unit sets a discount rate for the insurance premium depending on whether the content of the living will satisfies a preset condition, and calculates the insurance premium to which the discount rate is applied.
A medical item extraction system configured to: The medical item extraction system according to claim 5,
The insurance premium calculation unit sets the discount rate by applying a discount in a plurality of steps each time the content of the living will satisfies a plurality of conditions set in advance.
A medical item extraction system configured to:

Images