Method for estimating thyroid disorder, method for estimation, estimation program, and method for supporting diagnosis

Abstract

To provide a technique of supporting diagnosis of a thyroid disorder by analyzing ingredients in hairs.SOLUTION: A thyroid disorder estimation device (10) has an estimation unit (14) for estimating whether a subject is having a thyroid disorder, with the amount of the mineral component in a hair collected from the subject as an input, by using an estimation model caused to learn the correlation between the amount of the mineral component in the hair and whether a subject is having a thyroid disorder, using learning data representing the amount of the mineral in the hair and the presence or absence of a thyroid disorder.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a thyroid disease estimation method, an estimation program, and a diagnostic support method. Regarding. [Background technology]

[0002] There is a known technology for predicting or diagnosing health conditions or diseases by detecting specific components contained in blood, tissues, etc. collected from a living body as indicators of health or disease (biomarkers) using a mass spectrometer (MS). Mineral components among such biological components are essential to the living body as raw materials for constituting cells and tissues, and also as components for maintaining and regulating the functions of the living body. Therefore, when an abnormality occurs in the living body, the amount of mineral components in the living body may change. Using this, there is a known technology for using mineral components in the living body as indicators of various diseases.

[0003] Patent Documents 1 and 2 describe techniques for detecting cancer using the concentration of mineral components in hair. In the technique described in Patent Document 1, cancer is detected using the concentration of mineral components in hair and the age and sex of the subject as indicators. In the technique described in Patent Document 2, the possibility of the presence of cancer is determined by comparing the relationship between the concentrations of multiple elements contained in hair with standard values. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2011-47715 A [Patent Document 2] JP 2013-221866 A Summary of the Invention [Problem to be solved by the invention]

[0005] In the techniques described in Patent Documents 1 and 2, cancer is detected by combining the mineral component concentration in hair with other indicators, and various information other than the mineral component concentration must be obtained. In addition, the techniques described in Patent Documents 1 and 2 are aimed at detecting cancer based on the mineral component concentration in hair, and it is difficult to apply them directly to other diseases. Therefore, it would be useful if there was a technique that could detect other diseases by analyzing the components in hair. Furthermore, it would be particularly useful if it could not only diagnose whether or not a patient has a disease, but also observe the progress of the disease, such as the effect of treatment.

[0006] One aspect of the present invention has been made to solve the above-mentioned problems, and aims to realize a technology that assists in the diagnosis of thyroid diseases by analyzing components in hair. [Means for solving the problem]

[0007] In order to solve the above problems, a thyroid disease prediction method according to one embodiment of the present invention involves an information processing device executing an estimation step of predicting the presence or absence of thyroid disease in a subject using a prediction model that has learned the correlation between the amount of mineral components in hair and the presence or absence of thyroid disease, using training data representing the amount of mineral components in hair and the presence or absence of thyroid disease, and using the amount of mineral components in hair obtained from the subject as input.

[0008] In one embodiment of the present invention, a method for estimating thyroid disease involves an information processing device executing an estimation step of estimating the change in the stage of thyroid disease over time in a subject using an estimation model that has learned a correlation between the amount of mineral components in the hair at each distance from the scalp and the change in the stage of thyroid disease over time, using as input the amount of mineral components in the hair at each distance from the scalp obtained from the subject, using learning data representing the amount of mineral components in the hair at each distance from the scalp and the change in the stage of thyroid disease over time.

[0009] A thyroid disease prediction device according to one embodiment of the present invention includes an estimation unit that uses training data representing the amount of mineral components in hair and the presence or absence of thyroid disease to predict the presence or absence of thyroid disease in a subject, using an estimation model that has learned a correlation between the amount of mineral components in hair and the presence or absence of thyroid disease, based on the amount of mineral components in hair obtained from the subject as input.

[0010] A thyroid disease estimation device according to one embodiment of the present invention has an estimation unit that uses learning data representing the amount of mineral components in hair at each distance from the scalp and changes in the stage of thyroid disease over time to estimate changes in the stage of thyroid disease in a subject over time, using an estimation model that has learned a correlation between the amount of mineral components in hair at each distance from the scalp and changes in the stage of thyroid disease over time, as input, and estimates changes in the stage of thyroid disease in the subject over time using the amount of mineral components in hair at each distance from the scalp obtained from the subject.

[0011] A thyroid disease estimation program according to one embodiment of the present invention is a thyroid disease estimation program for causing a computer to function as the thyroid disease estimation device described in embodiment 5 or 6, and is a thyroid disease estimation program for causing a computer to function as the estimation unit.

[0012] A method for supporting the diagnosis of thyroid disease according to one embodiment of the present invention includes a preparation step of washing hair obtained from a subject with a surfactant and then drying to prepare a sample for mineral component analysis; an analysis step of mass spectrometry of the amount of mineral components in the sample prepared in the preparation step; and a provision step of using training data representing the amount of mineral components in hair and the presence or absence of thyroid disease to provide information regarding the presence or absence of thyroid disease in the subject, as an indicator of thyroid disease, by using an estimation model that has learned a correlation between the amount of mineral components in hair and the presence or absence of thyroid disease, using the amount of mineral components obtained in the analysis step as input.

[0013] A diagnostic support method for thyroid disease according to one embodiment of the present invention includes a preparation step of washing a sample group obtained from a subject and dividing the hair into a group of samples by distance from the scalp with a surfactant and then drying to prepare a group of samples for mineral component analysis; an analysis step of mass spectrometry of the mineral components contained in each of the sample groups prepared in the preparation step; and a providing step of providing, as an index of thyroid disease, information regarding the change in the stage of thyroid disease over time in the subject, which is output from an estimation model that has learned a correlation between the amount of mineral components in the hair at each distance from the scalp and the change in the stage of thyroid disease over time using learning data representing the amount of mineral components in the hair at each distance from the scalp and the change in the stage of thyroid disease over time. Effect of the Invention

[0014] According to one aspect of the present invention, a technology can be realized that assists in the diagnosis of thyroid diseases by analyzing components in hair. [Brief description of the drawings]

[0015] [Figure 1] FIG. 1 is a block diagram showing an example of a configuration of the main components of an inference system including a thyroid disease inference device according to an embodiment of the present invention. [Diagram 2] 1 is a flowchart showing an example of an inference process executed by a thyroid disease inference device according to an embodiment of the present invention. [Diagram 3] 1 is a flowchart showing an example of an estimation model generation process executed by an estimation model generation device that generates a thyroid disease estimation model used in a thyroid disease estimation device according to an embodiment of the present invention. [Figure 4] FIG. 13 is a diagram showing the results of an analysis of various mineral components extracted from hair in an example, comparing subjects who were not affected by thyroid disease with subjects who were affected by thyroid disease. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] [Thyroid Disease Prediction System 100] 1 is a block diagram showing an example of a configuration of a main part of a thyroid disease estimation system 100 according to one embodiment of the present invention. The estimation system 100 includes a thyroid disease estimation device 10 and a thyroid disease estimation model generation device 20. The estimation system 100 also includes an input device 30, an output device 40, and a storage device 50. The estimation device 10, the estimation model generation device 20, the input device 30, the output device 40, and the storage device 50 are connected to each other so as to be able to communicate with each other.

[0017] The input device 30 accepts an input operation by a user to the estimation system 100. As an example, the input device 30 accepts an input of information regarding the amount of mineral components in the subject's hair, which is used in the estimation device 10 to estimate the presence or absence of a thyroid disease or a change in the stage of a thyroid disease over time. The input device 30 accepts an input of data used in the estimation model generation device 20 to generate a thyroid disease estimation model 24. The input device 30 can output the received information to the estimation device 10, the estimation model generation device 20, and the output device 40, and can store it in the storage device 50.

[0018] As an example, the output device 40 outputs the result estimated by the estimation device 10. The manner of output by the output device 40 is not particularly limited. The output device 40 may be, for example, a display device that displays the estimation result as an image, a printing device that prints the estimation result, or an alarm device that outputs the estimation result as sound. Furthermore, the output device 40 may be a display of a mobile device such as a smartphone that displays the result estimated by the estimation device 10.

[0019] The storage device 50 stores programs and data used in the estimation system 100. The storage device 50 stores, as an example, various data input via the input device 30. The storage device 50 also stores, as an example, teacher data used for generating an estimation model in the estimation model generating device 20 and the generated thyroid disease estimation model 24. The storage device 50 also stores, as an example, the thyroid disease estimation model 24, input information, and output information used for estimation processing in the estimation device 10. The storage device 50 may also have a database on a cloud or a server that stores various data.

[0020] (Estimation device 10) The estimation device 10 estimates the presence or absence of a thyroid disease in a subject, or the change in the stage of a thyroid disease over time. Here, the presence or absence of a thyroid disease in a subject refers to whether or not the subject is affected by a thyroid disease, and may be the possibility or probability that the subject is affected by a thyroid disease. Furthermore, the change in the stage of a thyroid disease in a subject over time refers to the change in the stage over time, such as from the onset of a thyroid disease in a subject to recovery, and may be the progress of the thyroid disease.

[0021] As an example, the estimation device 10 estimates the presence or absence of thyroid disease in a subject, or the change in the stage of thyroid disease over time, using a thyroid disease estimation model 24 generated by the estimation model generation device 20. The inventors discovered that the amount of mineral components in hair differs between healthy individuals and individuals suffering from thyroid disease, and have completed the present invention. According to one aspect of the present invention, an index of thyroid disease estimated from the amount of mineral components in hair can be provided. This can assist in the diagnosis of thyroid disease. According to one aspect of the present invention, the presence or absence of thyroid disease in a subject, or the change in the stage of thyroid disease over time can be estimated based only on information on the amount of mineral components in hair.

[0022] Regarding the estimation by the estimation device 10, thyroid diseases include thyrotoxicosis and hypothyroidism as thyroid dysfunction, and examples thereof include Graves' disease and Hashimoto's disease.

[0023] The estimation device 10 includes a control unit (information processing device) 11. The control unit 11 controls each unit of the estimation device 10, and is realized by a processor and a memory, for example. In this example, the processor accesses a storage (not shown), loads a program (not shown) stored in the storage into the memory, and executes a series of instructions included in the program. This constitutes each unit of the control unit 11. As each unit, the control unit 11 includes a data acquisition unit 12, a model acquisition unit 13, and an estimation unit 14.

[0024] <Data Acquisition Section 12> The data acquisition unit 12 acquires the measurement results of the amount of mineral components in the hair acquired from the subject. The amount of mineral components in the hair acquired from the subject may be, for example, obtained by the following preparation process and analysis process.

[0025] First, in the preparation step, a specimen to be subjected to analysis is prepared. In the preparation step, a pretreatment is performed on hair obtained from a subject before the hair is subjected to analysis. In the preparation step, the hair obtained from the subject is washed with a surfactant and then dried to prepare a specimen for analysis of mineral components. Here, hair includes head hair and body hair collected or prepared from humans or animals. Hair may also include parts involved in hair production, such as the hair root sheath and hair matrix present around the hair root within the hair follicle (or hair follicle), in addition to the hair shaft part outside the skin and the hair root part inside the skin.

[0026] In the preparation step, the surfactant used to wash the hair obtained from the subject is, for example, SDS. In addition, in the preparation step, the hair may be washed using methanol, acetone, isopropanol, or the like.

[0027] In the preparation step, the hair obtained from the subject is cut to, for example, 0.5 cm or more and 1.5 cm or less; 0.5 cm or more and 1.2 cm or less; or 0.5 cm or more and 1.0 cm or less, and washed. The hair may be cut after washing or drying. The washing time of the hair in the preparation step may be, for example, 1 minute or more and 10 minutes or less; 3 minutes or more and 8 minutes or less; or 4 minutes or more and 6 minutes or less. The hair after washing with the surfactant may be further washed with water to rinse off the surfactant.

[0028] In the preparation step, the washed hair may be dried at room temperature, or may be dried by heating at a temperature of 20° C. or more and 80° C. or less. When drying the hair at room temperature, the drying time can be appropriately set, but the hair may be dried overnight.

[0029] In the preparation step, the washed and dried hair is prepared as a specimen for analysis of mineral components. The amount of hair in the specimen may be 2 mg or more and 10 mg or less, for example, 2 mg or more and 8 mg or less; or 2 mg or more and 5 mg or less. After weighing the hair, the specimen is subjected to analysis of mineral components.

[0030] The mineral components analyzed in the specimen are mineral components contained in the specimen, and an example thereof is at least one component selected from the group consisting of cadmium, molybdenum, magnesium, and iodine. As shown in the examples described later, the amounts of these mineral components contained in hair differ between healthy subjects and patients with thyroid disease.

[0031] In the preparation step, the hair obtained from the subject may be divided by distance to prepare a sample group consisting of a plurality of samples. The progress of the thyroid disease can be estimated from the results of measuring the amount of mineral components in the hair for each distance from the scalp. This allows the time-dependent change in stages such as from the onset of thyroid disease to the cure of the disease to be estimated. In the preparation step of these sample groups, as an example, the length of the hair that has grown in the number of days from the start of treatment for the thyroid disease to the collection date is calculated, and the collected hair is divided based on a position that is that length away from the scalp, and a sample group consisting of each sample is prepared. In the preparation step, the hair may be cut at a length corresponding to the period from the collection date, taking into account the length that the hair grows in one month, and a plurality of samples may be prepared for each period.

[0032] The specimen prepared in the preparation step may be subjected to mass spectrometry of mineral components in the analysis step. In the analysis step, as an example, the specimen prepared in the preparation step is subjected to acid decomposition treatment to be dissolved, and the concentration of mineral components in the decomposed solution is measured by a multi-element simultaneous measurement method (ICP-MS method). Conventionally known measurement conditions and measurement devices can be used as the measurement conditions and measurement devices for measuring the mineral components in the specimen by the ICP-MS method. In addition, the specimen prepared in the preparation step may be analyzed by other known analysis methods to measure the amount of mineral components in hair.

[0033] The data acquisition unit 12 outputs information indicating the amount of mineral components in the hair of the subject, acquired as described above, to the estimation unit 14.

[0034] <Model Acquisition Section 13> The model acquisition unit 13 acquires a thyroid disease estimation model 24 generated by the estimation model generation device 20, in which the amount of mineral components in hair acquired from the subject is input information, and information representing the presence or absence of thyroid disease in the subject or the change in the stage of thyroid disease over time is output information. The thyroid disease estimation model 24 may be an estimation model that has learned a correlation between the amount of mineral components in hair and the presence or absence of thyroid disease using learning data representing the amount of mineral components in hair and the presence or absence of thyroid disease. The thyroid disease estimation model 24 may be an estimation model that has learned a correlation between the amount of mineral components in hair at each distance from the scalp and the change in the stage of thyroid disease over time using learning data representing the amount of mineral components in hair at each distance from the scalp and the change in the stage of thyroid disease over time.

[0035] The model acquiring unit 13 may acquire the thyroid disease estimation model 24 generated by the estimation model generating device 20 and stored in the storage device 50. The model acquiring unit 13 outputs the acquired thyroid disease estimation model 24 to the estimation unit 14.

[0036] <Estimation part 14> The estimation unit 14 estimates the presence or absence of a thyroid disease in the subject or the change in the stage of the thyroid disease over time using the thyroid disease estimation model 24. The estimation unit 14 inputs the measurement results of the amount of mineral components in the hair obtained from the subject to the thyroid disease estimation model 24, and acquires information representing the presence or absence of a thyroid disease in the subject as output information. The estimation unit 14 may also acquire information representing the change in the stage of the thyroid disease over time as output information by dividing the hair obtained from the subject by distance and inputting the measurement results of the amount of mineral components in the specimens contained in a specimen group consisting of a plurality of specimens to the thyroid disease estimation model 24 for the change in the stage of the thyroid disease over time. The estimation unit 14 outputs the acquired output information to the output device 40. The estimation unit 14 may store the acquired output information in the storage device 50.

[0037] The estimation device 10 can estimate the presence or absence of a thyroid disease in a subject or changes in the stage of a thyroid disease over time from the measurement results of the amount of mineral components in hair obtained from the subject. In this way, the estimation device 10 can estimate the presence or absence of a thyroid disease in a subject or changes in the stage of a thyroid disease over time simply by measuring the amount of mineral components in hair obtained from the subject.

[0038] (Estimation model generating device 20) The estimation model generating device 20 generates a thyroid disease estimation model 24 for estimating the presence or absence of a thyroid disease in a subject or changes in the stage of a thyroid disease over time. The estimation model generating device 20 generates the thyroid disease estimation model 24 by training using training data including the amount of mineral components in hair, and the presence or absence of a thyroid disease or changes in the stage of a thyroid disease over time.

[0039] The estimation model generating device 20 includes a control unit (information processing device) 21. The control unit 21 controls each unit of the estimation model generating device 20, and is realized by a processor and a memory, for example. In this example, the processor accesses a storage (not shown), loads a program (not shown) stored in the storage into the memory, and executes a series of instructions included in the program. This constitutes each unit of the control unit 21. As each unit, the control unit 21 includes a data acquisition unit 22 and a model generating unit (generating unit) 23.

[0040] <Data Acquisition Unit 22> The data acquisition unit 22 acquires data for generating the thyroid disease estimation model 24. The data acquisition unit 22 acquires training data including the amount of mineral components in the hair and the presence or absence of thyroid disease or the change over time in the stage of thyroid disease. The training data may be data in which the amount of mineral components in the hair is associated with data representing the presence or absence of thyroid disease or the change over time in the stage of thyroid disease in the subject from whom the hair was obtained. The amount of mineral components in the hair included in the training data may be the measurement result obtained by the above-mentioned preparation step and analysis step. The data representing the presence or absence of thyroid disease or the change over time in the stage of thyroid disease included in the training data may be the result of a diagnosis of thyroid disease in the subject from whom the hair was obtained.

[0041] The data acquiring unit 22 may read out the teacher data from the storage device 50 based on an input signal representing an instruction to start generating an estimation model from the input device 30. The data acquiring unit 22 may also acquire the teacher data input via the input device 30. The data acquiring unit 22 outputs the acquired teacher data to the model generating unit 23.

[0042] <Model Generation Unit 23> The model generating unit 23 generates the thyroid disease estimation model 24 by learning the teacher data acquired by the data acquiring unit 22. The model generating unit 23 generates the thyroid disease estimation model 24 in which the amount of mineral components in the hair acquired from the subject is input information, and the presence or absence of thyroid disease or the change in the stage of thyroid disease over time in the subject is output information. The model generating unit 23 generates the thyroid disease estimation model 24 by learning the correlation between the amount of mineral components in the hair and the presence or absence of thyroid disease using teacher data including the amount of mineral components in the hair and the presence or absence of thyroid disease. In addition, the model generating unit 23 may generate the thyroid disease estimation model 24 for the change in the stage of thyroid disease over time by dividing the hair acquired from the subject by distance, and using teacher data including the amount of mineral components in the specimens included in a specimen group consisting of a plurality of specimens and the change in the stage of thyroid disease over time, learning the correlation between the amount of mineral components in the hair included in the specimen group and the change in the stage of thyroid disease over time.

[0043] The model generation unit 23 may generate the thyroid disease estimation model 24 using, for example, a known learning method such as a neural network, a decision tree, a random forest, or a support vector machine.

[0044] The estimation model generating device 20 can generate a thyroid disease estimation model 24 that estimates the presence or absence of thyroid disease in a subject or the change in the stage of thyroid disease over time from the amount of mineral components in hair obtained from the subject. By using the thyroid disease estimation model 24, it is possible to provide an index of thyroid disease estimated from the amount of mineral components in hair, thereby supporting the diagnosis of thyroid disease. The estimation model generating device 20 can generate an estimation model that estimates the presence or absence of thyroid disease in a subject or the change in the stage of thyroid disease over time based only on information on the amount of mineral components in hair.

[0045] (Estimation process) An estimation method according to one aspect of the present invention is executed by a control unit 11 of an information processing device such as the estimation device 10. The flow of the thyroid disease estimation process (thyroid disease estimation method) by the estimation device 10 will be described with reference to FIG. 2. FIG. 2 is a flowchart showing an example of the estimation process executed by the estimation device 10. As shown in FIG. 2, first, the data acquisition unit 12 acquires the amount of mineral components in the hair (step S11). Next, the model acquisition unit 13 acquires the thyroid disease estimation model 24 generated by the estimation model generation device 20 (step S12).

[0046] The estimation unit 14 estimates the presence or absence of a thyroid disease in the subject using the thyroid disease estimation model 24 acquired by the model acquisition unit 13. The estimation unit 14 inputs information representing the amount of mineral components in the hair of the subject to the thyroid disease estimation model 24, and acquires an output estimated value representing the presence or absence of a thyroid disease (step S13, estimation step). The estimation unit 14 outputs the acquired estimated value to the output device 40 as an estimation result (step S14), and ends the estimation process.

[0047] (Thyroid disease estimation model generation processing) A thyroid disease estimation model generating method according to one aspect of the present invention is executed by a control unit 21 of an information processing device such as an estimation model generating device 20. A flow of a process for generating a thyroid disease estimation model 24 by the estimation model generating device 20 (a thyroid disease estimation model generating method) will be described with reference to Fig. 3. Fig. 3 is a flowchart showing an example of the estimation model generating process executed by the estimation model generating device 20. As shown in Fig. 3, first, the data acquiring unit 22 acquires teacher data (step S21).

[0048] Next, the model generation unit 23 generates a thyroid disease estimation model 24 in which information representing the amount of mineral components in the subject's hair is input information and information representing the presence or absence of a thyroid disease in the subject is output information by learning using the acquired teacher data (step S22, generation step).The model generation unit 23 then outputs the generated thyroid disease estimation model 24 to the estimation device 10 (step S23), and the thyroid disease estimation model generation process ends.

[0049] [Method for supporting diagnosis of thyroid disease] A method for supporting the diagnosis of thyroid disease according to one embodiment of the present invention includes a preparation step of washing hair obtained from a subject with a surfactant and then drying to prepare a sample for mineral component analysis; an analysis step of mass spectrometry of the amount of mineral components in the sample prepared in the preparation step; and a provision step of using training data representing the amount of mineral components in hair and the presence or absence of thyroid disease to provide information regarding the presence or absence of thyroid disease in the subject, as an indicator of thyroid disease, by using an estimation model that has learned a correlation between the amount of mineral components in hair and the presence or absence of thyroid disease, using the amount of mineral components obtained in the analysis step as input.

[0050] A diagnostic support method for thyroid disease according to another aspect of the present invention includes a preparation step of washing a sample group obtained from a subject and dividing the hair into a group of samples by distance from the scalp with a surfactant and then drying to prepare a group of samples for mineral component analysis; an analysis step of mass spectrometry of the mineral components contained in each of the sample groups prepared in the preparation step; and a provision step of using an estimation model that has learned a correlation between the amount of mineral components in the hair at each distance from the scalp and the change in the stage of thyroid disease over time using learning data representing the amount of mineral components in the hair at each distance from the scalp and the change in the stage of thyroid disease over time, as an input, and providing information regarding the change in the stage of thyroid disease in the subject over time as an index of thyroid disease.

[0051] As an example, these diagnostic support methods for thyroid diseases may provide, as an index of thyroid diseases, the inference results obtained by the above-mentioned estimation device 10. That is, the preparation steps and analysis steps in these diagnostic support methods for thyroid diseases are similar to the preparation steps and analysis steps described for the estimation device 10.

[0052] In the providing step of the method for supporting diagnosis of thyroid disease according to one embodiment of the present invention, the relationship between the amount of mineral components in a sample and thyroid disease can be provided to a user. This can support the diagnosis of whether or not a subject has a thyroid disease. In the providing step, as an example, when the measured value of a mineral component in a sample is outside a predetermined reference range, an indicator is presented that indicates that the thyroid disease level is outside the reference range. Here, as an example, the thyroid disease level may indicate the possibility of having a thyroid disease in stages.

[0053] According to the method for supporting diagnosis of thyroid disease according to one aspect of the present invention, the amount of mineral components in hair measured at each distance from the scalp can be provided as an index showing the progress of thyroid disease. In this way, by providing information showing the time-dependent change in the stages of thyroid disease, such as from onset to recovery, it is possible to support the observation of the progress of thyroid disease.

[0054] The method of providing the user with the thyroid disease indicator in the providing step is not particularly limited. As an example, the measurement values ​​of the mineral components in the sample and the thyroid disease indicator may be displayed on a computer display, a display of a mobile device such as a smartphone, or the like. In addition, the thyroid disease indicator may be provided to the user by providing a medium on which the measurement values ​​of the mineral components in the sample and the thyroid disease indicator are printed.

[0055] [Software implementation example] The functions of the thyroid disease estimation device 10 and the estimation model generation device 20 (hereinafter referred to as "each device") can be realized by a program for causing a computer to function as each of the devices, and a program for causing a computer to function as each control block of each device (particularly each part included in the control unit 11 and the control unit 21) (a thyroid disease estimation program and an estimation model generation program).

[0056] In this case, each of the above devices includes a computer having at least one control device (e.g., a processor) and at least one storage device (e.g., a memory) as hardware for executing the above program. The above program is executed by the control device and the storage device, thereby realizing each function described in the above embodiment.

[0057] The above-mentioned program may be recorded in one or more computer-readable recording media, not in a temporary manner. The recording media may or may not be included in each of the above-mentioned devices. In the latter case, the above-mentioned program may be supplied to each of the above-mentioned devices via any wired or wireless transmission medium.

[0058] In addition, some or all of the functions of each of the control blocks can be realized by a logic circuit. For example, an integrated circuit in which a logic circuit that functions as each of the control blocks is formed is also included in the scope of the present invention. In addition, the functions of each of the control blocks can be realized by, for example, a quantum computer.

[0059] Furthermore, each process described in the above embodiment may be executed by AI (Artificial Intelligence). In this case, the AI ​​may be operated by the control device or another device (for example, an edge computer or a cloud server).

[0060] [Additional Note 1] In the method for predicting thyroid disease according to aspect 1 of the present invention, an information processing device executes an estimation step of predicting the presence or absence of thyroid disease in a subject using a prediction model that has learned the correlation between the amount of mineral components in hair and the presence or absence of thyroid disease using learning data representing the amount of mineral components in hair and the presence or absence of thyroid disease, and using the amount of mineral components in hair obtained from the subject as input.

[0061] In a method for estimating thyroid disease according to aspect 2 of the present invention, an information processing device executes an estimation step of estimating a change in the stage of thyroid disease over time in a subject using an estimation model that has learned a correlation between the amount of mineral components in hair at each distance from the scalp and the change in the stage of thyroid disease over time, using as input the amount of mineral components in hair at each distance from the scalp obtained from the subject, using learning data representing the amount of mineral components in hair at each distance from the scalp and the change in the stage of thyroid disease over time.

[0062] A method for predicting a thyroid disease according to a third aspect of the present invention is the method according to the first or second aspect, wherein the mineral component is at least one component selected from the group consisting of cadmium, molybdenum, magnesium, and iodine.

[0063] A fourth aspect of the present invention relates to a method for predicting a thyroid disease, in which the thyroid disease is hyperthyroidism or hypothyroidism in any one of the first to third aspects.

[0064] The thyroid disease prediction device of aspect 5 of the present invention includes an prediction unit that uses training data representing the amount of mineral components in hair and the presence or absence of thyroid disease to predict the presence or absence of thyroid disease in a subject, using an prediction model that has learned a correlation between the amount of mineral components in hair and the presence or absence of thyroid disease, based on the amount of mineral components in hair obtained from the subject as input.

[0065] The thyroid disease estimation device of aspect 6 of the present invention has an estimation unit that uses learning data representing the amount of mineral components in hair at each distance from the scalp and changes in the stage of thyroid disease over time to estimate changes in the stage of thyroid disease in a subject over time, using an estimation model that has learned a correlation between the amount of mineral components in hair at each distance from the scalp and changes in the stage of thyroid disease over time, as input, and estimates changes in the stage of thyroid disease in the subject over time using the amount of mineral components in hair at each distance from the scalp obtained from the subject.

[0066] A thyroid disease estimation program of aspect 7 of the present invention is a thyroid disease estimation program for causing a computer to function as the thyroid disease estimation device described in aspect 5 or 6, and is a thyroid disease estimation program for causing a computer to function as the estimation unit.

[0067] A diagnostic support method for thyroid disease according to aspect 8 of the present invention includes a preparation step of washing hair obtained from a subject with a surfactant and drying to prepare a sample for mineral component analysis; an analysis step of mass spectrometry of the amount of mineral components in the sample prepared in the preparation step; and a provision step of using training data representing the amount of mineral components in hair and the presence or absence of thyroid disease to provide information regarding the presence or absence of thyroid disease in the subject as an indicator of thyroid disease, using an estimation model that has learned a correlation between the amount of mineral components in hair and the presence or absence of thyroid disease, as input, using the amount of mineral components obtained in the analysis step as an input.

[0068] A diagnostic support method for thyroid disease according to a ninth aspect of the present invention includes a preparation step of washing with a surfactant and drying a group of samples obtained from a subject and dividing the hair into groups according to distance from the scalp to prepare a group of samples for mineral component analysis; an analysis step of mass spectrometry of the mineral components contained in each of the group of samples prepared in the preparation step; and a provision step of using learning data representing the amount of mineral components in the hair at each distance from the scalp and changes in the stage of thyroid disease over time to input an estimation model that has learned a correlation between the amount of mineral components in the hair at each distance from the scalp and changes in the stage of thyroid disease over time, and providing information regarding changes in the stage of thyroid disease in the subject over time as an index of thyroid disease.

[0069] [Additional Note 2] The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. Embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention. EXAMPLES

[0070] The mineral components in hair were measured as follows. Hair samples were taken from three healthy subjects and three patients with Graves' disease.

[0071] Hair was cut and collected from the head and scalp of each subject. The cut hair was cut from the root and used. The collected hair was washed with SDS and naturally dried at room temperature before being used for analysis.

[0072] The concentrations of cadmium (Cd), molybdenum (Mo), magnesium (Mg), sodium (Na), and iodine (I) in the samples pretreated as described above were measured by ICP-MS using a mass spectrometer. The concentrations of each mineral component in the samples of healthy subjects (HC, control) and subjects with Graves' disease are shown in Figure 4. Figure 4 shows the results of an analysis of various mineral components extracted from hair in an example, comparing subjects with and without thyroid disease.

[0073] As shown in Figure 4, the mineral concentration in hair of Graves' disease patients was measured during both the treatment and abnormal periods. Sample groups including samples from both the treatment and abnormal periods were prepared by calculating the length of hair that had grown over the number of days from the start of thyroid disease treatment to the date of collection, and dividing the collected hair at a position that was that length away from the scalp.

[0074] The hair division positions were determined based on the fact that the period from the first consultation date to the collection date of the Graves' disease patient was approximately 3 months, that hair grows approximately 1 cm per month, and that the depth from the root to the scalp is 4 mm. The hair up to 2.6 cm from the scalp was used as a sample for the treatment period, and the hair from 2.6 cm from the scalp to the tip was used as a sample for the abnormal period. The hair of the control healthy subject was also divided in the same way, and samples for the treatment period control and the abnormal period control were used.

[0075] As shown in Figure 4, the amount of mineral components in hair from both healthy subjects and patients with Graves' disease could be measured. Also, as shown in Figure 4, the concentrations of cadmium, molybdenum, magnesium, sodium, and iodine differed between samples from healthy subjects and samples from patients with Graves' disease. As a result, it was shown that analysis of these mineral components in hair can assist in the diagnosis of thyroid diseases. In addition, the values ​​of cadmium and molybdenum were high in the abnormal phase and tended to decrease after the start of treatment, while the values ​​of magnesium were low in the abnormal phase and tended to increase after the start of treatment. By monitoring the behavior of changes in the concentrations of these components, more detailed diagnostic support is possible. [Industrial Applicability]

[0076] The present invention can be used in the medical field, for example, to predict or diagnose health conditions or diseases. [Explanation of symbols]

[0077] 10 Estimation device 14 Estimation part

Claims

1. an estimation step of estimating the presence or absence of a thyroid disease in a subject using an estimation model that has learned a correlation between the amount of mineral components in hair and the presence or absence of a thyroid disease using training data representing the amount of mineral components in hair and the presence or absence of a thyroid disease, and using the amount of mineral components in hair obtained from the subject as an input; The method for predicting a thyroid disease is executed by an information processing device.

2. an estimation step of estimating a change in the stage of thyroid disease over time in a subject using an estimation model that has learned a correlation between the amount of mineral components in hair at each distance from the scalp and the change in the stage of thyroid disease over time, by using training data representing the amount of mineral components in hair at each distance from the scalp and the change in the stage of thyroid disease over time, and inputting the amount of mineral components in hair at each distance from the scalp obtained from the subject; The method for predicting a thyroid disease is executed by an information processing device.

3. The method for predicting a thyroid disease according to claim 1 or 2, wherein the mineral component is at least one component selected from the group consisting of cadmium, molybdenum, magnesium, and iodine.

4. The method for predicting a thyroid disease according to claim 1 or 2, wherein the thyroid disease is hyperthyroidism or hypothyroidism.

5. an estimation unit that estimates the presence or absence of a thyroid disease in a subject by using a learning data representing the amount of mineral components in hair and the presence or absence of a thyroid disease and an estimation model that has learned a correlation between the amount of mineral components in hair and the presence or absence of a thyroid disease, and inputs the amount of mineral components in hair obtained from the subject; A thyroid disease prediction device comprising:

6. an estimation unit that estimates a change in the stage of a thyroid disease over time in a subject using an estimation model that has learned a correlation between the amount of mineral components in the hair at each distance from the scalp and the change in the stage of a thyroid disease over time, and inputs the amount of mineral components in the hair at each distance from the scalp obtained from the subject, using learning data that indicates the amount of mineral components in the hair at each distance from the scalp and the change in the stage of a thyroid disease over time; A thyroid disease prediction device comprising:

7. A thyroid disease inference program for causing a computer to function as the thyroid disease inference device according to claim 5 or 6, the thyroid disease inference program causing a computer to function as the inference unit.

8. A preparation step of washing hair obtained from a subject with a surfactant and then drying the hair to prepare a sample for mineral component analysis; an analyzing step of performing mass spectrometry on the amount of mineral components in the sample prepared in the preparing step; a providing step of providing, as an index of thyroid disease, information on the presence or absence of thyroid disease in a subject, using an estimation model that has been trained to learn a correlation between the amount of mineral components in hair and the presence or absence of thyroid disease using training data representing the amount of mineral components in hair and the presence or absence of thyroid disease, with the amount of mineral components obtained in the analyzing step as an input; A method for supporting diagnosis of thyroid diseases, comprising:

9. A preparation step of washing the hair obtained from the subject with a surfactant and drying the hair to prepare a sample group for mineral component analysis; an analyzing step of performing mass spectrometry on mineral components contained in each of the sample groups prepared in the preparing step; a providing step of providing, as an index of thyroid disease, information on the change in the stage of thyroid disease over time in a subject using an estimation model that has learned a correlation between the amount of mineral components in hair at each distance from the scalp and the change in the stage of thyroid disease over time, using learning data representing the amount of mineral components in hair at each distance from the scalp and the change in the stage of thyroid disease over time, and inputting the amount of mineral components in hair at each distance from the scalp obtained from the subject; A method for supporting diagnosis of thyroid diseases, comprising:

Images