JP2021076437A - System, server device, client device, biosensor, biosensor set, data acquisition device, and program - Google Patents

Abstract

To provide a system with which it is possible to provide even an examinee in a region or an environment difficult to commute to a medical institution with a prediction result pertaining to the progress state of tooth decay and periodontal disease on the basis of data obtained by a simpler technique.SOLUTION: A system comprises a server device, a client device and a data acquisition device. The data acquisition device acquires training data that includes the electrochemical properties of a test liquid and the progress state of tooth decay and periodontal disease and transmits the training data to the server device. The server device performs learning on the basis of the feature quantity obtained from the electrochemical properties included in the training data and the progress state. The client device measures the electrochemical properties of a specimen, acquires test data and transmits the test data to the server device. The server device applies the test data to a sorter and predicts the progress state of tooth decay and periodontal disease, transmits the result of prediction to the client device, and prompts the client device to display it.SELECTED DRAWING: Figure 1

Description

The present invention relates to a system, a server device, a client device, a biosensor, a biosensor set, a data acquisition device, and a program.

In recent years, with the improvement of computer processing power, it has become possible to handle a large amount of data in various formats at high speed, and machine learning etc. based on the large amount of data obtained by analyzing samples obtained from humans. Technological development for predicting human health condition etc. is being promoted by the method.

For example, Patent Document 1 states, "Whether a biometric data acquisition means for acquiring biometric data to be predicted and biometric data for constructing a predictive model and a patient corresponding to the biometric data for constructing a predictive model have developed a disease. When the onset information acquisition means for acquiring the onset information indicating whether or not the disease has occurred and the onset information indicates that the disease has developed, the reference period at the time of onset of the onset of the disease among the biological data for constructing the prediction model A sample extraction means for extracting biometric data to be a positive sample from biometric data from before to the time of onset and extracting biometric data to be a negative sample from biometric data before the reference period at the time of onset. Prediction to build a prediction model by using the positive sample as training data to develop the disease within the reference period and using the negative sample as training data not to develop the disease within the reference period. Onset prediction that predicts whether or not a patient corresponding to the prediction target biological data will develop the disease within the reference period based on the model construction means, the prediction target biological data, and the prediction model. The sample extraction means includes means, and the sample extraction means uses the result of performing a predetermined calculation on the extracted biological data as a positive sample as the positive sample, and performs a predetermined calculation on the biological data as the negative sample. Is used as the negative sample, and the disease onset prediction device. ”Is described.

Japanese Unexamined Patent Publication No. 2019-16235

The disease onset prediction device described in Patent Document 1 can predict the occurrence of a disease in advance based on biological data such as so-called vital signs, but acquisition of biological data is generally complicated. Furthermore, it could not be applied to predict the progress of caries and periodontal disease.

Therefore, the present invention is a prediction result relating to the progress state of caries and periodontal disease based on the data acquired by a simpler method even for a subject in an area or environment where it is difficult to go to a medical institution. The challenge is to provide a system that can provide.
Another object of the present invention is to provide a server device, a client device, a biosensor, a biosensor set, a data acquisition device, and a program.

As a result of diligent studies to achieve the above-mentioned problems, the present inventors have found that the above-mentioned problems can be achieved by the following configurations.

[1] It has a server device, a client device, and a data acquisition device.
The client device and the data acquisition device are configured to be able to communicate with the server device via a computer network including the Internet, and the data acquisition device is a test solution containing an electron source and a patient's fluid. The training data including the electrochemical characteristics, the caries of the patient, and the progress state of the periodontal disease are acquired, and the training data is transmitted to the server device, and the server device converts the training data into the training data. Learning is performed based on the feature amount obtained from the included electrochemical characteristics and the progress state, and a classifier is generated in which the feature amount is used as an explanatory variable and the progress state is used as an objective variable. The electrochemical characteristics of the sample including the electron source and the fluid of the subject are measured, test data is acquired, the test data is transmitted to the server device, and the server device further classifies the test data into the above classification. A system that is applied to a vessel to predict the progress of caries and periodontal disease of the subject, transmit the prediction result to the client device, and prompt the client device to display the prediction result.
[2] Training data reception for receiving training data including the electrochemical characteristics of the test solution including the electron source and the patient's fluid, the caries of the patient, and the progress of periodontal disease from the data acquisition device. A classifier generator that generates a classifier that performs learning based on the unit, the feature amount obtained from the electrochemical characteristics, and the progress state, and uses the feature amount as an explanatory variable and the progress state as an objective variable. And the test data receiving unit that receives the test data including the electrochemical characteristics of the sample including the electron source and the subject's fluid from the client device, and the test data applied to the classifier to the subject. A server device having a prediction unit that predicts the progress of erosion and periodontal disease, and a result transmission unit that returns the result of the prediction to the client device and prompts the client device to display the result.
[3] In the sample measuring unit for measuring the electrochemical characteristics of the sample including the electron source and the fluid of the subject, the test data transmitting unit for transmitting the test data including the electrochemical characteristics to the server device, and the server device. The result receiving unit that receives the result of the prediction of the caries of the subject and the progress state of the periodontal disease obtained by applying the test data to the classifier from the server device, and the result are displayed. A client device having an output unit, wherein the classifier has the electrochemical properties of the test solution containing the electron source and the patient's fluid acquired by the data acquisition device, the caries of the patient, and the patient's body fluid. , Learning is performed based on the feature amount obtained from the electrochemical characteristics in the training data including the progress state of periodontal disease and the progress state, and the feature amount generated by the server device is used as an explanatory variable. , A client device that is a classifier whose objective variable is the progress state.
[4] The sample measuring unit is connected to the electrode of a biosensor having at least two electrodes arranged so as to be in contact with the sample, and measures the electrochemical characteristics of the sample obtained from the electrode [3]. ] The client device described in.
[5] The biosensor further has a sample capture unit for capturing the sample, and the electron source is arranged in advance in the sample capture unit, and when it comes into contact with the fluid, the electron source and the above are described. The client device according to [4], wherein a sample is generated by the solution.
[6] The client device according to [5], wherein the sample capture unit is made of a solid electrolyte containing the electron source.
[7] A biosensor used for measuring the electrochemical characteristics of a sample containing an electron source and a subject's fluid, and at least two electrodes arranged so as to be in contact with the sample, and the above-mentioned in advance. A biosensor in which an electron source is arranged and has a sample capture unit in which a sample is generated by the electron source and the solution in contact with the solution.
[8] A biosensor set having the biosensor according to [7], wherein a plurality of the above-mentioned electron sources are arranged in advance.
[9] In each of the plurality of test solution accommodating portions arranged in a two-dimensional array in which the test solution containing the electron source and the patient's fluid is accommodating, the test solution accommodating unit is in contact with the test solution. A training data measuring unit and a patient's caries, which are electrically connected to each of the electrode portions composed of at least two electrodes arranged in the above and measure the electrochemical characteristics of the test solution obtained from the electrode portions. A data acquisition device including an operation unit that receives an input of a progress state of periodontal disease, and a training data transmission unit that transmits training data including the electrochemical characteristics and the progress state to a server device.
[10] The data acquisition device according to [9], wherein the plurality of test liquid accommodating portions and the electrode portions form a multi-well plate with electrodes.
[11] The computer receives training data including the electrochemical characteristics of the test solution including the electron source and the patient's fluid, the caries of the patient, and the progress of periodontal disease from the data acquisition device. Learning is performed based on the steps, the feature amount obtained from the electrochemical characteristics, and the progress state, and a step of generating a classifier using the feature amount as an explanatory variable and the progress state as an objective variable, and the above. The stage of receiving test data including the electrochemical characteristics of the sample including the electron source and the subject's fluid from the client device, and applying the test data to the classifier to apply the test data to the subject's caries and periodontal period. A program that executes a stage of predicting the progress of a disease and a stage of transmitting the result of the prediction to the client device and prompting the client device to display the result.
[12]
The subject obtained by transmitting the test data including the electrochemical characteristics of the sample including the electron source and the subject's fluid to the computer to the server device and applying the test data to the classifier by the server device. The classifier is a program that executes a stage of receiving the result of prediction of the progress of dental caries and periodontal disease from the server device and a stage of displaying the result of the prediction. The electrochemical characteristics of the test solution including the electron source and the patient's fluid acquired by the data acquisition device, and the electrochemical characteristics in the training data including the caries of the patient and the progress of periodontal disease. A program that performs learning based on the feature amount obtained from the above and the progress state, and uses the feature amount generated by the server device as an explanatory variable and the progress state as an objective variable.

According to the present invention, even for a subject in an area or environment where it is difficult to go to a medical institution, a prediction result regarding the progress state of caries and periodontal disease based on the data acquired by a simpler method. Can provide a system that can provide.
Further, according to the present invention, a server device, a client device, a biosensor, a biosensor set, a data acquisition device, and a program can also be provided.

This is a configuration example of the system according to the embodiment of the present invention. It is a hardware block diagram of the data acquisition apparatus which concerns on embodiment of this invention. It is an exploded perspective view of the multi-well plate with an electrode. It is a top view of the multi-well plate with an electrode. It is a figure which shows the enlarged plan view of the well of the multi-well plate with an electrode, and the back surface | It is a functional block diagram of a data acquisition device. It is a hardware block diagram of the server apparatus which concerns on embodiment of this invention. It is a functional block diagram of a server device. This is the operation flow of the control unit of the server device that operates according to the program. It is a hardware block diagram of the client apparatus which concerns on embodiment of this invention. It is an exploded perspective view of a biosensor. It is a functional block diagram of a client device. This is the operation flow of the control unit of the client device that operates according to the program. It is an operation flow of the system which concerns on embodiment of this invention.

Hereinafter, the present invention will be described in detail.
The description of the constituent elements described below may be based on typical embodiments of the present invention, but the present invention is not limited to such embodiments.
In the present specification, the numerical range represented by using "~" means a range including the numerical values before and after "~" as the lower limit value and the upper limit value.

[system]
FIG. 1 is a configuration example of a system according to an embodiment of the present invention. The system 100 includes a data acquisition device 101, a server device 102, and a client device 103, and the data acquisition device 101 and the server device 102, and the server device 102 and the client device 103 form a computer network including the Internet. It is configured to be able to communicate with each other via.

[Data acquisition device]
FIG. 2 is a hardware configuration diagram of the data acquisition device 101 according to the embodiment of the present invention. The data acquisition device 101 includes a processor 201, a storage device 202, a measurement module 203, an input device 204, an output device 205, and a communication module 206. The processor 201, the storage device 202, the measurement module 203, the input device 204, the output device 205, and the communication module 206 are connected by the bus 207. The processor 201 controls the data acquisition device 101. The storage device 202 serves as a work area for the processor 201. Further, the storage device 202 is a non-temporary or temporary recording medium for storing various programs and data.

Examples of the storage device 202 include a ROM (Read Only Memory), a RAM (Random Access Memory), an HDD (Hard Disk Drive), a flash memory, and an SSD (Solid State Drive). Input device 204 inputs data. The input device 204 includes, for example, a keyboard, a mouse, a touch panel, a numeric keypad, a scanner, and the like. The output device 205 outputs data. The output device 205 includes, for example, a display and a printer. The communication module 206 connects to a computer network including the Internet and transmits / receives data.

The measurement module 203 has a plurality of wells arranged in a two-dimensional array in which a test solution containing an electron source and a patient's fluid is housed, and each of the wells is in the well so as to be in contact with the test solution. A connection terminal that is removable from a multi-well plate 208 with an electrode having at least two electrodes arranged in the above and for electrically connecting to the electrode at the time of measurement; and controls the temperature of the test solution at the time of measurement. It is a unit including a temperature control block for controlling the temperature; a potential / galvanostat in which the potential between the two electrodes can be adjusted; and the like.

FIG. 3 is an exploded perspective view of the multi-well plate 208 with electrodes, and FIG. 4 is a plan view of the multi-well plate with electrodes. The multi-well plate 208 with electrodes is composed of a plate top 301 and a plate bottom 308.
The plate upper 301 is composed of a wall 306 having an upper base material 302, an outer wall 303, an upper surface 304 (see FIG. 4), and an inner surface 305. The plate upper 301 of the multi-well plate 208 with electrodes has a cylindrical through hole H, and the side surface of the through hole H is defined by the inner surface 305 of the wall 306. The plate top 301 is combined with the plate top 301 with a plate bottom 308 defining wells 307.

The plate bottom 308 has a lower substrate 309, a working electrode 400, a counter electrode 401, and a reference electrode 402 arranged on the lower substrate 309, and when combined with the plate upper 301, the through hole H One opening is closed, resulting in the definition of a cylindrical well 307 with an opening at the top.

The multi-well plate 208 with electrodes is configured as a multi-well plate having 96 wells, but the number of wells contained in the multi-well plate used in this apparatus is not limited to the above, and for example, 6 or 24. It may be 96 pieces, 384 pieces, 1536 pieces, 6144 pieces, 9600 pieces, or the like. The volume of each well is preferably in the range of 10 nL to 10 mL.

As shown in FIG. 4, the well 307 has a wall 306 with an inner surface 305, a working electrode 400, a counter electrode 401, and a reference electrode 402. As shown in FIG. 4, the inner surface 305 of the wall 306 defines a cylindrical volume. The sample solution is contained in this cylindrical portion. The counter electrode 401 extends at the bottom of this cylindrical volume into an annular region between the inner surface 305 and the circular working electrode 400. The shape of the well is cylindrical, but is not limited to the above, and may be, for example, a quadrangular prism or a honeycomb-shaped polygonal prism.

FIG. 5 is an enlarged plan view of the well and a view showing the back surface thereof. The working electrode 400, the counter electrode 401, and the reference electrode 402 are provided with via holes provided at the bottom of the plate in the well, so that the working electrode 400, the counter electrode 401, and the back surface wiring 501 corresponding to each of the reference electrode 402 are provided. It is connected to 502 and 503, respectively, and each wiring on the back surface has connection terminals 504, 505 and 506, and is connected to the measurement module 203 via the electrode terminal.

FIG. 6 is a functional block diagram of the data acquisition device 101. The data acquisition device 101 includes a control unit 601, an operation unit 602, an output unit 603, a training data measurement unit 604, and a training data transmission unit 605.
The control unit 601 is configured to include the processor 201 and controls the operation of the data acquisition device 101. The training data transmission unit 605 is a function realized by the communication module 206 controlled by causing the processor 201 to execute the program stored in the storage device 202.
Further, the training data transmission unit 605 contains the electrochemical characteristics of the test solution measured by the training data measurement unit 604, the dental caries in the patient's oral cavity corresponding to the test solution, and the data on the progress of periodontal disease. This is a function of transmitting training data including the above to the server device 102.

The training data measurement unit 604 includes a measurement module 203, and includes a test liquid storage unit 606 that stores a test liquid containing an electron source and a patient's fluid, and an electrode for measuring the electrochemical characteristics of the test liquid. It is a function of being connected to a multi-well plate 208 with electrodes having a plurality of parts 607 and measuring the electrochemical characteristics of the test solution while controlling the temperature of the test solution, the potential between the electrodes, and the like. The test liquid accommodating portion 606 is configured to include the well 307 of the multi-well plate 208 with an electrode, and the electrode portion 607 includes a working electrode 400, a counter electrode 401, a reference electrode 402, and wiring of the multi-well plate 208 with an electrode. It is configured to include 501 to 503 and connection terminals 504 to 506.

The electron source is a compound capable of assisting or promoting an electron transfer function of bacteria or the like, and is not particularly limited, but the substance serving as an electron source is preferably an organic compound, for example, glucose, acetic acid, lactic acid or the like. Can be mentioned.
The present invention has found that the electrochemical response to various electron sources differs depending on the type of bacteria, and the training data shows the electricity of a plurality of test solutions to which a plurality of electron sources are added to one solution. If it includes chemical properties, it can be predicted more accurately.

The electrochemical characteristics are the characteristics of the test solution measured by an amperometry method, a voltammetry method, or the like.

The present inventor has been diligent in research for electrochemically detecting the activity of bacteria causing caries and / or periodontal disease in the oral cavity. As a result, the bacterium transfers electrons to the outside of the cell (typically, an extracellular solid) in a state where caries and / or periodontal disease can occur, thereby obtaining energy. I know that.

Specifically, the present inventors observed redox-specifically stained sections of Streptococcus mutans cultured under acidic conditions (pH 4) and the same bacteria cultured under neutral conditions with a transmission electron microscope. However, it has been found that the surface of the cell wall and the inner cell membrane are specifically stained when cultured under acidic conditions.

It has also been confirmed that Capnocytophaga ochracea, which is also known as the causative agent of periodontal disease, specifically expresses an electron transfer enzyme in an anaerobic environment.
The above shows that the formation of biofilm, metabolism in biofilm, and acidification of biofilm and generation of electric current (electron transfer to extracellular solid) are closely related.

When caries and / or periodontal disease occurs or progresses, it is presumed that the above-mentioned bacteria transfer excess electrons to the extracellular solid to obtain energy. The above-mentioned movement of electrons is reflected in the electrochemical properties of the test solution including the electron source and the solution. In other words, there is a close relationship between the electrochemical properties of the test solution and the progress of caries and periodontal disease.

In addition, in this specification, caries and the progress state of periodontal disease are the presence or absence of caries and / or periodontal disease (also referred to as "caries, etc."), and caries. If any of the above occurs, the progress of the occurrence shall be included. This is the data provided as a result of the dentist treating the oral cavity of the patient (and healthy person) who provided the fluid.

Here, the electron source is a compound capable of assisting or promoting the electron transfer function of bacteria or the like, and is not particularly limited, but the substance serving as the electron source is preferably an organic compound, for example, glucose, acetic acid, lactic acid or the like. Can be mentioned.
The present invention has found that the electrochemical response to various electron sources differs depending on the type of bacteria, and the training data shows the electricity of a plurality of test solutions to which a plurality of electron sources are added to one solution. If it includes chemical properties, it can be predicted more accurately.

Further, the operation unit 602 is configured to include the input device 204 and accepts user operations. Further, the output unit 603 is configured to include the output device 205, and displays various information and data.

Since the data acquisition device 101 can sequentially and automatically measure the electrochemical characteristics of a large number of test solutions contained in the multi-well plate 208 with electrodes, a large amount of training data can be acquired more quickly. When the multi-well plate with electrodes is a 96-well plate, the same electrons are in the vertical column (1 column x 8 rows) among the wells arranged in a two-dimensional array in an arrangement of 8 vertical × 12 horizontal. When a predetermined amount of source is added in advance and a maximum of 12 types are arranged, a test solution combined with a maximum of 12 types of electron sources is simply dispensed into the horizontal row (12 columns x 1 row). Can be prepared. In this case, the data of up to 8 kinds of liquids × 12 kinds of electron sources can be measured more quickly.

The present inventor has found that different electron sources (for example, glucose and lactate) and different electrochemical responses of caries and bacteria that cause periodontal disease are one type. It is preferable to use the response when various electron sources are added to the liquid as training data in that more accurate prediction can be made.

The training data may include data of patients (caries and / or persons with periodontal disease), but healthy persons (caries and / or periodontal disease) may occur. It is preferable that the data of those who do not have the data are also included. By including the data of healthy subjects in the training data, the presence or absence of caries and periodontal disease can be predicted with higher accuracy by using the obtained classifier.

[Server device]
FIG. 7 is a hardware configuration diagram of the server device 102 according to the embodiment of the present invention.
The server device 102 includes a processor 701, a storage device 702, an input device 703, an output device 704, and a communication module 705. The processor 701, the storage device 702, the input device 703, the output device 704, and the communication module 705 are connected by the bus 706. The processor 701 controls the server device 102. The storage device 702 serves as a work area for the processor 701. Further, the storage device 702 is a non-temporary or temporary recording medium for storing various programs and data. Examples of the storage device 702 include a ROM (Read Only Memory), a RAM (Random Access Memory), an HDD (Hard Disk Drive), a flash memory, and an SSD (Solid State Drive). The input device 703 inputs data. Examples of the input device 703 include a keyboard, a mouse, a touch panel, a numeric keypad, a scanner, and the like. The output device 704 outputs data. The output device 704 includes, for example, a display and a printer. The communication module 705 connects to a computer network including the Internet and transmits / receives data.

FIG. 8 is a functional block diagram of the server device 102. The server device 102 includes a control unit 801, an operation unit 802, an output unit 803, a training data reception unit 804, a classifier generation unit 805, a test data reception unit 806, a prediction unit 807, and a result transmission unit 808. And have.
The control unit 801 is configured to include the processor 701 and controls the operation of the server device 102. The training data receiving unit 804, the test data receiving unit 806, and the result transmitting unit 808 are functions realized by the controlled communication module 705 by causing the processor 701 to execute the program stored in the storage device 702. Further, the classifier generation unit 805 and the prediction unit 807 are functions realized by causing the processor 701 to execute the program stored in the storage device 702.
Further, the operation unit 802 is configured to include the input device 703 and accepts user operations. Further, the output unit 803 is configured to include an output device 704, and displays various information and data.

The training data receiving unit 804 receives training data including the electrochemical characteristics of the test solution including the electron source and the patient's fluid, the caries of the patient, and the progress of periodontal disease from the data acquisition device. And store it in the storage device 702.
The training data is obtained for each fluid obtained from the patient, the caries of the patient when the fluid was obtained, the progress of periodontal disease, and the test fluid containing the fluid and the electron source. It consists of a plurality of data having a structure in which the electrochemical characteristics of the above are combined.

The classifier generator 805 uses the training data to perform learning based on the feature amount obtained from the electrochemical characteristics and the progress state, and uses the feature amount as an explanatory variable and the progress state as an objective variable. Generate a classifier. The method for generating the classifier is not particularly limited, but known machine learning algorithms such as logistic regression, random forest, decision jungle, k-nearest neighbor method, and neural network can be used.

As a method of obtaining the feature quantity from the electrochemical characteristics, it is necessary to set a predetermined quantity in the measured value of the electrochemical characteristics (for example, the relationship between the time and the current response obtained by the amperometry method) as the feature quantity in advance. It may be a method of defining and extracting the above amount contained in the training data. Such quantities include, for example, the current value measured by the amperometry method, the change in the amount of current measured by the voltammetry method, and the value calculated from these values (eg, the differential value of the current value, etc.). Second-order differential value, average value, etc.) and the like.
Further, the feature amount generated by the multi-layer neural network or the like may be used by using the above feature amount or the actually measured value of the electrochemical characteristic itself as an input layer.

The test data receiving unit 806 receives test data including the electrochemical characteristics of the sample including the electron source and the fluid of the subject from the client device via a computer network including the Internet. The test data includes the electrochemical properties of the sample, including the subject's fluid and electron source.

The prediction unit 807 applies the test data to the classifier to generate a result of prediction of the caries of the subject and the progress state of periodontal disease. The result transmission unit 808 returns the result of the prediction to the client device and prompts the client device to display the result.

(Operation of server device)
The control unit 801 of the server device operates as follows according to the program.
FIG. 9 is an operation flow of the control unit 801 that operates according to the above program.

The control unit 801 controls the training data receiving unit 804 to receive the training data from the data acquisition device 101 (step S901). This training data includes the electrochemical properties of the test solution containing the electron source and the patient's fluid, the caries of the patient, and the progress of periodontal disease.
Next, the control unit 801 controls the classifier generation unit 805 to use the feature amount as an explanatory variable and the progress state as the objective variable based on the feature amount obtained from the electrochemical characteristics and the progress status. Generate a classifier (step S902).

Next, the control unit 801 controls the test data receiving unit 806 to receive the test data from the client device 103 (step S903). This test data includes the electrochemical properties of the specimen containing the same electron source and the subject's fluid as contained in the test fluid used in the training data.

Next, the control unit 801 controls the prediction unit 807, applies test data to the classifier, predicts the progress of caries and periodontal disease of the subject, and generates a prediction result. (Step S904).

Next, the control unit 801 controls the result transmission unit 808 to transmit the result of the prediction to the client device 103, and prompts the client device to display the result (step S905).
The above is executed by the control unit 801 including the processor 701.

According to this server device, the subject who uses the client device in a remote place, for example, lives in an area where it is difficult to see a medical institution, or is elderly, so the subject visits a medical institution. Even when it is difficult, it is possible to receive test data related to the subject's fluid and provide information for simply diagnosing the oral health condition of the subject.

[Client device]
FIG. 10 is a hardware configuration diagram of the client device 103 according to the embodiment of the present invention. The client device 103 includes a processor 1001, a storage device 1002, a measurement module 1003, an input device 1004, an output device 1005, and a communication module 1006. The processor 1001, the storage device 1002, the measurement module 1003, the input device 1004, the output device 1005, and the communication module 1006 are connected by the bus 1007. The processor 1001 controls the client device 103.

The storage device 1002 serves as a work area for the processor 1001. Further, the storage device 1002 is a non-temporary or temporary recording medium for storing various programs and data. Examples of the storage device 1002 include a ROM (Read Only Memory), a RAM (Random Access Memory), an HDD (Hard Disk Drive), a flash memory, and an SSD (Solid State Drive). The input device 1004 inputs data. The input device 1004 includes, for example, a keyboard, a mouse, a touch panel, a numeric keypad, a scanner, and the like. The output device 1005 outputs data. The output device 1005 includes, for example, a display and a printer. The communication module 1006 connects to a computer network including the Internet and transmits / receives data.

The client device 103 has a measurement module 1003. The measurement module 1003 is configured to accommodate a sample containing an electron source and a subject's fluid and to be connectable to a biosensor 1008 having at least two electrodes arranged in contact with the sample. The measurement module 1003 is a connection terminal for electrically connecting the electrodes of the biosensor 1008 to the client device 103 in a detachable manner; a temperature control block for controlling the temperature of the sample at the time of measurement; the above two. It is a unit containing a potentio / galvanostat whose potential between electrodes can be adjusted.

FIG. 11 is an exploded perspective view of the biosensor 1008. The biosensor 1008 is arranged at one end of the upper substrate 1101, the lower substrate 1102, the first electrode 1103 arranged on the lower substrate 1102, and the first electrode 1103, and the upper substrate 1101 and the lower substrate 1102 are laminated. In this state, the first connection terminal 1104 is exposed to the outside (because the upper board 1101 is shorter than the lower board 1102), and the second electrode 1105 arranged on the lower board 1102. It has a second connection terminal 1106 arranged at one end of the second electrode 1105. Electrochemical measurement is performed by connecting these connection terminals to the connection terminals of the measurement module 1003.

In the biosensor 1008, the upper substrate 1101 has a through hole in a part of the main surface, and the region CELL is partitioned by the surface of the through hole and the lower substrate 1102. The first electrode 1103 and the second electrode 1105 are exposed on the lower substrate 1102 side of the region CELL (bottom surface of the region CELL), and the biosensor 1008 is formed by applying and dropping a sample to the region CELL. The sample is captured in the cell, and the electrochemical characteristics of the sample can be measured.

If a solid electrolyte containing a predetermined electron source is arranged in advance in the region CELL, it is preferable that a sample can be prepared by applying the subject's solution to the solid electrolyte containing the electron source.

The electron source is the same as the electron source used in the test solution included in the training data. Since the electron source contained in the sample is the same as the electron source contained in the test solution, it is easy to obtain highly accurate prediction results.
If the training data includes the electrochemical characteristics of a plurality of test solutions prepared by adding a plurality of types of electron sources for each type of solution, the test data also includes a plurality of test solutions for one type of solution. It preferably contains the electrochemical properties of a plurality of specimens prepared by adding different types of electron sources. By doing so, it is easy to obtain the results of higher forecasts of the system.

When preparing a plurality of samples in combination with a plurality of types of electron sources for one type of fluid obtained from the same subject, a biosensor having a solid electrolyte containing a different electron source in the above region CELL is used as an electron. Multiple preparations may be made according to the type of source (biosensor set). The distillates collected from the same subject at the same time may be applied to each of these plurality of biosensors.

The first electrode 1103 and the second electrode 1105 in the biosensor 1008 can be formed by using, for example, a known photolithography technique, a known printing technique, or the like.
Although the biosensor 1008 has two electrodes, it is not limited to the above and may have three or more electrodes.

FIG. 12 is a functional block diagram of the client device 103. The client device 103 includes a control unit 1201, an operation unit 1202, an output unit 1203, a sample measurement unit 1204, a test data transmission unit 1205, and a result reception unit 1206.
The control unit 1201 includes a processor 1001 and controls the operation of the client device 103. The test data transmission unit 1205 and the result reception unit 1206 are functions realized by the controlled communication module 1006 by causing the processor 1001 to execute the program stored in the storage device 1002. The test data transmission unit 1205 has a function of transmitting test data including the electrochemical characteristics of the sample measured by the sample measurement unit 1204 to the server device. Further, the result receiving unit 1206 receives the result of the prediction regarding the progress state of the periodontal disease and the caries in the oral cavity of the subject transmitted from the server device 102, and outputs the result via the output unit 1203. Has a function.

The sample measurement unit 1204 includes a measurement module 1003, and has a sample capture unit 1207 for capturing the fluid of the subject and holding the sample, and an electrode having two electrodes arranged so as to be in contact with the sample. It is connected to a biosensor 1008 having a unit 1208, and has a function of measuring the electrochemical characteristics of the sample while controlling the temperature of the sample, the potential between the electrodes, and the like.

The sample capture unit 1207 is configured to include a region CELL partitioned by a through hole of the upper substrate 1101 of the biosensor 1008 and the surface of the lower substrate 1102, and captures a sample containing an electron source and a subject's fluid. It is a function. Further, the electrode portion 1208 is configured to include a first electrode 1103 of the biosensor 1008, a first connection terminal 1104, a second electrode 1105, and a second connection terminal 1106, and is connected to the connection terminal of the measurement module 1003. This is a function that enables electrochemical measurement of a sample.

Further, the operation unit 1202 is configured to include the input device 1004 and accepts user operations. Further, the output unit 1203 is configured to include the output device 1005, and displays various information and data.

(Operation of client device)
The control unit 1201 of the client device operates as follows according to the program. FIG. 13 is an operation flow of the control unit 1201 that operates according to the above program.

The control unit 1201 controls the test data transmission unit 1205 to control the test data including the electrochemical characteristics of the sample including the electron source and the fluid of the subject, which is obtained by controlling the sample measurement unit 1204, and is a server device. It is transmitted to 102 (step S1301).

Next, the control unit 1201 receives the result of the prediction of the caries in the oral cavity of the subject and the progress state of periodontal disease obtained by applying the test data to the classifier by the server device 102, and the result receiving unit 1206. Is controlled and received from the server device 102 (step S1302).
In addition, this classifier uses the electrochemical characteristics of the test solution including the electron source contained in the sample and the patient's fluid acquired by the data acquisition device 101, and the dental caries and periodontal disease of the patient. Learning is performed based on the feature amount and the progress state obtained from the electrochemical characteristics in the training data including the progress state, and the feature amount generated by the server device 102 is used as an explanatory variable and the progress state is used as an objective variable. It is a classifier.

Next, the control unit 1201 controls the output unit 1203 to display the result of the prediction received from the server device 102 (step S1303).
The above is executed by the control unit 1201 including the processor 1001.

(Operation flow of the entire system)
Next, the overall operation flow of the system 100 will be described with reference to FIGS. 1 and 14.
FIG. 14 is an example of the operation flow of the system 100.
First, the training data measuring unit 604 of the data acquisition device 101 measures the electrochemical characteristics of the test solution (104) including the electron source and the patient's fluid. In addition, dental caries and periodontal disease (hereinafter referred to as "dental caries, etc.") in the oral cavity of the patient who provided the fluid by user operation and access to the database, etc. The input of the diagnosis result (105) by the dentist in the progress state of (referred to as) is received via the operation unit 802 of the data acquisition device 101. The test solution may simply contain a solution provided by a healthy person. That is, the training data (106) may include data on the electrochemical characteristics of the test solution containing the fluid of a healthy person and the progress state of caries and the like (no caries and the like).

Next, the data acquisition device 101 generates training data (106) used for generating a classifier described later from the electrochemical characteristics obtained by measurement and the received diagnostic result (step S1401). ..
The training data (106) includes a plurality of vector data for each sample in which electrochemical characteristics and diagnostic results are combined.

Next, the data acquisition device 101 transmits the training data (106) to the server device 102, and the server device 102 receives the training data (106) from the data acquisition device 101 (step S1402). The server device 102 learns based on the feature amount obtained from the electrochemical characteristics included in the received training data (106) and the progress state, uses the feature amount as an explanatory variable, and aims at the progress state. A classifier (107) as a variable is generated (step S1403).

Next, in order to acquire the test data (108), the client device 103 obtains the electrochemical characteristics of the sample (109) containing the same electron source used in the data acquisition device 101 and the fluid of the subject. Measure. The client device 103 generates test data (108) from the measured electrochemical characteristics (step S1404). The test data (108) is vector data including electrochemical characteristics summarized for each sample.

Next, the client device 103 transmits the test data (108) to the server device 102, and the server device 102 receives the test data (108) from the client device 103 (step S1405). The server device 102 applies the test data (108) to the classifier (107) (step S1406), and as an output of the classifier (107), it relates to the caries in the oral cavity of the subject and the progress state of periodontal disease. Generate the prediction result (110).

Next, the server device 102 transmits the prediction result (110) to the client device 103, and the client device 103 receives the prediction result (110) from the server device 102 (step S1407). The server device 102 prompts the client device 103 to display the result of the prediction (110), and the client device 103 controls the output unit 1203 to display the prediction result.

According to the system according to the embodiment of the present invention, it is possible to obtain the prediction result regarding the progress state of caries and the like only by a non-invasive and simple procedure in which the subject provides the fluid. In addition, since the prediction results are provided via a computer network including the Internet, prediction results related to the progress of caries and periodontal disease can be obtained even for subjects in areas or environments where it is difficult to go to medical institutions. Can be provided.

The present invention is not limited to the above embodiment, and includes various modifications. For example, the above-described embodiment has been described in detail in order to explain the present invention in an easy-to-understand manner, and is not necessarily limited to those having all the described configurations. Further, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. Further, it is possible to add / delete / replace a part of the configuration of each embodiment with another configuration.

100: System 101: Data acquisition device 102: Server device 103: Client device 201: Processor 202: Storage device 203: Measurement module 204: Input device 205: Output device 206: Communication module 207: Bus 208: Multi-well plate with electrode 301 : Plate upper 302: Upper base material 303: Outer wall 304: Upper surface 305: Inner surface 306: Wall 307: Well 308: Plate bottom 309: Lower substrate 400: Action electrode 401: Counter electrode 402: Reference electrode 501: Wiring 502: Wiring 503: Wiring 504: Connection terminal 505: Connection terminal 506: Connection terminal 601: Control unit 602: Operation unit 603: Output unit 604: Training data measurement unit 605: Training data transmission unit 606: Test solution storage unit 607: Electrode unit 701 : Processor 702: Storage device 703: Input device 704: Output device 705: Communication module 706: Bus 801: Control unit 802: Operation unit 803: Output unit 804: Training data reception unit 805: Classifier generator 806: Test data reception Unit 807: Prediction unit 808: Result transmission unit 1001: Processor 1002: Storage device 1003: Measurement module 1004: Input device 1005: Output device 1006: Communication module 1007: Bus 1008: Biosensor 1101: Upper board 1102: Lower board 1103: 1st electrode 1104: 1st connection terminal 1105: 2nd electrode 1106: 2nd connection terminal 1201: Control unit 1202: Operation unit 1203: Output unit 1204: Specimen measurement unit 1205: Test data transmission unit 1206: Result reception unit 1207: Specimen capture unit 1208: Electrode unit

Claims (12)

It has a server device, a client device, and a data acquisition device.
The client device and the data acquisition device are configured to be able to communicate with the server device via a computer network including the Internet.
The data acquisition device acquires the electrochemical characteristics of the test solution including the electron source and the patient's fluid, and the training data including the caries of the patient and the progress state of periodontal disease. The training data is transmitted to the server device,
The server device performs learning based on the feature amount obtained from the electrochemical characteristics included in the training data and the progress state, and uses the feature amount as an explanatory variable and the progress state as an objective variable. To generate
The client device measures the electrochemical characteristics of a sample containing the electron source and the fluid of the subject, acquires test data, and transmits the test data to the server device.
The server device further applies the test data to the classifier, predicts the progress of caries and periodontal disease of the subject, and transmits the prediction result to the client device. A system that encourages client devices to display. A training data receiving unit that receives training data including the electrochemical characteristics of the test solution including the electron source and the patient's fluid, the caries of the patient, and the progress of periodontal disease from the data acquisition device.
A classifier generator that generates a classifier that performs learning based on the feature amount obtained from the electrochemical characteristics and the progress state, uses the feature amount as an explanatory variable, and uses the progress state as an objective variable.
A test data receiving unit that receives test data including the electrochemical characteristics of the sample including the electron source and the subject's fluid from the client device, and a test data receiving unit.
A prediction unit that applies the test data to the classifier to predict the progress of caries and periodontal disease of the subject, and
A server device having a result transmission unit that returns the result of the prediction to the client device and prompts the client device to display the result. A sample measuring unit that measures the electrochemical properties of a sample containing an electron source and the subject's fluid,
A test data transmitter that transmits test data including the electrochemical characteristics to the server device,
A result receiving unit that receives from the server device the results of prediction of the caries of the subject and the progress state of periodontal disease obtained by applying the test data to the classifier in the server device.
A client device having an output unit for displaying the result.
The classifier is a training data including the electrochemical characteristics of the test solution including the electron source and the patient's fluid, and the caries of the patient and the progress of periodontal disease acquired by the data acquisition device. With a classifier that performs learning based on the feature amount obtained from the electrochemical characteristics and the progress state in the above, and uses the feature amount generated by the server device as an explanatory variable and the progress state as an objective variable. There is a client device. The third aspect of claim 3, wherein the sample measuring unit is connected to the electrode of a biosensor having at least two electrodes arranged so as to be in contact with the sample, and measures the electrochemical property of the sample obtained from the electrode. Client device. The biosensor further has a sample capture unit for capturing the sample.
The client device according to claim 4, wherein the electron source is arranged in advance in the sample capturing unit, and a sample is generated by the electron source and the liquid when it comes into contact with the liquid. The client device according to claim 5, wherein the sample capture unit is made of a solid electrolyte containing the electron source. A biosensor used to measure the electrochemical properties of a sample, including the electron source and the subject's fluid.
With at least two electrodes arranged to contact the specimen,
A biosensor having a sample capture unit in which the electron source is arranged in advance and the sample is generated by the electron source and the solution in contact with the solution. A biosensor set having the biosensor according to claim 7, wherein a plurality of the electron sources are arranged in advance. Each of the plurality of test solution storage units arranged in a two-dimensional array in which the test liquid containing the electron source and the patient's fluid is stored is arranged in the test liquid storage unit so as to be in contact with the test liquid. A training data measuring unit that is electrically connected to each of the electrode portions composed of at least two electrodes and measures the electrochemical characteristics of the test solution obtained from the electrode portions.
An operation unit that accepts inputs of the patient's caries and periodontal disease progress status,
A data acquisition device having a training data transmission unit that transmits training data including the electrochemical characteristics and the progress state to a server device. The data acquisition device according to claim 9, wherein the plurality of test liquid accommodating portions and the electrode portions form a multi-well plate with electrodes. On the computer
The stage of receiving training data from the data acquisition device including the electrochemical characteristics of the test solution including the electron source and the patient's fluid, the caries of the patient, and the progress of periodontal disease.
A step of performing learning based on the feature amount obtained from the electrochemical characteristics and the progress state, and generating a classifier using the feature amount as an explanatory variable and the progress state as an objective variable.
The stage of receiving test data including the electrochemical properties of the sample including the electron source and the subject's fluid from the client device, and
A step of applying the test data to the classifier to predict the progress of caries and periodontal disease of the subject, and
A program that executes a step of transmitting the result of the prediction to the client device and prompting the client device to display the result. On the computer
The stage of transmitting test data including the electrochemical properties of the sample including the electron source and the subject's fluid to the server device, and
The stage of receiving the result of the prediction of the caries of the subject and the progress state of periodontal disease obtained by applying the test data to the classifier by the server device from the server device, and
A program that executes the stage of displaying the result of the prediction and the step of displaying the result of the prediction.
The classifier is a training data including the electrochemical characteristics of the test solution including the electron source and the patient's fluid, and the caries of the patient and the progress of periodontal disease acquired by the data acquisition device. With a classifier that performs learning based on the feature amount obtained from the electrochemical characteristics and the progress state in the above, and uses the feature amount generated by the server device as an explanatory variable and the progress state as an objective variable. There is a program.

Images