Saliva-based blood sugar prediction system and method using artificial intelligence deep learning technique

Abstract

To provide a saliva-based blood sugar prediction system and method using artificial intelligence deep learning technique.SOLUTION: In a postprandial blood sugar level prediction system, a processor 222 of a server includes: a learning modeling unit that learns a sugar change inference model so as to infer a difference in the pattern between a blood sugar level change and a saliva sugar change due to a meal while considering a physical index, and learns a postprandial blood sugar level inference model so as to infer the correlation between the postprandial saliva sugar and the postprandial blood sugar level in consideration of the pattern difference; a subject information acquisition unit that acquires the physical index and postprandial saliva sugar of the subject; a pattern difference estimation unit that uses the sugar change inference model, with the physical index of the subject as an input parameter, to estimate the pattern difference in the subject; and a postprandial blood sugar level prediction unit that uses the postprandial blood sugar level inference model, with the postprandial saliva sugar of the subject and the estimated pattern difference as input parameters, to predict a postprandial blood sugar level of the subject.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The following embodiments relate to a system and method for predicting blood glucose levels from salivary sugars. [Background technology]

[0002] For diagnosing diabetes, blood glucose, which is the concentration of glucose in the blood, and urinary glucose are generally used as indicators.

[0003] However, in urine glucose tests, people with impaired glucose tolerance, who are on the borderline between healthy people and diabetes patients, are classified as diabetes patients, which reduces the accuracy of diabetes diagnosis.

[0004] Therefore, blood glucose levels are mainly used to ensure the accuracy of diabetes diagnosis. However, measuring blood glucose requires blood sampling, which has the limitation that it can only be done by invasive methods, and therefore has the drawback of being time-consuming and painful to draw blood.

[0005] Furthermore, current blood glucose tests have the limitation that they must be performed in a fasting state before a meal, which means that the tester must maintain an empty stomach for the test.

[0006] Therefore, there is a need to propose a technology that can overcome the limitations and restrictions of blood glucose testing as described above while resolving the drawbacks. Summary of the Invention [Problem to be solved by the invention]

[0007] One embodiment proposes a system and method that can predict postprandial blood glucose levels from postprandial salivary sugar and diagnose diabetes from postprandial blood glucose levels, overcoming the limitation of blood glucose measurement that blood can only be collected by invasive methods, thereby resolving drawbacks such as the effort and pain involved in blood collection, while also resolving the constraint and drawback of having to maintain an empty stomach for the test.

[0008] More specifically, one embodiment proposes a system and method for predicting postprandial blood glucose levels from postprandial salivary sugars by utilizing a sugar change inference model that infers pattern differences in blood glucose level changes and salivary sugar changes due to meals while taking into account physical indicators, and a postprandial blood glucose level inference model that infers the correlation between postprandial salivary sugars and postprandial blood glucose levels while taking into account pattern differences.

[0009] In this regard, one embodiment proposes a system and method for improving the accuracy of inference and prediction by personalizing and training a glucose change inference model and a postprandial blood glucose level inference model to suit the subject.

[0010] The technical problem to be solved by the present invention should not be limited to the above, but may be variously expanded within the scope of the technical idea and scope of the present invention. [Means for solving the problem]

[0011] According to one embodiment, the postprandial blood glucose prediction system includes a learning modeling unit that trains a glucose change inference model to infer pattern differences in blood glucose changes and salivary sugar changes due to meals while taking into account physical indices, and trains a postprandial blood glucose inference model to infer correlations between postprandial salivary sugar and postprandial blood glucose while taking into account the pattern differences; The method may include a subject information acquisition unit that acquires the subject's information, a pattern difference estimation unit that uses the subject's physical index as input parameters and estimates the subject's pattern difference using the glucose change inference model, and a postprandial blood glucose level prediction unit that uses the subject's postprandial saliva sugar and the estimated pattern difference as input parameters and predicts the subject's postprandial blood glucose level using the postprandial blood glucose level inference model.

[0012] According to one embodiment, the pattern differences may be characterized by differences in time delay, maximum values, and rates of change between blood glucose and salivary sugar changes due to the meal.

[0013] According to another aspect, the learning modeling unit may be characterized by personalizing and learning the glucose change inference model and the postprandial blood glucose level inference model to suit the subject based on personal data including the subject's fasting blood glucose level.

[0014] According to another aspect, the learning modeling unit may construct the glucose change inference model by learning experimental data that infers the pattern difference while taking into account the physical indicators using an artificial intelligence deep learning technique, and may construct the postprandial glucose level inference model by learning experimental data that infers the correlation between the postprandial saliva glucose and the postprandial glucose level while taking into account the pattern difference using an artificial intelligence deep learning technique.

[0015] According to another embodiment, the physical indicators may include gender, age, weight, body mass index (BMI), and waist circumference.

[0016] According to another aspect, the learning modeling unit may be characterized by inferring correlations between the physical indicators and HOMA-IR and HOMA β-cell, and training the glucose change inference model to infer correlations between the pattern difference and the HOMA-IR and HOMA β-cell.

[0017] According to yet another aspect, the learning modeling unit may be characterized by training the glucose change inference model to infer a correlation between the insulin resistance and insulin secretion capacity calculated by the HOMA-IR and HOMAβ-cell and the pattern difference.

[0018] According to one embodiment, a method for predicting postprandial blood glucose levels may include the steps of acquiring a subject's physical indicators and postprandial salivary sugar, estimating a pattern difference for the subject using a sugar change inference model (the sugar change inference model is trained to infer a pattern difference between blood glucose changes and salivary sugar changes due to meals while taking into account the physical indicators) using the subject's physical indicators as input parameters, and predicting the subject's postprandial blood glucose level using a postprandial blood glucose level inference model (the postprandial blood glucose level inference model is trained to infer a correlation between postprandial salivary sugar and postprandial blood glucose while taking into account the pattern difference) using the subject's postprandial salivary sugar and the estimated pattern difference as input parameters.

[0019] According to one embodiment, the pattern differences may be characterized by differences in time delay, maximum values, and rates of change between blood glucose and salivary sugar changes due to the meal.

[0020] According to another aspect, the glucose change inference model and the postprandial blood glucose level inference model may be characterized in that they are personalized and trained for the subject based on personal data including the subject's fasting blood glucose level.

[0021] According to another aspect, the method for predicting postprandial blood glucose level includes the steps of: constructing the glucose change inference model by learning experimental data that infers the pattern difference while taking into account the physical index using an artificial intelligence deep learning technique; and P.3 The method may further include a step of constructing the postprandial blood glucose level inference model by learning experimental data using artificial intelligence deep learning technology to infer the correlation between blood glucose levels while taking into account the pattern differences.

[0022] According to another embodiment, the physical indicators may include gender, age, weight, body mass index (BMI), and waist circumference.

[0023] According to another aspect, the glucose change inference model may be characterized by being trained to infer correlations between the physical indicators and HOMA-IR and HOMA β-cell, and to infer correlations between the pattern difference and the HOMA-IR and HOMA β-cell.

[0024] According to yet another aspect, the glucose change inference model may be characterized by being trained to infer a correlation between the pattern difference and the insulin resistance and insulin secretory capacity calculated by the HOMA-IR and HOMA β-cell.

[0025] According to one embodiment, there is provided a computer-readable recording medium having recorded thereon a computer program for causing a computer device to execute a postprandial blood glucose prediction method, the postprandial blood glucose prediction method may include the steps of acquiring a subject's physical indicators and postprandial salivary sugar, estimating a pattern difference of the subject using a sugar change inference model (the sugar change inference model is trained to infer a pattern difference between blood glucose changes and salivary sugar changes due to meals while taking into account the physical indicators) using the subject's postprandial salivary sugar and the estimated pattern difference as input parameters and predicting the subject's postprandial blood glucose using a postprandial blood glucose inference model (the postprandial blood glucose inference model is trained to infer a correlation between postprandial salivary sugar and postprandial blood glucose while taking into account the pattern difference). [Effects of the Invention]

[0026] One embodiment proposes a system and method for predicting postprandial blood glucose levels from postprandial saliva sugars and diagnosing diabetes from postprandial blood glucose levels, thereby overcoming the limitation of blood glucose measurement that blood can only be collected by invasive methods, thereby resolving drawbacks such as the effort and pain involved in blood collection, while also resolving the constraint and drawback of having to maintain an empty stomach for the test.

[0027] More specifically, one embodiment proposes a system and method for predicting postprandial blood glucose levels from postprandial salivary sugar by utilizing a sugar change inference model that infers pattern differences in blood glucose level changes and salivary sugar changes due to meals while taking into account physical indicators, and a postprandial blood glucose level inference model that infers the correlation between postprandial salivary sugar and postprandial blood glucose levels while taking into account pattern differences.

[0028] In this case, one embodiment can propose a system and method that can improve the accuracy of inference and prediction by personalizing and training a glucose change inference model and a postprandial blood glucose level inference model to suit the subject.

[0029] The effects of the present invention should not be limited to those described above, but may be variously expanded within the scope of the technical idea and scope of the present invention. [Brief explanation of the drawings]

[0030] [Figure 1] FIG. 1 illustrates an example diabetes diagnostic environment, according to one embodiment. [Figure 2] FIG. 2 is a block diagram illustrating the internal configuration of an electronic device and a server according to an embodiment. [Figure 3] 1 is a block diagram illustrating an example of a postprandial blood glucose prediction system according to one embodiment. [Figure 4] 1 is a flowchart illustrating a method for predicting postprandial blood glucose levels in one embodiment. [Figure 5] FIG. 1 is a diagram for explaining a glucose change inference model used in a postprandial blood glucose level prediction method in one embodiment. [Figure 6] FIG. 1 is a diagram for explaining a glucose change inference model used in a postprandial blood glucose level prediction method in one embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0031] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, the present invention should not be limited to the embodiments. The same reference numerals in the drawings denote the same elements.

[0032] Furthermore, the terminology used herein is intended to appropriately describe preferred embodiments of the present invention, and may vary depending on the audience, operator's intentions, or the practices of the field to which the present invention pertains. Therefore, these terms should be defined based on the overall content of this specification. For example, the singular forms used herein also include the plural forms unless otherwise indicated by the context. Furthermore, the terms "comprise" and / or "comprising" used herein do not exclude the presence or addition of one or more other components, steps, operations, and / or elements to which a reference is made. Furthermore, although terms such as "first" and "second" are used herein to indicate various regions, directions, shapes, etc., these regions, directions, and shapes should not be construed as being limited by such terms. These terms are used to distinguish a given region, direction, or shape from other regions, directions, or shapes. Therefore, a portion referred to as a "first portion" in one embodiment may be referred to as a "second portion" in another embodiment.

[0033] It should also be understood that the various embodiments of the present invention, although different from one another, are not necessarily mutually exclusive. For example, a particular shape, structure, and characteristic described herein may be implemented in other embodiments without departing from the spirit and scope of the present invention. It should also be understood that the location, arrangement, or configuration of individual components within each category of presented embodiments may be changed without departing from the spirit and scope of the present invention.

[0034] In the following embodiments, a system and method for predicting postprandial blood glucose levels that can predict postprandial blood glucose levels from postprandial saliva sugar and diagnose diabetes from postprandial blood glucose levels will be described.

[0035] The postprandial blood glucose level prediction method may be executed by at least one computer device that realizes a server or electronic device (for example, a measurement device or a user terminal connected to a measurement device by communication) described below. As a result, at least one computer device included in the server or electronic device described below may constitute a postprandial blood glucose level prediction system that executes the postprandial blood glucose level prediction method. The computer device that realizes the postprandial blood glucose level prediction system may execute the postprandial blood glucose level prediction method according to the embodiment under the control of the executed computer program. The above-mentioned computer program may be recorded on a computer-readable recording medium in combination with the computer device to cause the computer device to execute the postprandial blood glucose level prediction method. The computer program described here may be in the form of an independent program package, or the form of an independent program package may be pre-installed on the computer device and integrated into an operating system or other program packages. It may be in a form that is linked to a cage.

[0036] Hereinafter, salivary sugar refers to glucose contained in saliva, and blood glucose level refers to the concentration of glucose contained in blood. Furthermore, hereinafter, obtaining salivary sugar means obtaining information about salivary sugar (e.g., salivary sugar concentration, index, numerical value, etc.), and predicting blood glucose means predicting information about blood glucose (e.g., blood glucose concentration, index, numerical value, etc.).

[0037] Fig. 1 is a diagram showing an example of a diabetes diagnosis environment in one embodiment. The diabetes diagnosis environment in Fig. 1 shows an example including multiple electronic devices 110, 120, 130, 140, and 150, a server 160, and a network 170. Fig. 1 is merely an example for explaining the present invention, and the number of electronic devices and the number of servers should not be limited to those shown in Fig. 1.

[0038] The multiple electronic devices 110, 120, 130, 140, and 150 are terminals such as smartphones, mobile phones, tablet PCs, navigation systems, computers, notebook PCs, digital broadcasting terminals, PDAs (Personal Digital Assistants), and PMPs (Portable Multimedia Players) that are respectively owned by the subject of the sample, and each of the multiple electronic devices 110, 120, 130, 140, and 150 may be equipped with a corresponding collection device 111, biosensor 112, and measurement device 113.

[0039] That is, each sample subject may possess an electronic device 110, a collection device 111, a biosensor 112, and a measurement device 113, and the biosensor 112 and the measurement device 113 may be connected to each of the multiple electronic devices 110, 120, 130, 140, 150 through dedicated applications installed on each of the multiple electronic devices 110, 120, 130, 140, 150.

[0040] The collection device 111 is a device for collecting a specimen (e.g., saliva) and may include a specimen collection section, a filter, and a compression tube. When the specimen is saliva, the collection device 111 may remove interfering substances from the collected saliva using a filter in the compression tube, and provide the saliva from which the interfering substances have been removed to the outside, i.e., to the biosensor 112.

[0041] Biosensor 112 is a component that performs the function of sensing the information to be measured (e.g., glucose) from the sample (e.g., saliva) collected by collection device 111 when the sample is inserted into collection device 111 and from which interfering substances have been removed. It may insert a biosensor strip into measurement device 113, recognize the sample based on a sample recognition signal received from measurement device 113, and provide measurement device 113 with a response signal for the glucose contained in the sample via a sample measurement signal applied separately from the sample recognition signal.

[0042] The measuring device 113 recognizes the insertion of the biosensor 112 and provides a sample recognition signal to the biosensor 112 to determine whether a sample has come into contact with the biosensor 112, and if it determines that a sample has come into contact, it provides a sample measurement signal and receives a response signal of the sample to be measured (a response signal to glucose in saliva (salivary sugar)), thereby measuring and displaying the response signal of the sample (a response signal to salivary sugar, for example, the concentration or index value of salivary sugar).

[0043] The measurement device 113 may also include a communication module for communicating with an electronic device (e.g., the first electronic device 110) or a server 160 via a network 170. In such a case, the measurement device 113 may receive a response signal of the analyte (a response signal to salivary sugars, e.g., salivary sugars). By transmitting the response signal (response signal to salivary sugar, for example, the concentration or index value of salivary sugar, etc.) to the electronic device 110 or the server 160, the electronic device 110 or the server 160 can predict the postprandial blood glucose level of the sample subject (hereinafter referred to as the subject) based on the response signal of the sample (response signal to salivary sugar, for example, the concentration or index value of salivary sugar, etc.), and diagnose diabetes in the subject based on the postprandial blood glucose level.

[0044] However, this is not limiting, and the measuring device 113 itself can predict the subject's postprandial blood glucose level based on the response signal of the sample (a response signal to salivary sugar, for example, the concentration or index value of salivary sugar), and diagnose the subject's diabetes based on the postprandial blood glucose level. In such a case, the measuring device 113 does not need to include a communication module.

[0045] The communication method is not limited, and may include not only communication methods using communication networks (for example, a mobile communication network, a wired Internet, a wireless Internet, and a broadcast network) that can be included in network 170, but also short-range wireless communication between devices. For example, network 170 may include any one or more of networks such as a personal area network (PAN), a local area network (LAN), a campus area network (CAN), a metropolitan area network (MAN), a wide area network (WAN), a broadband network (BBN), and the Internet. Furthermore, network 170 may include any one or more of network topologies including, but not limited to, a bus network, a star network, a ring network, a mesh network, a star-bus network, a tree or hierarchical network, etc.

[0046] The server 160 may be realized as a computer device or multiple computer devices that communicate with multiple electronic devices 110, 120, 130, 140, 150 via a network 170, receives response signals of samples measured from measuring devices 113 corresponding to each of the multiple electronic devices 110, 120, 130, 140, 150 or multiple electronic devices 110, 120, 130, 140, 150 (response signals to salivary sugars, for example, salivary sugar concentrations or index values), and provides a postprandial blood glucose prediction service that predicts a subject's postprandial blood glucose level and diagnoses diabetes based on the received response signals of the samples (response signals to salivary sugars, for example, salivary sugar concentrations or index values).

[0047] That is, a subject who possesses multiple electronic devices 110, 120, 130, 140, and 150 can use the postprandial blood glucose prediction service provided by server 160 by using measuring device 113 or electronic device 110 to connect to server 160 under the control of a pre-installed operating system (OS) or at least one program (e.g., a browser or the installed application).

[0048] 2 is a block diagram illustrating the internal configuration of the electronic devices and the server according to an embodiment. Each of the electronic devices 110, 120, 130, 140, and 150 and the server 160 may be implemented by the computer device 200 described with reference to FIG.

[0049] Such a computer device 200 may include, as shown in Figure 2, a memory 210, a processor 220, a communication interface 230, and an input / output interface 240. The memory 210 is a computer-readable recording medium and may include random access memory (RAM), read only memory (ROM), and a persistent mass storage device such as a disk drive. Here, a persistent mass storage device such as a ROM or a disk drive is a separate and permanent storage device that is distinct from the memory 210. The memory 210 may be included in the computer device 200 as a continuous storage device. The memory 210 may also store an operating system and at least one program code. Such software components may be loaded into the memory 210 from a computer-readable storage medium separate from the memory 210. Such a separate computer-readable storage medium may include a computer-readable storage medium such as a floppy drive, a disk, a tape, a DVD / CD-ROM drive, or a memory card. In other embodiments, the software components may be loaded into the memory 210 through the communication interface 230, which is not a computer-readable storage medium. For example, the software components may be loaded into the memory 210 of the computer device 200 based on a computer program installed by a file received over the network 170.

[0050] Processor 220 may be configured to process computer program instructions by performing basic arithmetic, logic, and input / output operations. The instructions may be provided to processor 220 by memory 210 or by communication interface 230. For example, processor 220 may be configured to execute instructions received according to program code stored in a storage device such as memory 210.

[0051] The communication interface 230 may provide a function for the computer device 200 to communicate with other devices (e.g., the above-mentioned storage device) via the network 170. For example, requests, instructions, data, files, etc. generated by the processor 220 of the computer device 200 in accordance with program code stored in a storage device such as the memory 210 may be transmitted to other devices via the network 170 under the control of the communication interface 230. Conversely, signals, instructions, data, files, etc. from other devices may be received by the computer device 200 via the communication interface 230 of the computer device 200 via the network 170. The signals, instructions, data, etc. received via the communication interface 230 may be transmitted to the processor 220 or the memory 210, and files, etc. may be recorded on a storage medium (e.g., the above-mentioned permanent storage device) that the computer device 200 may further include.

[0052] The input / output interface 240 may be a means for interfacing with the input / output device 250. For example, the input device may include a device such as a microphone, keyboard, or mouse, and the output device may include a device such as a display or speaker. As another example, the input / output interface 240 may be a means for interfacing with a device that integrates input and output functions into one, such as a touch screen. The input / output device 250 may be configured as a single device together with the computer device 200.

[0053] Also, in other embodiments, computing device 200 may include fewer or more components than those shown in Figure 2. However, most prior art components need not be explicitly shown in the figures. For example, computing device 200 may be implemented to include at least some of the input / output devices 250 described above, and may further include other components such as a transceiver, a database, etc.

[0054] Specific embodiments of a method and system for predicting a postprandial blood glucose level that provides a postprandial blood glucose level prediction service will be described below.

[0055] FIG. 3 is a block diagram showing an example of a postprandial blood glucose level prediction system in one embodiment, FIG. 4 is a flowchart showing a postprandial blood glucose level prediction method in one embodiment, and FIGS. 5 and 6 are diagrams for explaining a glucose change inference model used in the postprandial blood glucose level prediction method in one embodiment.

[0056] In the following embodiment, the computer device 200 executes the postprandial blood glucose level prediction method described below to predict the postprandial blood glucose level of a subject, and provides a service of diagnosing diabetes in the subject based on the prediction. To this end, the computer device 200 may be configured with a postprandial blood glucose level prediction system that executes the postprandial blood glucose level prediction method. As an example, the postprandial blood glucose level prediction system may be implemented in the form of an independently operating program, or may be configured in the form of an in-app for a dedicated application and implemented to operate on the dedicated application.

[0057] The processor 220 of the computer device 200 may be realized as components for executing the postprandial blood glucose level prediction method shown in Fig. 4. As an example, the processor 220 may include a learning modeling unit 310, a subject information acquisition unit 320, a pattern difference estimation unit 330, and a postprandial blood glucose level prediction unit 340, as shown in Fig. 3, to execute steps 410 to 430 shown in Fig. 4. Depending on the embodiment, the components of the processor 220 may be selectively included or excluded from the processor 220. Also, depending on the embodiment, the components of the processor 220 may be separated or combined to express the functions of the processor 220.

[0058] Such processor 222 and components of processor 222 may control server 160 to execute steps 410 to 430 included in the postprandial blood glucose level prediction method shown in Fig. 4. For example, processor 220 and components of processor 220 may be realized to execute instructions from operating system code and at least one program code included in memory 210.

[0059] Here, the components of the processor 220 may be representations of different functions of the processor 220 that are executed by the processor 220 according to instructions provided by the program code recorded on the server 160. For example, a postprandial blood glucose level prediction unit 340 may be used as a functional representation of the processor 220 that predicts the subject's postprandial blood glucose level.

[0060] The processor 220 may read necessary instructions from the memory 210, which has been loaded with instructions related to the control of the computer device 200. In this case, the read instructions may include instructions for controlling the processor 220 to execute steps 410 to 430 described below.

[0061] Steps 410 to 430 described below may be performed in an order different from that shown in FIG. 4, some of steps 410 to 430 may be omitted, or additional steps may be included.

[0062] Prior to step 410, the learning modeling unit 310 may pre-construct a glucose change inference model by training it using an AI deep learning technique with input data from experiments that infer pattern differences in blood glucose level changes and salivary glucose changes due to meals while taking into account physical indicators, and may pre-construct a postprandial glucose level inference model by training it using an AI deep learning technique with input data from experiments that infer the correlation between postprandial salivary glucose and postprandial blood glucose while taking into account pattern differences in blood glucose level changes and salivary glucose changes due to meals. The experimental data is clinical trial data from a large number of subject samples, and the glucose change inference model and postprandial blood glucose level inference model trained from the experimental data can be said to provide average inference results and prediction results that respectively satisfy a diverse group of subjects.

[0063] The difference in the patterns of blood glucose and salivary sugar changes is due to the speed at which sugar appears in the blood and saliva. This may include differences in time delay, maximum value, and rate of change between blood glucose and salivary sugar changes due to a meal. For example, due to differences in the speed and degree of expression of blood glucose and salivary sugar changes in blood and saliva, as shown in Figure 5, there may be differences in the time delay between increases and decreases, differences in maximum and minimum sugar values, and differences in the rate of increase and decrease.

[0064] Such pattern differences in blood glucose level changes and salivary sugar changes are indicators that have a significant impact on the algorithm for predicting postprandial blood glucose levels from postprandial salivary sugar, and therefore the postprandial blood glucose level inference model used in the postprandial blood glucose level prediction method may be trained to infer the correlation between postprandial salivary sugar and postprandial blood glucose levels while taking into account pattern differences in blood glucose level changes and salivary sugar changes due to meals.By using the postprandial blood glucose level inference model in the postprandial blood glucose level prediction method, the accuracy of postprandial blood glucose levels predicted from postprandial salivary sugar can be improved.

[0065] On the other hand, the pattern differences in blood glucose and salivary glucose changes due to meals are influenced by physical indices. This is because physical indices such as gender, age, weight, body mass index (BMI), and waist circumference are correlated with HOMA-IR and HOMA β-cell, as shown in Figure 6, and insulin resistance and insulin secretion capacity are calculated from HOMA-IR and HOMA β-cell. Therefore, the pattern differences in blood glucose and salivary glucose changes due to meals are correlated with insulin resistance and insulin secretion capacity.

[0066] Using this principle, the learning modeling unit 310 may infer the correlation between physical indices and HOMA-IR and HOMA β-cell, and train the glucose change inference model to infer the correlation between insulin resistance and insulin secretion capacity calculated from HOMA-IR and HOMA β-cell and the pattern differences in blood glucose changes and salivary glucose changes due to meals. This improves the accuracy with which the glucose change inference model infers the pattern differences in blood glucose changes and salivary glucose changes due to meals.

[0067] In particular, the learning modeling unit 310 personalizes and trains the sugar change inference model and postprandial blood glucose level inference model to suit the subject based on the subject's personal data (for example, blood glucose level change data and saliva sugar change data due to meals based on the subject's fasting blood glucose, the subject's postprandial saliva sugar data and postprandial blood glucose level data, etc.), thereby identifying the subject when the sugar change inference model and postprandial blood glucose level inference model are used, thereby enabling to provide inference results and prediction results with dramatically improved accuracy.

[0068] Various known machine learning algorithms may be used in the learning process of each of the above-mentioned glucose change inference model and postprandial blood glucose level inference model.

[0069] In step 410, the subject information acquisition unit 320 may acquire the subject's physical indices and postprandial salivary sugar. For example, as described above, the subject information acquisition unit 320 may receive and acquire information about postprandial saliva measured by the measuring device 113 corresponding to each of the plurality of electronic devices 110, 120, 130, 140, and 150 (e.g., a response signal to the postprandial salivary sugar, such as the concentration or index value of the postprandial salivary sugar), or may receive and acquire the subject's physical indices from each of the plurality of electronic devices 110, 120, 130, 140, and 150.

[0070] In step 420, the pattern difference estimation unit 330 may use the subject's physical index as input parameters and use a glucose change inference model to estimate pattern differences in the subject's blood glucose level changes and saliva glucose changes due to meals.

[0071] In step 430, the postprandial blood glucose level prediction unit 340 uses the subject's postprandial salivary sugar and the estimated blood glucose level change due to a meal and the pattern difference of the salivary sugar change as input parameters to generate a postprandial blood glucose level inference model. The predicted postprandial blood glucose level of the subject may be provided to an electronic device 110 associated with the subject or a measuring device 113 associated with the electronic device 110.

[0072] The postprandial blood glucose level prediction unit 340 may also diagnose diabetes of the subject based on the predicted postprandial blood glucose level and provide the result. Diagnosing diabetes of the subject may mean checking whether the subject is of the normal type, the borderline type, or the diabetic type.

[0073] In this way, the postprandial blood glucose prediction system and method of one embodiment predicts postprandial blood glucose from postprandial saliva sugar using a sugar change inference model that infers pattern differences in blood glucose changes and saliva sugar changes due to meals while taking physical indicators into account, and a postprandial blood glucose inference model that infers the correlation between postprandial saliva sugar and postprandial blood glucose while taking pattern differences into account.This overcomes the limitation of only being able to collect blood using invasive methods, solving disadvantages such as the effort and pain involved in blood collection, while also solving the constraint and disadvantage of having to maintain an empty stomach for the test.

[0074] In addition, the postprandial blood glucose level prediction system and method according to one embodiment can improve the accuracy of inference and prediction by personalizing and training the glucose change inference model and postprandial blood glucose level inference model to suit the subject.

[0075] The above describes how to predict postprandial blood glucose levels from postprandial salivary sugars, but the above-mentioned postprandial blood glucose level prediction system and method can also predict postprandial salivary sugars from postprandial blood glucose levels. In such cases, in response to the subject information acquisition unit 320 acquiring blood glucose instead of salivary sugar, the postprandial blood glucose level prediction unit 340 may predict postprandial salivary sugars from the acquired postprandial blood glucose levels.

[0076] The above-described devices may be implemented using hardware components, software components, and / or a combination of hardware and software components. For example, the devices and components described in the embodiments may be implemented using one or more general-purpose or special-purpose computers, such as a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA), a programmable logic unit (PLU), a microprocessor, or various devices capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications running on the OS. The processing device may also access, record, manipulate, process, and generate data in response to the execution of the software. For ease of understanding, a single processing device may be described. However, those skilled in the art will understand that a processing device may include multiple processing elements and / or multiple types of processing elements. For example, a processing device may include multiple processors or one processor and one controller. Other processing configurations, such as parallel processors, are also possible.

[0077] The software may include computer programs, codes, instructions, or a combination of one or more of these, and may configure a processing device to operate as desired or may independently or collectively instruct the processing device. The software and / or data may be embodied in any type of machine, component, physical device, virtual device, computer storage medium, or device to be interpreted by the processing device or to provide instructions or data to the processing device. The software may be distributed over computer systems connected by a network, and may be distributed in a distributed manner. The software and data may be stored on one or more computer-readable storage media.

[0078] Methods according to embodiments may be embodied in the form of program instructions executable by various computer means and stored on a computer-readable medium. The computer-readable medium may include, alone or in combination, program instructions, data files, data structures, and the like. The program instructions stored on the medium may be specially designed for the embodiments or may be readily available to those skilled in the art of computer software. Examples of computer-readable storage media include magnetic media such as hard disks, floppy disks, and magnetic tape; optical media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language code, such as that generated by a compiler, but also high-level language code executed by a computer using an interpreter, for example.

[0079] Although the embodiments have been described above based on limited examples and drawings, those skilled in the art will appreciate that various modifications and variations may be made from the above description. For example, the described techniques may be performed in an order different from that described, and / or the described system, structure, device, circuit, or other element may be coupled or combined in a manner different from that described, or may be substituted or replaced by other elements or equivalents, and still achieve suitable results.

[0080] Therefore, different embodiments are within the scope of the appended claims, provided that they are equivalent to the claims.

Claims

1. a learning modeling unit that trains a sugar change inference model to infer the pattern difference between blood glucose level changes and salivary sugar changes due to a meal while taking into account physical indicators, and trains a postprandial blood glucose level inference model to infer the correlation between postprandial salivary sugar and postprandial blood glucose level while taking into account the pattern difference; a subject information acquisition unit that acquires the subject's physical index and postprandial saliva sugar; a pattern difference estimation unit that estimates a pattern difference of the subject using the glucose change inference model with the subject's physical index as an input parameter; and a postprandial blood glucose level prediction section that predicts the postprandial blood glucose level of the subject using the postprandial saliva sugar of the subject and the estimated pattern difference as input parameters and that uses the postprandial blood glucose level inference model; A postprandial blood glucose level prediction system including:

2. The pattern difference is The postprandial blood glucose level prediction system of claim 1, characterized in that it includes the time delay difference, maximum value difference, and rate of change difference of blood glucose level change and salivary sugar change due to the meal.

3. The learning modeling unit The postprandial blood glucose level prediction system of claim 1, characterized in that the glucose change inference model and the postprandial blood glucose level inference model are personalized and trained to suit the subject based on personal data including the subject's fasting blood glucose level.

4. The learning modeling unit The postprandial blood glucose level prediction system of claim 1, characterized in that the sugar change inference model is constructed by learning experimental data that infers the pattern difference while taking into account the physical indicators using artificial intelligence deep learning technology, and the postprandial blood glucose level inference model is constructed by learning experimental data that infers the correlation between the postprandial saliva sugar and the postprandial blood glucose level while taking into account the pattern difference using artificial intelligence deep learning technology.

5. The physical index is The postprandial blood glucose level prediction system according to claim 1, characterized in that it includes gender, age, weight, body mass index (BMI), and waist circumference.

6. The learning modeling unit The postprandial blood glucose prediction system of claim 1, characterized in that the glucose change inference model is trained to infer the correlation between the physical index and HOMA-IR and HOMAβ-cell, and to infer the correlation between the pattern difference and HOMA-IR and HOMAβ-cell.

7. The learning modeling unit The postprandial blood glucose prediction system of claim 6, characterized in that the glucose change inference model is trained to infer a correlation between the insulin resistance and insulin secretion capacity calculated by the HOMA-IR and HOMAβ-cell and the pattern difference.

8. obtaining the subject's physical indices and postprandial salivary sugars; A step of estimating the pattern difference of the subject using a glucose change inference model (the glucose change inference model is trained to infer the pattern difference of blood glucose level change and salivary glucose change due to meals while taking the physical indices of the subject into consideration) using the subject's physical indices as input parameters; and The subject's postprandial saliva sugar and the estimated pattern difference are used as input parameters, and a postprandial blood glucose level inference model (the postprandial blood glucose level inference model is trained to infer the correlation between postprandial saliva sugar and postprandial blood glucose level while taking the pattern difference into consideration) is used to calculate the postprandial blood glucose level of the subject. Value prediction stage A method for predicting postprandial blood glucose levels, comprising:

9. The pattern difference is The method for predicting postprandial blood glucose levels according to claim 8, characterized in that it includes the difference in time delay, maximum value, and rate of change of blood glucose levels and salivary sugar levels due to the meal.

10. The sugar change inference model and the postprandial blood glucose level inference model are The method for predicting postprandial blood glucose levels according to claim 8, characterized in that the method is personalized and learned for the subject based on personal data including the subject's fasting blood glucose level.

11. constructing the glucose change inference model by learning experimental data that infers the pattern difference while taking the physical index into consideration using an artificial intelligence deep learning technique; and constructing the postprandial blood glucose level inference model by learning experimental data that infers the correlation between the postprandial saliva sugar and the postprandial blood glucose level while taking into account the pattern difference using an artificial intelligence deep learning technique; The method for predicting a postprandial blood glucose level according to claim 8, further comprising:

12. The physical index is The method for predicting postprandial blood glucose levels according to claim 8, characterized in that the information includes gender, age, weight, body mass index (BMI), and waist circumference.

13. The sugar change inference model is The method for predicting postprandial blood glucose levels according to claim 8, characterized in that the method is trained to infer correlations between the physical indicators and HOMA-IR and HOMA β-cell, and to infer correlations between the pattern difference and HOMA-IR and HOMA β-cell.

14. The sugar change inference model is The method for predicting postprandial blood glucose levels according to claim 13, characterized in that the method is trained to infer a correlation between the insulin resistance and insulin secretion capacity calculated by the HOMA-IR and HOMAβ-cell and the pattern difference.

15. A computer-readable recording medium having a computer program recorded thereon for causing a computer device to execute a postprandial blood glucose level prediction method, The postprandial blood glucose level prediction method includes: obtaining the subject's physical indices and postprandial salivary sugars; A step of estimating the pattern difference of the subject using a glucose change inference model (the glucose change inference model is trained to infer the pattern difference between blood glucose level changes and salivary glucose changes due to meals while taking the body indices into account) using the subject's physical indices as input parameters; and A step of predicting the postprandial blood glucose level of the subject using the subject's postprandial saliva sugar and the estimated pattern difference as input parameters and utilizing a postprandial blood glucose level inference model (the postprandial blood glucose level inference model is trained to infer the correlation between the postprandial saliva sugar and the postprandial blood glucose level while taking the pattern difference into consideration). A computer-readable recording medium comprising:

Images