Calibration methods for non-invasive biological information

Abstract

The present invention relates to a method for calibrating non-invasive biological information measured by a non-invasive biological information measuring device, and more specifically, to a method for calibrating non-invasive biological information by learning non-invasive biological information measured by a non-invasive blood glucose meter from continuous biological information measured by a continuous blood glucose meter, and by calibrating the non-invasive biological information personalized for a user, and by calibrating uncertain non-invasive biological information measured by a non-invasive blood glucose meter personalized for a user using the continuous biological information measured by the continuous blood glucose meter, thereby accurately determining even an increase / decrease pattern of a user's biological information from the still uncertain non-invasive biological information.

Description

Technical Field

[0001] The present invention relates to a calibration method for non-invasive biological information measured by a non-invasive biological information measuring device. More specifically, the non-invasive biological information measured by a non-invasive blood glucose measuring device is learned using continuous biological information measured via a continuous blood glucose measuring device, and the non-invasive biological information can be personalized and calibrated for the user. By calibrating the non-invasive biological information measured by the non-invasive blood glucose measuring device in a personalized manner using the continuous biological information measured via the continuous blood glucose measuring device, it is possible to accurately determine even the increase and decrease patterns of the user's biological information from the still uncertain non-invasive biological information. The present invention relates to a calibration method for non-invasive biological information.

Background Art

[0002] Diabetes is a chronic disease that frequently occurs in modern people. According to the International Diabetes Federation (IDF), the number of diabetes patients worldwide reaches 400 million.

[0003] Diabetes is caused by various factors such as obesity, stress, incorrect eating habits, and congenital inheritance, resulting in an absolute or relative deficiency of insulin produced by the pancreas, which cannot immediately balance the sugar in the blood. As a result, the sugar component in the blood becomes absolutely increased, leading to the onset of the disease.

[0004] Normally, the blood contains a certain concentration of glucose, and tissue cells obtain energy from it.

[0005] However, if the glucose increases more than necessary, it cannot be properly stored in the liver, muscles, or fat cells, etc., and accumulates in the blood. As a result, diabetes patients maintain a much higher blood glucose level than normal people. The excessive blood glucose passes through the tissues and is excreted in the urine, causing a deficiency of the sugar absolutely necessary for each tissue of the body and bringing abnormalities to each tissue of the body.

[0006] Diabetes is characterized by having almost no noticeable symptoms in its early stages. However, as the disease progresses, specific symptoms such as excessive thirst, increased appetite, frequent urination, weight loss, general fatigue, itching, and persistent wounds on the hands and feet may appear. If the disease progresses even further, complications such as vision impairment, hypertension, kidney disease, stroke, periodontal disease, muscle spasms and neuralgia, and gangrene may develop.

[0007] To diagnose diabetes and manage it to prevent it from progressing to complications, systematic blood glucose monitoring and treatment must be carried out in parallel.

[0008] Diabetes requires continuous monitoring of blood glucose levels for management, and the demand for blood glucose monitoring devices is steadily increasing. Numerous studies have confirmed that when diabetic patients strictly control their blood glucose levels, the incidence of diabetic complications decreases significantly. Therefore, it is crucial for diabetic patients to regularly monitor their blood glucose levels for blood glucose control.

[0009] For managing blood glucose levels in diabetic patients, finger prick blood glucose meters are commonly used. While these meters are helpful for managing blood glucose levels in diabetic patients, they only display the result at the time of measurement, making it difficult to accurately grasp frequently changing blood glucose levels. Furthermore, using a finger prick blood glucose meter requires drawing blood every time to measure blood glucose levels at any given time, which places a significant burden on diabetic patients.

[0010] To overcome the limitations of blood-sampling blood glucose meters, continuous glucose monitoring systems (CGMS) have been developed that are inserted into the body to measure blood glucose levels at intervals of several minutes. This allows for easier management of diabetic patients and quicker responses to emergencies.

[0011] A continuous glucose monitor is a device that extracts bodily fluids (intersitital fluid) through a sensor partially inserted into a part of the user's body, and measures the user's blood glucose level in real time from the extracted fluid. The sensor for the continuous glucose monitor, which is partially inserted into the human body, extracts the user's bodily fluids while inserted for a certain period, for example, about 15 days, and the continuous glucose monitor periodically generates and provides blood glucose biometric information to the user from the extracted fluid.

[0012] However, even with continuous glucose monitors, there is the inconvenience of having to replace the sensor every week, every 15 days, or every month and insert it into the body, and furthermore, there is the inconvenience of having to calibrate the blood glucose levels measured using a separate blood sampling device at set intervals.

[0013] On the other hand, non-invasive blood glucose meters that can measure blood glucose levels using portable devices without drawing blood are currently being researched and developed. Non-invasive blood glucose testing is harmless to the human body, painless, has no side effects, and offers excellent reproducibility, making it the blood glucose testing method that all diabetic patients dream of.

[0014] Research into non-invasive methods for measuring blood glucose levels has been steadily conducted since the 1990s, and various approaches utilizing different principles have been attempted.

[0015] The representative non-invasive blood glucose measurement methods currently under research and development can be broadly divided into four principles. The first method attempted involved attaching a consumable patch to the skin to measure blood glucose levels in the subcutaneous tissue (Transdermal), but this method has limitations in accuracy and reproducibility due to its restriction to the thin skin layer. Subsequently, methods using optical principles were attempted, where light is incident on the skin, and the spectrum of the reflected light is measured and analyzed to determine blood glucose concentration (Optical), and much research has been conducted in this field to date. On the other hand, methods that measure blood glucose levels by applying electrical stimulation to induce an electrochemical reaction have been attempted to some extent (Electrochemical), or methods that measure blood glucose concentration using ultrasound that penetrates deep into the body have been proposed (Ultrasound).

[0016] However, while non-invasive blood glucose meters have the advantage of being able to measure blood glucose levels painlessly by applying them to or outside the skin, they are less accurate than conventional methods that measure blood glucose levels by directly drawing blood from the body or extracting bodily fluids. Furthermore, because each user has different physical or physiological conditions, it is difficult to accurately measure blood glucose levels uniformly for all users using the same method. [Overview of the project] [Problems that the invention aims to solve]

[0017] The present invention aims to solve the problems of the conventional non-invasive biological information measurement methods described above. The objective of the present invention is to provide a method for learning non-invasive biological information measured by a non-invasive biological information measuring device using continuous biological information measured through a continuous biological information measuring device, thereby personalizing and calibrating the non-invasive biological information for the user.

[0018] Another objective of the present invention is to provide a method for calibrating non-invasive biological information that can be personalized and calibrated for a user using continuous biological information measured through a continuous biological information measuring device, and that can be accurately determined from the increase or decrease pattern of the user's biological information based on the non-invasive biological information.

[0019] Another objective of the present invention is to provide a method for calibrating non-invasive biological information that compares non-invasive biological information and continuous biological information based on input event information, and learns the continuous biological information corresponding to the non-invasive biological information when an event occurs, thereby enabling accurate determination of the user's biological information using only event information and non-invasive biological information, even without continuous biological information.

[0020] Another objective of the present invention is to provide a method for calibrating non-invasive biological information that compares non-invasive biological information with continuous biological information based on input event information, learns the continuous biological information corresponding to the non-invasive biological information when an event occurs and then uses only the event information and non-invasive biological information to determine or predict the user's biological information, and provides the user with an alarm in the event of a crisis. [Means for solving the problem]

[0021] To achieve the objectives of the present invention, a method for calibrating non-invasive biological information according to one embodiment of the present invention is characterized by comprising the steps of: measuring the continuous biological information of a user via a continuous biological information measuring device in which a part of the sensor is inserted into the user's body and measures the user's biological information over a certain period of time; measuring the non-invasive biological information of a user over a certain period of time via a non-invasive biological information measuring device that measures the user's biological information in a non-invasive manner; and comparing the non-invasive biological information and continuous biological information measured over a certain period of time, and learning the continuous biological information corresponding to the non-invasive biological information in a personalized manner for the user.

[0022] Preferably, a method for calibrating non-invasive biological information according to one embodiment of the present invention is characterized by further including the step of calibrating additional non-invasive biological information by determining the continuous biological information learned in accordance with the additional non-invasive biological information when additional non-invasive biological information of the user is acquired via a non-invasive biological information measuring device after a certain period of time has elapsed.

[0023] Here, the calibration method of non-invasive biological information according to an embodiment of the present invention compares the increase / decrease pattern of non-invasive biological information with the increase / decrease pattern of continuous biological information, personalizes and learns the increase / decrease pattern of continuous biological information corresponding to the increase / decrease pattern of non-invasive biological information for the user, determines the increase / decrease pattern of continuous biological information learned corresponding to the increase / decrease pattern of additional non-invasive biological information from the additional non-invasive biological information, and calibrates the increase / decrease pattern of the additional non-invasive biological information based on the determined increase / decrease pattern of the continuous biological information.

[0024] Preferably, the calibration method of non-invasive biological information according to an embodiment of the present invention further includes a step of acquiring event information about an event that occurred to the user during a certain period, comparing the non-invasive biological information and the continuous biological information based on the event information, and personalizing and learning the continuous biological information corresponding to the non-invasive biological information at the time of event occurrence for the user.

[0025] Here, the event information is characterized by being input via an interface screen for event input to be displayed.

[0026] Preferably, the calibration method of non-invasive biological information according to an embodiment of the present invention further includes a step of acquiring the occurrence condition information of the event that occurred to the user together when acquiring the event information, comparing the non-invasive biological information and the continuous biological information according to the event information and the occurrence condition information, and personalizing and learning the continuous biological information corresponding to the non-invasive biological information at the time of event occurrence for the user according to the occurrence condition information.

[0027] Here, the occurrence condition information is characterized by being at least any one of season information, time information, location information, position information, temperature information, and humidity information in which the event occurred.

[0028] Here, the occurrence condition information is characterized by being input via an interface screen for occurrence condition input to be displayed.

[0029] Here, the method for calibrating non-invasive biological information measures the continuous biological information of a user over a number of fixed periods via a plurality of continuous biological information measurement devices, compares the non-invasive biological information measured over a number of fixed periods with the continuous biological information, and personalizes and learns the continuous biological information corresponding to the non-invasive biological information for the user during a number of fixed periods.

[0030] Here, the number of fixed periods are set apart from each other. As an example, the number of fixed periods are set apart at the same time interval. As another example, the number of fixed periods are set apart depending on at least one of environmental conditions, seasonal conditions, the physical conditions of the user, and the physiological conditions of the user.

[0031] Preferably, the method for calibrating non-invasive biological information according to an embodiment of the present invention further includes a step of calibrating the additional non-invasive biological information by determining the continuous biological information learned corresponding to the future event information and the additional non-invasive biological information when the future event information occurring to the user after a fixed period and the additional non-invasive biological information of the user are acquired.

[0032] Preferably, in another embodiment of the present invention, the method for calibrating non-invasive biological information further includes a step of calibrating the additional non-invasive biological information by determining the continuous biological information learned corresponding to the future event information, the occurrence condition information of the future event when the future event occurs, and the additional non-invasive biological information when the future event information occurring to the user after a fixed period, the occurrence condition information of the future event when the future event occurs, and the additional non-invasive biological information of the user are acquired.

[0033] The method for calibrating non-invasive biological information according to the present invention compares the increase / decrease pattern of the non-invasive biological information with the increase / decrease pattern of the continuous biological information, personalizes and learns the increase / decrease pattern of the continuous biological information corresponding to the increase / decrease pattern of the non-invasive biological information for the user, determines the increase / decrease pattern of the continuous biological information learned corresponding to the increase / decrease pattern of the additional non-invasive biological information and the additional event information from the additional non-invasive biological information, and calibrates the increase / decrease pattern of the additional non-invasive biological information.

[0034] Another embodiment of the present invention provides a method for calibrating non-invasive biological information, comprising the steps of: measuring a user's continuous biological information via a continuous biological information measuring device in which a part of a sensor is inserted into the user's body and measures the user's biological information over a certain period of time; measuring a user's non-invasive biological information via a non-invasive biological information measuring device that measures the user's biological information in a non-invasive manner over a certain period of time; acquiring event information about events that occurred to the user during the measurement of continuous biological information; comparing the increase / decrease patterns of the event information and non-invasive biological information with the increase / decrease patterns of the continuous biological information, and learning the increase / decrease patterns of the continuous biological information that correspond to the increase / decrease patterns of the event information and non-invasive biological information, personalized to the user; acquiring additional non-invasive biological information of the user and additional event information that occurs to the user via the non-invasive biological information measuring device after a certain period of time has elapsed; and calibrating the increase / decrease patterns of the additional non-invasive biological information by determining the increase / decrease patterns of the additional non-invasive biological information determined from the additional non-invasive biological information and the increase / decrease patterns of the continuous biological information learned in accordance with the additional event information.

[0035] Here, event information and additional event information are entered via a displayable interface screen for event input.

[0036] Preferably, a method for calibrating non-invasive biological information according to another embodiment of the present invention further includes the step of acquiring event information together with event occurrence condition information that occurred to the user when acquiring event information, and is characterized by comparing the increase / decrease pattern of non-invasive biological information corresponding to the event information and occurrence condition information with the increase / decrease pattern of continuous biological information, and learning the increase / decrease pattern of continuous biological information that corresponds to the increase / decrease pattern of non-invasive biological information when an event occurs, using the occurrence condition information, in a personalized manner for the user.

[0037] Here, the event occurrence condition information or additional event occurrence condition information is entered via a displayable interface screen for inputting occurrence conditions.

[0038] Preferably, a method for calibrating non-invasive biological information according to another embodiment of the present invention further comprises the steps of: determining the rate of increase or decrease based on the increase or decrease pattern of the calibrated additional non-invasive biological information; and providing an alarm to the user when the determined rate of increase or decrease exceeds the critical rate of change.

[0039] Preferably, a method for calibrating non-invasive biological information according to another embodiment of the present invention further comprises the steps of: predicting future rates of increase or decrease based on the increase or decrease pattern of the calibrated additional non-invasive biological information; and providing an alarm to the user if the predicted future rates of increase or decrease exceed a critical rate of change. [Effects of the Invention]

[0040] The non-invasive biological information calibration method according to the present invention has the following effects.

[0041] Firstly, the calibration method for non-invasive biological information according to the present invention allows for the personalization and calibration of non-invasive biological information for each user, even if the physical and biological conditions differ for each user, by learning the non-invasive biological information measured by a non-invasive biological information measuring device with accurate continuous biological information measured through a continuous biological information measuring device.

[0042] Secondly, the calibration method for non-invasive biological information according to the present invention allows for accurate determination of the increase or decrease pattern of user biological information, even if the non-invasive biological information measured by a non-invasive biological information measuring device is personalized to the user using continuous biological information measured through a continuous biological information measuring device, thereby enabling accurate determination of the increase or decrease pattern of user biological information from still inaccurate non-invasive biological information.

[0043] Third, the calibration method for non-invasive biological information according to the present invention compares non-invasive biological information based on input event information with continuous biological information, and learns the continuous biological information corresponding to the non-invasive biological information for the user when an event occurs, thereby enabling accurate determination of the user's biological information using only event information and non-invasive biological information, even without continuous biological information.

[0044] Fourth, the non-invasive biological information calibration method according to the present invention compares non-invasive biological information based on input event information with continuous biological information, and learns the continuous biological information corresponding to the non-invasive biological information when an event occurs, thereby enabling the system to judge or predict the user's biological information using only event information and non-invasive biological information, and to provide the user with an alarm in the event of a crisis. [Brief explanation of the drawing]

[0045] [Figure 1] This diagram illustrates a non-invasive biological information calibration system according to one embodiment of the present invention. [Figure 2] This figure illustrates a non-invasive biological information calibration system according to another embodiment of the present invention. [Figure 3] This is a functional block diagram illustrating a non-invasive biological information calibration device according to the present invention. [Figure 4] This is a diagram illustrating the operation of the learning unit according to the present invention. [Figure 5] This is a diagram illustrating the operation of the calibration unit according to the present invention. [Figure 6] This is a flowchart illustrating a method for calibrating non-invasive biological information according to one embodiment of the present invention. [Figure 7] This is a flowchart illustrating a method for calibrating non-invasive biological information according to another embodiment of the present invention. [Figure 8] This shows examples of non-invasive blood glucose information measured with a non-invasive vital signs monitor and continuous blood glucose information measured with a continuous vital signs monitor. [Figure 9] This explains the period during which a continuous biometric information measuring device is worn on the body. [Figure 10] This invention describes an example of an interface screen for inputting event information. [Figure 11] This section describes an example of event information entered via an input interface screen. [Figure 12]This diagram illustrates an example of displaying blood glucose information measured using only non-invasive vital signs monitoring devices after the removal of continuous vital signs monitoring devices. [Figure 13] This flowchart illustrates an example of providing an alarm to a user based on the pattern of increase or decrease in blood glucose information. [Figure 14] This flowchart illustrates an example of providing an alarm to a user based on the future rate of increase or decrease in blood glucose information. [Figure 15] This shows an example of an alarm message provided to the user. [Modes for carrying out the invention]

[0046] It should be noted that the technical terms used in this invention are used merely to describe specific embodiments and are not intended to limit the invention. Furthermore, unless otherwise defined in this invention, the technical terms used in this invention should be interpreted as generally understood by a person of ordinary skill in the art to which this invention pertains, and should not be interpreted in an overly comprehensive or overly restrictive sense. In addition, if any technical terms used in this invention are incorrect and do not accurately express the spirit of the invention, they should be understood as equivalent to technical terms that a person skilled in the art would correctly understand.

[0047] Furthermore, the singular expressions used in this invention include plural expressions unless the context clearly indicates otherwise. In this invention, terms such as "constitutes" or "includes" should not be interpreted as necessarily including all of the multiple components or steps described in this invention, and may not include some of the components or steps, or may further include additional components or steps.

[0048] It should be noted that the attached drawings are intended to facilitate understanding of the concept of the present invention, and should not be interpreted as limiting the concept of the present invention.

[0049] Figure 1 is a diagram illustrating a non-invasive biological information calibration system according to one embodiment of the present invention, and Figure 2 is a diagram illustrating a non-invasive biological information calibration system according to another embodiment of the present invention.

[0050] The non-invasive biological information calibration system according to the present invention comprises a continuous biological information measuring instrument 10 and a non-invasive biological information measuring instrument 30. The non-invasive biological information can be calibrated via a separate user terminal 50, or the non-invasive biological information can be directly calibrated using the non-invasive biological information measuring instrument 30 without a separate user terminal 50.

[0051] In the following description, the continuous biometric device 10 and the non-invasive biometric device 30 will be explained as devices that measure the user's blood glucose level. However, depending on the field to which the present invention is applied, the continuous biometric device 10 and the non-invasive biometric device 30 may be devices capable of measuring various types of biological information.

[0052] First, referring to Figure 1, the system for calibrating non-invasive biological information via the user terminal 50 will be explained. The continuous biological information measuring device 10 is equipped with a sensor, the sensor being partially inserted into and attached to the user's body, and capable of measuring the user's blood glucose information by extracting bodily fluids over a certain period of time. The non-invasive biological information measuring device 30 is a device that can measure the user's blood glucose information in a non-invasive manner, either in contact with or away from the user's skin while worn by the user.

[0053] The user terminal 50 is wirelessly or wiredly connected to the continuous vital signs monitor 10 and receives continuous blood glucose information of the user measured periodically or upon request from the continuous vital signs monitor 10. The user terminal 50 is wirelessly or wiredly connected to the non-invasive vital signs monitor 30 and receives non-invasive blood glucose information of the user measured periodically or upon request from the non-invasive vital signs monitor 30.

[0054] Preferably, the user can input information about events that occur to the user over a certain period of time while the continuous biometric information measuring device 10 is attached to the user terminal 50. The user terminal 50 displays an interface screen on which event information can be input, and the user can input information about events that will occur to the user in the future or events that have occurred in the past via the interface screen. Depending on the field to which the present invention is applied, it may further include a plurality of sensors for sensing events that occur to the user, for example, the user terminal 50 can determine events that have occurred to the user via activity level sensors, location sensors, etc., and can automatically input the determined events by user confirmation.

[0055] Here, an event is something that can affect the user's blood glucose level. For example, it could be an event that increases the user's blood glucose level, such as when the user eats breakfast, lunch, dinner, or a snack, or an event that decreases the user's blood glucose level, such as exercise, work, or studying. To more accurately determine the effect of an event on the user's blood glucose level, detailed event information such as the type and amount of food consumed by the user, or detailed event information such as the type and duration of exercise performed by the user, can be entered along with the event information.

[0056] Preferably, the user can input further information about the conditions for an event to occur into the user terminal 50 when an event occurs. The user terminal 50 displays an interface screen on which event condition information can be input, and the user can input event condition information via the interface screen. Depending on the field to which the present invention is applied, it may further include a plurality of sensors for sensing the conditions for an event that has occurred to the user. For example, the user terminal 50 may determine the event condition information via a location sensor, a temperature sensor, a humidity sensor, etc., or may acquire the event condition information via a network. The event condition information can be automatically input by user confirmation.

[0057] The user terminal 50 includes a storage means capable of storing continuous blood glucose information, non-invasive blood glucose information, event information, and occurrence condition information over a certain period of time, and a processor means capable of learning the continuous blood glucose information corresponding to the non-invasive blood glucose information from the continuous blood glucose information, non-invasive blood glucose information, event information, and occurrence condition information in a way that is personalized to the user, using a learning model stored in the storage means.

[0058] The user terminal 50 receives continuous blood glucose information from the continuous vital signs monitor 10 and non-invasive blood glucose information from the non-invasive vital signs monitor 30. By comparing the received non-invasive blood glucose information and continuous blood glucose information over a certain period, the terminal can learn the continuous blood glucose information corresponding to the non-invasive blood glucose information in a personalized manner for the user.

[0059] Preferably, the user terminal 50 acquires, in addition to continuous blood glucose information and non-invasive blood glucose information, information on events that occurred to the user during a certain period and information on the conditions under which the events occurred. By comparing the non-invasive blood glucose information and continuous blood glucose information corresponding to the event information and the conditions under which the events occurred, the terminal can learn personalized continuous blood glucose information that corresponds to the non-invasive blood glucose information when an event occurs based on the conditions under which the events occurred.

[0060] In other words, using the continuous vital signs measuring device 10 and the user terminal 50, the continuous vital signs measuring device 10 can continuously measure the user's blood glucose information for a certain period, for example, one week, 15 days, or one month, and the user terminal 50 can learn personalized continuous blood glucose information corresponding to non-invasive blood glucose information using the continuous blood glucose information measured over a certain period.

[0061] The continuous biometric information measuring device 10 is attached to the user's body for a certain period of time and then removed, and the user's blood glucose information is determined using only the non-invasive biometric information measuring device 30. However, when the user terminal 50 receives non-invasive blood glucose information, it applies the non-invasive blood glucose information, event information, and event occurrence condition information to the calibration model using a calibration model generated from the learning results, and calibrates the non-invasive blood glucose information measured by the non-invasive biometric information measuring device 30.

[0062] This overcomes the drawbacks of using conventional non-invasive biometric devices to measure a user's blood glucose information, such as inaccurate readings or different blood glucose levels being measured for each user. It allows for the personalization of non-invasive biometric information to the user, enabling accurate measurement of blood glucose information, or at least accurate determination of the user's blood glucose fluctuation patterns.

[0063] Next, referring to Figure 2, a system for calibrating non-invasive biological information without the mediation of a user terminal 50 will be described. The continuous biological information measuring device 10 has a part of its sensor inserted into and attached to the user's body, and measures the user's blood glucose information by extracting bodily fluids over a certain period of time, while the non-invasive biological information measuring device 30 measures the user's blood glucose information in a non-invasive manner while the user wears it over a certain period of time.

[0064] The continuous vital signs monitor 10 and the non-invasive vital signs monitor 30 are connected by wireless or wired communication, and the non-invasive vital signs monitor 30 receives continuous blood glucose information of the user measured periodically or upon request from the continuous vital signs monitor 10.

[0065] Preferably, the user can input information about events that occur to the user over a certain period of time while the continuous biometric information measuring device 10 is attached to the non-invasive biometric information measuring device 30. The non-invasive biometric information measuring device 30 displays an interface screen on which event information can be input, and the user can input information about events that will occur to the user in the future or events that have occurred in the past via the interface screen. Depending on the field to which the present invention is applied, it may further include a plurality of sensors for sensing events that occur to the user, for example, the non-invasive biometric information measuring device 30 can determine events that have occurred to the user via activity level sensors, position sensors, etc., and can automatically input the determined events upon user confirmation.

[0066] Preferably, the user can input further information regarding the conditions for an event to occur into the non-invasive biometric information measuring device 30 when an event occurs. The non-invasive biometric information measuring device 30 displays an interface screen into which event occurrence condition information can be input, and the user can input event occurrence condition information via the interface screen. Depending on the field to which the present invention is applied, it may further include a plurality of sensors for sensing the conditions for an event that has occurred to the user. For example, the non-invasive biometric information measuring device 30 may determine the event occurrence condition information via a position sensor, temperature sensor, humidity sensor, etc., or may acquire the event occurrence condition information via a network. The event occurrence condition information can be automatically input by user confirmation.

[0067] The non-invasive biometric information measuring device 30 includes a storage means capable of storing continuous blood glucose information, non-invasive blood glucose information, event information, and occurrence condition information over a certain period of time, and a processor means capable of learning the continuous blood glucose information corresponding to the non-invasive blood glucose information from the continuous blood glucose information, non-invasive blood glucose information, event information, and occurrence condition information in a way that is personalized to the user, using a learning model stored in the storage means.

[0068] The non-invasive vital signs measuring device 30 compares continuous blood glucose information received from the continuous vital signs measuring device 10 with non-invasive blood glucose information, and can learn the continuous blood glucose information corresponding to the non-invasive blood glucose information in a personalized manner for the user.

[0069] Preferably, the non-invasive biometric information measuring device 30 compares non-invasive blood glucose information and continuous blood glucose information corresponding to event information and occurrence condition information, and can learn personalized continuous blood glucose information corresponding to non-invasive blood glucose information when an event occurs based on occurrence condition information.

[0070] In other words, by using the continuous vital signs measuring device 10 together with the non-invasive vital signs measuring device 30, the continuous vital signs measuring device 10 can measure the user's continuous blood glucose information for a certain period, for example, one week, 15 days, or one month, and the non-invasive vital signs measuring device 30 can learn personalized continuous blood glucose information corresponding to the non-invasive blood glucose information using the continuous blood glucose information measured over a certain period.

[0071] The continuous biometric device 10 is attached to the user's body for a certain period of time and then removed, and the user's blood glucose information is measured using only the non-invasive biometric device 30. However, when the non-invasive biometric device 30 acquires non-invasive blood glucose information using a calibration model generated from the learning results, it calibrates the non-invasive blood glucose information by applying the non-invasive blood glucose information, event information, and event occurrence condition information to the calibration model.

[0072] This overcomes the drawbacks of conventional non-invasive biometric devices, where measuring a user's blood glucose information results in inaccurate information or differs from user to user. Instead, it allows for accurate measurement of blood glucose information by personalizing non-invasive biometric data for each user, or at least accurately determining the user's blood glucose fluctuation patterns.

[0073] Figure 3 is a functional block diagram illustrating the non-invasive biological information calibration device according to the present invention. The non-invasive biological information calibration device described in Figure 3 can be implemented in a user terminal in the case of Figure 1, and in a non-invasive biological information measuring instrument in the case of Figure 2.

[0074] Referring to Figure 3 for a more detailed explanation, the communication unit 110 communicates with an external terminal and sends and receives data. Here, when the non-invasive bio-information calibration device is implemented as a user terminal, the communication unit 110 sends and receives data with the continuous bio-information measuring instrument and the non-invasive bio-information measuring instrument. When the non-invasive bio-information calibration device is implemented as a non-invasive bio-information measuring instrument, the communication unit 110 sends and receives data with the continuous bio-information measuring instrument. The communication unit 110 can send and receive data with an external terminal by wired or wireless means, for example, by methods such as Bluetooth, NFC (Near Field Communication), infrared communication, Wi-Fi communication, and USB cable communication.

[0075] The continuous vital signs monitor attaches to the user's body and continuously measures the user's continuous blood glucose information for a certain period of time. The non-invasive vital signs monitor is also attached to the user and measures non-invasive blood glucose information in a non-invasive manner. The memory unit 130 stores the measured continuous blood glucose information and non-invasive blood glucose information. The learning unit 120 applies the continuous blood glucose information measured through the continuous vital signs monitor and the non-invasive blood glucose information measured through the non-invasive vital signs monitor over a certain period of time to a learning model and learns the continuous blood glucose information corresponding to the non-invasive blood glucose information, personalizing it for the user.

[0076] Preferably, while measuring continuous blood glucose information for a certain period, event information that occurs to the user or event occurrence condition information can be acquired at the time an event occurs, and the acquired continuous blood glucose information, non-invasive blood glucose information, event information, and event occurrence condition information are stored in the memory unit 130. When the learning unit 120 learns continuous blood glucose information corresponding to non-invasive blood glucose information using continuous blood glucose information, it compares the continuous blood glucose information, non-invasive blood glucose information, event information, and event occurrence condition information stored in the memory unit 130 with the non-invasive blood glucose information and continuous blood glucose information corresponding to the event information and occurrence condition information, and learns continuous blood glucose information corresponding to non-invasive blood glucose information when an event occurs using the occurrence condition information.

[0077] In other words, the learning unit 120 uses a training dataset consisting of continuous blood glucose information, non-invasive blood glucose information, event information that occurred during the measurement of continuous blood glucose information, and event occurrence condition information when an event occurs to learn how to personalize continuous blood glucose information corresponding to non-invasive blood glucose information when an event occurs based on the occurrence condition information, and generates a calibration model for calibrating non-invasive blood glucose information from the learning results.

[0078] Here, the learning unit 120 can perform learning using various learning model algorithms, such as Generalized Linear Models (GLM), Decision Trees, Random Forests, Gradient Boosting Machines (GBM), and Deep Learning. Depending on the field to which the present invention is applied, continuous blood glucose information corresponding to non-invasive blood glucose information can be personalized for the user, and various learning model algorithms can be used during learning, which falls within the scope of the present invention.

[0079] Here, event information can be directly input by the user via an event input interface screen output to the user interface unit 150, and event occurrence condition information can be directly input by the user via an occurrence condition input interface screen output to the user interface unit 150.

[0080] Depending on the field to which the present invention is applied, an event that has occurred to the user can be determined via the event determination unit 160. The event determination unit 160 can determine an event that has occurred to the user based on information received from an activity sensor, a location sensor, etc. Preferably, when the event determination unit 160 determines that an event has occurred, it outputs information about the event determined by the user interface unit 150, and if confirmed by the user, it is confirmed as an event that has occurred to the user.

[0081] The present invention can determine event occurrence conditions via the occurrence condition determination unit 170 depending on the field to which the present invention is applied. The occurrence condition determination unit 170 can determine event occurrence conditions based on information received from location sensors, activity level sensors, temperature sensors, humidity sensors, etc., or seasonal information, location information, position information, temperature information, humidity information, etc., obtained via a network. Preferably, when the occurrence condition determination unit 170 determines the occurrence conditions of an event, the user interface unit 150 outputs the determined event occurrence conditions, and if confirmed by the user, the event occurrence conditions are finalized.

[0082] The continuous biometrics measuring device is used to generate a personalized calibration model for non-invasive blood glucose information and is removed from the user's body after the calibration model has been generated. The calibration unit 140 calibrates the non-invasive blood glucose information by applying the non-invasive blood glucose information measured by the non-invasive biometrics measuring device to the calibration model. Preferably, if event information, event occurrence condition information, and non-invasive blood glucose information that occur in the user after the continuous biometrics measuring device has been removed are acquired by the calibration unit 140, the calibration unit 140 calibrates the non-invasive blood glucose information by applying the event information, event occurrence condition information, and non-invasive biometric information to the calibration model.

[0083] The alarm unit 180 determines the rate of increase or decrease of non-invasive blood glucose information from the calibrated non-invasive blood glucose information and provides an alarm to the user if the rate of increase or decrease of non-invasive blood glucose information exceeds the critical rate of change, or it predicts the future rate of increase or decrease of non-invasive blood glucose information from the calibrated non-invasive blood glucose information and provides an alarm to the user if the future rate of increase or decrease of non-invasive blood glucose information exceeds the critical rate of change.

[0084] Figure 4 is a diagram illustrating the operation of the learning unit according to the present invention, and Figure 5 is a diagram illustrating the operation of the calibration unit according to the present invention.

[0085] First, referring to Figure 4, the operation of the learning unit is explained as follows: When continuous blood glucose information and non-invasive blood glucose information are input to the learning unit 120, the learning unit 120 applies the continuous blood glucose information for the same time period as the non-invasive blood glucose information to the learning model algorithm, learns the continuous blood glucose information corresponding to the non-invasive blood glucose information in a way that is personalized to the user, and generates a calibration model for calibrating the non-invasive blood glucose information with the learning results.

[0086] Preferably, the learning unit 120 can receive continuous blood glucose information, non-invasive blood glucose information, event information, and event occurrence condition information. The learning unit 120 applies the continuous blood glucose information, non-invasive blood glucose information, event information, and event occurrence condition information to a learning model algorithm, learns the continuous blood glucose information corresponding to the non-invasive blood glucose information when an event occurs, personalizes it for the user based on the event occurrence condition information, and generates a calibration model for calibrating the non-invasive blood glucose information based on the learning results.

[0087] On the other hand, referring to Figure 5, the operation of the calibration unit can be explained as follows: After the continuous vital signs measuring device is removed from the body, the user's blood glucose information is measured using only the non-invasive vital signs measuring device. The calibration unit 140 can then apply the additional non-invasive blood glucose information obtained after the continuous vital signs measuring device is removed to the calibration model to calibrate the additional non-invasive blood glucose information.

[0088] Preferably, the calibration unit 140 can input additional event information and additional event occurrence condition information in addition to non-invasive blood glucose information, and the calibration unit 140 can calibrate the additional non-invasive blood glucose information by applying the non-invasive blood glucose information, event information, and event occurrence condition information to the calibration model.

[0089] Figure 6 is a flowchart illustrating a method for calibrating non-invasive biological information according to one embodiment of the present invention.

[0090] To explain in more detail with reference to Figure 6, when continuous blood glucose information is measured over a certain period of time using a continuous vital signs measuring device, the continuous blood glucose information is received and acquired (S111), and when non-invasive blood glucose information is measured over a certain period of time using a non-invasive vital signs measuring device, the non-invasive blood glucose information is acquired (S113).

[0091] The continuous vital signs measuring device is attached to the body for a certain period to measure continuous blood glucose information, and is removed from the body after that period has elapsed. The continuous blood glucose information measured over the period and the non-invasive blood glucose information measured at the same time as the continuous blood glucose information over the period are applied to a learning model algorithm to personalize and learn for the user, and a calibration model is generated from the learning results (S115).

[0092] A calibration model can be generated by comparing non-invasive blood glucose information and continuous blood glucose information measured at the same time, extracting features from the non-invasive blood glucose information, and learning the continuous blood glucose information corresponding to the extracted features. Alternatively, the non-invasive blood glucose information and continuous blood glucose information measured at the same time can be applied to the input nodes of an artificial neural network model, and the weights of the hidden nodes can be calculated using a linear regression method to generate a calibration model. Since learning methods based on features extracted from machine learning and learning methods based on artificial neural network models are widely known, a detailed explanation of these will be omitted.

[0093] Depending on the field to which the present invention is applied, in addition to generating a calibration model by learning sequential biological information corresponding to non-invasive blood glucose information in a personalized manner for the user, a calibration model can also be generated by learning the increase / decrease patterns of sequential blood glucose information corresponding to the increase / decrease patterns of non-invasive blood glucose information in a personalized manner for the user. A calibration model for increase / decrease patterns may be more accurate than calibrating the user's blood glucose level from non-invasive blood glucose information by simply calibrating whether the user's blood glucose is increasing or decreasing based on the non-invasive blood glucose information.

[0094] Referring again to Figure 6, when additional non-invasive blood glucose information is obtained from a non-invasive blood glucose measuring device after the continuous vital signs measuring device has been removed from the body (S117), the additional non-invasive blood glucose information is applied to the calibration model to calibrate the non-invasive blood glucose information (S119).

[0095] Figure 7 is a flowchart illustrating a non-invasive biological information calibration method according to another embodiment of the present invention.

[0096] Another embodiment of the present invention, as illustrated in Figure 7, is a method for calibrating non-invasive biological information by using, in addition to continuous blood glucose information acquired over a certain period, event information that occurred over a certain period or information on the conditions under which an event occurred, together with the non-invasive blood glucose information to be personalized learning for the user.

[0097] Referring to Figure 7, when continuous blood glucose information is measured over a certain period of time using a continuous vital signs measuring device, the continuous blood glucose information is received and acquired (S131). When non-invasive blood glucose information is measured over a certain period of time using a non-invasive vital signs measuring device, the non-invasive blood glucose information is acquired (S132).

[0098] If an event occurs during a certain period, information about the event and the conditions under which the event occurred are obtained (S133). Here, the event information includes everything that may affect the user's biometric information, such as the exercise performed by the user, the type and duration of the exercise, the food consumed, the type and amount of food consumed, the degree of stress the user is experiencing, and the user's physical condition (sleep duration and quality, whether they have an illness, etc.).

[0099] On the other hand, information regarding the conditions under which an event occurs may include the season, weather, time, location, temperature, and humidity at the time the event occurred.

[0100] While users can directly input this event information and event occurrence condition information via an input interface screen, it can also be automatically determined based on information acquired through various sensors or via the network.

[0101] Continuous blood glucose information measured over a certain period, non-invasive blood glucose information measured at the same time as the continuous blood glucose information over a certain period, and in addition, event information and event occurrence condition information are applied to a learning model algorithm to personalize it for the user and train it, and a calibration model is generated from the training results (S134).

[0102] By comparing non-invasive blood glucose information measured using event information and event occurrence condition information with continuous blood glucose information measured at the same time, features of the non-invasive blood glucose information can be extracted, and a calibration model can be generated by learning the continuous blood glucose information corresponding to the extracted features. Alternatively, the event information, event occurrence condition information, and non-invasive blood glucose information and continuous blood glucose information measured at the same time can be applied to the input nodes of an artificial neural network model, and a calibration model can be generated by calculating the weights of the hidden nodes using a linear regression method. Since learning methods based on features extracted from machine learning and learning methods based on artificial neural network models are widely known, a detailed explanation of these will be omitted.

[0103] Depending on the field to which the present invention is applied, in addition to generating a calibration model by learning and personalizing continuous biological information corresponding to non-invasive biological information at the time of event occurrence under event occurrence conditions, a calibration model can also be generated by learning and personalizing the increase / decrease pattern of continuous blood glucose information corresponding to the increase / decrease pattern of non-invasive blood glucose information at the time of event occurrence under event occurrence conditions.

[0104] Referring again to Figure 7, after the continuous vital signs monitor is removed from the body, additional non-invasive blood glucose information is obtained from the non-invasive vital signs monitor (S135). If event information and event occurrence conditions are obtained for the user (S137), the additional non-invasive blood glucose information, event information, and event occurrence conditions are applied to the calibration model to calibrate the non-invasive blood glucose information (S139).

[0105] Figure 8 shows an example of non-invasive blood glucose information measured with a non-invasive vital signs monitor and continuous blood glucose information measured with a continuous vital signs monitor.

[0106] As shown in Figure 8(a), non-invasive blood glucose information measured by a non-invasive biometric device may yield meaningless values ​​that repeatedly show increases and decreases in blood glucose levels, such as noise or static, rather than increasing or decreasing equally when the user's blood glucose increases or decreases equally when blood glucose decreases, due to various reasons such as the inaccuracy of the blood glucose measurement principle or the difficulty in personalizing the non-invasive biometric device for the user.

[0107] In contrast, as shown in Figure 8(b), while continuous blood glucose information measured by a continuous vital signs monitor can guarantee a certain degree of accuracy, it measures continuous blood glucose information that increases equally when the user's blood glucose increases and decreases equally when their blood glucose decreases.

[0108] In this way, continuous blood glucose information measured over a certain period of time through a continuous vital signs measuring device can be used to personalize and calibrate non-invasive blood glucose information for the user.

[0109] Such continuous biometric measurement devices are inserted into the user's body for a certain period to measure continuous blood glucose information, and the continuous blood glucose information acquired over that period can be used to personalize and calibrate non-invasive blood glucose information measured by a non-invasive biometric measurement device for the user.

[0110] Figure 9 illustrates the period during which a continuous biological information measuring device is worn on the body.

[0111] Referring to Figure 9(a), the continuous biological information measuring device can be inserted into the user's body and attached during any of the T1, T2, or T3 time periods.

[0112] Here, T1, T2, and T3 can be attached to the user's body for the duration of use of the continuous biometric measurement device, for example, one week, 15 days, or one month.

[0113] However, even if the continuous vital signs monitor is intended for use for one month, it may be worn on the user's body for shorter periods (T1, T2, T3) if necessary. Here, the period during which the continuous vital signs monitor is inserted into the body and measures continuous blood glucose information may be set to the time required for the calibration model to be completed. That is, even if the continuous vital signs monitor is intended for use for one month, if a calibration model with the required accuracy is generated from continuous blood glucose information measured over 10 days, the continuous vital signs monitor may be removed after 10 days.

[0114] In this invention, by using multiple continuous biological information measuring devices to measure continuous blood glucose information over numerous fixed periods, a more accurate calibration model can be generated from the continuous blood glucose information measured over numerous fixed periods.

[0115] Multiple fixed periods for inserting and attaching multiple continuous vital signs measuring devices to measure continuous blood glucose information can be set apart from each other. However, as shown in Figure 9(b), the multiple fixed periods T1, T2, T3, and T4 can be set apart at the same time interval, or as shown in Figure 9(c), the multiple fixed periods T1, T2, T3, and T4 can be set apart at different intervals depending on at least one of the following: environmental conditions, seasonal conditions, the user's physical conditions, or the user's physiological conditions.

[0116] Figure 10 illustrates an example of an interface screen for inputting event information in the present invention.

[0117] As shown in Figure 10(a), an input interface screen for entering event information is activated on the user terminal's display, allowing the user to input event type, detail type, event history, etc., via the input interface screen.

[0118] As shown in Figure 10(b), when a non-invasive biological information measuring device is equipped with a display unit, an input interface screen for inputting event information is activated on the display unit, and the user can input event type, detail type, event history, etc. via the input interface screen.

[0119] Figure 11 illustrates an example of event information entered via an input interface screen.

[0120] As shown in Figure 11, when events that occurred to the user during a certain period and information about the conditions under which those events occurred are entered via the input interface screen, an event icon E is displayed together to indicate that the event information and the conditions under which the event occurred have been entered as time progresses. When the event icon is selected, the detailed history of information about that event and the conditions under which it occurred is activated.

[0121] Figure 12 illustrates an example of displaying blood glucose information measured using only non-invasive vital signs monitoring devices after the removal of continuous vital signs monitoring devices.

[0122] After the continuous vital signs monitor is removed, the user is provided with blood glucose information measured using only non-invasive vital signs monitors. As shown in Figure 12(a), blood glucose information R, which has been calibrated by applying future input event information, future event occurrence condition information, and future non-invasive blood glucose information to a calibration model, is displayed.

[0123] As shown in Figure 12(b), information on the increase / decrease pattern of the user's blood glucose, determined from blood glucose information calibrated by applying future input event information, future event occurrence conditions information, and future non-invasive blood glucose information to the calibration model, is displayed.

[0124] Figure 13 is a flowchart illustrating an example of providing an alarm to the user based on the pattern of increase or decrease in blood glucose information.

[0125] Referring to Figure 13 for a more detailed explanation, additional non-invasive blood glucose information, additional event information, and information on the conditions for the occurrence of additional events are applied to the calibration model to determine the increase / decrease pattern of the user's blood glucose information (S151), and the rate of change from the determined increase / decrease pattern is determined (S153).

[0126] The system determines whether the determined rate of increase or decrease is greater than the critical rate of change (S155). If the determined rate of increase or decrease is greater than the critical rate of change, an alarm message is generated and provided to the user (S157).

[0127] Figure 14 is a flowchart illustrating an example of providing an alarm to the user based on the future rate of increase or decrease in blood glucose information.

[0128] Referring to Figure 14 for a more detailed explanation, additional non-invasive blood glucose information, additional event information, and information on the conditions for the occurrence of additional events are applied to the calibration model to determine the increase / decrease pattern of the user's blood glucose information (S171), and the expected future rate of increase / decrease is determined based on the determined increase / decrease pattern (S173). Here, the expected future rate of increase / decrease can be determined based on the increase / decrease pattern that is expected to be observed in the future, which has been determined through learning and personalized for the user.

[0129] The system determines whether the determined future rate of increase or decrease is greater than the critical rate of change (S175). If the determined future rate of increase or decrease is greater than the critical rate of change, an alarm message is generated and provided to the user (S177).

[0130] Figure 15 shows an example of an alarm message provided to the user.

[0131] As shown in Figure 15(a), event icons E, which indicate that an event has occurred for the user along with the calibrated blood glucose information R, and alarm icons A, which indicate that an alarm message exists, are displayed in chronological order. If alarm icon A2 is selected, a specific alarm message can be activated.

[0132] As shown in Figure 15(b), event icons E, which indicate that an event has occurred for the user, and alarm icons A, which indicate that an alarm message exists, are displayed in chronological order along with information about the user's blood glucose increase / decrease pattern. If alarm icon A1 is selected, a specific alarm message can be activated.

[0133] On the other hand, the embodiments of the present invention described above can be created as programs executable by a computer and can be implemented in a general-purpose digital computer that operates the program using a computer-readable recording medium.

[0134] The recording media readable by the aforementioned computer include storage media such as magnetic storage media (e.g., ROM, floppy disks, hard disks, etc.), optical reading media (e.g., CD-ROMs, DVDs, etc.), and carrier waves (e.g., transmission over the Internet).

[0135] Although the present invention has been described with reference to embodiments shown in the drawings, these are merely illustrative, and it will be understood by those with ordinary skill in the art that various modifications and equivalent other embodiments are possible therefrom. Accordingly, the true scope of technical protection of the present invention should be determined by the technical idea of ​​the appended claims.

Claims

1. A continuous biological information measuring device that includes a sensor configured to be inserted into the user's body in at least part and to measure the user's continuous biological information over a certain period of time, A step of measuring the user's non-invasive biological information over a certain period of time via a non-invasive biological information measuring device that measures the user's non-invasive biological information by being separated from or in contact with the user's skin, The steps include: acquiring event information and occurrence condition information for events that occur to the user while measuring continuous biological information; In order to generate a model personalized for the user, the process involves comparing the increase / decrease pattern of the non-invasive biological information based on the event information and the occurrence condition information with the increase / decrease pattern of the continuous biological information, thereby learning the increase / decrease pattern of the continuous biological information corresponding to the increase / decrease pattern of the non-invasive biological information for each event that matches the occurrence condition information. The steps include obtaining additional non-invasive biometric information of the user, additional event information, and additional occurrence condition information regarding additional events that occur to the user after the personalized model for the user has been generated, A step of calibrating the additional non-invasive biological information based on the personalized model to which the additional non-invasive biological information, the additional event information, and the additional occurrence condition information have been applied. A method for calibrating non-invasive biological information, characterized by including [a specific element].

2. The increase / decrease pattern of continuous biological information learned in correspondence with the increase / decrease pattern of additional non-invasive biological information is determined to calibrate the increase / decrease pattern of the additional non-invasive biological information. The method for calibrating non-invasive biological information according to feature 1.

3. The aforementioned event information is, The method for calibrating non-invasive biological information according to claim 1, characterized in that the information is input via an interface screen for event input that is displayed.

4. The aforementioned occurrence condition information is, The method for calibrating non-invasive biological information according to claim 1, characterized in that the information is at least one of the following: seasonal information, time information, location information, position information, temperature information, and humidity information in which the aforementioned event occurred.

5. The aforementioned occurrence condition information is, The method for calibrating non-invasive biological information according to claim 4, characterized in that the information is input via an interface screen for inputting occurrence conditions that is displayed.

6. The continuous biological information is measured over a number of fixed periods via multiple continuous biological information measuring devices, including the continuous biological information measuring device. A method for calibrating non-invasive biological information according to any one of claims 1 to 5.

7. The method for calibrating non-invasive biological information according to claim 6, characterized in that the aforementioned number of fixed periods are set to be spaced apart from each other.

8. The method for calibrating non-invasive biological information according to claim 7, characterized in that the aforementioned number of fixed periods are set to be separated by the same time intervals.

9. The method for calibrating non-invasive biological information according to claim 7, characterized in that the aforementioned numerous fixed periods are set apart by at least one of environmental conditions, seasonal conditions, the user's physical conditions, and the user's physiological conditions.

10. A method for calibrating non-invasive biological information according to Claim 1, The pattern of increase or decrease of the additional non-invasive biological information is calibrated by determining the pattern of increase or decrease of the additional non-invasive biological information and the pattern of increase or decrease of the continuous biological information learned in accordance with the additional event information. Calibration methods for non-invasive biological information.

Images