Method and computing device for non-invasively estimating blood glucose level, device for non-invasively measuring electrocardiogram signal, and non-temporary computer-readable recording medium

Abstract

To provide a method for non-invasively estimating a glucose level, suitable for correctly and non-invasively estimating a glucose level of a user by a computing device.SOLUTION: A method includes: receiving a plurality of electrocardiogram waveforms of a user; extracting at least two first electrocardiogram features from each of the plurality of electrocardiogram waveforms of the user; determining first feature peak positions respectively corresponding to the first electrocardiogram features; calculating at least one peak distance among the plurality of first feature peak positions; and estimating a glucose level of the user on the basis of the peak distance. The first electrocardiogram feature is selected from the group consisting of P wave, Q wave, R wave, S wave, T wave, and U wave. There are also provided a computing device for non-invasively estimating a glucose level, a device for non-invasively measuring an electrocardiogram signal, and a non-transient computer readable recording medium that are used in the method.SELECTED DRAWING: Figure 6

Description

Technical Field

[0001] The present disclosure relates to a method and a computing device for estimation, a device for measurement, and a non-transitory computer-readable recording medium. Specifically, it relates to a method and a computing device for non-invasively estimating a blood glucose level, a device for non-invasively measuring an electrocardiogram (ECG) signal, and a non-transitory computer-readable recording medium.

Background Art

[0002] According to the World Health Organization (WHO), the number of diabetes patients increased from 108 million in 1980 to 422 million in 2014. In the United States alone, 30 million adults are suffering from diabetes, and among them, 7.2 million are not aware that they have diabetes. Therefore, the measurement and monitoring of blood glucose levels have become important issues in the medical field.

[0003] Conventional blood glucose level measurements are invasive measurements, regardless of whether they are continuously measured. For example, when a user measures their blood glucose level using a blood glucose meter or a continuous glucose monitor, they need to prick a needle into their body. The fact that the user experiences physical pain and causes a great deal of discomfort is one of the problems of invasive measurements. Therefore, users of blood glucose level measurement devices are often limited to diabetes patients who need to manage their blood glucose levels. In addition to the cost of treatment drugs, the costs of test strips, measurement devices, and readers also need to be added to the total cost. Therefore, developing a method and a device for non-invasively estimating blood glucose levels would be very helpful for diabetes patients.

[0004] In conventional non-invasive blood glucose measurement, in order to acquire a single-lead electrocardiogram signal, the user's both hands need to contact the measurement device as electrodes to form a V1 lead. To acquire the electrocardiogram signal, a person who is eating has to put down the tableware every 15 minutes during the meal and spend 1 minute to acquire the electrocardiogram signal with both hands. If the signal quality is not satisfactory during this period, the user has to spend another 1 minute to measure the signal until the signal quality is sufficiently improved. Although it is a non-invasive measurement, the purpose of continuous measurement cannot be achieved, so the trend of blood glucose change cannot be known in detail.

[0005] In addition, for the method of estimating the user's blood glucose value using the electrocardiogram signal, very high requirements are imposed on the signal quality. In order to obtain accurate results, it is further necessary to accurately detect the P wave, Q wave, R wave, S wave, and T wave for feature extraction.

[0006] There are mainly two methods for measuring the change in blood glucose value based on the electrocardiogram signal. One is the method of extracting features such as the QT interval from the form of the electrocardiogram waveform, and as analysis methods, statistical analysis and machine learning are often used. The results of statistical analysis show that the QT interval and the ST interval have a high correlation with hypoglycemia, and the PR interval and the ST interval have a high correlation with hyperglycemia. The other is the method of using the electrocardiogram signal itself as a feature. However, it has been shown that the performance varies greatly from person to person in such a method. Therefore, using the form of the electrocardiogram signal as a feature has more stable performance than directly using the electrocardiogram signal.

[0007] The method based on extracting features from the form of the electrocardiogram waveform is an excellent method, but it only shows the relationship between hypoglycemia and the QT interval (and / or ST interval) and the relationship between hyperglycemia and the PR interval (and / or ST interval). That is, since the ST interval has a high correlation with both hypoglycemia and hyperglycemia, the user's blood glucose value cannot be estimated by this method either.

[0008] Therefore, in order to estimate the user's blood glucose level based on the characteristics of the electrocardiogram extracted from the electrocardiogram waveform, an improvement in non-invasive blood glucose measurement methods is expected.

SUMMARY OF THE INVENTION

PROBLEMS TO BE SOLVED BY THE INVENTION

[0009] An object of the present disclosure is to provide a technique advantageous for non-invasively estimating a user's blood glucose level using a plurality of electrocardiogram (ECG) waveforms. In this technique, the accuracy and non-invasiveness of estimating the user's blood glucose level using a plurality of electrocardiograms are improved.

MEANS FOR SOLVING THE PROBLEMS

[0010] To achieve at least the above object, the present disclosure provides a method for non-invasively estimating a blood glucose level, which estimates the user's blood glucose level by a computing device. The method includes receiving a plurality of electrocardiogram (ECG) waveforms of a user, extracting at least two first electrocardiogram features from each of the plurality of electrocardiogram waveforms of the user, determining a first feature peak position corresponding to each of these first electrocardiogram features, respectively, calculating at least one peak distance between these first feature peak positions for each of the plurality of electrocardiogram waveforms, and estimating the user's blood glucose level based on at least one peak distance. These first electrocardiogram features are selected from the group consisting of P wave, Q wave, R wave, S wave, T wave, and U wave.

[0011] In one embodiment, the method for non-invasively estimating the blood glucose level further includes calculating a corrected blood glucose level of the user by an equation for correcting the blood glucose level based on the user's blood glucose level.

[0012] In one embodiment, the method for non-invasively estimating the blood glucose level further includes calculating, for each of a plurality of electrocardiogram waveforms, at least one peak-to-peak slope between these first characteristic peak positions. The step of estimating the user's blood glucose level is further performed based on at least one peak-to-peak slope.

[0013] In one embodiment, the method for non-invasively estimating the blood glucose level further includes extracting at least three second electrocardiogram characteristics from each of a plurality of electrocardiogram waveforms of the user, and calculating at least one incidence rate based on these second electrocardiogram characteristics. The step of estimating the user's blood glucose level is further performed based on at least one incidence rate. These second electrocardiogram characteristics are selected from the group consisting of P wave, Q wave, R wave, S wave, T wave, and U wave.

[0014] In one embodiment, the method for non-invasively estimating the blood glucose level further includes extracting at least four third electrocardiogram characteristics from each of a plurality of electrocardiogram waveforms of the user, and calculating at least one amplitude ratio based on these third electrocardiogram characteristics. The step of estimating the user's blood glucose level is further performed based on at least one amplitude ratio. These third electrocardiogram characteristics are selected from the group consisting of P wave, Q wave, R wave, S wave, T wave, and U wave.

[0015] In one embodiment, the method for non-invasively estimating the blood glucose level further includes extracting at least one fourth electrocardiogram characteristic from each of a plurality of electrocardiogram waveforms of the user, and calculating at least one sharpness result based on at least one fourth electrocardiogram characteristic. The step of estimating the user's blood glucose level is further performed based on at least one sharpness result. At least one fourth electrocardiogram characteristic is selected from the group consisting of P wave, Q wave, R wave, S wave, T wave, and U wave.

[0016] In one embodiment, the method for non-invasively estimating the blood glucose level described above includes evaluating the quality of a plurality of electrocardiogram waveforms of a user, outputting the real-time blood glucose level of the user when the plurality of electrocardiogram waveforms of the user are evaluated as normal electrocardiogram signals, and outputting the past blood glucose level of the user when the plurality of electrocardiogram waveforms of the user are evaluated as noise signals. The real-time blood glucose level of the user is the blood glucose level estimated in real time. The past blood glucose level of the user is the blood glucose level estimated previously.

[0017] In one embodiment, the method for non-invasively estimating the blood glucose level described above further includes determining whether the number of a plurality of electrocardiogram waveforms of the user is less than a preset value, and re-receiving the plurality of electrocardiogram waveforms of the user when the number of the plurality of electrocardiogram waveforms of the user is less than the preset value.

[0018] Also, the present disclosure provides a computing device for non-invasively estimating a blood glucose level, which is signal-connected to an electrocardiogram (ECG) measurement device and receives a plurality of electrocardiogram waveforms of a user from the electrocardiogram measurement device. The computing device includes a storage module and a blood glucose level estimation module. The blood glucose level estimation module is configured to be electrically connected to the storage module. A plurality of program codes are stored in the storage module. After the blood glucose level estimation module executes the plurality of program codes stored in the storage module, the blood glucose level estimation module executes the steps of any one of the methods for non-invasively estimating the blood glucose level described above.

[0019] In addition, the present disclosure provides a device for non-invasively measuring an electrocardiogram (ECG) signal that is electrically connected to a computing device and outputs a plurality of electrocardiogram waveforms of a user to the computing device. The device includes a measurement module and a signal transmission module. The measurement module has a plurality of electrodes that are electrically connected to the user. The signal transmission module is configured to be electrically connected to the measurement module. The measurement module is configured to measure a plurality of electrocardiogram waveforms of the user. The signal transmission module is configured to transmit a plurality of electrocardiogram waveforms of the user to the computing device and cause the computing device to execute steps of any one of the methods for non-invasively estimating the blood glucose value described above.

[0020] After a plurality of stored program codes are loaded and executed by a computing device, a non-transitory computer-readable recording medium can implement any one of the methods for non-invasively estimating the blood glucose value described above.

Advantages of the Invention

[0021] Since the present disclosure provides an improvement in the technical field related to non-invasively estimating a user's blood glucose value using a plurality of electrocardiogram waveforms, according to the present disclosure, the user's blood glucose value can be accurately estimated by a non-invasive method.

Brief Description of the Drawings

[0022]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7A

Figure 7B

Figure 8A

Figure 8B

Figure 8C

Figure 8D

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14A

Figure 14B

Figure 15

Figure 16

Figure 17A

Figure 17B

Figure 18A

Figure 18B

Figure 19

Figure 20

Figure 21

Figure 22A

Figure 22B

Figure 23

Figure 24

Figure 25

Figure 26

Figure 27

Figure 28

Figure 29A

Figure 29B

Mode for Carrying Out the Invention

[0023] The present disclosure will be described in detail by the following embodiments and the accompanying drawings so that those having ordinary knowledge in the technical field to which the present disclosure pertains can understand the object, features, and effects of the present disclosure.

[0024] Before the present disclosure is described in detail, it should be noted that in the following description, the same components or steps may be represented by the same reference numerals.

[0025] Also, in the context of the present disclosure, terms such as "first", "second", "third", "fourth", "fifth", "sixth", etc. are not used to limit the components themselves or indicate a specific order of the components, but should also be noted that they are used to distinguish the components.

[0026] Also, it should be noted that each step described in this specification may be executed in order, in reverse order, or the steps may be appropriately changed or skipped during the control and processing.

[0027] Also, the expression "the first step may be executed following the execution of the second step" may mean that the first step may be directly executed after the execution of the second step, or alternatively, the first step may be executed after another step (for example, the third step) is executed first.

[0028] Since many different physiological signals are generated by the human body itself and many medical devices and sensors can convert physiological signals into specific electrical signals, it is possible to obtain important clinical information by signal - processing physiological signals with an electronic device.

[0029] For example, since the heart itself is composed of multiple muscle groups that spontaneously pulsate and contract regularly, when the multiple muscle groups of the heart pulsate or contract, a minute amount of electric current is generated, and the current is reflected on the surface of the body through the conductive tissues and body fluids around the heart. Therefore, it is possible to convert the voltage changes caused by the multiple muscle groups of the heart into corresponding electrocardiograms by related electronic devices. More specifically, it is possible to receive and record the electrical signals caused by the multiple muscle groups of the heart by a plurality of electrodes in contact with the skin of the human body.

[0030] The method provided by the present disclosure is suitable for signal processing of a plurality of electrocardiogram waveforms in order to estimate blood glucose levels based on the plurality of electrocardiogram waveforms. Since a plurality of electrocardiogram waveforms of a user can be received by a non-invasive measurement device, the method provided by the present disclosure can estimate the blood glucose level of the user in a non-invasive manner.

[0031] Please refer to FIG. 1. FIG. 1 is a waveform diagram illustrating two examples of a plurality of electrocardiogram waveforms received from a user. The horizontal axis of FIG. 1 represents time (in seconds), and the vertical axis of FIG. 1 represents the amplitude value of the electrocardiogram waveform (in millivolts).

[0032] As shown in FIG. 1, in the first example, it is shown that the number of a plurality of electrocardiogram waveforms received from the user in 10 seconds is 20. That is, the first example shown in FIG. 1 shows that the heart rate of the user is 120 beats per minute. As shown in FIG. 1, in the second example, it is shown that the number of a plurality of electrocardiogram waveforms received from the user in 10 seconds is 30. That is, the second example shown in FIG. 1 shows that the heart rate of the user is 180 beats per minute.

[0033] The number of multiple electrocardiogram waveforms received from a user within a certain period may vary from user to user. For example, in the first example and the second example shown in FIG. 1, multiple electrocardiogram waveforms received from two different users are respectively shown. Furthermore, the number of electrocardiogram waveforms received from a user within a certain period may also vary according to different states of the same user. For example, the first example may show multiple electrocardiogram waveforms received from a user in a normal state, and the second example may show multiple electrocardiogram waveforms received from the same user in a state where emotions such as tension, excitement, or fear are heightened.

[0034] Please refer to FIG. 2. FIG. 2 is a waveform diagram for explaining an example of multiple electrocardiogram features extracted from each of multiple electrocardiogram waveforms. The horizontal axis of FIG. 2 represents time (in seconds), and the vertical axis of FIG. 2 represents the amplitude values (in millivolts) of multiple electrocardiogram waveforms. As shown in FIG. 2, each of the received multiple electrocardiogram waveforms may have multiple electrocardiogram features such as P wave, Q wave, R wave, S wave, T wave, and U wave. The multiple electrocardiogram features may be used to represent voltage changes by multiple muscle groups of the heart.

[0035] Since multiple muscle groups of the heart may be contracted by atrial depolarization, a first deviation (i.e., a first voltage change) called a P wave occurs in the electrocardiogram waveform. Since multiple muscle groups of the heart may be contracted by left ventricular depolarization and right ventricular depolarization, a second deviation (i.e., a second voltage change) called a QRS waveform occurs in the electrocardiogram waveform. The change amount of the second deviation is clearly larger than that of the first deviation. Usually, there are three deviations in the QRS complex, namely a first downward deviation, a first upward deviation, and a second downward deviation. The first downward deviation is called the Q wave, the first upward deviation is called the R wave, and the second downward deviation is called the S wave. Since multiple muscle groups of the heart may be contracted by ventricular repolarization, a third deviation (i.e., a third voltage change) called a T wave occurs in the electrocardiogram waveform. The deviation immediately after the T wave (i.e., the fourth voltage change) may be called a U wave.

[0036] Please refer to FIG. 3. FIG. 3 is a waveform diagram for explaining calculating at least one peak distance between a plurality of first characteristic peak positions for each of a plurality of electrocardiogram waveforms. Note that in FIG. 3, as an example, how the peak distance is calculated using only one electrocardiogram waveform is shown, but the peak distance of each of the received plurality of electrocardiogram waveforms can be calculated in the same manner as shown in FIG. 3.

[0037] As shown in FIG. 3, since each of the received plurality of electrocardiogram waveforms has a plurality of electrocardiogram characteristics such as, for example, P wave, Q wave, R wave, S wave, T wave, and U wave, the peak distance can be calculated based on the plurality of electrocardiogram characteristics. In order to calculate the peak distance, it is necessary to define the specific positions of each of the plurality of electrocardiogram characteristics.

[0038] In some embodiments, the specific position of the P wave refers to the point where the amplitude value changes most significantly in the first voltage change, that is, the maximum value of the P wave; the specific position of the Q wave refers to the point where the amplitude value changes most significantly in the first downward deviation in the second voltage change, that is, the minimum value of the Q wave; the specific position of the R wave refers to the point where the amplitude value changes most significantly in the first upward deviation in the second voltage change, that is, the maximum value of the R wave; the specific position of the S wave refers to the point where the amplitude value changes most significantly in the second downward deviation in the second voltage change, that is, the minimum value of the S wave; the specific position of the T wave refers to the point where the amplitude value changes most significantly in the third voltage change, that is, the maximum absolute value of the T wave; and the specific position of the U wave may refer to the point where the amplitude value changes most significantly in the fourth voltage change, that is, the maximum value of the U wave.

[0039] As described above, the peak distance between the P wave and the Q wave is the difference between the specific positions of the P wave and the Q wave, that is, the time difference between the maximum value of the P wave and the minimum value of the Q wave. The peak distance between the Q wave and the S wave is the difference between the specific positions of the Q wave and the S wave, that is, the time difference between the minimum value of the Q wave and the minimum value of the S wave. The peak distance between the S wave and the T wave may be the difference between the specific positions of the S wave and the T wave, that is, the time difference between the minimum value of the S wave and the maximum value of the absolute value of the T wave.

[0040] For example, as shown in FIG. 3, when the time point at which the maximum value of the R wave occurs is 13:44:20.234 and the time point at which the maximum value of the absolute value of the T wave occurs is 13:44:20.385, the calculation result of the peak distance between the R wave and the T wave is 0.151 seconds (that is, the time difference between the maximum value of the R wave and the maximum value of the absolute value of the T wave).

[0041] Therefore, the peak distance between each of the plurality of electrocardiogram features can be calculated by calculating the time difference between two electrocardiogram features, and the unit of the peak distance may be seconds or milliseconds.

[0042] Please refer to FIG. 4. FIG. 4 is a schematic diagram illustrating an example of a computing device 400A that is signal-connected to an electrocardiogram measurement device 300 and signal-connected to at least one of a machine learning model 500, a neural network model 510, and a convolutional neural network model 520 according to an embodiment of the present disclosure.

[0043] Since the computing device 400A is electrically connected to the electrocardiogram measurement device 300, the computing device 400A may receive a plurality of electrocardiogram waveforms of the user from the electrocardiogram measurement device 300. In some embodiments, the computing device 400A may receive electrocardiogram waveforms of a plurality of users from the electrocardiogram measurement device 300 via a physical signal line connection. For example, the physical signal line connection may be, but is not limited to, a network signal line connection compliant with the Internet Protocol (IP). Further, in some embodiments, the computing device 400A may receive electrocardiogram waveforms of a plurality of users from the electrocardiogram measurement device 300 via a virtual signal line connection. For example, the virtual signal line connection may be, but is not limited to, a Wi-Fi connection compliant with a wireless network protocol.

[0044] In some embodiments, the computing device 400A may be electrically connected to another device (e.g., a wearable measurement device) that can provide a plurality of electrocardiogram waveforms, and receive a plurality of electrocardiogram waveforms of the user from the other device (not shown).

[0045] The computing device 400A includes a receiving unit 402, an extracting unit 404, a determining unit 406, a calculating unit 408, and an estimating unit 410, and the computing device 400A may perform a series of specific signal processes on the received plurality of electrocardiogram waveforms of the user. That is, after the computing device 400A receives a plurality of electrocardiogram waveforms of the user, the computing device 400A may estimate the blood glucose level of the user based on the plurality of electrocardiogram waveforms of the user.

[0046] The receiving unit 402 may be configured to receive a plurality of electrocardiogram waveforms of the user. That is, the computing device 400A may receive a plurality of electrocardiogram waveforms of the user from the electrocardiogram measurement device 300 or another device that can provide a plurality of electrocardiogram waveforms via a wired or wireless signal transmission path by the receiving unit 402.

[0047] The extraction unit 404 may be configured to extract a plurality of electrocardiogram features from each of the plurality of electrocardiogram waveforms. That is, the extraction unit 404 may extract a plurality of electrocardiogram features from each of the plurality of electrocardiogram waveforms in time series for each of the plurality of electrocardiogram waveforms. The plurality of extracted electrocardiogram features may include, but are not limited to, P waves, Q waves, R waves, S waves, T waves, and U waves (as shown in FIG. 2). In some embodiments, the plurality of first electrocardiogram features, the plurality of second electrocardiogram features, and / or the plurality of third electrocardiogram features may be respectively extracted by the extraction unit 404 from each of the plurality of electrocardiogram waveforms. Therefore, taking the plurality of first electrocardiogram features as an example, the extraction unit 404 may extract the plurality of first electrocardiogram features from each of the plurality of electrocardiogram waveforms, and the plurality of first electrocardiogram features may include P waves, Q waves, R waves, S waves, T waves, and U waves. In some embodiments, the extraction unit 404 may input a plurality of electrocardiogram waveforms of a user into the convolutional neural network model 520, and each of the plurality of electrocardiogram features may be extracted by the convolutional neural network model 520.

[0048] The determination unit 406 may be configured to determine a feature peak position corresponding to each of the plurality of electrocardiogram features. In some embodiments, the determination unit 406 may respectively determine a first feature peak position corresponding to each of the plurality of first electrocardiogram features, and the plurality of first electrocardiogram features may include P waves, Q waves, R waves, S waves, T waves, and U waves. For example, the determination unit 406 may respectively determine the position of the maximum value of the P wave, the position of the minimum value of the Q wave, the position of the maximum value of the R wave, the position of the minimum value of the S wave, the position of the maximum absolute value of the T wave, and the position of the maximum value of the U wave.

[0049] The calculation unit 408 may be configured to calculate the peak distances between a plurality of characteristic peak positions. In some embodiments, the calculation unit 408 calculates the peak distances between two of the plurality of first characteristic peak positions, for example, the peak distance between the position of the maximum value of the P wave and the position of the minimum value of the Q wave, the peak distance between the position of the minimum value of the Q wave and the position of the minimum value of the S wave, and / or the peak distance between the position of the minimum value of the S wave and the position of the maximum value of the absolute value of the T wave, but is not limited thereto. The calculation unit 408 may calculate the peak distances between any two of the plurality of first characteristic peak positions. For example, when the plurality of first characteristic peak positions include the position of the maximum value of the P wave, the position of the minimum value of the Q wave, the position of the maximum value of the R wave, the position of the minimum value of the S wave, the position of the maximum value of the absolute value of the T wave, and the position of the maximum value of the U wave, the calculation unit 408 may select any two of them to calculate the peak distance.

[0050] The estimation unit 410 may be configured to estimate the blood glucose level of the user based on at least one peak distance. That is, the estimation unit 410 may estimate the blood glucose level of the user based on the calculation results of the respective peak distances. Specifically, by inputting the calculation results of the respective peak distances into an estimation model, the blood glucose level of the user is estimated by the estimation model. In some embodiments, in the estimation unit 410, the calculation results of the respective peak distances may be input into the machine learning model 500, and the blood glucose level of the user may be estimated by the machine learning model 500. In other embodiments, the estimation unit 410 may input the calculation results of the respective peak distances into the neural network model 510, and the blood glucose level of the user may be estimated by the neural network model 510.

[0051] In some embodiments, the estimation unit 410 may be configured to estimate the user's blood glucose level based on at least one peak distance and at least one incidence rate. In other embodiments, the estimation unit 410 may be configured to estimate the user's blood glucose level based on at least one peak distance and at least one amplitude ratio. In other embodiments, the estimation unit 410 may be configured to estimate the user's blood glucose level based on at least one peak distance, at least one incidence rate, and at least one amplitude ratio.

[0052] The machine learning model 500 may be configured to estimate the user's blood glucose level based on at least one peak distance. In some embodiments, the machine learning model 500 may be further configured to estimate the blood glucose level based on at least one incidence rate and / or at least one amplitude ratio. The machine learning model 500 may include or utilize a machine learning algorithm. In some embodiments, the machine learning model 500 may be a supervised learning model, and since the machine learning model 500 is trained with a plurality of electrocardiogram data, the user's blood glucose level can be estimated by the machine learning model 500. In some embodiments, each of the plurality of electrocardiogram data may include at least one peak distance of each of the plurality of electrocardiogram waveforms and a corresponding blood glucose value, but is not limited thereto. In some embodiments, in order for the machine learning model 500 to be able to estimate the user's blood glucose level based on at least one peak distance, the machine learning model 500 may be implemented by a decision tree algorithm, a random forest algorithm incorporating bagging, an extreme gradient boosting (XGBoost) algorithm, or a support vector machine algorithm.

[0053] In some embodiments, the machine learning model 500 may be a semi-supervised learning model or an unsupervised learning model. For example, the machine learning model 500 may be implemented by CycleGAN.

[0054] In some embodiments, the machine learning model 500 may be integrated into the computing device 400A. That is, the computing device 400A may store algorithms related to the machine learning model 500 and implement the functions of the machine learning model 500 by executing the algorithms stored in the computing device 400A.

[0055] The neural network model 510 may be configured to estimate a user's blood glucose level based on at least one peak distance. In some embodiments, the neural network model 510 may be further configured to estimate the blood glucose level based on at least one incidence rate and / or at least one amplitude ratio. The neural network model 510 may include an input layer, a hidden layer, and an output layer. The input layer of the neural network model 510 may be configured to receive a plurality of electrocardiogram features input from outside the neural network model 510, and the number of nodes in the input layer may depend on the plurality of extracted electrocardiogram features. For example, when the plurality of electrocardiogram features extracted from each of the plurality of electrocardiogram waveforms include P wave, Q wave, R wave, S wave, T wave, and U wave, the number of nodes in the input layer may be, for example, 6. The hidden layer of the neural network model 510 is located between the input layer and the output layer and may be configured to perform a series of operation processes on the variables of the nodes in the input layer. In some embodiments, in order to increase the complexity of the calculation of the neural network model 510, the number of layers of the hidden layer may be configured to be at least 3 layers. The number of nodes in each layer of the hidden layer may be more than, less than, or the same as the number of nodes in the previous layer, which depends on the complexity of the calculation. For example, when the neural network model 510 is configured to have an input layer, three hidden layers, and an output layer, the number of nodes in the first hidden layer may be more than the number of nodes in the input layer, the number of nodes in the second hidden layer may be less than the number of nodes in the first hidden layer, and the number of nodes in the third hidden layer may be less than the number of nodes in the second hidden layer. More specifically, when the number of nodes in the input layer is arranged to be 6, the number of nodes in the first hidden layer may be greater than 6. In some embodiments, the activation function of the hidden layer may be, but is not limited to, a ReLU function, a Tanh function, a sigmoid function, etc. In some embodiments, the hidden layer may be normalized so that the value input to the hidden layer is guaranteed to be between the sensing regions of the activation function.The output layer of the neural network model 510 is configured to output the final calculation result, and the final calculation result may be obtained by executing a series of calculation processes by the hidden layer. That is, after receiving each of a plurality of electrocardiogram features, the neural network model 510 executes a series of calculation processes by the hidden layer between the input layer and the output layer, so the output layer of the neural network model 510 can be configured to output the user's blood glucose value (that is, the final calculation result).

[0056] In some embodiments, the neural network model 510 may be integrated into the computing device 400A. That is, the computing device 400A may store the algorithm related to the neural network model 510, and may realize the function of the neural network model 510 by executing the algorithm stored in the computing device 400A.

[0057] The convolutional neural network model 520 may be configured to extract a plurality of electrocardiogram features based on a plurality of electrocardiogram waveforms. In some embodiments, the convolutional neural network model 520 may include a convolutional layer configured to extract a plurality of electrocardiogram features from a plurality of electrocardiogram waveforms, and a fully connected layer configured to have substantially the same configuration and purpose as the neural network model 510. In some embodiments, the number of convolutional layers may be configured to be at least three layers. In some embodiments, since the fully connected layer of the convolutional neural network model 520 has substantially the same configuration as the neural network model 510, the user's blood glucose value may also be estimated by the convolutional neural network model 520 (that is, each of the plurality of electrocardiogram features is extracted by the convolutional layer, and the user's blood glucose value is estimated by the fully connected layer).

[0058] In some embodiments, the convolutional neural network model 520 may be integrated into the computing device 400A. That is, the computing device 400A may store algorithms related to the convolutional neural network model 520, and may implement the functions of the convolutional neural network model 520 by executing the algorithms stored in the computing device 400A.

[0059] In some embodiments, the computing device 400A may include an electrocardiogram waveform database 452 and a blood glucose level database 454. The electrocardiogram waveform database 452 may be configured to store a plurality of electrocardiogram waveforms of the user. That is, the computing device 400A may first store the plurality of electrocardiogram waveforms of the user received from the electrocardiogram measurement device 300 in the electrocardiogram waveform database 452, and then the receiving unit 402 may receive the plurality of electrocardiogram waveforms of the user from the electrocardiogram waveform database 452. The blood glucose level database 454 may be configured to store the blood glucose levels of the user. That is, after the estimated blood glucose level of the user and / or the calculated corrected blood glucose level of the user are obtained, the computing device 400A may store the estimated blood glucose level of the user and / or the corrected blood glucose level of the user in the blood glucose level database 454.

[0060] Thus, the computing device 400A provided by the present disclosure is used to implement the method for non-invasively estimating blood glucose levels described below. After receiving a plurality of electrocardiogram waveforms of the user, the computing device 400A can estimate the blood glucose level of the user and / or calculate the corrected blood glucose level of the user based on the plurality of electrocardiogram waveforms of the user.

[0061] According to the computing device 400A shown in FIG. 4, not only can means for non-invasively estimating the user's blood glucose level be provided to the user, but the user's blood glucose level can also be estimated more accurately. Furthermore, the computing device 400A shown in FIG. 4 can improve the technical field related to non-invasively estimating the user's blood glucose level by utilizing a plurality of electrocardiogram waveforms.

[0062] Please refer to FIG. 5. FIG. 5 is a schematic diagram for explaining an example of the calculation unit 408 that is signal-connected to the determination unit 406 and the estimation unit 410 according to an embodiment of the present disclosure. In some embodiments, the calculation unit 408 may be configured to calculate at least one peak distance between a plurality of characteristic peak positions. Furthermore, the calculation unit 408 may be configured to calculate results such as at least one incidence rate, at least one amplitude ratio, a corrected blood glucose value, a peak-to-peak slope, and / or at least one sharpness, but is not limited thereto. Therefore, the calculation unit 408 may include a peak interval calculator 502, and may further include an incidence rate calculator 504, an amplitude ratio calculator 506, a corrected blood glucose value calculator 508, a peak-to-peak slope calculator, and / or a sharpness result calculator.

[0063] The peak distance calculator 502 may be configured to calculate at least one peak distance between a plurality of characteristic peak positions. That is, the peak distance calculator 502 may be configured to calculate the peak distance between two of the plurality of characteristic peak positions, for example, the time difference between the position of the maximum value of the P wave and the position of the minimum value of the Q wave (as shown in FIG. 3), but is not limited thereto.

[0064] The incidence rate calculator 504 may be configured to calculate at least one incidence rate of each electrocardiogram feature of a plurality of electrocardiogram waveforms respectively. That is, the incidence rate calculator 504 may be configured to calculate the ratio of the occurrence of the second electrocardiogram feature in each electrocardiogram waveform, for example, the incidence rate of the P wave, but is not limited thereto.

[0065] The amplitude ratio calculator 506 may be configured to calculate the ratio of the amplitude values between each of a plurality of electrocardiogram features. That is, the amplitude ratio calculator 506 may be configured to calculate the peak-to-peak ratio between each of a plurality of third electrocardiogram features, for example, the amplitude ratio of the P wave, but is not limited thereto.

[0066] The corrected blood glucose value calculator 508 may be configured to calculate the corrected blood glucose value of the user. That is, after the blood glucose value of the user is estimated by the estimation unit 410, the corrected blood glucose value calculator 508 of the calculation unit 408 receives the estimated blood glucose value of the user from the estimation unit 410, and further performs a correction operation on the estimated blood glucose value of the user by an equation for correcting the blood glucose value, and may calculate the corrected blood glucose value of the user.

[0067] Please refer to FIG. 6. FIG. 6 is a flowchart for explaining a first example of a method for non-invasively estimating a blood glucose value according to an embodiment of the present disclosure. The method shown in FIG. 6 may include steps S610, S620, S630, S640, and S650.

[0068] In step S610, a plurality of electrocardiogram waveforms of the user are received. Step S610 may be executed by the receiving unit 402 of the computing device 400A shown in FIG. 4. In some embodiments, by executing step S610, a plurality of electrocardiogram waveforms of the user may be received in real time from, for example, an electrocardiogram measurement device 300, an electrocardiogram sensor (not shown), or a wearable device (not shown), but are not limited thereto. In other embodiments, by executing step S610, a plurality of electrocardiogram waveforms of the user may be received from the electrocardiogram waveform database 452, and the plurality of electrocardiogram waveforms of the user may be received indirectly.

[0069] In some embodiments, the number of a plurality of electrocardiogram waveforms may depend on the electrocardiogram reception duration. That is, in step S610, a plurality of electrocardiogram waveforms of the user may be received during the electrocardiogram reception duration, and the electrocardiogram reception duration is set to at least 1 minute so that a sufficient number of a plurality of electrocardiogram waveforms are surely received, whereby the blood glucose level of the user can be estimated, and the blood glucose level can be estimated and / or calculated more accurately.

[0070] In step S620, at least two first electrocardiogram features are extracted from each of the plurality of electrocardiogram waveforms of the user. Step S620 may be executed by the extraction unit 404 of the computing device 400A shown in FIG. 4. In some embodiments, step S620 may be executed following the execution of step S610. By executing step S620, at least two first electrocardiogram features may be extracted from each of the plurality of electrocardiogram waveforms, and the first electrocardiogram features may be selectable from the group consisting of P wave, Q wave, R wave, S wave, T wave, and U wave.

[0071] In some examples, by executing step S620, the P wave and Q wave of each of the plurality of electrocardiogram waveforms may be extracted as the first electrocardiogram features, respectively. In other embodiments, by executing step S620, the S wave and T wave of each of the plurality of electrocardiogram waveforms may be extracted as the first electrocardiogram features, respectively. In other embodiments, by executing step S620, the P wave, Q wave, R wave, S wave, T wave, and U wave of each of the plurality of electrocardiogram waveforms may be extracted as the first electrocardiogram features, respectively. That is, by executing step S620, at least two of the P wave, Q wave, R wave, S wave, T wave, and U wave of each of the plurality of electrocardiogram waveforms may be extracted as the first electrocardiogram features.

[0072] In some embodiments, the first electrocardiogram feature may be extracted by the convolutional neural network model 520. That is, after a plurality of electrocardiogram waveforms of a user are input into the convolutional neural network model 520, the first electrocardiogram feature may be extracted by the convolutional layer of the convolutional neural network model 520.

[0073] In step S630, a first feature peak position corresponding to each of the plurality of first electrocardiogram features is determined respectively. Step S630 may be executed by the determination unit 406 of the computing device 400A shown in FIG. 4. In some embodiments, step S630 may be executed following the execution of step S620. The first feature peak position corresponding to each of the plurality of first electrocardiogram features may be determined respectively by executing step S630, and the first feature peak position is substantially the same as that described in FIG. 3.

[0074] In some examples, when the extracted first electrocardiogram features are P waves and Q waves, the position of the maximum value of the P wave and the position of the minimum value of the Q wave in each of the plurality of electrocardiogram waveforms may be determined respectively by executing step S630. In other embodiments, when the extracted first electrocardiogram features are S waves and T waves, the position of the minimum value of the S wave and the position of the maximum value of the absolute value of the T wave in each of the plurality of electrocardiogram waveforms may be determined respectively by executing step S630. In other embodiments, when the extracted first electrocardiogram features are P waves, Q waves, R waves, S waves, T waves, and U waves, the position of the maximum value of the P wave, the position of the minimum value of the Q wave, the position of the maximum value of the R wave, the position of the minimum value of the S wave, the position of the maximum value of the absolute value of the T wave, and the position of the maximum value of the U wave in each of the plurality of electrocardiogram waveforms may be determined respectively by executing step S630.

[0075] Since the first characteristic peak position is the position of the maximum or minimum value of the first electrocardiogram characteristic, in some embodiments, the method for finding the first characteristic peak position may be realized by continuously comparing the numerical values of two adjacent points. In other embodiments, the method for finding the first characteristic peak position may be realized by calculating the slope of the tangent line of each point.

[0076] Compared with the existing technology, the determination and utilization of the first characteristic peak position become easier. This contributes to the estimation of the user's blood glucose level.

[0077] In step S640, for each of the plurality of electrocardiogram waveforms, at least one peak distance between the plurality of first characteristic peak positions is calculated. Step S640 may be executed by the peak distance calculator 502 of the calculation unit 408 shown in FIG. 5. In some embodiments, step S640 may be executed following the execution of step S630. By executing step S640, the time difference between any two of the plurality of first characteristic peak positions is calculated as the peak distance, and the peak distance is substantially the same as that described in FIG. 3.

[0078] In some examples, when the extracted first electrocardiogram features are the P wave and the Q wave, the time difference between the position of the maximum value of the P wave and the position of the minimum value of the Q wave is calculated as the peak distance by executing step S640. In other embodiments, when the extracted first electrocardiogram features are the S wave and the T wave, the time difference between the position of the minimum value of the S wave and the position of the maximum value of the absolute value of the T wave is calculated as the peak distance by executing step S640. In other embodiments, when the extracted first electrocardiogram features are the P wave, Q wave, R wave, S wave, T wave, and U wave, the time difference between any two positions among the position of the maximum value of the P wave, the position of the minimum value of the Q wave, the position of the maximum value of the R wave, the position of the minimum value of the S wave, the position of the maximum value of the absolute value of the T wave, and the position of the maximum value of the U wave is calculated as the peak distance by executing step S640. That is, the peak distance may include, but is not limited to, the time difference between the P wave and the Q wave, the time difference between the P wave and the R wave, the time difference between the P wave and the S wave, the time difference between the P wave and the T wave, the time difference between the P wave and the U wave, the time difference between the Q wave and the R wave, the time difference between the Q wave and the S wave, the time difference between the Q wave and the T wave, the time difference between the Q wave and the U wave, the time difference between the R wave and the S wave, the time difference between the R wave and the T wave, the time difference between the R wave and the U wave, the time difference between the S wave and the T wave, the time difference between the S wave and the U wave, and / or the time difference between the T wave and the U wave.

[0079] In step S650, the user's blood glucose level is estimated based on the peak distance. Step S650 may be executed by the estimation unit 410 of the computing device 400A shown in FIG. 4. In some embodiments, step S650 may be executed subsequently after step S640 is executed. By executing step S650, the user's blood glucose level is estimated based on the calculation of the peak distance. More specifically, after the calculation of the peak distance is input into the estimation model, the user's blood glucose level is estimated by the estimation model.

[0080] According to the method for non-invasively estimating a blood glucose level shown in FIG. 6, not only can a means for non-invasively estimating the user's blood glucose level be provided to the user, but also the user's blood glucose level can be estimated more accurately based on the peak distance as described above. Furthermore, the method for non-invasively estimating a blood glucose level shown in FIG. 6 can improve the technical field related to non-invasively estimating a user's blood glucose level by using a plurality of electrocardiogram waveforms.

[0081] Please refer to FIG. 7A. FIG. 7A is a detailed flowchart illustrating a method for estimating a user's blood glucose level based on a peak distance according to an embodiment of the present disclosure. That is, step S650 shown in FIG. 6 includes steps S710A, S720A, S730A, and S740A, and may be completed by executing steps S710A, S720A, S730A, and S740A, and steps S710A, S720A, S730A, and S740A may be executed by the estimation unit 410 of the computing device 400A shown in FIG. 4.

[0082] In step S710A, the peak distance is normalized. In some embodiments, step S710A may be executed subsequent to the execution of step S640. By executing step S710A, the calculation of the peak distance may be normalized to corresponding values within a range (for example, between 0 and 1).

[0083] In step S720A, the normalized peak distance is input into the machine learning model 500. In some embodiments, step S720A may be executed subsequent to the execution of step S710A. By executing step S720A, the normalized peak distance may be input into the machine learning model 500 such as, for example, a decision tree algorithm, a random forest algorithm incorporating bagging, an extreme gradient boosting (XGBoost) algorithm, or a support vector machine algorithm.

[0084] In step S730A, the user's blood glucose level is estimated by the machine learning model 500. In some embodiments, step S730A may be executed following the execution of step S720A. By executing step S730A, after the normalized peak distance is input into the machine learning model 500, the user's blood glucose level may be estimated based on the normalized peak distance. Since the machine learning model 500 is trained with a plurality of data and verified with prediction results, the user's blood glucose level can be estimated by the machine learning model 500 with a prediction accuracy of 0.80 or higher, preferably 0.90 or higher.

[0085] In step S740A, the user's blood glucose level is output by the machine learning model 500. In some embodiments, step S740A may be executed following the execution of step S730A. By executing step S740A, the user's estimated blood glucose level may be output by the machine learning model 500 to the display device and / or the database. In some examples, the user's estimated blood glucose level may be output to the blood glucose level database 454 of the computing device 400A and stored in the blood glucose level database 454 synchronously.

[0086] Please refer to FIG. 8A. FIG. 8A is a schematic diagram illustrating an example of the machine learning model 500 according to an embodiment of the present disclosure. As shown in FIG. 8A, the machine learning model 500 may be implemented by an Extreme Gradient Boosting (XGBoost) algorithm. The XGBoost algorithm is designed by a plurality of decision trees, and the result of the previous decision tree affects the subsequent decision tree. Therefore, each of the plurality of decision trees is related to each other, thereby making the prediction result of the machine learning model 500 more accurate. In some embodiments, the machine learning model 500 may be implemented by other algorithms such as, for example, a decision tree algorithm, a random forest algorithm incorporating Bagging, an Extreme Gradient Boosting (XGBoost) algorithm, or a support vector machine algorithm.

[0087] Refer to FIG. 7B. FIG. 7B is a detailed flowchart illustrating a method for estimating a user's blood glucose level based on peak distance according to an embodiment of the present disclosure. That is, step S650 shown in FIG. 6 includes steps S710B, S720B, S730B, and S740B, and may be completed by executing steps S710B, S720B, S730B, and S740B. Steps S710B, S720B, S730B, and S740B may be executed by the estimation unit 410 of the computing device 400A shown in FIG. 4.

[0088] In step S710B, the peak distance is normalized. In some embodiments, step S710B may be executed following the execution of step S640. In some embodiments, step S710B may be substantially the same as step S710A.

[0089] In step S720A, the normalized peak distance is input into the neural network model 510. In some embodiments, step S720B may be executed following the execution of step SS710B. By executing step S720A, the normalized peak distance is input into the neural network model 510.

[0090] In step S730B, the user's blood glucose level is estimated by the neural network model 510. In some embodiments, step S730B may be executed following the execution of step S720B. By executing step S730B, after the normalized peak distance is input into the neural network model 510, the user's blood glucose level may be estimated based on the normalized peak distance. Since the neural network model 510 is trained with a plurality of data and verified with prediction results, the user's blood glucose level can be estimated by the neural network model 510 with a prediction accuracy of 0.80 or higher, preferably 0.90 or higher.

[0091] In step S740B, the user's blood glucose level is output to the neural network model 510. In some embodiments, step S740B may be executed subsequent to the execution of step S730B. By executing step S740B, the estimated blood glucose level of the user may be output by the neural network model 510 to the display device and / or the database. In some examples, the estimated blood glucose level of the user may be output to the blood glucose level database 454 of the computing device 400A and stored in the blood glucose level database 454 synchronously.

[0092] Please refer to FIG. 8B. FIG. 8B is a schematic diagram illustrating an example of the neural network model 510 according to an embodiment of the present disclosure. The neural network model 510 may include an input layer IL, at least three hidden layers HL, and an output layer OL. In some embodiments, after receiving the peak distance, the neural network model 510 may execute a series of arithmetic processes by each node of the hidden layer HL to estimate the user's blood glucose level, and output the estimated blood glucose level of the user by the output layer OL. In some embodiments, the neural network model 510 may directly receive information of each of a plurality of electrocardiogram features, and estimate the user's blood glucose level by executing a series of arithmetic processes on the information of each of the plurality of electrocardiogram features by the hidden layer HL.

[0093] In some embodiments, each electrocardiogram feature of the plurality of electrocardiogram waveforms may be extracted by the convolutional neural network model 520 respectively. Also, since the fully connected layer of the convolutional neural network model 520 has substantially the same configuration as the neural network model 510, after extracting each electrocardiogram feature of the plurality of electrocardiogram waveforms, the convolutional neural network model 520 may directly estimate the user's blood glucose level based on each of the plurality of electrocardiogram waveforms, particularly based on the peak distance between the plurality of electrocardiogram features.

[0094] Please refer to FIG. 8C. FIG. 8C is a schematic diagram illustrating an example of a convolutional neural network model 520 according to an embodiment of the present disclosure. After receiving a plurality of electrocardiogram waveforms EW of a user, the convolutional neural network model 520 can extract each of a plurality of electrocardiogram features by a convolutional layer CL. Please refer to FIG. 8D simultaneously. FIG. 8D is a schematic diagram illustrating an example of the convolutional layer CL according to an embodiment of the present disclosure. As shown in FIG. 8D, the convolutional neural network model 520 may include at least three (e.g., three) convolutional layers CL.

[0095] Please refer to FIG. 9. FIG. 9 is a flowchart illustrating a second example of a method for non-invasively estimating a blood glucose level according to an embodiment of the present disclosure. The method shown in FIG. 9 may include steps S610, S620, S630, S640, S650, and S910, wherein steps S610, S620, S630, S640, and S650 are substantially the same as the steps shown in FIG. 6.

[0096] In step S910, a corrected blood glucose level of the user is calculated based on the blood glucose level of the user. Step S910 may be executed by a corrected blood glucose level calculator 508 of the calculator 408 shown in FIG. 5. In some embodiments, step S910 may be executed subsequent to the execution of step S650. By executing step S910, the corrected blood glucose level of the user is calculated by an equation using the corrected blood glucose level based on the blood glucose level of the user estimated by the estimator 410.

[0097] In some embodiments, the equation for the corrected blood glucose level is, for example,

[0098]

Equation

[0099] It may be a cubic equation of one variable such as this, but it is not limited to this. Each parameter (a, b, c, d) in the corrected blood glucose value equation may be determined based on the estimated blood glucose value (variable x) of the user and the actual blood glucose value (variable y) of the user measured by the blood glucose meter.

[0100] As an example, when the estimated blood glucose values of the user are 90, 135, and 180 in sequence, and the actual blood glucose values of the user measured by the blood glucose meter are 85, 120, and 190 in sequence, the values of each parameter (a, b, c, d) are known by solving a system of simultaneous equations.

[0101]

Number

[0102] Therefore, the corrected blood glucose value calculator 508 may correct the estimated blood glucose value of the user by performing a calculation operation on the estimated blood glucose value of the user according to

[0103]

Number

[0104] the corrected blood glucose value equation.

[0105] In some embodiments, the corrected blood glucose value equation may be, for example,

[0106]

Number

[0107] a quadratic equation of one variable such as this, but it is not limited to this. Each parameter (a, b, c) in the equation may be determined based on the estimated blood glucose value (variable x) of the user and the actual blood glucose value (variable y) of the user measured by the blood glucose meter.

[0108]

Number

[0109] Therefore, the corrected blood glucose value calculator 508 may correct the estimated blood glucose value of the user by performing a calculation operation on the estimated blood glucose value of the user according to the equation of the corrected blood glucose value

[0110] [Number]

[0111] By performing a calculation operation on the estimated blood glucose value of the user according to [Equation], the estimated blood glucose value of the user may be corrected.

[0112] According to the method for non-invasively estimating the blood glucose value shown in FIG. 9, not only can means for non-invasively estimating the blood glucose value of the user be provided to the user, but also a blood glucose value (i.e., the corrected blood glucose value of the user) that is the same as or close to the actual measurement result by the blood glucose meter can be calculated using the equation of the corrected blood glucose value. Thus, in the method shown in FIG. 9, the difference between the estimated result of the blood glucose value and the actual measurement result by the blood glucose meter can be compensated, and a blood glucose value that is substantially the same as or close to the actual measurement result by the blood glucose meter can be provided to the user.

[0113] In some embodiments, when the blood glucose value estimated by the estimation unit 410 is sufficiently accurate (for example, when comparing the estimated blood glucose value with the actual measurement result by the blood glucose meter, the range of the difference is within 0.05), step S910 may be omitted. That is, whether to execute step S910 may depend on the user's requirement for the accuracy of the estimated result of the blood glucose value. When the user desires a more accurate result of the blood glucose value, a more accurate result of the blood glucose value (i.e., the corrected blood glucose value of the user) is provided by the method shown in FIG. 9.

[0114] Please refer to FIG. 10. FIG. 10 is a flowchart for explaining a third example of a method for non-invasively estimating a blood glucose level according to an embodiment of the present disclosure. The method shown in FIG. 10 may include steps S610, S620, S630, S640, S1010, and S1020, wherein steps S610, S620, S630, and S640 are substantially the same as the steps shown in FIG. 6.

[0115] In step S1010, at least one peak-to-peak slope between a plurality of first characteristic peak positions of each of the plurality of electrocardiogram waveforms is calculated. Step S1010 may be executed by the arithmetic unit 408 of the computing device 400A shown in FIG. 4. In some embodiments, step S910 may be executed subsequent to the execution of step S630. Further, step S1010 may be executed simultaneously with step S640.

[0116] In some examples, when the extracted first electrocardiogram features are the P wave and the Q wave, the slope between the position of the maximum value of the P wave and the position of the minimum value of the Q wave is calculated as the peak-to-peak slope by executing step S1010. In other embodiments, when the extracted first electrocardiogram features are the S wave and the T wave, the slope between the position of the minimum value of the S wave and the position of the maximum value of the absolute value of the T wave is calculated as the peak-to-peak slope by executing step S640. In other embodiments, when the extracted first electrocardiogram features are the P wave, Q wave, R wave, S wave, T wave, and U wave, the slope between any two positions among the position of the maximum value of the P wave, the position of the minimum value of the Q wave, the position of the maximum value of the R wave, the position of the minimum value of the S wave, the position of the maximum value of the absolute value of the T wave, and the position of the maximum value of the U wave is calculated as the peak-to-peak slope by executing step S640. That is, the peak-to-peak slope may include, but is not limited to, the slope between the P wave and the Q wave, the slope between the P wave and the R wave, the slope between the P wave and the S wave, the slope between the P wave and the T wave, the slope between the P wave and the U wave, the slope between the Q wave and the R wave, the slope between the Q wave and the S wave, the slope between the Q wave and the T wave, the slope between the Q wave and the U wave, the slope between the R wave and the S wave, the slope between the R wave and the T wave, the slope between the R wave and the U wave, the slope between the S wave and the T wave, the slope between the S wave and the U wave, and / or the slope between the T wave and the U wave.

[0117] In step S1020, the user's blood glucose value is estimated based on the peak distance and the peak-to-peak slope. Step S1020 may be executed by the estimation unit 410 of the computing device 400A shown in FIG. 4. In some embodiments, step S1130 may be executed subsequent to the execution of steps S640 and S1010. By executing step S1020, the user's blood glucose value is estimated based on the calculation result of the peak distance and the calculation result of the peak-to-peak slope. More specifically, after the calculation results of the peak distance and the peak-to-peak slope are input into the estimation model, the user's blood glucose value is estimated by the estimation model.

[0118] According to the method for non-invasively estimating blood glucose levels shown in FIG. 10, not only can a means for non-invasively estimating the user's blood glucose level be provided to the user, but the user's blood glucose level can also be estimated more accurately based on the peak distance and peak-to-peak slope as described above. Furthermore, the method for non-invasively estimating blood glucose levels shown in FIG. 10 can improve the technical field related to non-invasively estimating the user's blood glucose level by using a plurality of electrocardiogram waveforms.

[0119] Also, in some embodiments, the method for non-invasively estimating blood glucose levels shown in FIG. 10 may further include step S910 shown in FIG. 9, and the corrected blood glucose level of the user may be calculated by executing step S910.

[0120] Please refer to FIG. 11. FIG. 11 is a flowchart illustrating a fourth example of a method for non-invasively estimating blood glucose levels according to an embodiment of the present disclosure. The method shown in FIG. 11 may include steps S610, S620, S630, S640, S1110, S1120, and S1130, where steps S610, S620, S630, and S640 are substantially the same as the steps shown in FIG. 6.

[0121] In step S1110, at least three second electrocardiogram features are extracted from each of the user's plurality of electrocardiogram waveforms. Step S1110 may also be executed by the extraction unit 404 of the computing device 400A shown in FIG. 4. In some embodiments, step S1110 may be executed following the execution of step S610. Furthermore, step S1110 may be executed simultaneously with step S620. By executing step S1110, at least three second electrocardiogram features may be extracted from each of the plurality of electrocardiogram waveforms, and the second electrocardiogram features may be selectable from the group consisting of P wave, Q wave, R wave, S wave, T wave, and U wave.

[0122] In some examples, by executing step S1110, each of the P waves, Q waves, and R waves of a plurality of electrocardiogram waveforms is extracted as a second electrocardiogram feature. In other embodiments, by executing step S1110, each of the S waves, T waves, and U waves of a plurality of electrocardiogram waveforms may be extracted as a second electrocardiogram feature. In other embodiments, by executing step S1110, each of the P waves, Q waves, R waves, S waves, T waves, and U waves of a plurality of electrocardiogram waveforms may be extracted as a second electrocardiogram feature. That is, by executing step S1110, at least three of the P waves, Q waves, R waves, S waves, T waves, and U waves of a plurality of electrocardiogram waveforms may be extracted as second electrocardiogram features.

[0123] In some embodiments, the second electrocardiogram feature may be extracted by the convolutional neural network model 520. That is, after a plurality of electrocardiogram waveforms of a user are input into the convolutional neural network model 520, the second electrocardiogram feature may be extracted by the convolutional layer of the convolutional neural network model 520.

[0124] In step S1120, at least one occurrence rate is calculated based on the second electrocardiogram feature. Step S1120 may be executed by the occurrence rate calculator 504 of the calculation unit 408 shown in FIG. 5. In some embodiments, step S1120 may be executed following the execution of step S1110. By executing step S1120, at least one occurrence rate, for example, the occurrence rate of the P wave, may be calculated based on a plurality of second electrocardiogram features, but is not limited thereto. Step S1120 may be a step of calculating at least one appearance rate based on the amplitude values of a plurality of second electrocardiogram features (for example, the maximum value, the minimum value, the root mean square value, or the peak-to-peak value, but is not limited thereto). More specifically, at least one occurrence rate may be calculated based on comparing the amplitude values of a plurality of second electrocardiogram features with a preset value so as to determine whether each of the plurality of second electrocardiogram features occurs in each of the plurality of electrocardiogram waveforms. Further, by comparing the amplitude values of the plurality of second electrocardiogram features with each other, it may be determined whether each of the plurality of second electrocardiogram features occurs in each of the plurality of electrocardiogram waveforms, and at least one occurrence rate may be calculated based on the determined result.

[0125] In some examples, when the extracted second electrocardiogram features are the P wave, Q wave, and R wave, the occurrence rate of the P wave, the occurrence rate of the Q wave, and / or the occurrence rate of the R wave may be calculated by executing step S1120. In other embodiments, when the extracted second electrocardiogram features are the S wave, T wave, and U wave, the occurrence rate of the S wave, the occurrence rate of the T wave, and / or the occurrence rate of the U wave may be calculated by executing step S1120. In other embodiments, when the extracted second electrocardiogram features are the P wave, Q wave, R wave, S wave, T wave, and U wave, the occurrence rate of the P wave, the occurrence rate of the Q wave, the occurrence rate of the R wave, the occurrence rate of the S wave, the occurrence rate of the T wave, and / or the occurrence rate of the U wave may be calculated by executing step S1120.

[0126] In step S1130, the user's blood glucose level is estimated based on the peak distance and the incidence rate. Step S1130 may be executed by the estimation unit 410 of the computing device 400A shown in FIG. 4. In some embodiments, step S1130 may be executed subsequent to the execution of step S640 and step S1120. By executing step S1130, the user's blood glucose level may be estimated based on the calculation result of the peak distance and the calculation result of the incidence rate. More specifically, after the calculation results of the peak distance and the incidence rate are input into the estimation model, the user's blood glucose level is estimated by the estimation model.

[0127] According to the method for non-invasively estimating the blood glucose level shown in FIG. 11, not only can means for non-invasively estimating the user's blood glucose level be provided to the user, but also the user's blood glucose level can be estimated more accurately based on the peak distance and the incidence rate as described above. Furthermore, the method for non-invasively estimating the blood glucose level shown in FIG. 11 can improve the technical field related to non-invasively estimating the user's blood glucose level by using a plurality of electrocardiogram waveforms.

[0128] Also, in some embodiments, the method for non-invasively estimating the blood glucose level shown in FIG. 11 may further include step S910 shown in FIG. 9, and the corrected blood glucose level of the user may be calculated by executing step S910.

[0129] Please refer to FIG. 12. FIG. 12 is a detailed flowchart illustrating a method for calculating at least one incidence rate based on a second electrocardiogram feature according to an embodiment of the present disclosure. That is, step S1120 shown in FIG. 11 includes step S1210, S1220, S1230, S1240, S1250, and S1260, and may be completed by executing step S1210, S1220, S1230, S1240, S1250, and S1260, among which step S1210, S1220, S1230, S1240, S1250, and S1260 may be executed by the incidence rate calculator 504 of the calculation unit 408 shown in FIG. 5.

[0130] In step S1210, the occurrence interval is determined. In some embodiments, step S1210 may be executed following the execution of step S1110. By executing step S1210, the occurrence interval may be determined based on one of the second electrocardiogram features.

[0131] Step S1210 may be a step of determining the occurrence interval by moving forward or backward a specific occurrence interval distance with respect to one of the second electrocardiogram features as a reference point. In some embodiments, the occurrence interval distance may depend on the distance between two adjacent electrocardiogram waveforms, that is, the RR interval.

[0132] In some embodiments, by executing step S1210, the occurrence interval may be determined by using the position of the maximum value of the R wave as a reference point and moving forward by 0.33 times the RR interval. In other embodiments, by executing step S1210, the occurrence interval may be determined by using the position of the maximum value of the R wave as a reference point and moving backward by 0.67 times the RR interval. In other embodiments, the occurrence interval may be determined by using the position of the maximum absolute value of the T wave as a reference point and moving backward by a specific occurrence interval distance (i.e., until encountering a new P wave of the electrocardiogram waveform).

[0133] In step S1220, the first peak-to-peak and the second peak-to-peak within the occurrence interval are calculated respectively. In some embodiments, step S1220 may be executed following the execution of step S1210. By executing step S1220, the first peak-to-peak and the second peak-to-peak within the occurrence interval of each of the plurality of electrocardiogram waveforms may be calculated.

[0134] In some examples, when the extracted second electrocardiogram feature is a P wave, a Q wave, or an R wave, by executing step S1220, for each of the plurality of electrocardiogram waveforms, the peak-to-peak between the maximum value of the P wave and the minimum value of the Q wave within the occurrence interval (that is, forward by 0.33 times the RR interval from the position of the maximum value of the P wave) may be calculated as the first peak-to-peak, and the peak-to-peak between the maximum value of the R wave and the minimum value of the Q wave may be calculated as the second peak-to-peak. In other embodiments, when the extracted second electrocardiogram feature is an R wave, an S wave, or a T wave, by executing step S1220, for each of the plurality of electrocardiogram waveforms, the peak-to-peak between the maximum absolute value of the T wave and the minimum value of the S wave within the occurrence interval (that is, backward by 0.67 times the RR interval from the position of the maximum value of the R wave) may be calculated as the first peak-to-peak, and the peak-to-peak between the maximum value of the R wave and the minimum value of the S wave may be calculated as the second peak-to-peak. In other embodiments, when the extracted second electrocardiogram feature is an S wave, a T wave, or a U wave, by executing step S1220, for each of the plurality of electrocardiogram waveforms, the peak-to-peak between the maximum value of the U wave and the minimum value of the S wave within the occurrence interval (that is, backward by a specific occurrence interval distance from the position of the maximum absolute value of the T wave until a new P wave of the electrocardiogram waveform is encountered) may be calculated as the first peak-to-peak, and the peak-to-peak between the maximum absolute value of the T wave and the minimum value of the S wave may be calculated as the second peak-to-peak.

[0135] In step S1230, the ratio of the first peak-to-peak to the second peak-to-peak is calculated. In some embodiments, step S1230 may be executed subsequent to the execution of step S1220. The ratio of the first peak-to-peak to the second peak-to-peak may be calculated as the ratio of the peak-to-peak for each of the plurality of electrocardiogram waveforms.

[0136] In some examples, when the extracted second electrocardiogram features are the P wave, Q wave, and R wave, the ratio of the peak-to-peak between the P wave and the Q wave to the peak-to-peak between the R wave and the Q wave may be calculated as the peak-to-peak ratio (which may also be referred to as the first peak-to-peak ratio) respectively by executing step S1230. In other embodiments, when the extracted second electrocardiogram features are the R wave, S wave, and T wave, the ratio of the peak-to-peak between the T wave and the S wave to the peak-to-peak between the R wave and the S wave may be calculated as the peak-to-peak ratio (which may also be referred to as the second peak-to-peak ratio) respectively by executing step S1230. In other embodiments, when the extracted second electrocardiogram features are the S wave, T wave, and U wave, the ratio of the peak-to-peak between the U wave and the S wave to the peak-to-peak between the T wave and the S wave may be calculated as the peak-to-peak ratio (which may also be referred to as the third peak-to-peak ratio) respectively by executing step S1230.

[0137] In step S1240, the peak-to-peak ratios are respectively compared with a preset ratio. In some embodiments, step S1240 may be executed following the execution of step S1230. For each of the plurality of electrocardiogram waveforms, the peak-to-peak ratio may be compared with the preset ratio respectively (for example, comparing the values of the two ratios).

[0138] In some examples, when the extracted second electrocardiogram feature is a P wave, a Q wave, or an R wave, by executing step S1240, the ratio of the first peak-to-peak of each of the plurality of electrocardiogram waveforms may be compared with a preset ratio respectively. To determine whether the peak-to-peak between the P wave and the Q wave is greater than the peak-to-peak between the R wave and the Q wave multiplied by the preset ratio, the preset ratio may be a ratio value between 0.05 and 0.15, specifically set to 0.1. In other embodiments, when the extracted second electrocardiogram feature is an R wave, an S wave, or a T wave, by executing step S1240, the ratio of the second peak-to-peak of each of the plurality of electrocardiogram waveforms may be compared with a preset ratio respectively. To determine whether the peak-to-peak between the T wave and the S wave is greater than the peak-to-peak between the R wave and the S wave multiplied by the preset ratio, the preset ratio may be a ratio value between 0.2 and 0.4, specifically set to 0.25. In other embodiments, when the extracted second electrocardiogram feature is an S wave, a T wave, or a U wave, by executing step S1240, the ratio of the third peak-to-peak of each of the plurality of electrocardiogram waveforms may be compared with a preset ratio respectively. To determine whether the peak-to-peak between the U wave and the S wave is greater than the peak-to-peak between the T wave and the S wave multiplied by the preset ratio, the preset ratio may be a ratio value between 0.12 and 0.18, specifically set to 0.15.

[0139] In step S1250, comparison results are generated respectively. In some embodiments, step S1250 may be executed following the execution of step S1240.

[0140] In some examples, when the extracted second electrocardiogram feature is a P wave, a Q wave, or an R wave, by executing step S1250, comparison results of each of a plurality of electrocardiogram waveforms may be generated. Among them, the comparison result may be a result in which the ratio of the first peak-to-peak is greater than a preset ratio, or may be a result in which the ratio of the first peak-to-peak is less than or equal to the preset ratio. In other embodiments, when the extracted second electrocardiogram feature is an R wave, an S wave, or a T wave, by executing step S1250, comparison results of each of a plurality of electrocardiogram waveforms may be generated. Among them, the comparison result may be a result in which the ratio of the second peak-to-peak is greater than a preset ratio, or may be a result in which the ratio of the second peak-to-peak is less than or equal to the preset ratio. In other embodiments, when the extracted second electrocardiogram feature is an S wave, a T wave, or a U wave, by executing step S1250, comparison results of each of a plurality of electrocardiogram waveforms may be generated. Among them, the comparison result may be a result in which the ratio of the third peak-to-peak is greater than a preset ratio, or may be a result in which the ratio of the third peak-to-peak is less than or equal to the preset ratio.

[0141] In step S1260, an incidence rate is calculated based on a plurality of comparison results. In some embodiments, step S1260 may be executed subsequent to the execution of step S1250. By executing step S1260, the incidence rate may be calculated based on the comparison results of each of a plurality of electrocardiogram waveforms.

[0142] In some examples, when the extracted second electrocardiogram feature is a P wave, a Q wave, or an R wave, by executing step S1260, the occurrence rate of the P wave (i.e., the ratio when the result of the first peak-to-peak ratio is greater than a preset ratio) may be calculated based on the result where the first peak-to-peak ratio is greater than the preset ratio and the result where the first peak-to-peak ratio is less than or equal to the preset ratio. In other embodiments, when the extracted second electrocardiogram feature is an R wave, an S wave, or a T wave, by executing step S1260, the occurrence rate of the T wave (i.e., the ratio when the result of the second peak-to-peak ratio is greater than a preset ratio) may be calculated based on the result where the second peak-to-peak ratio is greater than the preset ratio and the result where the second peak-to-peak ratio is less than or equal to the preset ratio. In other embodiments, when the extracted second electrocardiogram feature is an S wave, a T wave, or a U wave, by executing step S1260, the occurrence rate of the U wave (i.e., the ratio when the result of the third peak-to-peak ratio is greater than a preset ratio) may be calculated based on the result where the third peak-to-peak ratio is greater than the preset ratio and the result where the third peak-to-peak ratio is less than or equal to the preset ratio.

[0143] By executing the steps shown in FIG. 12, in order to calculate the occurrence rate more accurately, a plurality of comparison results may be generated by comparing the peak-to-peak ratio with a preset ratio, and the occurrence rate may include, but is not limited to, the occurrence rate of the P wave, the occurrence rate of the Q wave, the occurrence rate of the R wave, the occurrence rate of the S wave, the occurrence rate of the T wave, and / or the occurrence rate of the U wave.

[0144] Please refer to FIG. 13. FIG. 13 is a schematic diagram for explaining a method of calculating at least one incidence rate based on a second electrocardiogram feature according to an embodiment of the present disclosure. As shown in FIG. 13, taking one electrocardiogram waveform as an example, when the extracted second electrocardiogram features are P wave, Q wave, R wave, S wave, T wave, and U wave, by executing each step shown in FIG. 12, in order to calculate the incidence rate, the peak-to-peak between a plurality of second electrocardiogram features and the ratio of the peak-to-peak between a plurality of peak-to-peaks are respectively calculated, and the ratio of the plurality of peak-to-peaks may be compared with a preset ratio. Among them, the incidence rate may be, but is not limited to, the incidence rate of the P wave, the incidence rate of the Q wave, the incidence rate of the R wave, the incidence rate of the S wave, the incidence rate of the T wave, and / or the incidence rate of the U wave.

[0145] Please refer to FIG. 14A. FIG. 14A is a detailed flowchart for explaining a method of estimating a user's blood glucose level based on peak distance and incidence rate according to an embodiment of the present disclosure. That is, step S1130 shown in FIG. 11 includes steps S1410A, S1420A, S1430A, and S1440A, and may be completed by executing steps S1410A, S1420A, S1430A, and S1440A. Among them, steps S1410A, S1420A, S1430A, and S1440A may be executed by the estimation unit 410 of the computing device 400A shown in FIG. 4.

[0146] In step S1410A, the peak distance is normalized. In some embodiments, step S1410A may be executed following the execution of steps S640 and S1120. In some embodiments, step S1410A may be substantially the same as step S710A shown in FIG. 7A.

[0147] In S1420A, the normalized peak distance and the occurrence rate are input into the machine learning model 500. In some embodiments, step S1420A may be executed following the execution of step S1410A. By executing step S1420A, the normalized peak distance and the occurrence rate may be input into a machine learning model 500 such as, for example, a decision tree algorithm, a random forest algorithm incorporating bagging, an extreme gradient boosting (XGBoost) algorithm, or a support vector machine algorithm.

[0148] In step S1430A, the user's blood glucose level is estimated by the machine learning model 500. In some embodiments, step S1430A may be executed following the execution of step S1420A. By executing step S1430A, after the normalized peak distance and the occurrence rate are input into the machine learning model 500, the user's blood glucose level may be estimated based on the normalized peak distance and the occurrence rate. Since the machine learning model 500 is trained with a plurality of data and verified with prediction results, the user's blood glucose level can be estimated by the machine learning model 500 with a prediction accuracy of 0.80 or higher, preferably 0.90 or higher.

[0149] In step S1440A, the user's blood glucose level is output by the machine learning model 500. In some embodiments, step S1440A may be executed following the execution of step S1430A. In some embodiments, step S1440A may be substantially the same as step S740A shown in FIG. 7A.

[0150] Please refer to FIG. 14B. FIG. 14B is a detailed flowchart illustrating a method for estimating a user's blood glucose level based on peak distance and incidence rate according to an embodiment of the present disclosure. That is, step S1130 shown in FIG. 11 includes steps S1410B, S1420B, S1430B, and S1440B, and may be completed by executing steps S1410B, S1420B, S1430B, and S1440B, among which steps S1410B, S1420B, S1430B, and S1440B may be executed by the estimation unit 410 of the computing device 400A shown in FIG. 4.

[0151] In step S1410B, the peak distance is normalized. In some embodiments, step S1410B may be executed subsequent to the execution of steps S640 and S1120. In some embodiments, step S1410B may be substantially the same as step S710A shown in FIG. 7A.

[0152] In step S1420B, the normalized peak distance and incidence rate are input into the neural network model 510. In some embodiments, step S1420B may be executed subsequent to the execution of step S1410B. By executing step S1420B, the normalized peak distance and incidence rate may be input into the neural network model 510.

[0153] In step S1430B, the user's blood glucose level is estimated by the neural network model 510. In some embodiments, step S1430B may be executed following the execution of step S1420B. By executing step S1430B, after the normalized peak distance and occurrence rate are input into the neural network model 510, the user's blood glucose level may be estimated based on the normalized peak distance and occurrence rate. Since the neural network model 510 is trained with a plurality of data and verified with prediction results, the user's blood glucose level can be estimated by the neural network model 510 with a prediction accuracy of 0.80 or higher, preferably 0.90 or higher.

[0154] In step S1440B, the user's blood glucose level is output by the neural network model 510. In some embodiments, step S1440B may be executed following the execution of step S1430B. In some embodiments, step S1440B may be substantially the same as step S740B shown in FIG. 7B.

[0155] Please refer to FIG. 15. FIG. 15 is a flowchart illustrating a fifth example of a method for non-invasively estimating a blood glucose level according to an embodiment of the present disclosure. The method shown in FIG. 15 may include steps S610, S620, S630, S640, S1510, S1520, and S1530, wherein steps S610, S620, S630, and S640 are substantially the same as the steps shown in FIG. 6.

[0156] In step S1510, at least four third electrocardiogram features are extracted from each of the plurality of electrocardiogram waveforms of the user. Step S1510 may also be executed by the extraction unit 404 of the computing device 400A shown in FIG. 4. In some embodiments, step S1510 may be executed following the execution of step S610. Further, step S1510 may be executed simultaneously with step S620. By executing step S1510, at least four third electrocardiogram features may be extracted from each of the plurality of electrocardiogram waveforms, and the third electrocardiogram features may be selectable from the group consisting of P wave, Q wave, R wave, S wave, T wave, and U wave.

[0157] In some examples, by executing step S1510, the P wave, Q wave, R wave, and S wave of each of the plurality of electrocardiogram waveforms may be extracted as the third electrocardiogram features, respectively. In other embodiments, by executing step S1510, the Q wave, R wave, S wave, and T wave of each of the plurality of electrocardiogram waveforms may be extracted as the third electrocardiogram features, respectively. In other embodiments, by executing step S1510, the Q wave, R wave, S wave, and U wave of each of the plurality of electrocardiogram waveforms may be extracted as the third electrocardiogram features, respectively. In other embodiments, by executing step S1510, the P wave, Q wave, R wave, S wave, T wave, and U wave of each of the plurality of electrocardiogram waveforms may be extracted as the third electrocardiogram features, respectively. That is, by executing step S1510, at least four of the P wave, Q wave, R wave, S wave, T wave, and U wave of each of the plurality of electrocardiogram waveforms may be extracted as the third electrocardiogram features.

[0158] In some embodiments, the third electrocardiogram features may be extracted by the convolutional neural network model 520. That is, after the plurality of electrocardiogram waveforms of the user are input into the convolutional neural network model 520, the third electrocardiogram features may be extracted by the convolutional layer of the convolutional neural network model 520.

[0159] In step S1520, at least one amplitude ratio is calculated based on a plurality of third electrocardiogram features. Step S1520 may be executed by the amplitude ratio calculator 506 of the calculation unit 408 shown in FIG. 5. In some embodiments, step S1520 may be executed subsequent to the execution of step S1510. By executing step S1520, at least one amplitude ratio, for example, the amplitude ratio of the P wave, may be calculated based on a plurality of third electrocardiogram features, but is not limited thereto. Step S1520 may be a step of calculating at least one amplitude ratio based on the amplitude values (for example, but not limited to, the maximum value, the minimum value, the root mean square value, or the peak-to-peak value) of a plurality of third electrocardiogram features. More specifically, at least one amplitude ratio is calculated by performing a series of calculation operations on a plurality of third electrocardiogram features.

[0160] In some examples, when the extracted third electrocardiogram features are the P wave, Q wave, R wave, and S wave, the amplitude ratio of the P wave may be calculated by executing step S1520. In other embodiments, when the extracted third electrocardiogram features are the Q wave, R wave, S wave, and T wave, the amplitude ratio of the T wave may be calculated by executing step S1520. In other embodiments, when the extracted third electrocardiogram features are the Q wave, R wave, S wave, and U wave, the amplitude ratio of the U wave may be calculated by executing step S1520. In other embodiments, when the extracted third electrocardiogram features are the P wave, Q wave, R wave, S wave, T wave, and U wave, the amplitude ratio of the P wave, the amplitude ratio of the T wave, and / or the amplitude ratio of the U wave may be calculated by executing step S1520, but is not limited thereto.

[0161] In step S1530, the user's blood glucose level is estimated based on the peak distance and the amplitude ratio. Step S1530 may be executed by the estimation unit 410 of the computing device 400A shown in FIG. 4. In some embodiments, step S1530 may be executed subsequent to the execution of step S640 and step S1520. By executing step S1530, the user's blood glucose level may be estimated based on the calculation result of the peak distance and the calculation result of the amplitude ratio. More specifically, after the calculation results of the peak distance and the amplitude ratio are input into the estimation model, the user's blood glucose level is estimated by the estimation model.

[0162] According to the method for non-invasively estimating the blood glucose level shown in FIG. 15, not only can a means for non-invasively estimating the user's blood glucose level be provided to the user, but also the user's blood glucose level can be estimated more accurately based on the peak distance and the amplitude ratio as described above. Furthermore, the method for non-invasively estimating the blood glucose level shown in FIG. 15 can improve the technical field related to non-invasively estimating the user's blood glucose level by using a plurality of electrocardiogram waveforms.

[0163] Also, in some embodiments, the method for non-invasively estimating the blood glucose level shown in FIG. 15 further includes step S910 shown in FIG. 9, and the corrected blood glucose level of the user may be calculated by executing step S910.

[0164] Please refer to FIG. 16. FIG. 16 is a detailed flowchart illustrating a method for calculating at least one amplitude ratio based on a third electrocardiogram feature according to an embodiment of the present disclosure. That is, step S1520 shown in FIG. 15 includes step S1610, S1620, S1630, and S1640, and may be completed by executing step S1610, S1620, S1630, and S1640A. Steps S1610, S1620, S1630, and S1640 may be executed by the amplitude ratio calculator 506 of the calculation unit 408 shown in FIG. 5.

[0165] In step S1610, the third peak-to-peak, the fourth peak-to-peak, the fifth peak-to-peak, and the sixth peak-to-peak are respectively calculated. In some embodiments, step S1610 may be executed subsequent to the execution of step S1510. By executing step S1610, for each of a plurality of electrocardiogram waveforms, the third peak-to-peak, the fourth peak-to-peak, the fifth peak-to-peak, and the sixth peak-to-peak may be calculated.

[0166] In some embodiments, when the extracted third electrocardiogram features are the P wave, Q wave, R wave, and S wave, by executing step S1610, for each of a plurality of electrocardiogram waveforms, the peak-to-peak between the maximum value of the P wave and the minimum value of the Q wave is taken as the third peak-to-peak, the peak-to-peak between the maximum value of the P wave and the minimum value of the S wave is taken as the fourth peak-to-peak, the peak-to-peak between the maximum value of the R wave and the minimum value of the Q wave is taken as the fifth peak-to-peak, and the peak-to-peak between the maximum value of the R wave and the minimum value of the S wave is taken as the sixth peak-to-peak and calculated. In other embodiments, when the extracted third electrocardiogram features are the Q wave, R wave, S wave, and T wave, by executing step S1610, for each of a plurality of electrocardiogram waveforms, the peak-to-peak between the maximum absolute value of the T wave and the minimum value of the Q wave is taken as the third peak-to-peak, the peak-to-peak between the maximum absolute value of the T wave and the minimum value of the S wave is taken as the fourth peak-to-peak, the peak-to-peak between the maximum value of the R wave and the minimum value of the Q wave is taken as the fifth peak-to-peak, and the peak-to-peak between the maximum value of the R wave and the minimum value of the S wave is taken as the sixth peak-to-peak and calculated. In other embodiments, when the extracted third electrocardiogram features are the Q wave, R wave, S wave, and U wave, by executing step S1610, for each of a plurality of electrocardiogram waveforms, the peak-to-peak between the maximum value of the U wave and the minimum value of the Q wave is taken as the third peak-to-peak, the peak-to-peak between the maximum value of the U wave and the minimum value of the S wave is taken as the fourth peak-to-peak, the peak-to-peak between the maximum value of the R wave and the minimum value of the Q wave is taken as the fifth peak-to-peak, and the peak-to-peak between the maximum value of the R wave and the minimum value of the S wave is taken as the sixth peak-to-peak and calculated.

[0167] In some embodiments, the third peak-to-peak values of each of the plurality of electrocardiogram waveforms may be further summed, and the summed calculation result may be divided by the number of the third peak-to-peak values to calculate the average value of the third peak-to-peak values. Similarly, the fourth peak-to-peak value, the fifth peak-to-peak value, and / or the sixth peak-to-peak value may be calculated by the same calculation operation, and the average value of the fourth peak-to-peak value, the average value of the fifth peak-to-peak value, and / or the average value of the sixth peak-to-peak value may be calculated respectively. Thereby, steps S1620, S1630, and S164 described below may be continuously executed based on the average value of the third peak-to-peak value, the average value of the fourth peak-to-peak value, the average value of the fifth peak-to-peak value, and the average value of the sixth peak-to-peak value, so that the amplitude ratio is calculated with fewer calculation operations (that is, to avoid performing calculations for each of the plurality of electrocardiogram waveforms in steps S1620, S1630, and S164).

[0168] In step S1620, a first average value of the third peak-to-peak value and the fourth peak-to-peak value is calculated. In some embodiments, step S1620 may be executed subsequent to the execution of step S1610. By executing step S1620, the third peak-to-peak value and the fourth peak-to-peak value may be summed, and the summed calculation result may be divided by 2 to calculate the first average value. In some embodiments, by executing step S1620, the average value of the third peak-to-peak value and the average value of the fourth peak-to-peak value may be summed, and the summed calculation result may be divided by 2 to calculate the first average value.

[0169] In some embodiments, when the extracted second electrocardiogram features are the P wave, Q wave, R wave, and S wave, by executing step S1620, the peak-to-peak between the maximum value of the P wave and the minimum value of the Q wave (i.e., the third peak-to-peak) and the peak-to-peak between the maximum value of the P wave and the minimum value of the S wave (i.e., the fourth peak-to-peak) may be summed, and the summed calculation result may be divided by 2 to calculate the first average value. In one embodiment, when the extracted second electrocardiogram features are the Q wave, R wave, S wave, and T wave, by executing step S1620, the peak-to-peak between the maximum absolute value of the T wave and the minimum value of the Q wave (i.e., the third peak-to-peak) and the peak-to-peak between the maximum absolute value of the T wave and the minimum value of the S wave (i.e., the fourth peak-to-peak) may be summed, and the summed calculation result may be divided by 2 to calculate the first average value. In one embodiment, when the extracted second electrocardiogram features are the Q wave, R wave, S wave, and U wave, by executing step S1620, the peak-to-peak between the maximum value of the U wave and the minimum value of the Q wave (i.e., the third peak-to-peak) and the peak-to-peak between the maximum value of the U wave and the minimum value of the S wave (i.e., the fourth peak-to-peak) may be summed, and the summed calculation result may be divided by 2 to calculate the first average value.

[0170] In step S1630, a second average value of the fifth peak-to-peak and the sixth peak-to-peak is calculated. In some embodiments, step S1630 may be executed following the execution of step S1620. Further, step S1630 may be executed simultaneously with step S1620. By executing step S1630, the fifth peak-to-peak and the sixth peak-to-peak may be summed, and the summed calculation result may be divided by 2 to calculate the second average value. In some embodiments, by executing step S1630, the average value of the fifth peak-to-peak and the average value of the sixth peak-to-peak may be summed, and the summed calculation result may be divided by 2 to calculate the second average value.

[0171] In some embodiments, when the extracted second electrocardiogram features are P wave, Q wave, R wave, and S wave, by executing step S1630, the peak-to-peak between the maximum value of the R wave and the minimum value of the Q wave (i.e., the fifth peak-to-peak), and the peak-to-peak between the maximum value of the R wave and the minimum value of the S wave (i.e., the sixth peak-to-peak) may be summed, and the summed calculation result may be divided by 2 to calculate the second average value. In one embodiment, when the extracted second electrocardiogram features are Q wave, R wave, S wave, and T wave, by executing step S1630, the peak-to-peak between the maximum value of the R wave and the minimum value of the Q wave (i.e., the fifth peak-to-peak), and the peak-to-peak between the maximum value of the R wave and the minimum value of the S wave (i.e., the sixth peak-to-peak) may be summed, and the summed calculation result may be divided by 2 to calculate the second average value. In one embodiment, when the extracted second electrocardiogram features are Q wave, R wave, S wave, and U wave, by executing step S1630, the peak-to-peak between the maximum value of the R wave and the minimum value of the Q wave (i.e., the fifth peak-to-peak), and the peak-to-peak between the maximum value of the R wave and the minimum value of the S wave (i.e., the sixth peak-to-peak) may be summed, and the summed calculation result may be divided by 2 to calculate the second average value.

[0172] In step S1640, the ratio of the first average value to the second average value is calculated as the amplitude ratio. In some embodiments, step S1640 may be executed subsequent to the execution of step S1630. By executing step S1640, the ratio of the first average value to the second average value is calculated as the amplitude ratio.

[0173] By executing the steps shown in FIG. 16, at least one amplitude ratio can be calculated more accurately based on the ratio of the first average value to the second average value by converting the peak-to-peak between a plurality of third electrocardiogram features into the first average value and the second average value. The amplitude ratio may include, but is not limited to, the amplitude ratio of the P wave, the amplitude ratio of the T wave, and / or the amplitude ratio of the U wave.

[0174] Please refer to FIGS. 17A and 17B simultaneously. FIG. 17A is a waveform diagram illustrating an example of calculating at least one amplitude ratio based on a third electrocardiogram feature according to an embodiment of the present disclosure, and FIG. 17B is a waveform diagram illustrating another example of calculating at least one amplitude ratio based on a third electrocardiogram feature according to an embodiment of the present disclosure. As shown in FIGS. 17A and 17B, for example, taking one electrocardiogram waveform as an example, when the extracted third electrocardiogram features are P wave, Q wave, R wave, S wave, T wave, and U wave, by executing the steps shown in FIG. 16, in order to calculate at least one amplitude ratio, the peak-to-peak between a plurality of third electrocardiogram features may be calculated respectively, and the first average value and the second average value may be calculated respectively based on the peak-to-peak.

[0175] Please refer to FIG. 18A. FIG. 18A is a detailed flowchart illustrating a method for estimating a user's blood glucose value based on peak distance and amplitude ratio according to an embodiment of the present disclosure. That is, step S1530 shown in FIG. 15 includes steps S1810A, S1820A, S1830A, and S1840A, and may be completed by executing steps S1810A, S1820A, S1830A, and S1840A, among which steps S1810A, S1820A, S1830A, and S1840A may be executed by the estimation unit 410 of the computing device 400A shown in FIG. 4.

[0176] In step S1810A, the peak distance and the amplitude ratio are each normalized. In some embodiments, step S1810A may be executed subsequent to the execution of step S640 and step S1520. By executing step S1810A, the calculation result of the peak distance may be normalized to a corresponding value within a range (for example, between 0 and 1), and the calculation result of the amplitude ratio may be normalized to a value within a range (for example, between 0 and 1).

[0177] In step S1820A, the normalized peak distance and the normalized amplitude ratio are input into the machine learning model 500. In some embodiments, step S1820A may be executed subsequent to the execution of step S1810A. By executing step S1820A, the normalized peak distance and the normalized amplitude ratio may be input into the machine learning model 500, such as, for example, a decision tree algorithm, a random forest algorithm incorporating bagging, an extreme gradient boosting (XGBoost) algorithm, or a support vector machine algorithm.

[0178] In step S1830A, the user's blood glucose level is estimated by the machine learning model 500. In some embodiments, step S1830A may be executed subsequent to the execution of step S1820A. By executing step S1830A, after the normalized peak distance and the normalized amplitude ratio are input into the machine learning model 500, the user's blood glucose level may be estimated based on the normalized peak distance and the normalized amplitude ratio. Since the machine learning model 500 is trained with a plurality of data and verified with prediction results, the user's blood glucose level can be estimated by the machine learning model 500 with a prediction accuracy of 0.80 or higher, preferably 0.90 or higher.

[0179] In step S1840A, the user's blood glucose level is output by the machine learning model 500. In some embodiments, step S1840A may be executed subsequent to the execution of step S1830A. In some embodiments, step S1840A may be substantially the same as step S740A shown in FIG. 7A.

[0180] Please refer to FIG. 18B. FIG. 18B is a detailed flowchart illustrating a method for estimating a user's blood glucose level based on peak distance and amplitude ratio according to an embodiment of the present disclosure. That is, step S1530 shown in FIG. 15 includes steps S1810B, S1820B, S1830B, and S1840B, and may be completed by executing steps S1810B, S1820B, S1830B, and S1840B, among which steps S1810B, S1820B, S1830B, and S1840B may be executed by the estimation unit 410 of the computing device 400A shown in FIG. 4.

[0181] In step S1810B, the peak distance and the amplitude ratio are each normalized. In some embodiments, step S1810B may be executed subsequent to the execution of steps S640 and S1520. In some embodiments, step S1810B may be substantially the same as step S1810A shown in FIG. 18A.

[0182] In step S1420B, the normalized peak distance and the normalized amplitude ratio are input into the neural network model 510. In some embodiments, step S1820B may be executed subsequent to the execution of step S1810B. By executing step S1820B, the normalized peak distance and the normalized amplitude ratio may be input into the neural network model 510.

[0183] In step S1830B, the user's blood glucose level is estimated by the neural network model 510. In some embodiments, step S1830B may be executed subsequent to the execution of step S1820B. By executing step S1430B, after the normalized peak distance and the normalized amplitude ratio are input into the neural network model 510, the user's blood glucose level may be estimated based on the normalized peak distance and the normalized amplitude ratio. Since the neural network model 510 is trained with a plurality of data and verified with prediction results, the user's blood glucose level can be estimated by the neural network model 510 with a prediction accuracy of 0.80 or higher, preferably 0.90 or higher.

[0184] In step S1840B, the user's blood glucose level is output by the neural network model 510. In some embodiments, step S1840B may be executed subsequent to the execution of step S1830B. In some embodiments, step S1840B may be substantially the same as step S740B shown in FIG. 7B.

[0185] Please refer to FIG. 19. FIG. 19 is a flowchart illustrating a sixth example of a method for non-invasively estimating a blood glucose level according to an embodiment of the present disclosure. The method shown in FIG. 19 may include steps S610, S620, S630, S640, S1910, S1920, and S1930, and steps S610, S620, S630, and S640 are substantially the same as the steps shown in FIG. 6.

[0186] In step S1910, at least one fourth electrocardiogram feature is extracted from each of the plurality of electrocardiogram waveforms of the user. Step S1910 may be executed by the extraction unit 404 of the computing device 400A shown in FIG. 4. In some embodiments, step S1910 may be executed following the execution of step S610. Further, step S1910 may be executed simultaneously with step S620. The fourth electrocardiogram feature can be selected from the group consisting of P wave, Q wave, R wave, S wave, T wave, and U wave. In some embodiments, step S1910 may be substantially similar to step S620 shown in FIG. 6.

[0187] In some embodiments, the fourth electrocardiogram feature may be extracted by the convolutional neural network model 520. That is, after a plurality of electrocardiogram waveforms of the user are input into the convolutional neural network model 520, the fourth electrocardiogram feature may be extracted by the convolutional layer of the convolutional neural network model 520.

[0188] In step S1920, at least one sharpness result is calculated based on the fourth electrocardiogram feature. Step S1920 may be executed by the calculation unit 408 of the computing device 400A shown in FIG. 4. In some embodiments, step S1920 may be executed following the execution of step S1910. Step S1920 may be a step of calculating the sharpness result based on the amplitude value of the fourth electrocardiogram feature (for example, but not limited to, the maximum value or the minimum value). More specifically, the sharpness result is calculated by performing a series of calculation operations on the fourth electrocardiogram feature.

[0189] In step S1930, the user's blood glucose level is estimated based on the peak distance and the sharpness result. Step S1930 may be executed by the estimation unit 410 of the computing device 400A shown in FIG. 4. In some embodiments, step S1930 may be executed subsequent to the execution of step S640 and step S1920. By executing step S1930, the user's blood glucose level may be estimated based on the calculation result of the peak distance and the calculation result of the sharpness result. More specifically, after the calculation results of the peak distance and the sharpness result are input into the estimation model, the user's blood glucose level is estimated by the estimation model.

[0190] According to the method for non-invasively estimating the blood glucose level shown in FIG. 19, not only can means for non-invasively estimating the user's blood glucose level be provided to the user, but also the user's blood glucose level can be estimated more accurately based on the peak distance and the sharpness result as described above. Furthermore, the method for non-invasively estimating the blood glucose level shown in FIG. 19 can improve the technical field related to non-invasively estimating the user's blood glucose level by using a plurality of electrocardiogram waveforms.

[0191] Furthermore, according to the method for non-invasively estimating the blood glucose level shown in FIG. 19, when a plurality of electrocardiogram waveforms having inverted T waves are received, the user's blood glucose level can be estimated more accurately by using the sharpness result.

[0192] Also, in some embodiments, the method for non-invasively estimating the blood glucose level shown in FIG. 19 may further include step S910 shown in FIG. 9, and the corrected blood glucose level of the user may be calculated by executing step S910.

[0193] Please refer to FIG. 20. FIG. 20 is a detailed flowchart illustrating a method for calculating sharpness results based on fourth electrocardiogram features according to an embodiment of the present disclosure. That is, step S1920 shown in FIG. 19 includes steps S2010, S2020, S2030, and S2040, and may be completed by executing steps S2010, S2020, S2030, and S2040, among which steps S2010, S2020, S2030, and S2040 may be executed by the calculation unit 408 of the computing device 400A shown in FIG. 4.

[0194] In step S2010, fourth feature peak positions corresponding to the fourth electrocardiogram features are respectively determined. In some embodiments, step S2010 may be executed following the execution of step S1910. In some embodiments, step S2010 may be substantially the same as step S630 shown in FIG. 6. In some embodiments, the fourth feature peak position may refer to the position of the maximum value of the P wave, the position of the minimum value of the Q wave, the position of the maximum value of the R wave, the position of the minimum value of the S wave, the position of the maximum absolute value of the T wave, and / or the position of the maximum value of the U wave. In some embodiments, by executing step S2010, when the extracted fourth electrocardiogram feature is the T wave, the position of the maximum absolute value of each T wave of a plurality of electrocardiogram waveforms (that is, "T" shown in FIG. 21) may be respectively determined. P ") may be determined respectively.

[0195] In step S2020, a first slope and a second slope are respectively calculated based on the fourth feature peak positions. In some embodiments, step S2020 may be executed following the execution of step S2010.

[0196] The first slope is calculated based on the fourth feature peak positions. More specifically, the first slope is calculated by an equation. In some embodiments, when the extracted fourth electrocardiogram feature is the T wave, the above equation is

[0197]

Equation

[0198] where "SL1" is the first slope, "T0" is the amplitude value of the T-wave position, and "N" is the number of sampling points. For example, "N" is 10.

[0199] The second slope is calculated based on the fourth characteristic peak position. More specifically, the second slope is calculated by an equation. In some embodiments, when the extracted fourth electrocardiogram feature is the T-wave, the above equation is

[0200]

Equation

[0201] where "SL2" is the second slope, "T0" is the amplitude value of the fourth characteristic peak position (i.e., the amplitude value of "T" P "), and "N" is the number of sampling points. For example, "N" is 10.

[0202] In step S2030, the first sharpness is calculated based on the first slope. In some embodiments, step S2030 may be executed subsequent to the execution of step S2020. The first sharpness is calculated by an equation. In some embodiments, when the extracted fourth electrocardiogram feature is the T-wave, the above equation is

[0203]

Equation

[0204] where "SP1" is the first sharpness, "T0" is the amplitude value of the fourth characteristic peak position (i.e., the amplitude value of "T" P "), "SL1" is the first slope, and "N" is the number of sampling points. For example, "N" is 10.

[0205] In step S2040, the second sharpness is calculated based on the second slope. In some embodiments, step S2040 may be executed following the execution of step S2030. Further, step S2040 may be executed simultaneously with step S2030. The second sharpness is calculated by an equation. In some embodiments, when the extracted fourth electrocardiogram feature is the T wave, the above equation is

[0206] [Number]

[0207] where "SP2" is the second sharpness, "T0" is the amplitude value of the fourth feature peak position (i.e., the amplitude value of "T P "), "SL2" is the second slope, and "N" is the number of sampling points. For example, "N" is 10.

[0208] The first sharpness and the second sharpness are regarded as the sharpness result. Therefore, the sharpness result can be calculated by executing steps S2010, S2020, S2030, and S2040.

[0209] Please refer to FIG. 21. FIG. 21 is a waveform diagram for explaining a method of calculating a sharpness result based on a fourth electrocardiogram feature according to an embodiment of the present disclosure. In some embodiments, the sharpness result may be calculated by differential calculation and integral calculation.

[0210] The first slope is calculated based on the position of the maximum value of the absolute value of the T wave. More specifically, the first slope is calculated by a differential equation. The above-mentioned differential equation is

[0211] [Number]

[0212] where "SL1" is the first slope, "T P" is the amplitude value at the fourth characteristic peak position (i.e., the amplitude value of "T P "), and "T S " is the amplitude value at the start point of the T wave.

[0213] The second slope is calculated based on the position of the maximum absolute value of the T wave. More specifically, the second slope is calculated by a differential equation. The above-mentioned differential equation is

[0214]

Equation

[0215] where "SL2" is the second slope, "T P " is the amplitude value at the fourth characteristic peak position (i.e., the amplitude value of "T P "), and "T e " is the amplitude value at the end point of the T wave.

[0216] The first sharpness is calculated based on the position of the maximum absolute value of the T wave. More specifically, the first sharpness is calculated by an integral equation. The above-mentioned integral equation is

[0217]

Equation

[0218] where "SP1" is the first sharpness, "T P " is the amplitude value at the fourth characteristic peak position (i.e., the amplitude value of "T P "), "T S " is the amplitude value at the start point of the T wave, and "ECG(x)" is a plurality of electrocardiogram waveforms.

[0219] The second sharpness is calculated based on the position of the maximum absolute value of the T wave. More specifically, the second sharpness is calculated by an integral equation. The above-mentioned integral equation is

[0220]

Equation

[0221] where "SP2" is the second sharpness, "T P " is the amplitude value of the fourth characteristic peak position (i.e., the amplitude value of "T P "), "T e " is the amplitude value at the end point of the T wave, and "ECG(x)" is a plurality of electrocardiogram waveforms.

[0222] Please refer to FIG. 22A. FIG. 22A is a detailed flowchart illustrating a method for estimating a user's blood glucose level based on peak distance and sharpness results according to an embodiment of the present disclosure.

[0223] In step S2210A, the peak distance and the sharpness result are each normalized. In some embodiments, step S2210A may be executed subsequent to the execution of step S640 and step S1920. By executing step S2210A, the calculated result of the peak distance may be normalized to a corresponding value within a range (for example, between 0 and 1), and the calculated result of the sharpness result may be normalized to a corresponding value within a range (for example, between -2,000 and 2,000).

[0224] In step S2220A, the normalized peak distance and the normalized sharpness result are input into the machine learning model 500. In some embodiments, step S2220A may be executed subsequent to the execution of step S2210A. By executing step S2220A, the normalized peak distance and the normalized sharpness result may be input into the machine learning model 500, such as a decision tree algorithm, a random forest algorithm incorporating bagging, an extreme gradient boosting (XGBoost) algorithm, or a support vector machine algorithm.

[0225] In step S2230A, the user's blood glucose level is estimated by the machine learning model 500. In some embodiments, step S2230A may be executed subsequent to the execution of step S2220A. By executing step S2230A, after the normalized peak distance and the normalized sharpness result are input into the machine learning model 500, the user's blood glucose level may be estimated based on the normalized peak distance and the normalized sharpness result rate. Since the machine learning model 500 is trained with a plurality of data and verified with prediction results, the user's blood glucose level can be estimated by the machine learning model 500 with a prediction accuracy of 0.80 or higher, preferably 0.90 or higher.

[0226] In step S2240A, the user's blood glucose level is output by the machine learning model 500. In some embodiments, step S2240A may be executed subsequent to the execution of step S2230A. In some embodiments, step S2240A is substantially the same as step S740A shown in FIG. 7A.

[0227] Please refer to FIG. 22B. FIG. 22B is a detailed flowchart illustrating a method for estimating a user's blood glucose level based on a peak distance and a sharpness result according to an embodiment of the present disclosure.

[0228] In step S2210B, the peak distance and the sharpness result are each normalized. In some embodiments, step S2210B may be executed subsequent to the execution of steps S640 and S1920. In some embodiments, step S2210B is substantially the same as step S2210A shown in FIG. 22A.

[0229] In step S2220B, the normalized peak distance and the normalized sharpness result are input into the neural network model 510. In some embodiments, step S2220B may be executed following the execution of step S2210B. By executing step S2220B, the normalized peak distance and the normalized sharpness result may be input into the neural network model 510.

[0230] In step S2230B, the user's blood glucose level is estimated by the neural network model 510. In some embodiments, step S2230B may be executed following the execution of step S2220B. By executing step S2230B, after the normalized peak distance and the normalized sharpness result are input into the neural network model 510, the user's blood glucose level may be estimated based on the normalized peak distance and the normalized sharpness result. Since the neural network model 510 is trained with a plurality of data and verified with prediction results, the user's blood glucose level can be estimated by the neural network model 510 with a prediction accuracy of 0.80 or higher, preferably 0.90 or higher.

[0231] In S1840B, the user's blood glucose level is output by the neural network model 510. In some embodiments, step S1840B may be executed following the execution of step S1830B. In some embodiments, step S1840B is substantially the same as step S740B shown in FIG. 7B.

[0232] Please refer to FIG. 23. FIG. 23 is a schematic diagram illustrating another example of a computing device 400B that is electrically connected to an electrocardiogram measurement device 300 and electrically connected to at least one of a machine learning model 500, a neural network model 510, and a convolutional neural network model 520 according to an embodiment of the present disclosure. The computing device 400B may have the same configuration and functions as the computing device 400A shown in FIG. 4, and may further include an evaluation unit 412, a counting unit 414, and an output unit 416.

[0233] The evaluation unit 412 may be configured to evaluate the quality of a plurality of electrocardiogram waveforms. That is, the computing device 400B may evaluate the quality of the received plurality of electrocardiogram waveforms by the evaluation unit 412. More specifically, the evaluation unit 412 may evaluate whether there is noise in the plurality of electrocardiogram waveforms. When the result of the evaluation of the noise in the plurality of electrocardiogram waveforms is smaller than a preset noise threshold, the quality of the plurality of electrocardiogram waveforms may be determined to be good electrocardiogram waveforms (that is, normal electrocardiogram signals). Conversely, when the result of the evaluation of the noise in the plurality of electrocardiogram waveforms is equal to or greater than the preset noise threshold, the quality of the plurality of electrocardiogram waveforms may be determined to be poor electrocardiogram waveforms (that is, there is noise). In some embodiments, the evaluation unit 412 may evaluate the quality of the received plurality of electrocardiogram waveforms by calculating the distribution of the values of the RR intervals between adjacent electrocardiogram waveforms, the duration of each of the plurality of electrocardiogram waveforms, the heart rate of the user, the distribution of each of the plurality of electrocardiogram features, the amplitude value of each of the plurality of electrocardiogram features, and / or the similarity between the plurality of electrocardiogram waveforms.

[0234] In some embodiments, when the values of the RR intervals between adjacent electrocardiogram waveforms are slightly different from each other (for example, the error rate is within 0.1) and the percentage of outliers is less than a preset value (for example, 0.1), the quality of the plurality of electrocardiogram waveforms may be determined to be good electrocardiogram waveforms. In some embodiments, when the respective durations of the plurality of electrocardiogram waveforms are slightly different from each other (for example, the error rate is within 0.1) and the percentage of outliers is less than a preset value (for example, 0.1), the quality of the plurality of electrocardiogram waveforms may be determined to be good electrocardiogram waveforms. In some embodiments, when the user's heart rate is within the normal range, the quality of the plurality of electrocardiogram waveforms may be determined to be good electrocardiogram waveforms. In some embodiments, when the respective distributions of the plurality of electrocardiogram features are slightly different from each other (for example, the error rate of the time difference between the P wave and the R wave of each of the plurality of electrocardiogram waveforms is within 0.1) and the percentage of outliers is less than a preset value (for example, 0.1), the quality of the plurality of electrocardiogram waveforms may be determined to be good electrocardiogram waveforms. In some examples, when the respective amplitude values of the plurality of electrocardiogram features are slightly different from each other (for example, the error rate is within 0.1) and the percentage of outliers is less than a preset value (for example, 0.1), the quality of the plurality of electrocardiogram waveforms may be determined to be good electrocardiogram waveforms. In some embodiments, when the similarity between the plurality of electrocardiogram waveforms is large (for example, the similarity is greater than 0.9) and the percentage of outliers is less than a preset value (for example, 0.1), the quality of the plurality of electrocardiogram waveforms may be determined to be good electrocardiogram waveforms.

[0235] The counting unit 414 may be configured to count the number of a plurality of electrocardiogram waveforms of the received user. That is, when the computing device 400B starts receiving a plurality of electrocardiogram waveforms, the computing device 400B may start receiving the amplitude values of each point for the plurality of electrocardiogram waveforms, convert the amplitude values of each point into corresponding waveforms, and calculate the number of the corresponding waveforms.

[0236] In some embodiments, the counting unit 414 may be further configured to compare the number of corresponding waveforms (i.e., a plurality of received electrocardiogram waveforms) with a preset value to determine whether the computing device 400B has received a sufficient number of a plurality of electrocardiogram waveforms. In some embodiments, the preset value may be set to 60 or more, but is not limited thereto.

[0237] The output unit 416 may be configured to output the blood glucose level of the user. That is, the output unit 416 may output the blood glucose level of the user estimated by the estimation unit 410 and / or the corrected blood glucose level of the user calculated by the corrected blood glucose calculator 508. In some embodiments, the blood glucose level of the user estimated by the estimation unit 410 and / or the corrected blood glucose level of the user calculated by the corrected blood glucose calculator 508 may be stored in synchronization with the blood glucose level database 454, so that the output unit 416 can output the blood glucose level stored in the blood glucose level database 454.

[0238] Thereby, the computing device 400B provided by the present disclosure can be used to implement a method for non-invasively estimating the blood glucose level. Therefore, after receiving a plurality of electrocardiogram waveforms of the user, the computing device 400B can estimate the blood glucose level of the user and / or calculate the corrected blood glucose level of the user based on the plurality of electrocardiogram waveforms of the user.

[0239] Furthermore, the computing device 400B provided by the present disclosure further evaluates the quality and quantity of the received plurality of electrocardiogram waveforms, and by ensuring that the method for non-invasively estimating the blood glucose level is realized by a plurality of electrocardiogram waveforms of a certain quantity and / or a certain quality, the blood glucose level of the user can be estimated more accurately and / or the corrected blood glucose level of the user can be calculated.

[0240] According to the computing device 400B shown in FIG. 23, not only can means for non-invasively estimating the user's blood glucose level be provided to the user, but the user's blood glucose level can also be estimated more accurately. Furthermore, the computing device 400B shown in FIG. 23 can improve the technical field related to non-invasively estimating the user's blood glucose level by utilizing a plurality of electrocardiogram waveforms.

[0241] Please refer to FIG. 24. FIG. 24 is a flowchart illustrating a seventh example of a method for non-invasively estimating a blood glucose level according to an embodiment of the present disclosure. The method shown in FIG. 24 may include steps S610, S2410, S2420, and S2430, and step S610 is substantially the same as the step shown in FIG. 6.

[0242] In step S2410, the quality of a plurality of electrocardiogram waveforms is evaluated. Step S2410 is executed by the evaluation unit 412 of the computing device 400B shown in FIG. 23. In some embodiments, step S2410 may be executed subsequent to the execution of step S610. By executing step S2410, the quality of the plurality of electrocardiogram waveforms received by the receiving unit 402 is evaluated.

[0243] If a plurality of electrocardiogram waveforms are evaluated as noise NS (that is, if the evaluation result of the plurality of electrocardiogram waveforms with respect to noise is equal to or greater than a preset noise threshold), the user's past blood glucose level may be output (that is, step S2420 is executed). If a plurality of electrocardiogram waveforms are evaluated as a normal electrocardiogram signal NM (that is, if the evaluation result of the plurality of electrocardiogram waveforms with respect to noise is less than a preset noise threshold), the user's real-time blood glucose level may be output (that is, step S2430 is executed).

[0244] In step S2420, the user's past blood glucose values are output. Step S2420 may be executed by the output unit 416 of the computing device 400B shown in FIG. 23. In some embodiments, step S2420 may be executed following the execution of step S2410. By executing step S2420, the user's past blood glucose values may be output.

[0245] In some embodiments, the method provided by the present disclosure estimates the user's blood glucose values over a certain period (e.g., every second). When multiple electrocardiogram waveforms are evaluated as noise NS at time point A and step S2420 is executed, the user's past blood glucose values may be output. The past blood glucose values may refer to the user's blood glucose values obtained before time point A (e.g., the blood glucose value estimated at time point A - 1, or the blood glucose value estimated at the time when the electrocardiogram signal NM was last evaluated as normal). In some embodiments, the past blood glucose values may refer to the blood glucose values stored in the blood glucose database 454, and the past blood glucose values may refer to the user's blood glucose values estimated by the estimation unit 410 and / or the user's corrected blood glucose values calculated by the corrected blood glucose value calculator 508.

[0246] In step S2430, the user's real - time blood glucose values are output. Step S2430 is executed by the output unit 416 of the computing device 400B shown in FIG. 23. In some embodiments, step S2430 may be executed following the execution of step S2410. In some embodiments, the user's blood glucose values estimated by the estimation unit 410 and / or the user's corrected blood glucose values calculated by the corrected blood glucose value calculator 508 may be output in real - time. In other embodiments, the user's blood glucose values stored in synchronization with the blood glucose database 454 may be output in real - time.

[0247] According to the method for non-invasively estimating a blood glucose level shown in FIG. 24, not only can means for non-invasively estimating the user's blood glucose level be provided to the user, but the user's blood glucose level can also be estimated more accurately based on a plurality of electrocardiogram waveforms of a specific quality. That is, by implementing the method for non-invasively estimating the blood glucose level shown in FIG. 24, according to the above method, it is possible to ensure that the user's blood glucose level is estimated based on a plurality of electrocardiogram waveforms of a specific quality (that is, the estimated result of the user's blood glucose level will not be significantly biased by a plurality of electrocardiogram waveforms with noise). Furthermore, by the method for non-invasively estimating the blood glucose level shown in FIG. 24, the technical field related to non-invasively estimating the user's blood glucose level by using a plurality of electrocardiogram waveforms can be improved.

[0248] Please refer to FIG. 25. FIG. 25 is a flowchart illustrating an eighth example of a method for non-invasively estimating a blood glucose level according to an embodiment of the present disclosure. The method shown in FIG. 25 may include steps S610, S620, S630, S640, S650, and S2510, and steps S610, S620, S630, S640, and S650 are substantially the same as the steps shown in FIG. 6.

[0249] In step S2510, it is determined whether the number of a plurality of electrocardiogram waveforms of the user is less than a preset value. Step S2510 may be executed by the counting unit 414 of the computing device 400B shown in FIG. 23. In some embodiments, step S2510 may be executed subsequent to the execution of step S610. Step S2510 may be executed to compare the number of received electrocardiogram waveforms with a preset value to determine whether the number of a plurality of electrocardiogram waveforms of the user is less than the preset value.

[0250] When the number of received electrocardiogram waveforms is equal to or greater than a preset value (i.e., "NO" shown in FIG. 25), step S620 is executed, and the next step in the method for non-invasively estimating the blood glucose level is completed. When the number of received electrocardiogram waveforms is less than the preset value (i.e., "YES" shown in FIG. 25), multiple electrocardiogram waveforms of the user may be received again until the number of received electrocardiogram waveforms exceeds the preset value (i.e., return to step S610).

[0251] According to the method for non-invasively estimating the blood glucose level shown in FIG. 25, not only can means for non-invasively estimating the user's blood glucose level be provided to the user, but also the user's blood glucose level can be estimated more accurately with a sufficient number of multiple electrocardiogram waveforms. That is, by implementing the method for non-invasively estimating the blood glucose level shown in FIG. 25, it is possible to ensure that the user's blood glucose level is estimated based on a sufficient number of multiple electrocardiogram waveforms according to the above method (i.e., the estimation result of the user's blood glucose level will not be greatly biased due to insufficient multiple electrocardiogram waveforms). Furthermore, the method for non-invasively estimating the blood glucose level shown in FIG. 25 can improve the technical field related to non-invasively estimating the user's blood glucose level by using multiple electrocardiogram waveforms.

[0252] In some embodiments, according to the method for non-invasively estimating the blood glucose level, by estimating the user's blood glucose level based on at least one peak distance, at least one incidence rate, at least one amplitude ratio, and sharpness result, not only can means for non-invasively estimating the user's blood glucose level be provided to the user, but also the user's blood glucose level can be estimated more accurately by using the peak distance, incidence rate, amplitude ratio, and sharpness result. Furthermore, the above method for non-invasively estimating the blood glucose level can improve the technical field related to non-invasively estimating the user's blood glucose level by using multiple electrocardiogram waveforms.

[0253] Also, in some embodiments, the method for non-invasively estimating the above blood glucose level may further include step S910 shown in FIG. 9, and the corrected blood glucose level of the user may be calculated by executing step S910.

[0254] Please refer to FIG. 26. FIG. 26 is a schematic diagram illustrating an example of a computing device 2600 for non-invasively estimating a blood glucose level, which is electrically connected to at least one of an electrocardiogram sensor 310, an electrocardiogram monitoring device 320, and a server 330, and is also electrically connected to at least one of a display device 380 and a server 390.

[0255] Since the computing device 2600 is electrically connected to at least one of the electrocardiogram sensor 310, the electrocardiogram monitoring device 320, and the server 330, the computing device 2600 may receive a plurality of electrocardiogram waveforms of the user from at least one of the electrocardiogram sensor 310, the electrocardiogram monitoring device 320, and the server 330. In some embodiments, the computing device 2600 may receive a plurality of electrocardiogram waveforms of the user from at least one of the electrocardiogram sensor 310, the electrocardiogram monitoring device 320, and the server 330 via a physical signal line connection. For example, the physical signal line connection may be, but is not limited to, a network signal line connection compliant with the Internet Protocol (IP). Further, in some embodiments, the computing device 2600 may receive a plurality of electrocardiogram waveforms of the user from at least one of the electrocardiogram sensor 310, the electrocardiogram monitoring device 320, and the server 330 via a virtual signal line connection. For example, the virtual signal line connection may be, but is not limited to, a Wi-Fi connection compliant with a wireless network protocol.

[0256] In some embodiments, computing device 2600 may be electrically connected to another device (e.g., a wearable measurement device) that can provide multiple electrocardiogram waveforms to receive multiple electrocardiogram waveforms of the user from the other device (not shown).

[0257] The computing device 2600 may include a memory module 2610 and a blood glucose estimation module 2620 so as to be able to perform specific signal processing on the received multiple electrocardiogram waveforms. That is, after the computing device 2600 receives multiple electrocardiogram waveforms of the user, the computing device 2600 may estimate the user's blood glucose level and / or calculate the user's corrected blood glucose level based on the multiple electrocardiogram waveforms of the user, and output the estimated blood glucose level and / or the corrected blood glucose level of the user.

[0258] Since the computing device 2600 is electrically connected to at least one of the display device 380 and the server 390, the computing device 2600 may output the estimated blood glucose level and / or the corrected blood glucose level of the user to at least one of the display device 380 and the server 390.

[0259] In some embodiments, the computing device 2600 may be electrically connected to another device that can receive, utilize, store, and / or display the estimated blood glucose level and / or the corrected blood glucose level of the user for subsequent use by another device (not shown).

[0260] The memory module 2610 may be configured to store a program, a series of program codes, and / or a series of instruction sets. In some embodiments, the memory module 2610 may include, but is not limited to, one or more non-volatile memories and one or more volatile memories. The non-volatile memory may be, for example, a read-only memory, a flash memory, or a non-volatile random access memory, but is not limited thereto. The volatile memory may be, for example, a dynamic random access memory or a static random access memory, but is not limited thereto. In some embodiments, the memory module 2610, particularly the non-volatile memory, can store a program, a series of program codes, and / or a series of instruction sets related to a method for non-invasively estimating a blood glucose level as described above.

[0261] The blood glucose level estimation module 2620 may be configured to be electrically connected to the memory module 2610 and may be configured to estimate the user's blood glucose level based on a plurality of electrocardiogram waveforms of the user. That is, after a plurality of electrocardiogram waveforms of the user are received, the blood glucose level estimation module 2620 can estimate the user's blood glucose level based on the plurality of electrocardiogram waveforms of the user by executing the steps in any one of the methods for non-invasively estimating a blood glucose level as described above. In some embodiments, the blood glucose level estimation module 2620 can also calculate the user's corrected blood glucose level.

[0262] In some embodiments, the blood glucose level estimation module 2620 may include, but is not limited to, one or more processors. The processor may be, for example, a central processing unit, but is not limited thereto.

[0263] The memory module 2610 stores a program, a series of program codes, and / or a series of instruction sets related to the method for non-invasively estimating blood glucose levels as described above. Therefore, after loading and executing the program, code, and / or instruction set, the blood glucose level estimation module 2620 can implement any one of the methods for non-invasively estimating blood glucose levels as described above by the blood glucose level estimation module, particularly by the processor.

[0264] In some embodiments, the computing device 2600 may further include a signal output module (not shown). The signal output module may be configured to be electrically connected to the blood glucose level estimation module 2620, and may be configured to output the user's blood glucose level estimated by the blood glucose level estimation module 2620 and / or the calculated corrected blood glucose level of the user.

[0265] Thus, the computing device 2600 provided by the present disclosure is used to execute any step of the method for non-invasively estimating blood glucose levels as described above. After receiving a plurality of electrocardiogram waveforms of the user, the computing device 2600 can estimate the user's blood glucose level based on the plurality of electrocardiogram waveforms of the user.

[0266] According to the computing device 2600 shown in FIG. 26, not only can the user be provided with means for non-invasively estimating the user's blood glucose level, but also the user's blood glucose level can be estimated more accurately. Furthermore, the computing device 2600 shown in FIG. 26 can improve the technical field related to non-invasively estimating the user's blood glucose level by using a plurality of electrocardiogram waveforms.

[0267] Please refer to FIG. 27. FIG. 27 is a schematic diagram illustrating an example of a device 2700 for non-invasively measuring an electrocardiogram signal electrically connected to at least one of a computing device 2200, a computing device 400A, a computing device 400B, and a server 350 according to an embodiment of the present disclosure. Please refer to FIG. 27.

[0268] Since the device 2700 is signal-connected to at least one of the computing device 2600, the computing device 400A, the computing device 400B, and the server 350, the device 2700 may provide a plurality of electrocardiogram waveforms of a user to at least one of the computing device 2600, the computing device 400A, the computing device 400B, and the server 350. In some embodiments, the device 2700 may provide a plurality of electrocardiogram waveforms of a user to at least one of the computing device 2600, the computing device 400A, the computing device 400B, and the server 350 via a physical signal line connection. For example, the physical signal line connection may be, but is not limited to, a network signal line connection compliant with the Internet Protocol (IP). Further, in some embodiments, the device 2700 may provide a plurality of electrocardiogram waveforms of a user to at least one of the computing device 2600, the computing device 400A, the computing device 400B, and the server 350. For example, the virtual signal line connection may be, but is not limited to, a Wi-Fi connection compliant with a wireless network protocol.

[0269] Device 2700 includes a measurement module 2710 and a signal transmission module 2720. Device 2700 can provide a plurality of electrocardiogram waveforms of a user to at least one of computing device 2600, computing device 400A, computing device 400B, and server 350, so that at least one of computing device 2600, computing device 400A, computing device 400B, and server 350 can execute any step of the method for non-invasively estimating the blood glucose value described above, and estimate the blood glucose value of the user based on the plurality of electrocardiogram waveforms measured by device 2700.

[0270] The measurement module 2710 may be configured to have a plurality of electrodes electrically connected to the user and configured to measure a plurality of electrocardiogram waveforms of the user. Since the depolarization process of the heart causes voltage changes on the body surface, the voltage changes on the body surface can be measured by a plurality of electrodes electrically connected to the skin surface of the user, and a plurality of electrocardiogram waveforms of the user can be obtained.

[0271] The signal transmission module 2720 may be configured to be electrically connected to the measurement module 2710, transmit a plurality of electrocardiogram waveforms of the user to at least one of computing device 2600, computing device 400A, computing device 400B, and server 350, and be configured to cause at least one of computing device 2600, computing device 400A, computing device 400B, and server 350 to execute the steps of any one of the methods for non-invasively estimating the blood glucose value as described above.

[0272] As a result, the device 2700 provided by the present disclosure is used to transmit a plurality of measured electrocardiogram waveforms to at least one of the computing device 2600, the computing device 400A, the computing device 400B, and the server 350. At least one of the computing device 2600, the computing device 400A, the computing device 400B, and the server 350 can estimate the blood glucose level of the user based on the plurality of electrocardiogram waveforms measured by the device 2700.

[0273] Please refer to FIG. 28. FIG. 28 is a schematic diagram for explaining that the blood glucose level of a user is estimated based on a plurality of electrocardiogram waveforms received from the user according to an embodiment of the present disclosure.

[0274] The method for non-invasively estimating the blood glucose level may be implemented by the computing device described above. Specifically, the computing device as described above can estimate the blood glucose level of the user based on a plurality of electrocardiogram waveforms of the user, particularly the peak distance. For example, the peak distance is calculated based on the plurality of electrocardiogram waveforms shown in FIG. 28, and the blood glucose level of the user may be estimated based on the peak distance. As shown in FIG. 28, the peak distance is input into a machine learning model, that is, an Extreme Gradient Boosting (XGBoost) algorithm, and the blood glucose level of the user may be estimated by the machine learning model. As shown in FIG. 28, the estimated blood glucose level of the user is 124 mg / dL. In this case, the actual blood glucose level of the user measured by a blood glucose meter is 128 mg / dL. Since the estimated result of the blood glucose level is close to the actual blood glucose level, the present disclosure can not only provide the user with a means for non-invasively estimating the blood glucose level of the user, but also estimate the blood glucose level of the user more accurately by using a plurality of electrocardiogram waveforms, particularly the peak distance as described above.

[0275] Please refer to FIG. 29A. FIG. 29A is a schematic diagram showing a comparison result of the blood glucose level estimated by the present disclosure with other devices.

[0276] As shown in FIG. 29A, "UUID:48" is the subject's identification (ID), "CGM MARD" is the error between the true blood glucose value and the blood glucose value monitored by another device, and "XGB MARD" is the error between the true blood glucose value and the estimated blood glucose value by the device provided by the present disclosure. In the comparison result, it is shown that the estimated result of "XGB MARD" is more accurate than the monitored result of "CGM MARD". In some embodiments, "MARD" may be calculated by the following equation:

[0277]

Number

[0278] where "BGi" is the estimated blood glucose value, "Compi" is the true blood glucose value, and "N" is the number of sampling points. For example, when the estimated blood glucose values are {105, 95, 85} and the true blood glucose values are {100, 90, 90}, "MARD" is calculated and the calculation result of "MARD" is 5.37%.

[0279]

Number

[0280] Please refer to FIG. 29B. FIG. 29B is a schematic diagram showing another comparison result of comparing the blood glucose value estimated by the present disclosure with other devices.

[0281] As shown in FIG. 29B, "UUID:9" is the identification (ID) of the subject (i.e., another subject), "CGM MARD" is another device for monitoring the user's blood glucose value, and "XGB MARD" is the device provided by the present disclosure. In the comparison result, it is also shown that the estimated result of "XGB MARD" is more accurate than the monitored result of "CGM MARD".

[0282] In some embodiments, the steps of the method for non-invasively estimating the blood glucose level described above may be stored in a non-transitory computer-readable recording medium as a series of specific program codes or a series of specific instruction sets. The non-transitory computer-readable recording medium may be, for example, a hard disk, a compact disc (CD-ROM), a magnetic disk, or a flash drive (USB), but is not limited thereto. When the built-in program code or instruction set is loaded and executed by a computer, it enables the method for non-invasively estimating the blood glucose level as described above.

[0283] Although the present disclosure has been described by means of specific embodiments, modifications and changes may be made by those of ordinary skill in the art to which the present disclosure pertains without departing from the scope and spirit of the present disclosure described in the claims. Therefore, the scope of protection of this application should be limited by the claims rather than the content disclosed in the specification.

Description of Reference Numerals

[0284] 300 Electrocardiogram measurement device 310 Electrocardiogram sensor 320 Electrocardiogram monitoring device 330 Server 350 Server 380 Display device 390 Server 400A, 400B Computing devices 402 Receiver 404 Extraction unit 406 Decision unit 408 Calculation unit 410 Estimation unit 412 Evaluation unit 414 Counting unit 416 Output unit 452 Electrocardiogram waveform database 454 Blood glucose level database 500 Machine learning model 502 Peak distance calculator 504 Incidence rate calculator 506 Amplitude ratio calculator 508 Corrected blood glucose value calculator 510 Neural network model 520 Convolutional neural network model 2600 Computing device 2610 Memory module 2620 Blood glucose value estimation module 2700 Device 2710 Measurement module 2720 Signal transmission module BGV Blood glucose value CL Convolutional layer EF Electrocardiogram feature EW Electrocardiogram waveform HL Hidden layer IL Input layer NM Normal electrocardiogram signal NS Noise OL Output layer S610,S620,S630 Steps S640,S650 Steps S710A,S710B Steps S720A,S720B Steps S730A,S730B Steps S740A,S740B Steps S910 Step S1010,S1020 Steps S1110,S1120,S1130 Steps S1210,S1220,S1230 Steps S1240,S1250,S1260 Steps S1410A,S1410B Steps S1420A,S1420B Steps S1430A,S1430B Steps S1440A,S1440B Steps S1510,S1520,S1530 Steps Steps S1610 and S1620 Steps S1630 and S1640 Steps S1810A and S1810B Steps S1820A and S1820B Steps S1830A and S1830B Steps S1840A and S1840B Steps S1910, S1920, and S1930 Steps S2010 and S2020 Steps S2030 and S2040 Steps S2210A and S2210B Steps S2220A and S2220B Steps S2230A and S2230B Steps S2240A and S2240B Steps S2410, S2420, and S2430 Step S2510

Claims

1. A method for non-invasively estimating a user's blood glucose level, which estimates the user's blood glucose level by a computing device, comprising: receiving a plurality of electrocardiogram (ECG) waveforms of the user; extracting at least two first ECG features from each of the plurality of ECG waveforms of the user; determining a first feature peak position corresponding to each of these first ECG features; calculating at least one peak distance between these first feature peak positions for each of the plurality of ECG waveforms; estimating the user's blood glucose level based on the at least one peak distance; wherein these first ECG features are selected from the group consisting of P wave, Q wave, R wave, S wave, T wave, and U wave, and it is a method for non-invasively estimating a blood glucose level.

2. Estimating the user's blood glucose level based on the at least one peak distance comprises: normalizing the at least one peak distance; inputting the normalized at least one peak distance into a machine learning model; outputting the user's blood glucose level by the machine learning model; wherein the user's blood glucose level is estimated by the machine learning model based on the normalized at least one peak distance. The method for non-invasively estimating a blood glucose level according to Claim 1.

3. Estimating the user's blood glucose level based on the at least one peak distance comprises: normalizing the at least one peak distance; inputting the normalized at least one peak distance into a neural network model; outputting the user's blood glucose level by the neural network model; wherein the user's blood glucose level is estimated by the neural network model based on the normalized at least one peak distance. The method for non-invasively estimating a blood glucose level according to Claim 1.

4. further comprising calculating a corrected blood glucose level of the user by an equation for correcting a blood glucose level based on the user's blood glucose level. The method for non-invasively estimating a blood glucose level according to Claim 1.

5. further comprising calculating at least one peak-to-peak slope between these first feature peak positions for each of the plurality of ECG waveforms. wherein The estimation of the user's blood glucose level is further performed based on the at least one peak-to-peak slope, the method for non-invasively estimating the blood glucose level according to claim 1.

6. extracting at least three second electrocardiogram features from each of the plurality of electrocardiogram waveforms of the user; calculating at least one incidence rate based on these second electrocardiogram features; further comprising the estimation of the user's blood glucose level is further performed based on the at least one incidence rate, and these second electrocardiogram features are selected from the group consisting of P wave, Q wave, R wave, S wave, T wave, and U wave, the method for non-invasively estimating the blood glucose level according to claim 1.

7. Calculating the at least one incidence rate based on these second electrocardiogram features includes determining the occurrence interval of each of the plurality of electrocardiogram waveforms of the user; calculating a first peak-to-peak and a second peak-to-peak within the occurrence interval for each of the plurality of electrocardiogram waveforms based on these second electrocardiogram features; calculating the peak-to-peak ratio of the first peak-to-peak to the second peak-to-peak for each of the plurality of electrocardiogram waveforms; comparing the peak-to-peak ratio of each of the plurality of electrocardiogram waveforms with a preset ratio and generating a comparison result; calculating the at least one incidence rate based on these comparison results; The method for non-invasively estimating the blood glucose level according to claim 6, including.

8. Estimating the user's blood glucose level includes normalizing the at least one peak distance; inputting the normalized at least one peak distance and the at least one incidence rate into a machine learning model; outputting the user's blood glucose level by the machine learning model; including the user's blood glucose level is estimated by the machine learning model based on the normalized at least one peak distance and the at least one incidence rate, the method for non-invasively estimating the blood glucose level according to claim 6.

9. Estimating the user's blood glucose level includes normalizing the at least one peak distance; inputting the normalized at least one peak distance and the at least one incidence rate into a neural network model; Outputting the blood glucose level of the user by the neural network model; including; The method for non-invasively estimating the blood glucose level according to claim 6, wherein the blood glucose level of the user is estimated by the neural network model based on the at least one normalized peak distance and the at least one incidence rate.

10. Extracting at least four third electrocardiogram features from each of the plurality of electrocardiogram waveforms of the user; Calculating at least one amplitude ratio based on these third electrocardiogram features; further including; The estimation of the blood glucose level of the user is performed further based on the at least one amplitude ratio, and These third electrocardiogram features are selected from the group consisting of P wave, Q wave, R wave, S wave, T wave, and U wave. The method for non-invasively estimating the blood glucose level according to claim 1.

11. Calculating the at least one amplitude ratio based on these third electrocardiogram features Based on these third electrocardiogram features, for each of the plurality of electrocardiogram waveforms, calculating a third peak-to-peak, a fourth peak-to-peak, a fifth peak-to-peak, and a sixth peak-to-peak; Calculating a first average value between the third peak-to-peak and the fourth peak-to-peak; Calculating a second average value between the fifth peak-to-peak and the sixth peak-to-peak; Calculating the at least one amplitude ratio of the first average value to the second average value; The method for non-invasively estimating the blood glucose level according to claim 10, including.

12. Estimating the blood glucose level of the user Normalizing the at least one peak distance and the at least one amplitude ratio respectively; Inputting the normalized at least one peak distance and the normalized at least one amplitude ratio into a machine learning model; Outputting the blood glucose level of the user by the machine learning model; including; The method for non-invasively estimating the blood glucose level according to claim 10, wherein the blood glucose level of the user is estimated by the machine learning model based on the normalized at least one peak distance and the normalized at least one amplitude ratio.

13. Estimating the blood glucose level of the user Normalizing the at least one peak distance and the at least one amplitude ratio respectively; Inputting the at least one normalized peak distance and the at least one normalized amplitude ratio into a neural network model; Outputting the blood glucose value of the user by the neural network model; comprising; The blood glucose value of the user is estimated by the neural network model based on the at least one normalized peak distance and the at least one normalized amplitude ratio. A method for non-invasively estimating the blood glucose value according to claim 10.

14. Extracting at least one fourth electrocardiogram feature from each of the plurality of electrocardiogram waveforms of the user; Calculating at least one sharpness result based on the at least one fourth electrocardiogram feature; further comprising; The estimation of the blood glucose value of the user is performed further based on the at least one sharpness result, and The at least one fourth electrocardiogram feature is selected from the group consisting of P wave, Q wave, R wave, S wave, T wave, and U wave. A method for non-invasively estimating the blood glucose value according to claim 1.

15. Calculating the at least one sharpness result based on the at least one fourth electrocardiogram feature includes: Determining the fourth feature peak positions corresponding to the at least one fourth electrocardiogram feature respectively; Calculating a first slope and a second slope respectively based on the fourth feature peak positions; Calculating a first sharpness based on the first slope; Calculating a second sharpness based on the second slope; A method for non-invasively estimating the blood glucose value according to claim 14, comprising.

16. Estimating the blood glucose value of the user includes: Normalizing the at least one peak distance and the at least one sharpness result respectively; Inputting the at least one normalized peak distance and the at least one normalized sharpness result into a machine learning model; Outputting the blood glucose value of the user by the machine learning model; comprising; The blood glucose value of the user is estimated by the machine learning model based on the at least one normalized peak distance and the at least one normalized sharpness result. A method for non-invasively estimating the blood glucose value according to claim 14.

17. Estimating the blood glucose value of the user includes normalizing the at least one peak distance and the at least one sharpness result respectively; inputting the normalized at least one peak distance and the normalized at least one sharpness result into a neural network model; outputting the blood glucose value of the user by the neural network model; comprising; The method for non-invasively estimating the blood glucose value according to claim 14, wherein the blood glucose value of the user is estimated by the neural network model based on the normalized at least one peak distance and the normalized at least one sharpness result.

18. evaluating the quality of a plurality of electrocardiogram waveforms of the user; when the plurality of electrocardiogram waveforms of the user are evaluated as normal electrocardiogram signals, outputting the real-time blood glucose value of the user; when the plurality of electrocardiogram waveforms of the user are evaluated as noise signals, outputting the past blood glucose value of the user; further comprising; The real-time blood glucose value of the user is a blood glucose value estimated in real time, and The method for non-invasively estimating the blood glucose value according to claim 1, wherein the past blood glucose value of the user is a blood glucose value estimated previously.

19. determining whether the number of a plurality of electrocardiogram waveforms of the user is less than a preset value; when the number of the plurality of electrocardiogram waveforms of the user is less than the preset value, receiving the plurality of electrocardiogram waveforms of the user again; The method for non-invasively estimating the blood glucose value according to claim 1, further comprising.

20. A computing device for non-invasively estimating a blood glucose value, which is signal-connected to an electrocardiogram (ECG) measuring device and receives a plurality of electrocardiogram waveforms of a user from the electrocardiogram measuring device, comprising: a memory module; a blood glucose value estimation module configured to be signal-connected to the memory module; comprising; a plurality of program codes are stored in the memory module, and after the blood glucose value estimation module executes the plurality of program codes stored in the memory module, the blood glucose value estimation module executes the steps of the method for non-invasively estimating the blood glucose value according to any one of claims 1 to 19. A computing device for non-invasively estimating a blood glucose value.

21. A device for non-invasively measuring an electrocardiogram signal, electrically connected to a computing device and outputting a plurality of electrocardiogram waveforms of a user to the computing device, comprising: A measurement module having a plurality of electrodes electrically connected to the user; A signal transmission module configured to be electrically connected to the measurement module; Including: The measurement module is configured to measure a plurality of electrocardiogram waveforms of the user, and The signal transmission module transmits a plurality of electrocardiogram waveforms of the user to the computing device, and the computing device is configured to execute the steps of the method for non-invasively estimating the blood glucose value according to any one of claims 1 to 19. A device for non-invasively measuring an electrocardiogram (ECG) signal.

22. A non-transitory computer-readable recording medium on which program code is recorded, which causes a computing device to execute the method for non-invasively estimating the blood glucose according to any one of claims 1 to 19 after executing the program code. Non-transitory computer-readable recording medium.

Images