Blood glucose detection method for noninvasive glucometer and blood glucose detector

Abstract

The invention relates to a noninvasive blood glucose detection technology, in particular to a blood glucose detection method for a noninvasive blood glucose meter and a blood glucose detector. The invention discloses a blood glucose detection method for a noninvasive blood glucose meter. The blood glucose detection method comprises the following steps: acquiring near-infrared light data and mid-infrared light data and calculating corresponding weight coefficients; acquiring a predicted blood glucose value by using the blood glucose detection model, calculating credibility, comparing the credibility with a preset credibility threshold, and executing corresponding operation; and acquiring a blood glucose health assessment result by using the blood glucose health assessment model. The near-infrared spectrum technology and the mid-infrared spectrum technology are combined, the influence of environmental factors is considered in real time, the weight coefficients of the near-infrared light data and the mid-infrared light data are calculated respectively, analysis of the spectrum data is adjusted and optimized, a more accurate blood glucose detection result is provided, and the accuracy of blood glucose detection is improved. And reliable blood glucose measurement values can be obtained under different environmental conditions.

Description

Technical Field

[0001] The present invention relates to a non-invasive blood glucose detection technology, and in particular to a blood glucose detection method and a blood glucose detector for a non-invasive blood glucose meter. Background Art

[0002] The incidence of metabolic diseases such as diabetes has increased year by year, and blood glucose monitoring has become an indispensable part of the daily management of diabetic patients. Traditional blood glucose monitoring methods mainly rely on invasive fingertip blood sampling, which not only causes pain and inconvenience to patients, but also may lead to infection risks. In addition, for patients with brain diseases, such as stroke or brain injury, they need to test their blood glucose regularly at home to ensure safety and adjust treatment plans in time. These patients may have difficulty in moving and are more sensitive to pain. Therefore, traditional invasive blood glucose monitoring methods are particularly inconvenient for them.

[0003] In order to solve this problem, the development of a non-invasive, accurate and convenient blood glucose detection technology has become an urgent need in the medical and health field. At present, there are many non-invasive blood glucose detection technologies on the market, such as optical method, ultrasonic method, electromagnetic method, etc. Among them, the optical method has attracted widespread attention due to its non-invasiveness, rapid response and high sensitivity. The optical method mainly infers the blood glucose concentration by analyzing the propagation characteristics of light in biological tissues, such as absorption, scattering, etc.; however, the existing optical non-invasive blood glucose detection technology still faces some challenges, including the impact of environmental factors on the measurement results, and how to improve the accuracy and reliability of the measurement; in order to overcome these challenges, the present invention aims to provide a more advanced, accurate and user-friendly non-invasive blood glucose detection method to meet the needs of diabetic patients, especially patients with brain diseases who have difficulty in moving, for daily blood glucose monitoring, and bring innovation to the medical and health field. Summary of the invention

[0004] The present invention combines near-infrared spectroscopy technology and mid-infrared spectroscopy technology, and considers the impact of environmental factors such as temperature, humidity and light intensity on the measurement results in real time, and calculates the weight coefficients of near-infrared light data and mid-infrared light data respectively, thereby adjusting and optimizing the analysis of spectral data, providing more accurate blood glucose detection results, ensuring that reliable blood glucose measurement values ​​can be obtained under different environmental conditions, and making blood glucose value prediction and health assessment more accurate.

[0005] A blood glucose detection method for a non-invasive blood glucose meter, comprising:

[0006] When the user uses the blood glucose tester to test blood glucose, the blood glucose tester collects near-infrared light data and mid-infrared light data respectively through near-infrared spectroscopy technology and mid-infrared spectroscopy technology, and simultaneously obtains temperature data, humidity data, and light interference intensity data of the current collection environment, and uses the currently obtained temperature data, humidity data, and light interference intensity data as environmental interference data; and uses the environmental interference data to calculate the weight coefficients of the near-infrared light data and the mid-infrared light data respectively;

[0007] The near-infrared light data, the mid-infrared light data and the corresponding weight coefficients are input into the blood glucose detection model, and the predicted blood glucose value is output; at the same time, the obtained environmental interference data is used to calculate the credibility of the current predicted blood glucose value; it is determined whether the credibility of the current predicted blood glucose value is greater than the preset credibility threshold, and if so, the predicted blood glucose value is displayed through the blood glucose detector; if not, the predicted blood glucose value and the credibility are displayed simultaneously through the blood glucose detector, and the user is prompted that the credibility of the current predicted blood glucose value is low, and it is recommended to change the detection environment;

[0008] The obtained predicted blood glucose value and the corresponding credibility are transmitted to the cloud server, and the detection time of the predicted blood glucose value, the user's most recent meal time and the user's exercise status before obtaining the predicted blood glucose value are recorded; several recently obtained predicted blood glucose values ​​and the corresponding credibility, detection time, most recent meal time and exercise status are arranged in chronological order to form a blood glucose health assessment data set, and the blood glucose health assessment data set is used as the input of the blood glucose health assessment model to output the blood glucose health assessment result.

[0009] Preferably, the weight coefficients of the near infrared light data and the mid-infrared light data are calculated respectively using the environmental interference data, and the specific operations are as follows:

[0010] Based on the temperature data T, humidity data H and light interference intensity data L in the environmental interference data; for the temperature data T, set the appropriate temperature range to to , if the temperature data T is located at to The temperature influence index is The value is 0; if the temperature data T is less than or equal to , then use the formula Calculate the temperature impact index ; If the temperature data T is greater than or equal to , then use the formula Calculate the temperature impact index ;

[0011] For humidity data H, set the appropriate humidity range to to , if the humidity data H is located at to The humidity influence index is The value is 0; if the humidity data H is less than or equal to , then use the formula Calculate the humidity impact index ; If the humidity data H is greater than or equal to , then use the formula Calculate the humidity impact index ;

[0012] Set the standard light intensity for the light interference intensity data L , if the light interference intensity data L is less than or equal to the standard light intensity , then the light intensity affects the index The value is 0; if the light interference intensity data L is greater than the standard light intensity , then use the formula Calculate the light intensity impact index ;

[0013] The relative contribution coefficients of temperature, humidity and light interference intensity to near-infrared light data are set as , and , and set the relative contribution coefficients of temperature, humidity and light interference intensity to the mid-infrared light data as , and ,in, < , < , > ; Using the formula Calculate the impact index of near infrared light data , using the formula Calculate the influence index of mid-infrared light data ; Finally, use the formula Calculate the weight coefficient for obtaining near-infrared light data , weight coefficient of mid-infrared data By formula Calculate acquisition.

[0014] Preferably, the blood glucose detection model is established based on a fully connected neural network, including a first input layer, a first fully connected layer, a feature weighted layer, a second fully connected layer and a first output layer; the first input layer is used to input near-infrared light data, mid-infrared light data and corresponding weight coefficients; the first fully connected layer is used to extract features of near-infrared light data and mid-infrared light data; the feature weighted layer is used to apply the calculated weight coefficients to weighted processing of near-infrared light data features and mid-infrared light data features; the second fully connected layer is used to perform further feature extraction on the weighted near-infrared light data features and mid-infrared light data features; the first output layer is used to output predicted blood glucose values.

[0015] Preferably, the specific operations for training the blood glucose detection model are as follows:

[0016] Obtain several labeled blood glucose detection training samples, where the content of the label is the actual blood glucose value corresponding to the blood glucose detection training sample, and each blood glucose detection training sample contains a set of near-infrared light data, mid-infrared light data, and weight coefficients of infrared light data and mid-infrared light data; divide all blood glucose detection training samples into a blood glucose detection training set and a blood glucose detection verification set; train a blood glucose detection model using the blood glucose detection training set, and verify the blood glucose detection model using the blood glucose detection verification set to obtain a first verification result; determine whether the first verification result meets a preset first training condition, and if so, output the trained blood glucose detection model; if not, continue to use the blood glucose detection training set to train the blood glucose detection model.

[0017] Preferably, the obtained environmental interference data is used to calculate the credibility of the current predicted blood sugar value, and the specific operation is as follows:

[0018] For the temperature data T, in the suitable temperature range to Based on the temperature range, set the temperature range to to ; When the temperature data T is within the appropriate temperature range to When the temperature is between The value is 1; when the temperature data T exceeds the temperature range to When the temperature score The value is 0; when the temperature data T is to When , use the formula Calculate the temperature score ; When the temperature data T is located at to When between Calculate the temperature score ;

[0019] For humidity data H, in the suitable humidity range to Based on the humidity, set the humidity range to to ; When the humidity data H is within the appropriate humidity range to When the humidity score is between The value is 1; when the humidity data H exceeds the humidity usage range to When the humidity score The value is 0; when the humidity data H is to When , use the formula Calculate the moisture score ; When the humidity data H is located at to When between Calculate the moisture score ;

[0020] For the light interference intensity data L, at standard light intensity Based on the setting, limit the light intensity ; When the light interference intensity data L is less than or equal to the standard light intensity When the light intensity score The value is 1; when the light interference intensity data L is at the standard light intensity With limited light intensity When between Calculate the light intensity score ; When the light interference intensity data L is greater than or equal to the limit light intensity When the light intensity score The value is 0;

[0021] Finally, using the formula Calculate the credibility S of the currently acquired predicted blood sugar value; , and are the weights of the influence of temperature, humidity and light interference intensity on the prediction accuracy of blood glucose value, and + + =1.

[0022] Preferably, the blood glucose health assessment model is established based on the LSTM model, including a second input layer, an LSTM layer, a third fully connected layer and a second output layer; the second input layer is used to receive the input blood glucose health assessment data set; the LSTM layer is used to extract the temporal characteristics of the blood glucose health assessment data set; the third fully connected layer is used to further extract features from the output of the LSTM layer; the second output layer is used to output the blood glucose health assessment results.

[0023] Preferably, the specific operations for training the blood sugar health assessment model are as follows:

[0024] Obtain several health assessment training samples with labeled blood glucose health assessment results, each blood glucose health assessment training sample contains a set of predicted blood glucose values ​​sorted by time and corresponding credibility, detection time, recent meal time and exercise status; divide all health assessment training samples into a health assessment training set and a health assessment verification set, use the predicted blood glucose values ​​sorted by time and corresponding credibility, detection time, recent meal time and exercise status in the health assessment training set as the input of the blood glucose health assessment model, and the blood glucose health assessment results as the output of the blood glucose health assessment model, train the blood glucose health assessment model, and use the health assessment verification set to verify the blood glucose health assessment model to obtain a second verification result; determine whether the second verification result meets the preset second training condition, and if so, output the trained blood glucose health assessment model; if not, continue to train the blood glucose health assessment model through the health assessment training set.

[0025] Preferably, a blood glucose detector is provided, wherein the blood glucose detector is applied to any of the above-mentioned blood glucose detection methods for a non-invasive blood glucose meter.

[0026] The present invention has the following advantages:

[0027] 1. The present invention combines near-infrared spectroscopy technology and mid-infrared spectroscopy technology, and considers the impact of environmental factors such as temperature, humidity and light intensity on the measurement results in real time, and calculates the weight coefficients of near-infrared light data and mid-infrared light data respectively, thereby adjusting and optimizing the analysis of spectral data, providing more accurate blood glucose test results, ensuring that reliable blood glucose measurement values ​​can be obtained under different environmental conditions, and making blood glucose value prediction and health assessment more accurate.

[0028] 2. The present invention records and analyzes the user's blood sugar data through a cloud server, performs analysis through a blood sugar health assessment model, obtains blood sugar health assessment results, provides users with personalized health management suggestions, and enhances the user's understanding and management capabilities of their own health conditions; this convenience and user-friendly design make the present invention very suitable for daily home use, improving the patient's quality of life and the efficiency of diabetes management. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A schematic diagram of a flow chart of a blood glucose detection method for a non-invasive blood glucose meter according to an embodiment of the present invention;

[0030] Figure 2 It is a back view of the blood glucose detector of the present invention;

[0031] Figure 3 It is a bottom view of the blood glucose detector of the present invention;

[0032] Figure 4 It is a top view of the blood glucose detector of the present invention;

[0033] Figure 5 is a first perspective view of the blood glucose detector of the present invention;

[0034] Figure 6 is a second perspective view of the blood glucose detector of the present invention;

[0035] Figure 7 It is a right side view of the blood glucose detector of the present invention;

[0036] Figure 8 This is a front view of the blood glucose detector of the present invention;

[0037] Fig. 9 It is a left view of the blood glucose detector of the present invention. DETAILED DESCRIPTION

[0038] In order to enable persons skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0039] Embodiment, a blood glucose detection method for a non-invasive blood glucose meter, such as Figure 1 As shown, including:

[0040] For patients with brain diseases, such as stroke or brain injury, they need to test their blood sugar regularly at home to ensure safety and adjust treatment plans in time; these patients may have difficulty in moving and are more sensitive to pain, so the blood sugar detection method for non-invasive blood glucose meter of the present invention can be applied; when the user uses the blood glucose detector to detect blood sugar, the blood glucose detector collects near-infrared light data and mid-infrared light data respectively through near-infrared spectroscopy technology and mid-infrared spectroscopy technology. Near-infrared light can penetrate the surface of the skin and interact with glucose molecules in the subcutaneous tissue, thereby carrying information about blood sugar levels; mid-infrared light is more sensitive to the absorption of glucose molecules, so it can provide more direct The blood glucose concentration information is obtained; at the same time, the temperature data, humidity data, and light interference intensity data of the current collection environment are obtained, and the currently obtained temperature data, humidity data, and light interference intensity data are used as environmental interference data. The ambient temperature will affect the propagation of light in the skin and the activity of glucose molecules, so the temperature data is crucial for calibrating the measurement results; the humidity on the skin surface will affect the scattering and absorption of light, thereby affecting the accuracy of blood glucose measurement; the interference of external light sources may affect the collection of spectral data, so it is necessary to measure and consider the light interference intensity; the environmental interference data is used to calculate the weight coefficients of near-infrared light data and mid-infrared light data, respectively, to reduce the impact of environmental factors on blood glucose measurement results;

[0041] Near-infrared light data, mid-infrared light data and corresponding weight coefficients are input into the blood glucose detection model to output the predicted blood glucose value; at the same time, the obtained environmental interference data is used to calculate the credibility of the current predicted blood glucose value. Credibility is a quantitative indicator that reflects the reliability of blood glucose measurement results under current environmental conditions. At the same time, the credibility of the predicted blood glucose value plays a very important role in the subsequent analysis of the blood glucose health assessment model. It not only affects the judgment of blood glucose status, but also helps the model to more accurately understand the credibility of different measurement data. Blood glucose values ​​with higher credibility have a greater impact on the blood glucose health assessment model, while blood glucose values ​​with lower credibility have a smaller impact on the results; this helps to avoid the impact of low-credibility data on blood glucose health The interference of the training and evaluation of the health assessment model can be eliminated, so as to better evaluate and predict the blood sugar health status; determine whether the credibility of the current predicted blood sugar value is greater than the preset credibility threshold, if so, the predicted blood sugar value is displayed through the blood glucose tester; if not, the predicted blood sugar value and credibility are displayed simultaneously through the blood glucose tester, and the user is prompted that the credibility of the current predicted blood sugar value is low, and it is recommended to change the testing environment; in this way, the blood glucose tester can not only provide blood sugar prediction values, but also evaluate the reliability of the measurement results, ensuring that the user obtains accurate and reliable blood sugar monitoring information; this design improves the practicality of non-invasive blood sugar testing and the user's trust, and also provides users with suggestions for environmental adjustments to obtain better measurement results;

[0042] The obtained predicted blood glucose values ​​and corresponding credibility are transmitted to the cloud server for further analysis and storage; the cloud server provides a centralized platform for collecting and processing data from multiple users, and records the detection time of the predicted blood glucose value, the user's most recent meal time, and the user's exercise status before obtaining the predicted blood glucose value. This information is recorded together with the predicted blood glucose value because they may affect blood glucose levels; for example, blood glucose levels usually rise after eating, but may fall after exercise; several recently obtained predicted blood glucose values ​​and the corresponding credibility, detection time, most recent meal time, and exercise status are arranged in chronological order to form a blood glucose health assessment dataset. This dataset provides a comprehensive perspective for analyzing the user's blood glucose change trends and patterns. The blood glucose health assessment dataset is used as the input of the blood glucose health assessment model, and the blood glucose health assessment result is output. This result is an overall evaluation of blood glucose control, which helps users understand their own blood glucose management and adjust treatment plans in a timely manner.

[0043] The weight coefficients of near-infrared light data and mid-infrared light data are calculated using the environmental interference data. The specific operations are as follows:

[0044] Based on the temperature data T, humidity data H and light interference intensity data L in the environmental interference data; for the temperature data T, set the appropriate temperature range to to , if the temperature data T is located at to The temperature influence index is The value is 0; if the temperature data T is less than or equal to , then use the formula Calculate the temperature impact index ; If the temperature data T is greater than or equal to , then use the formula Calculate the temperature impact index ;

[0045] For humidity data H, set the appropriate humidity range to to , if the humidity data H is located at to The humidity influence index is The value is 0; if the humidity data H is less than or equal to , then use the formula Calculate the humidity impact index ; If the humidity data H is greater than or equal to , then use the formula Calculate the humidity impact index ;

[0046] Set the standard light intensity for the light interference intensity data L , if the light interference intensity data L is less than or equal to the standard light intensity , then the light intensity affects the index The value is 0; if the light interference intensity data L is greater than the standard light intensity , then use the formula Calculate the light intensity impact index ;

[0047] The relative contribution coefficients of temperature, humidity and light interference intensity to near-infrared light data are set as , and , and set the relative contribution coefficients of temperature, humidity and light interference intensity to the mid-infrared light data as , and , where NIR stands for near infrared and MIR stands for mid infrared. < , < , > ; Because temperature affects the absorption and scattering characteristics of infrared light; mid-infrared light is generally more sensitive to temperature changes, so under high temperature or unstable temperature conditions, the weight of mid-infrared light data may need to be reduced; changes in humidity may affect both near-infrared and mid-infrared spectra, but the impact on mid-infrared light is usually more significant because the absorption peak of moisture appears in the mid-infrared region; changes in ambient light may have a greater impact on near-infrared light because near-infrared light has a shorter wavelength and is easily interfered by visible light;

[0048] Using the formula Calculate the impact index of near infrared light data , using the formula Calculate the influence index of mid-infrared light data ; Finally, use the formula Calculate the weight coefficient for obtaining near-infrared light data , weight coefficient of mid-infrared data By formula Calculate acquisition.

[0049] The blood glucose detection model is based on a fully connected neural network, which is a basic deep learning model consisting of multiple fully connected layers, in which each neuron is connected to all neurons in the previous layer to form a dense network structure; this network structure allows information to flow freely in the network without being restricted by spatial relationships, enabling it to learn complex nonlinear relationships and patterns; fully connected neural networks are usually used to process structured data, such as tabular data, and can also be used as part of other complex network structures for feature extraction and classification tasks; during the training process, the prediction results are calculated through the forward propagation of the network, and then the weights of the network are adjusted through back propagation to Minimize the prediction error, thereby improving the accuracy and generalization ability of the model; the blood glucose detection model includes a first input layer, a first fully connected layer, a feature weighted layer, a second fully connected layer and a first output layer; the first input layer is used to input near-infrared light data, mid-infrared light data and corresponding weight coefficients; the first fully connected layer is used to extract the features of near-infrared light data and mid-infrared light data; the feature weighted layer is used to apply the calculated weight coefficients to weight the features of near-infrared light data and mid-infrared light data; the second fully connected layer is used to further extract features from the weighted near-infrared light data features and mid-infrared light data features; the first output layer is used to output the predicted blood glucose value.

[0050] The specific operations for training the blood glucose detection model are as follows:

[0051] Obtain several labeled blood glucose detection training samples, where the content of the label is the actual blood glucose value corresponding to the blood glucose detection training sample, and each blood glucose detection training sample contains a set of near-infrared light data, mid-infrared light data, and weight coefficients of infrared light data and mid-infrared light data. The pre-calculated weight coefficients remain unchanged during the training process and are only used to weight the features; divide all blood glucose detection training samples into a blood glucose detection training set and a blood glucose detection verification set; train the blood glucose detection model using the blood glucose detection training set, and verify the blood glucose detection model using the blood glucose detection verification set to obtain a first verification result; determine whether the first verification result meets the preset first training condition, and if so, output the trained blood glucose detection model; if not, continue to use the blood glucose detection training set to train the blood glucose detection model.

[0052] Using the acquired environmental interference data, the credibility of the current predicted blood sugar value is calculated. The specific operations are as follows:

[0053] For the temperature data T, in the suitable temperature range to Based on the temperature range, set the temperature range to to ; When the temperature data T is within the appropriate temperature range to When the temperature is between The value is 1; when the temperature data T exceeds the temperature range to When the temperature score The value is 0; when the temperature data T is to When , use the formula Calculate the temperature score ; When the temperature data T is located at to When between Calculate the temperature score ;

[0054] For humidity data H, in the suitable humidity range to Based on the humidity, set the humidity range to to ; When the humidity data H is within the appropriate humidity range to When the humidity score is between The value is 1; when the humidity data H exceeds the humidity usage range to When the humidity score The value is 0; when the humidity data H is to When , use the formula Calculate the moisture score ; When the humidity data H is located at to When between Calculate the moisture score ;

[0055] For the light interference intensity data L, at standard light intensity Based on the setting, limit the light intensity ; When the light interference intensity data L is less than or equal to the standard light intensity When the light intensity score The value is 1; when the light interference intensity data L is at the standard light intensity With limited light intensity When between Calculate the light intensity score ; When the light interference intensity data L is greater than or equal to the limit light intensity When the light intensity score The value is 0;

[0056] Finally, using the formula Calculate the credibility S of the currently acquired predicted blood sugar value; , and are the weights of the influence of temperature, humidity and light interference intensity on the prediction accuracy of blood glucose value, and + + =1.

[0057] The blood sugar health assessment model is established based on the LSTM model, which is a special type of recurrent neural network that can learn and memorize long-term dependent information. LSTM controls the flow of information by introducing input gates, forget gates, and output gates. These gating mechanisms allow the network to selectively retain or discard information, thereby effectively solving the gradient vanishing and gradient exploding problems of traditional RNNs when processing long sequence data. The blood sugar health assessment model includes a second input layer, an LSTM layer, a third fully connected layer, and a second output layer. The second input layer is used to receive the input blood sugar health assessment data set. The LSTM layer is used to extract the temporal characteristics of the blood sugar health assessment data set. The third fully connected layer is used to further extract features from the output of the LSTM layer. The second output layer is used to output the blood sugar health assessment results.

[0058] The specific operations for training the blood sugar health assessment model are as follows:

[0059] Obtain several health assessment training samples with labeled blood glucose health assessment results, each blood glucose health assessment training sample contains a set of predicted blood glucose values ​​sorted by time and corresponding credibility, detection time, recent meal time and exercise status; divide all health assessment training samples into a health assessment training set and a health assessment verification set, use the predicted blood glucose values ​​sorted by time and corresponding credibility, detection time, recent meal time and exercise status in the health assessment training set as the input of the blood glucose health assessment model, and the blood glucose health assessment results as the output of the blood glucose health assessment model, train the blood glucose health assessment model, and use the health assessment verification set to verify the blood glucose health assessment model to obtain a second verification result; determine whether the second verification result meets the preset second training condition, and if so, output the trained blood glucose health assessment model; if not, continue to train the blood glucose health assessment model through the health assessment training set.

[0060] A blood glucose tester, such as Figure 2-Figure 9 As shown, the blood glucose detector is applied to any of the above-mentioned blood glucose detection methods for non-invasive blood glucose meters.

[0061] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all such improvements and changes should fall within the scope of protection of the appended claims of the present invention. Parts not described in detail in this specification belong to the prior art known to those skilled in the art.

Claims

1. A blood glucose detection method for a non-invasive blood glucose meter, characterized in that: include: When the user uses the blood glucose tester to test blood glucose, the blood glucose tester collects near-infrared light data and mid-infrared light data respectively through near-infrared spectroscopy technology and mid-infrared spectroscopy technology, and simultaneously obtains temperature data, humidity data, and light interference intensity data of the current collection environment, and uses the currently obtained temperature data, humidity data, and light interference intensity data as environmental interference data; and uses the environmental interference data to calculate the weight coefficients of the near-infrared light data and the mid-infrared light data respectively; The near-infrared light data, the mid-infrared light data and the corresponding weight coefficients are input into the blood glucose detection model, and the predicted blood glucose value is output; at the same time, the obtained environmental interference data is used to calculate the credibility of the current predicted blood glucose value; it is determined whether the credibility of the current predicted blood glucose value is greater than the preset credibility threshold, and if so, the predicted blood glucose value is displayed through the blood glucose detector; if not, the predicted blood glucose value and the credibility are displayed simultaneously through the blood glucose detector, and the user is prompted that the credibility of the current predicted blood glucose value is low, and it is recommended to change the detection environment; The obtained predicted blood glucose value and the corresponding credibility are transmitted to the cloud server, and the detection time of the predicted blood glucose value, the user's most recent meal time and the user's exercise status before obtaining the predicted blood glucose value are recorded; several recently obtained predicted blood glucose values ​​and the corresponding credibility, detection time, most recent meal time and exercise status are arranged in chronological order to form a blood glucose health assessment data set, and the blood glucose health assessment data set is used as the input of the blood glucose health assessment model to output the blood glucose health assessment result.

2. A blood glucose detection method for a non-invasive blood glucose meter according to claim 1, characterized in that: The weight coefficients of near-infrared light data and mid-infrared light data are calculated using the environmental interference data. The specific operations are as follows: Based on the temperature data T, humidity data H and light interference intensity data L in the environmental interference data; for the temperature data T, set the appropriate temperature range to to , if the temperature data T is located at to The temperature influence index is The value is 0; if the temperature data T is less than or equal to , then use the formula Calculate the temperature impact index ; If the temperature data T is greater than or equal to , then use the formula Calculate the temperature impact index ; For humidity data H, set the appropriate humidity range to to , if the humidity data H is located at to The humidity influence index is The value is 0; if the humidity data H is less than or equal to , then use the formula Calculate the humidity impact index ; If the humidity data H is greater than or equal to , then use the formula Calculate the humidity impact index ; Set the standard light intensity for the light interference intensity data L , if the light interference intensity data L is less than or equal to the standard light intensity , then the light intensity affects the index The value is 0; if the light interference intensity data L is greater than the standard light intensity , then use the formula Calculate the light intensity impact index ; The relative contribution coefficients of temperature, humidity and light interference intensity to near-infrared light data are set as , and , and set the relative contribution coefficients of temperature, humidity and light interference intensity to the mid-infrared light data as , and ,in, < , < , > ; Using the formula Calculate the impact index of near infrared light data , using the formula Calculate the influence index of mid-infrared light data ; Finally, use the formula Calculate the weight coefficient for obtaining near-infrared light data , weight coefficient of mid-infrared data By formula Calculate and obtain.

3. A blood glucose detection method for a non-invasive blood glucose meter according to claim 2, characterized in that: The blood glucose detection model is established based on a fully connected neural network, including a first input layer, a first fully connected layer, a feature weighted layer, a second fully connected layer and a first output layer; the first input layer is used to input near-infrared light data, mid-infrared light data and corresponding weight coefficients; the first fully connected layer is used to extract features of near-infrared light data and mid-infrared light data; The feature weighting layer is used to apply the calculated weight coefficients to perform weighted processing on the near-infrared light data features and the mid-infrared light data features; the second fully connected layer is used to perform further feature extraction on the weighted near-infrared light data features and the mid-infrared light data features; the first output layer is used to output the predicted blood glucose value.

4. A blood glucose detection method for a non-invasive blood glucose meter according to claim 3, characterized in that: The specific operations for training the blood glucose detection model are as follows: Obtain several labeled blood glucose detection training samples, where the content of the label is the actual blood glucose value corresponding to the blood glucose detection training sample, and each blood glucose detection training sample contains a set of near-infrared light data, mid-infrared light data, and weight coefficients of infrared light data and mid-infrared light data; divide all blood glucose detection training samples into a blood glucose detection training set and a blood glucose detection verification set; train a blood glucose detection model using the blood glucose detection training set, and verify the blood glucose detection model using the blood glucose detection verification set to obtain a first verification result; determine whether the first verification result meets a preset first training condition, and if so, output the trained blood glucose detection model; if not, continue to use the blood glucose detection training set to train the blood glucose detection model.

5. A blood glucose detection method for a non-invasive blood glucose meter according to claim 4, characterized in that: Using the acquired environmental interference data, the credibility of the current predicted blood sugar value is calculated. The specific operations are as follows: For the temperature data T, in the suitable temperature range to Based on the temperature range, set the temperature range to to ; When the temperature data T is within the appropriate temperature range to When the temperature is between The value is 1; when the temperature data T exceeds the temperature range to When the temperature score The value is 0; when the temperature data T is to When , use the formula Calculate the temperature score ; When the temperature data T is located at to When between Calculate the temperature score ; For humidity data H, in the suitable humidity range to Based on the humidity, set the humidity range to to ; When the humidity data H is within the appropriate humidity range to When the humidity score is between The value is 1; when the humidity data H exceeds the humidity usage range to When the humidity score The value is 0; when the humidity data H is to When , use the formula Calculate the moisture score ; When the humidity data H is located at to When between Calculate the moisture score ; For the light interference intensity data L, at standard light intensity Based on the setting, limit the light intensity ; When the light interference intensity data L is less than or equal to the standard light intensity When the light intensity score The value is 1; when the light interference intensity data L is at the standard light intensity With limited light intensity When between Calculate the light intensity score ; When the light interference intensity data L is greater than or equal to the limit light intensity When the light intensity score The value is 0; Finally, using the formula Calculate the credibility S of the currently acquired predicted blood sugar value; , and are the weights of the influence of temperature, humidity and light interference intensity on the prediction accuracy of blood glucose value, and + + =1.

6. A blood glucose detection method for a non-invasive blood glucose meter according to claim 5, characterized in that: The blood sugar health assessment model is established based on the LSTM model, including the second input layer, the LSTM layer, the third fully connected layer and the second output layer; the second input layer is used to receive the input blood sugar health assessment data set; the LSTM layer is used to extract the temporal characteristics of the blood sugar health assessment data set; The third fully connected layer is used to further extract features from the output of the LSTM layer; The second output layer is used to output blood sugar health assessment results.

7. A blood glucose detection method for a non-invasive blood glucose meter according to claim 6, characterized in that: The specific operations for training the blood sugar health assessment model are as follows: Obtain several health assessment training samples with labeled blood glucose health assessment results, each blood glucose health assessment training sample contains a set of predicted blood glucose values ​​sorted by time and corresponding credibility, detection time, recent meal time and exercise status; divide all health assessment training samples into a health assessment training set and a health assessment verification set, use the predicted blood glucose values ​​sorted by time and corresponding credibility, detection time, recent meal time and exercise status in the health assessment training set as the input of the blood glucose health assessment model, and the blood glucose health assessment results as the output of the blood glucose health assessment model, train the blood glucose health assessment model, and use the health assessment verification set to verify the blood glucose health assessment model to obtain a second verification result; determine whether the second verification result meets the preset second training condition, and if so, output the trained blood glucose health assessment model; if not, continue to train the blood glucose health assessment model through the health assessment training set.

8. A blood glucose detector, characterized in that: The blood glucose detector is applied to the blood glucose detection method for a non-invasive blood glucose meter as described in any one of claims 1 to 7 above.

Images