A method and system for dynamic monitoring of blood sugar in diabetes

By dynamically dividing the subsequence of blood glucose data and adjusting the K value, and dynamic monitoring of blood glucose is solved by combining historical data, the problem of low accuracy in abnormal identification of blood glucose data in traditional methods is solved, and the accuracy of blood glucose monitoring is improved.

CN118806274BActive Publication Date: 2024-11-22QIANJIANGYUE HEALTH IND (GUANGZHOU) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411302845.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-19
Publication Date
2024-11-22
Estimated Expiration
2044-09-19

AI Technical Summary

Technical Problem

Due to measurement errors and environmental factors, traditional blood sugar dynamic monitoring methods have low accuracy in identifying abnormal blood sugar data, which affects monitoring accuracy.

Method used

By obtaining blood sugar data in real time, dividing blood sugar data subsequences of different properties, analyzing data fluctuations and blood sugar changes, dynamically adjusting K values, combining historical data for abnormal detection, and using the KNN algorithm for data correction.

Benefits of technology

It improves the accuracy and accuracy of blood sugar dynamic monitoring, adapts to blood sugar control under different states, and reduces the problems of overfitting and underfitting.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118806274B_ABST
    Figure CN118806274B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of medical care informatics technology, and specifically to a method and system for dynamic monitoring of blood sugar in diabetes, including: obtaining the first and second blood sugar data subsequences in the initial blood sugar data sequence, determining the stability index of the first and second blood sugar data subsequences; determining the wearer's control of blood sugar according to the stability index, using the blood sugar control degree to correct the initial K value, and determining the corrected K value of each blood sugar data in the initial blood sugar data sequence; performing KNN algorithm abnormality detection on the initial blood sugar data sequence to obtain distorted blood oxygen data, and updating the distorted blood oxygen data to obtain a real blood sugar data sequence. The present invention analyzes the wearer's blood sugar control degree under different states, combines the actual fluctuation situation, and adaptively determines the abnormal detection K value for each blood sugar data, thereby improving the accuracy of abnormal detection of blood sugar data and further improving the accuracy of dynamic blood sugar monitoring.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of medical care informatics, and in particular to a method and system for dynamic monitoring of blood sugar in diabetes. Background Art

[0002] Diabetes is a common chronic metabolic disease that seriously affects the quality of life and health of patients. Accurate monitoring and control of blood sugar levels are crucial for the treatment and management of diabetic patients. Traditional blood sugar monitoring methods mainly rely on intermittent sampling and cannot monitor the dynamic changes of blood sugar in real time. The continuous glucose monitoring (CGM) system collects continuous blood sugar data through non-invasive or minimally invasive means, providing more comprehensive and accurate blood sugar information for diabetic patients. With the continuous advancement of science and technology and people's emphasis on health management, CGM systems have been widely used in the field of diabetes treatment. CGM systems transmit blood sugar data to monitoring devices or mobile applications in real time through sensors, helping patients understand their blood sugar levels in real time and take timely measures to adjust them.

[0003] However, due to factors such as sensor distortion and patient movement, the blood glucose data collected by the CGM system may have measurement errors. The abnormal data generated by the measurement errors will have an impact on blood glucose detection and overall trend analysis, which can easily reduce the accuracy and reliability of dynamic blood glucose monitoring. Therefore, before analyzing the blood glucose data, data anomaly identification and preprocessing are required.

[0004] Traditional abnormal data detection methods are mainly based on the K-Nearest Neighbor (KNN) algorithm. The K value in the K-Nearest Neighbor algorithm is a fixed parameter, which represents the number of nearest neighbor data considered when calculating sample similarity. However, when the K-Nearest Neighbor algorithm is actually applied to abnormal blood glucose data detection, different patients, different time periods, and even different environmental factors may have different degrees of influence on blood glucose data, which leads to low accuracy of abnormal data detection results under a fixed K value, and overfitting and underfitting problems of blood glucose data, which affects the accuracy of dynamic blood glucose monitoring. Summary of the invention

[0005] In order to solve the technical problem that the accuracy of abnormal recognition of blood glucose data is low, which affects the accuracy of dynamic blood glucose monitoring, the purpose of the present invention is to provide a method and system for dynamic blood glucose monitoring of diabetes. The technical solutions adopted are as follows:

[0006] An embodiment of the present invention provides a method for dynamic monitoring of blood glucose in diabetes, the method comprising the following steps:

[0007] Acquire the blood sugar data of the wearer of the blood sugar monitoring device at each preset time in a day in real time to form an initial blood sugar data sequence; wherein the initial blood sugar data sequence is composed of a first blood sugar data subsequence and a second blood sugar data subsequence;

[0008] Analyzing the data fluctuation degree of the first blood sugar data subsequence and the degree of change of the overall blood sugar level according to the first blood sugar data subsequence, and determining the stability index of the first blood sugar data subsequence;

[0009] Determine the rising phase and the falling phase of each abnormal blood sugar data sequence according to the second blood sugar data subsequence; determine the stability index of the second blood sugar data subsequence according to each abnormal blood sugar data sequence and the time ratio of the rising phase and the falling phase, and the abnormal blood sugar threshold;

[0010] Obtaining stability indicators of a first preset number of historical pending blood sugar data subsequences, and determining the wearer's blood sugar control degree within the time range corresponding to the pending blood sugar data subsequence according to the pending blood sugar data subsequence and the stability indicators of each historical pending blood sugar data subsequence; wherein the pending blood sugar data subsequence is the first blood sugar data subsequence or the second blood sugar data subsequence;

[0011] Determine a revised K value for each blood sugar data in the initial blood sugar data sequence according to the pre-obtained initial K value of the blood sugar data, each blood sugar data in the initial blood sugar data sequence, and the degree of blood sugar control at the time point of each blood sugar data within its corresponding time range;

[0012] According to the corrected K value of each blood sugar data, the initial blood sugar data sequence is subjected to KNN algorithm abnormality detection to obtain distorted blood oxygen data; each distorted blood oxygen data in the initial blood sugar data sequence is corrected to obtain a true blood sugar data sequence.

[0013] Furthermore, the step of acquiring the first blood sugar data subsequence and the second blood sugar data subsequence comprises:

[0014] Performing second-order difference processing on the initial blood sugar data sequence to obtain a second-order difference sequence of the initial blood sugar data sequence;

[0015] Determine the maximum value and the minimum value in the second-order difference sequence, and form each blood glucose data in the time period corresponding to the maximum value and the minimum value into a blood glucose data subsequence as a second blood glucose data subsequence;

[0016] The blood sugar data in the initial blood sugar data sequence except the second blood sugar data subsequence are combined into a blood sugar data subsequence as a first blood sugar data subsequence.

[0017] Further, analyzing the data fluctuation degree of the first blood sugar data subsequence and the degree of change of the overall blood sugar level according to the first blood sugar data subsequence to determine the stability index of the first blood sugar data subsequence includes:

[0018] Performing straight line fitting on the first blood sugar data subsequence to obtain a fitted straight line; determining a residual sequence between the first blood sugar data subsequence and the fitted straight line, and taking the average value of each residual in the residual sequence as the degree of data fluctuation;

[0019] Determine the blood glucose data at the starting point and the ending point in the first blood glucose data subsequence, calculate the difference between the blood glucose data at the starting point and the ending point, and then calculate the ratio of the difference to the length of the corresponding time period of the first blood glucose data subsequence, and use the ratio as the degree of change of the overall blood glucose level;

[0020] The product of the data fluctuation degree and the overall blood sugar level change degree is normalized in inverse proportion to obtain a stability index of the first blood sugar data subsequence.

[0021] Further, determining the rising phase and the falling phase of each abnormal blood sugar data sequence according to the second blood sugar data subsequence includes:

[0022] Acquire blood glucose data greater than the abnormal blood glucose threshold in the second blood glucose data subsequence, record the blood glucose data greater than the abnormal blood glucose threshold as abnormal blood glucose data, and take a sequence consisting of continuous adjacent abnormal blood glucose data as an abnormal blood glucose data sequence;

[0023] For any abnormal blood sugar data sequence, the largest maximum value point corresponding to the abnormal blood sugar data sequence is determined, and the abnormal blood sugar data sequence is divided into an ascending phase and a descending phase using the largest maximum value point.

[0024] Furthermore, the step of determining the stability index of the second blood glucose data subsequence according to each abnormal blood glucose data sequence and the time ratio of the rising phase and the falling phase, and the abnormal blood glucose threshold, includes:

[0025] Determine the maximum abnormal blood sugar data in each abnormal blood sugar data sequence, calculate the cumulative sum of the differences between each maximum abnormal blood sugar data and the abnormal blood sugar threshold, and use the cumulative sum of all the differences as the first stability factor of the second blood sugar data subsequence;

[0026] Determine a second stability factor of a second blood glucose data subsequence according to a time ratio of a rising phase to a falling phase of each abnormal blood glucose data sequence;

[0027] The product of the first stability factor and the second stability factor is normalized in inverse proportion to obtain a stability index of the second blood glucose data subsequence.

[0028] Furthermore, the determining of the second stability factor of the second blood glucose data subsequence according to the time ratio of the rising phase to the falling phase of each abnormal blood glucose data sequence includes:

[0029] The calculation formula of the second stability factor of the second blood glucose data subsequence is:

[0030] ; In the formula, is the second stability factor of the second blood glucose data subsequence, U is the number of abnormal blood glucose data sequences corresponding to the second blood glucose data subsequence, u is the sequence number of the abnormal blood glucose data sequence corresponding to the second blood glucose data subsequence, is the time ratio of the rising phase and the falling phase of the u-th abnormal blood sugar data sequence, It is the average value of the time ratio between the rising phase and the falling phase of all abnormal blood glucose data series.

[0031] Furthermore, the step of determining the wearer's blood sugar control degree within the time range corresponding to the pending blood sugar data subsequence according to the pending blood sugar data subsequence and the stability index of each historical pending blood sugar data subsequence includes:

[0032] Calculate the average value of the stability index of the pending blood sugar data subsequence and each historical pending blood sugar data subsequence, and then calculate the standard deviation of the stability index of the pending blood sugar data subsequence and each historical pending blood sugar data subsequence;

[0033] The ratio of the average value of the stability index to the standard deviation is used as the blood sugar control factor, and the blood sugar control factor is normalized to obtain the wearer's blood sugar control degree within the corresponding time range of the pending blood sugar data subsequence.

[0034] Further, the method of determining the revised K value of each blood sugar data in the initial blood sugar data sequence according to the pre-obtained initial K value of the blood sugar data, each blood sugar data in the initial blood sugar data sequence, and the degree of blood sugar control at the time point of each blood sugar data within the time range to which it belongs, comprises:

[0035] For any blood sugar data in the initial blood sugar data sequence, a correction coefficient of the blood sugar data is determined according to the blood sugar data, a second preset number of blood sugar data closest to the blood sugar data, and the degree of blood sugar control at the time point of the blood sugar data within the time range to which it belongs;

[0036] The product of the correction coefficient of the blood sugar data and the initial K value is calculated as the corrected K value of the blood sugar data.

[0037] Further, determining the correction coefficient of the blood sugar data according to the blood sugar data, a second preset number of blood sugar data closest to the blood sugar data, and the degree of blood sugar control at the time point of the blood sugar data within the time range to which it belongs, includes:

[0038] The blood glucose data and a second preset number of blood glucose data closest to the blood glucose data are combined into an array, and the first-order difference absolute values ​​corresponding to the array are determined; the average values ​​of all the first-order difference absolute values ​​are normalized in inverse proportion, and the product of the normalized average value and the degree of blood glucose control at the time point of the blood glucose data within the time range to which it belongs is used as the correction coefficient of the blood glucose data.

[0039] An embodiment of the present invention further provides a diabetic blood glucose dynamic monitoring system, comprising a processor and a memory, wherein the processor is used to process instructions stored in the memory to implement a diabetic blood glucose dynamic monitoring method.

[0040] The present invention has the following beneficial effects:

[0041] The present invention provides a method and system for dynamic monitoring of blood sugar in diabetes, which relates to the technical field of measuring blood sugar and is mainly applied to the field of daily monitoring of blood sugar. Firstly, the initial blood sugar data sequence obtained is divided into a first blood sugar data subsequence and a second blood sugar data subsequence of different properties in combination with different manifestations of blood sugar in different states, and then the blood sugar data subsequences of different properties are subjected to targeted analysis and processing, which helps to make the subsequent correction K value more accurate; secondly, when analyzing the degree of control of blood sugar in different states, based on the principle of insulin secretion difference between normal people and diabetic patients, combined with the stability indicators of multiple different historical pending blood sugar data subsequences, it helps to improve the calculation reliability and accuracy of the degree of control of blood sugar, and provides data conditions with more reference value for the subsequent determination of the correction K value; then, the possible changes of blood sugar data are predicted by the degree of control of blood sugar, so as to perform the adaptation of the abnormal detection K value, which can ensure that each blood sugar data has its corresponding correction K value, which is more in line with the specific situation of the individual wearer of the device; finally, the initial blood sugar data sequence is subjected to KNN algorithm abnormal detection based on the correction K value, which can improve the accuracy of abnormal detection and further improve the accuracy of dynamic monitoring of blood sugar in diabetes. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 This is a flow chart of a method for dynamic monitoring of blood glucose in diabetes mellitus according to an embodiment of the present invention;

[0044] Figure 2 This is a flow chart of determining a stability index of a first blood glucose data subsequence in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the technical solutions proposed by the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0046] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0047] The application scenarios targeted by the present invention may be:

[0048] Due to the influence of different patients, collection periods and environmental factors, blood glucose data will fluctuate to varying degrees. Each blood glucose data needs to be combined with the changes in other blood glucose data around it to analyze its own degree of distortion. When using the K nearest neighbor algorithm to detect anomalies in blood glucose data, the traditional abnormal data detection results under a fixed K value are not accurate enough. Blood glucose data is prone to overfitting and underfitting problems, resulting in low accuracy of anomaly detection and affecting the accuracy of dynamic blood glucose monitoring.

[0049] Before adapting the corrected K value of each blood sugar data, it is necessary to understand the causes and characteristics of diabetes and the changes in blood sugar data at different times of the day, so as to analyze the blood sugar regulation or control ability, quantify the changes in blood sugar data, and determine the correction coefficient of the K value when identifying abnormal blood sugar data.

[0050] Diabetes is a group of metabolic diseases characterized by problems with the production or utilization of insulin in the body, leading to abnormally high blood sugar levels. Diabetes can be divided into different types, the two most common of which are type 1 diabetes and type 2 diabetes. Type 1 diabetes is caused by an autoimmune reaction in which the immune system attacks insulin-producing cells, which results in a decrease or cessation of insulin secretion. Patients usually need to inject exogenous insulin because their insulin secretion is disrupted and blood sugar cannot be effectively regulated. Type 2 diabetes is usually associated with decreased insulin efficiency and insulin resistance. Patients' insulin secretion may be normal or increased, and insulin injections may not be required at first, but as the disease progresses, some patients may need insulin treatment.

[0051] In addition, normal people have two modes of insulin secretion: basal insulin secretion and mealtime insulin secretion. Basal insulin secretion refers to the 24-hour continuous release of small doses of insulin by pancreatic beta cells to counteract the release of glucose from liver glycogen to maintain normal blood sugar levels in the non-meal state; mealtime insulin secretion refers to the physiological process in which the pancreas releases additional insulin to regulate blood sugar levels after eating. Generally speaking, the peak secretion of insulin is between 30 and 60 minutes, and then returns to normal levels in 2 to 3 hours.

[0052] Because different patients have different conditions and causes of disease, and different patients have different insulin secretion abilities, the degree of blood sugar control is different, and the corresponding drug dosage is also different. Therefore, different patients have different degrees of blood sugar control under different conditions.

[0053] Normal state: The stronger the blood sugar's self-regulation ability is, the stronger the blood sugar data's volatility is, and the closer the overall trend of the blood sugar data is to a straight line, the stronger the wearer's blood sugar control ability is under normal conditions. When identifying abnormal data for blood sugar data, a larger K value needs to be set to avoid overfitting problems caused by local data changes.

[0054] Postprandial state: Postprandial blood sugar changes of ordinary people usually change within a reasonable range. Only when eating some foods with very high sugar content will high blood sugar be produced, and the secretion of insulin in the body usually changes with the blood sugar content. The speed of blood sugar regulation and the consistency between insulin and blood sugar are relatively high. However, due to the poor insulin secretion ability of diabetic patients, they may need to inject insulin once or multiple times to regulate the blood sugar level in the body. After the injection of insulin, the absorption of insulin, the effect of insulin and the glucose-raising effect of food are different, resulting in different degrees of blood sugar control in patients. Therefore, in terms of blood sugar control, it is strongly correlated with the patient's own situation, medication situation and medication habits. If the patient's blood sugar control is better, then the closer it is to the blood sugar data of ordinary people; if the patient's blood sugar data has a larger extreme value and the more drastic the change, then it means that the patient's blood sugar control is worse. The weaker the patient's blood sugar control, the more likely the patient's blood sugar will fluctuate greatly. Therefore, for the patient's blood sugar data in the postprandial state, a smaller K value needs to be set to reduce underfitting.

[0055] Specifically, the embodiment of the present invention provides a method for dynamic monitoring of blood glucose in diabetes, such as Figure 1 As shown, the following steps are included:

[0056] S1, real-time acquisition of blood glucose data of the wearer of the blood glucose monitoring device at each preset time of the day to form an initial blood glucose data sequence.

[0057] In this embodiment, the blood glucose data of the device wearer at each preset time of the day is collected through the CGM system to form a sequence, and the Gaussian filter kernel with a window size of 5 is set to perform data smoothing on the sequence, and the smoothed sequence is used as the initial blood glucose data sequence. Among them, the implementation process of the CGM system and the Gaussian filter kernel are both prior arts, which are not within the scope of protection of the present invention and will not be elaborated in detail here; the device wearer generally refers to a diabetic patient.

[0058] It should be noted that, first of all, CGM is a minimally invasive blood glucose monitoring system, which reflects the blood glucose level by detecting the glucose concentration of the interstitial fluid of the subcutaneous tissue, and can continuously monitor the changes in the wearer's blood glucose data. Secondly, when collecting blood glucose data, the CGM wearable device is fixed on the user's upper arm or abdomen, and then the probe head is inserted into the subcutaneous tissue of the upper arm or abdomen, and the probe head and the chip in the wearable device are connected by a connecting line, so as to detect the wearer's blood glucose data through the probe head. Finally, when collecting blood glucose data, the collection starts from the first moment, and the time interval can be set to 1 minute; a day refers to 24 hours a day, and a day can be a day of a certain date. The blood glucose data is dynamically monitored once a day, and the blood glucose data of two days is subsequently divided and processed with 0 o'clock every day as the dividing line; of course, the collection frequency and collection period of blood glucose data can be set by the implementer according to the specific actual situation, and no specific limitation is made here.

[0059] The initial blood sugar data sequence of the day is segmented into a first blood sugar data sequence with a slow change and a second blood sugar data sequence with a fast change, so as to facilitate the subsequent quantification of the wearer's blood sugar control degree under different conditions and determine the correction coefficient of each blood sugar data abnormality detection K value. Specifically, the steps of obtaining the first blood sugar data subsequence and the second blood sugar data subsequence may include:

[0060] The first step is to perform second-order difference processing on the initial blood sugar data sequence to obtain a second-order difference sequence of the initial blood sugar data sequence.

[0061] Among them, the implementation process of the second-order difference processing is a prior art and is not within the protection scope of the present invention, and will not be elaborated here.

[0062] The second step is to determine the maximum and minimum values ​​in the second-order difference sequence, and to group the blood glucose data in the time periods corresponding to the maximum and minimum values ​​into a blood glucose data subsequence as the second blood glucose data subsequence.

[0063] The third step is to combine the blood sugar data in the initial blood sugar data sequence except the second blood sugar data subsequence into a blood sugar data subsequence as the first blood sugar data subsequence.

[0064] In this embodiment, based on the difference in the change characteristics of blood glucose data in the postprandial state and the normal state, the second blood glucose data sequence that changes rapidly in the initial blood glucose data sequence can be screened out by second-order difference processing, that is, the blood glucose data in the time period corresponding to the maximum value and the minimum value in the second-order difference sequence are composed into a blood glucose data subsequence as the second blood glucose data subsequence. The second blood glucose data subsequence usually refers to the blood glucose data situation during the day, and the first blood glucose data usually refers to the blood glucose data situation at night. Of course, the implementer can also adopt other methods to roughly divide the initial blood glucose data sequence into subsequences with different change characteristics, that is, the initial blood glucose data sequence is composed of the first blood glucose data subsequence and the second blood glucose data subsequence.

[0065] S2. Analyze the data fluctuation degree of the first blood glucose data subsequence and the change degree of the overall blood glucose level according to the first blood glucose data subsequence, and determine the stability index of the first blood glucose data subsequence.

[0066] The stability index provides data conditions for subsequent analysis of the patient's blood sugar control level. The stability index of the first blood sugar data subsequence can characterize the stability of the blood sugar data changes of patients under normal conditions on a certain date. The worse the stability of the blood sugar data changes, the worse the patient's basal insulin secretion ability, and the weaker the patient's blood sugar control ability under normal conditions.

[0067] The flow chart for determining the stability index of the first blood glucose data subsequence is as follows: Figure 2 As shown, specifically including:

[0068] In the first step, a straight line fitting is performed on the first blood sugar data subsequence to obtain a fitting straight line; a residual sequence between the first blood sugar data subsequence and the fitting straight line is determined, and the average value of each residual in the residual sequence is taken as the degree of data fluctuation.

[0069] In this embodiment, the least square method is used to perform straight line fitting. The implementation process of the least square method is a prior art and will not be elaborated in detail here.

[0070] The second step is to determine the blood glucose data at the starting point and the ending point in the first blood glucose data subsequence, calculate the difference between the blood glucose data at the starting point and the ending point, and then calculate the ratio of the difference to the length of the corresponding time period of the first blood glucose data subsequence, and use the ratio as the degree of change of the overall blood glucose level.

[0071] In this embodiment, the difference between the blood glucose data at the starting point and the blood glucose data at the ending point may be the absolute value of the difference between the two blood glucose data; the time period length may refer to the time length of the stable state.

[0072] The third step is to perform inverse proportional normalization processing on the product of the data fluctuation degree and the overall blood sugar level change degree to obtain the stability index of the first blood sugar data subsequence.

[0073] In this embodiment, the means for implementing the inverse proportional normalization process may be exp(-). Of course, the implementer may also use other methods to implement the inverse proportional normalization process of data.

[0074] As an example, the calculation formula for the stability index of the first blood glucose data subsequence on date i may be:

[0075] ; In the formula, is the stability index of the first blood glucose data subsequence on date i, exp is an exponential function with a natural constant as the base, is the data fluctuation degree of the first blood glucose data subsequence on date i, is the blood sugar data at the starting point of the first blood sugar data subsequence on date i, is the blood sugar data at the end point of the first blood sugar data subsequence on date i, To find the absolute value function, is the length of the period corresponding to the first blood glucose data subsequence on date i, is the overall blood sugar level change degree of the first blood sugar data subsequence on date i.

[0076] In the calculation formula of the stability index of the first blood glucose data subsequence, It is also the average value of all residuals corresponding to the first blood sugar data subsequence, which can characterize the data fluctuation degree of the blood sugar data under normal conditions. The greater the data fluctuation degree, the greater the fluctuation of the blood sugar data in the first blood sugar data subsequence, and the worse the stability; It can characterize the overall change in the patient's blood sugar level under normal conditions, that is, the increase in the patient's blood sugar under normal conditions. The greater the change in the overall blood sugar level, the more the patient's blood sugar will gradually increase under normal conditions, the worse the patient's basal insulin secretion ability is, and the weaker the patient's blood sugar control ability is under normal conditions; the degree of data fluctuation and the degree of change in the overall blood sugar level are negatively correlated with the stability of blood sugar data, so it is necessary to perform inverse proportional processing on the product of the data fluctuation degree and the degree of change in the overall blood sugar level; in addition, in order to facilitate subsequent data processing and improve calculation efficiency, the value range of the stability index is limited to between 0 and 1; analyzing the stability of the first blood sugar data subsequence from two angles can effectively improve the numerical accuracy of the stability index.

[0077] S3, determining the rising phase and falling phase of each abnormal blood glucose data sequence according to the second blood glucose data subsequence; determining the stability index of the second blood glucose data subsequence according to each abnormal blood glucose data sequence and the time ratio of its rising phase and falling phase, and the abnormal blood glucose threshold.

[0078] After obtaining the stability index of the blood sugar data change under the normal state, it is also necessary to analyze the stability of the blood sugar data change under the postprandial state, that is, to determine the stability index of the second blood sugar data subsequence.

[0079] In the first step, the rising phase and the falling phase of each abnormal blood sugar data sequence are determined according to the second blood sugar data subsequence.

[0080] The first sub-step is to determine each abnormal blood sugar data sequence corresponding to the second blood sugar data sub-sequence.

[0081] In this embodiment, in order to reduce the amount of subsequent calculations to a certain extent, only the blood glucose data exceeding the normal threshold line in the second blood glucose data subsequence is analyzed, and the blood glucose data greater than the abnormal blood glucose threshold in the second blood glucose data subsequence is obtained, and the blood glucose data greater than the abnormal blood glucose threshold is recorded as abnormal blood glucose data, and the sequence composed of continuous adjacent abnormal blood glucose data is used as the abnormal blood glucose data sequence. Among them, the abnormal blood glucose threshold can be set to 6.1 according to the blood glucose range of ordinary people, that is, the normal threshold line is set to 6.1. The implementer can set the abnormal blood glucose threshold according to the specific actual situation, and no specific limitation is made here.

[0082] The second sub-step is to determine the rising phase and the falling phase of each abnormal blood sugar data sequence.

[0083] In this embodiment, for any abnormal blood glucose data sequence, the least squares method is first used to perform curve fitting on the abnormal blood glucose data sequence to obtain a fitting curve corresponding to the abnormal blood glucose data sequence, and then the maximum point with the largest value in the fitting curve is determined. Finally, the largest maximum point is used to divide the abnormal blood glucose data sequence into an ascending stage and a descending stage.

[0084] Among them, the horizontal axis of the fitting curve is the collection time, and the vertical axis is the blood glucose data; the rising stage refers to the blood glucose data stage corresponding to when the starting point of the fitting curve gradually increases to the maximum maximum point, and the falling stage refers to the blood glucose data stage corresponding to when the maximum maximum point of the fitting curve gradually decreases to the end point of the fitting curve; the determination process of the rising stage and the falling stage of each abnormal blood glucose data sequence is the same, so each abnormal blood glucose data sequence will no longer be explained and analyzed one by one.

[0085] It should be noted that the purpose of determining the rising phase and the falling phase of each abnormal blood sugar data sequence is to provide data conditions for the subsequent calculation of the stability index of the second blood sugar data subsequence.

[0086] The second step is to determine the stability index of the second blood glucose data subsequence according to each abnormal blood glucose data sequence and the time ratio of its rising phase and falling phase, and the abnormal blood glucose threshold.

[0087] The first sub-step is to determine the maximum abnormal blood glucose data in each abnormal blood glucose data sequence, calculate the cumulative sum of the differences between each maximum abnormal blood glucose data and the abnormal blood glucose threshold, and use the cumulative sum of all the differences as the first stability factor of the second blood glucose data sub-sequence.

[0088] As an example, the calculation formula for the first stability factor of the second blood glucose data subsequence on date i may be:

[0089] ; In the formula, is the first stability factor of the second blood glucose data subsequence on date i, is the number of abnormal blood sugar data sequences corresponding to the second blood sugar data subsequence on date i, is the abnormal blood sugar data sequence number corresponding to the second blood sugar data subsequence of date i, max is the maximum value function, is each abnormal blood sugar data in the u-th abnormal blood sugar data sequence on date i, is the maximum abnormal blood glucose data in the u-th abnormal blood glucose data sequence on date i, and Y is the abnormal blood glucose threshold.

[0090] In the calculation formula of the first stability factor, Can be characterized The degree to which the highest blood sugar data exceeds the normal range, the larger the first stability factor is, the more times the abnormal data segments appear in the patient's single-day data, the larger the extreme value is, that is, the higher the blood sugar data of the second blood sugar data subsequence is, the greater the fluctuation of the blood sugar data is, the worse the stability of the patient's blood sugar data change in the postprandial state on date i, and the worse the patient's single-day blood sugar regulation ability is.

[0091] The second sub-step is to determine a second stability factor of the second blood glucose data sub-sequence according to the time ratio of the rising phase and the falling phase of each abnormal blood glucose data sequence.

[0092] First of all, it should be explained that the normal human body can secrete insulin in response to blood glucose concentration due to its own blood glucose regulation mechanism. Therefore, for each abnormal blood glucose data sequence, the time ratio of the rising phase and the falling phase of the abnormal blood glucose data sequence should be relatively close. If the time ratio of the rising phase and the falling phase is larger, it means that the patient's daily blood glucose regulation ability is worse. The time ratio refers to the ratio of the time used in the rising phase to the time used in the falling phase.

[0093] As an example, the calculation formula for the second stability factor of the second blood glucose data subsequence on date i may be:

[0094] ; In the formula, is the second stability factor of the second blood glucose data subsequence on date i, is the number of abnormal blood sugar data sequences corresponding to the second blood sugar data subsequence on date i, is the abnormal blood sugar data sequence number corresponding to the second blood sugar data subsequence on date i, is the time ratio of the rising phase and the falling phase of the u-th abnormal blood sugar data sequence on date i, It is the average value of the time ratio between the rising phase and the falling phase of all abnormal blood glucose data series on date i.

[0095] In the calculation formula of the second stability factor, the second stability factor can characterize the difference in insulin action efficiency corresponding to each abnormal blood glucose data sequence in the second blood glucose data subsequence. The larger the second stability factor is, the different insulin action efficiency is for different abnormal blood glucose data sequences, which indicates that the patient's blood glucose data stability is lower.

[0096] The third sub-step is to perform inverse proportional normalization processing on the product of the first stability factor and the second stability factor to obtain a stability index of the second blood glucose data sub-sequence.

[0097] In this embodiment, since the first stability factor, the second stability factor and the stability index are in inverse proportion, it is necessary to perform inverse proportion processing on the product of the first stability factor and the second stability factor to correct the calculation logic; in addition, in order to facilitate subsequent data analysis and reduce unnecessary data calculation, the inverse proportional product is normalized to finally obtain the stability index of the second blood glucose data subsequence. The calculation means of the inverse proportional normalization processing can be exp (-), of course, the implementer can also use other means to achieve the inverse proportional normalization processing of the data.

[0098] S4, obtaining stability indicators of a first preset number of historical pending blood sugar data subsequences, and determining the wearer's blood sugar control degree within the time range corresponding to the pending blood sugar data subsequences based on the pending blood sugar data subsequences and the stability indicators of each historical pending blood sugar data subsequence.

[0099] It should be noted that the pending blood sugar data subsequence refers to the first blood sugar data subsequence or the second blood sugar data subsequence. Since the calculation process of the first blood sugar data subsequence and the second blood sugar data subsequence for the degree of blood sugar control is the same, in order to avoid repeated description of the same technical features, the first blood sugar data subsequence and the second blood sugar data subsequence are collectively referred to as pending blood sugar data subsequences to determine the wearer's degree of blood sugar control within the time range corresponding to the pending blood sugar data subsequence.

[0100] The first step is to obtain stability indicators of a first preset number of historical pending blood sugar data subsequences.

[0101] In order to analyze the degree of blood sugar control of the wearer under different conditions on a certain day and quantify the differences between the pending blood sugar data subsequences on different days, it is necessary to obtain the stability index of the pending blood sugar data subsequences of multiple consecutive days before that day.

[0102] In this embodiment, the first preset number can be set to 15. The implementer can set the number of historical pending blood glucose data subsequences according to the actual situation without specific limitation. For example, the date of the pending blood glucose data subsequence is i, and the pending blood glucose data subsequences and their corresponding stability indicators for the 15th day are extracted, and the pending blood glucose data subsequences for the 15th day are used as the historical pending blood glucose data subsequences.

[0103] The second step is to determine the wearer's blood sugar control degree within the time range corresponding to the pending blood sugar data subsequence based on the pending blood sugar data subsequence and the stability index of each historical pending blood sugar data subsequence.

[0104] It should be noted that the worse the control of blood sugar is, the more likely it is that the pending blood sugar data subsequence will have a larger change. Therefore, when abnormally identifying each blood sugar data in the pending blood sugar data subsequence, it is necessary to consider a smaller range of blood sugar data, and the correction coefficient of the K value for abnormal identification should be smaller to adjust the fixed initial K value. Therefore, the control of blood sugar can be used to adjust the K value for abnormal identification, providing data conditions for the subsequent determination of the corrected K value.

[0105] The first sub-step is to calculate the average value of the stability index of the pending blood glucose data sub-sequence and each historical pending blood glucose data sub-sequence, and then calculate the standard deviation of the stability index of the pending blood glucose data sub-sequence and each historical pending blood glucose data sub-sequence.

[0106] In the second sub-step, the ratio of the mean value to the standard deviation of the stability index is used as the blood sugar control factor, and the blood sugar control factor is normalized to obtain the wearer's blood sugar control degree within the corresponding time range of the pending blood sugar data subsequence.

[0107] As an example, when the pending blood sugar data subsequence is the first blood sugar data subsequence, the calculation formula for the wearer's blood sugar control degree within the time range corresponding to the first blood sugar data subsequence at date i may be:

[0108] ; In the formula, is the wearer's blood sugar control degree within the time range corresponding to the first blood sugar data subsequence on date i, Norm is the normalization function, is the average value of the stability index of the first blood glucose data subsequence and each historical first blood glucose data subsequence, A is the number of days corresponding to all historical first blood glucose data subsequences, e is the e-th day before date i, It is the stability index of the historical first blood glucose data subsequence on the eth day before date i.

[0109] In the calculation formula of the blood sugar control degree, when e is 0, the historical first blood sugar data subsequence of the e-th day before date i refers to the first blood sugar data subsequence of date i; It can represent the average stability of the first blood glucose data subsequence and each historical pending blood glucose data subsequence, The larger the value, the more stable the wearer's blood sugar changes are during the normal and stable phase, and the greater the degree of control over blood sugar in the normal state. The stability of the stability index change of the first blood sugar data subsequence of the wearer on different dates can be characterized. The larger the value, the weaker the patient's control over blood sugar under normal conditions. By analyzing the control degree of blood sugar through the two aspects of the data characteristics of the first blood sugar data subsequence, the numerical accuracy of the control degree of blood sugar can be effectively enhanced, and reliable data conditions can be provided for the subsequent calculation of the corrected K value. At the same time, when analyzing the control degree of blood sugar of the wearer on date i within the corresponding time range of the first blood sugar data subsequence, the change of blood sugar data of the wearer before date i under normal conditions can be taken into consideration, which can further enhance the accuracy of the numerical accuracy of the control degree of blood sugar. reference value.

[0110] Similarly, when the pending blood sugar data subsequence is the second blood sugar data subsequence, the calculation formula for the wearer's blood sugar control degree within the time range corresponding to the second blood sugar data subsequence at date i can be:

[0111] ; In the formula, is the degree of blood sugar control of the wearer on date i within the time range corresponding to the second blood sugar data subsequence, Norm is the normalization function, is the average value of the stability index of the second blood glucose data subsequence and each historical second blood glucose data subsequence, B is the number of days corresponding to all historical second blood glucose data subsequences, e is the e-th day before date i, It is the stability index of the historical second blood glucose data subsequence on the eth day before date i.

[0112] It should be noted that the days corresponding to all historical first blood sugar data subsequences and all historical second blood sugar data subsequences should be the same.

[0113] S5, determining a revised K value for each blood sugar data in the initial blood sugar data sequence according to the pre-obtained initial K value of the blood sugar data, each blood sugar data in the initial blood sugar data sequence, and the degree of blood sugar control at the time point of each blood sugar data within its corresponding time range.

[0114] The degree of blood sugar control of the wearer in the time range corresponding to the pending blood sugar data subsequence can be used to obtain the degree of blood sugar control of each blood sugar data point in the initial blood sugar data sequence within its time range. According to the degree of blood sugar control corresponding to each blood sugar data, the possibility of blood sugar fluctuation of the wearer in different states is analyzed, and the initial K value of the KNN algorithm is adjusted in combination with the actual blood sugar fluctuation.

[0115] In this embodiment, each blood glucose data in the initial blood glucose data sequence has its corresponding corrected K value, and the calculation process of the corrected K value of each blood glucose data is consistent. For the convenience of description, only the corrected K value is determined for any blood glucose data in the initial blood glucose data sequence. Any blood glucose data may be the pth blood glucose data in the initial blood glucose data sequence. The calculation process of the corrected K value of the pth blood glucose data may include:

[0116] The first step is to determine the correction coefficient of the pth blood sugar data according to the pth blood sugar data, the second preset number of blood sugar data closest to the pth blood sugar data, and the degree of blood sugar control at the time point of the pth blood sugar data within the time range to which it belongs.

[0117] In this embodiment, the pth blood glucose data and the second preset number of blood glucose data closest to the pth blood glucose data are formed into an array, and the first-order difference absolute values ​​corresponding to the array are determined; the average value of all the first-order difference absolute values ​​is normalized in inverse proportion, and the product of the normalized average value and the degree of blood glucose control at the time point of the pth blood glucose data within the time range to which it belongs is used as the correction coefficient of the pth blood glucose data.

[0118] Among them, the second preset number can be set to 5, and the size of the second preset number can be set by the implementer according to the specific actual situation and is not specifically limited.

[0119] As an example, the calculation formula for the correction coefficient of the pth blood glucose data may be:

[0120] ; In the formula, is the correction coefficient of the pth blood sugar data on date i, is the degree of blood sugar control at the time point of the p-th blood sugar data on date i within its time range, X is the number of order difference absolute values ​​corresponding to the array corresponding to the p-th blood sugar data on date i, s is the sequence number of the order difference absolute values ​​corresponding to the array corresponding to the p-th blood sugar data on date i, is the absolute value of the sth first-order difference of the array corresponding to the pth blood sugar data on date i, is the average of all first-order difference absolute values ​​of the array corresponding to the p-th blood sugar data on date i, and exp is an exponential function with a natural constant as the base.

[0121] In the calculation formula of the correction coefficient, when the pth blood glucose data is in the first blood glucose data subsequence, equal ; When the pth blood sugar data is in the second blood sugar data subsequence, equal ; The larger it is, the greater the wearer's control over blood sugar, the smaller the possible changes in the wearer's blood sugar data, the larger the range of data that needs to be considered for the pth blood sugar data, and the larger the correction coefficient of the pth blood sugar data; It can represent the actual fluctuation around the pth blood sugar data. The larger it is, the greater the fluctuation of blood sugar data around the pth blood sugar data is, the smaller the range of data that the pth blood sugar data needs to consider, and the smaller the correction coefficient of the pth blood sugar data is.

[0122] The second step is to calculate the product of the correction coefficient of the p-th blood sugar data and the initial K value as the corrected K value of the p-th blood sugar data.

[0123] In this embodiment, when the pth blood glucose data is in a first blood glucose data subsequence of steady changes, the initial K value can be set to 10; when the pth blood glucose data is in a second blood glucose data subsequence of rapid changes, the initial K value can be set to 5. The specific size of the initial K value can be set by the implementer according to the actual situation and is not specifically limited here.

[0124] S6, according to the corrected K value of each blood sugar data, perform KNN algorithm abnormality detection on the initial blood sugar data sequence to obtain distorted blood oxygen data; correct each distorted blood oxygen data in the initial blood sugar data sequence to obtain a true blood sugar data sequence.

[0125] In this embodiment, by calculating the corrected K value of each blood sugar data, the KNN algorithm abnormality detection is performed on the corresponding blood sugar data in the initial blood sugar data sequence, and the distorted blood oxygen data in the initial blood sugar data sequence can be detected; each distorted blood oxygen data is recorded and deleted, and then the blood oxygen data at the position of each distorted blood oxygen data is interpolated and supplemented by the least squares method, and finally the real blood sugar data sequence is obtained. Among them, the implementation process of the KNN algorithm is a prior art, which is not within the protection scope of the present invention and will not be elaborated here.

[0126] An embodiment of the present invention further provides a diabetic blood glucose dynamic monitoring system, comprising a processor and a memory, wherein the processor is used to process instructions stored in the memory to implement a diabetic blood glucose dynamic monitoring method.

[0127] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for dynamic monitoring of blood sugar in diabetes, It is characterized in that The following steps are involved: Acquire the blood sugar data of the wearer of the blood sugar monitoring device at each preset time in a day in real time to form an initial blood sugar data sequence; wherein the initial blood sugar data sequence is composed of a first blood sugar data subsequence and a second blood sugar data subsequence; Analyzing the data fluctuation degree of the first blood sugar data subsequence and the degree of change of the overall blood sugar level according to the first blood sugar data subsequence, and determining the stability index of the first blood sugar data subsequence; Determine the rising phase and the falling phase of each abnormal blood sugar data sequence according to the second blood sugar data subsequence; determine the stability index of the second blood sugar data subsequence according to each abnormal blood sugar data sequence and the time ratio of the rising phase and the falling phase, and the abnormal blood sugar threshold; Obtaining stability indicators of a first preset number of historical pending blood sugar data subsequences, and determining the wearer's blood sugar control degree within the time range corresponding to the pending blood sugar data subsequence according to the pending blood sugar data subsequence and the stability indicators of each historical pending blood sugar data subsequence; wherein the pending blood sugar data subsequence is the first blood sugar data subsequence or the second blood sugar data subsequence; Determine a revised K value for each blood sugar data in the initial blood sugar data sequence according to the pre-obtained initial K value of the blood sugar data, each blood sugar data in the initial blood sugar data sequence, and the degree of blood sugar control at the time point of each blood sugar data within its corresponding time range; According to the corrected K value of each blood sugar data, the initial blood sugar data sequence is subjected to KNN algorithm abnormality detection to obtain distorted blood oxygen data; each distorted blood oxygen data in the initial blood sugar data sequence is corrected to obtain a true blood sugar data sequence.

2. A method for dynamic monitoring of blood sugar in diabetes according to claim 1, It is characterized in that The steps of acquiring the first blood sugar data subsequence and the second blood sugar data subsequence include: Performing second-order difference processing on the initial blood sugar data sequence to obtain a second-order difference sequence of the initial blood sugar data sequence; Determine the maximum value and the minimum value in the second-order difference sequence, and form each blood glucose data in the time period corresponding to the maximum value and the minimum value into a blood glucose data subsequence as a second blood glucose data subsequence; The blood sugar data in the initial blood sugar data sequence except the second blood sugar data subsequence are combined into a blood sugar data subsequence as a first blood sugar data subsequence.

3. A method for dynamic monitoring of blood sugar in diabetes according to claim 1, It is characterized in that The step of analyzing the data fluctuation degree of the first blood sugar data subsequence and the overall blood sugar level change degree according to the first blood sugar data subsequence to determine the stability index of the first blood sugar data subsequence includes: Performing straight line fitting on the first blood sugar data subsequence to obtain a fitted straight line; determining a residual sequence between the first blood sugar data subsequence and the fitted straight line, and taking the average value of each residual in the residual sequence as the degree of data fluctuation; Determine the blood glucose data at the starting point and the ending point in the first blood glucose data subsequence, calculate the difference between the blood glucose data at the starting point and the ending point, and then calculate the ratio of the difference to the length of the corresponding time period of the first blood glucose data subsequence, and use the ratio as the degree of change of the overall blood glucose level; The product of the data fluctuation degree and the overall blood sugar level change degree is normalized in inverse proportion to obtain a stability index of the first blood sugar data subsequence.

4. A method for dynamic monitoring of blood sugar in diabetes according to claim 1, It is characterized in that The step of determining the rising phase and the falling phase of each abnormal blood sugar data sequence according to the second blood sugar data subsequence comprises: Acquire blood glucose data greater than the abnormal blood glucose threshold in the second blood glucose data subsequence, record the blood glucose data greater than the abnormal blood glucose threshold as abnormal blood glucose data, and take a sequence consisting of continuous adjacent abnormal blood glucose data as an abnormal blood glucose data sequence; For any abnormal blood sugar data sequence, the largest maximum value point corresponding to the abnormal blood sugar data sequence is determined, and the abnormal blood sugar data sequence is divided into an ascending phase and a descending phase using the largest maximum value point.

5. A method for dynamic monitoring of blood sugar in diabetes according to claim 1, It is characterized in that The step of determining the stability index of the second blood glucose data subsequence according to each abnormal blood glucose data sequence and the time ratio of the rising phase and the falling phase, and the abnormal blood glucose threshold, includes: Determine the maximum abnormal blood sugar data in each abnormal blood sugar data sequence, calculate the cumulative sum of the differences between each maximum abnormal blood sugar data and the abnormal blood sugar threshold, and use the cumulative sum of all the differences as the first stability factor of the second blood sugar data subsequence; Determine a second stability factor of a second blood glucose data subsequence according to a time ratio of a rising phase to a falling phase of each abnormal blood glucose data sequence; The product of the first stability factor and the second stability factor is normalized in inverse proportion to obtain a stability index of the second blood glucose data subsequence.

6. A method for dynamic monitoring of blood sugar in diabetes according to claim 5, It is characterized in that The step of determining the second stability factor of the second blood glucose data subsequence according to the time ratio of the rising phase to the falling phase of each abnormal blood glucose data sequence comprises: The calculation formula of the second stability factor of the second blood glucose data subsequence is: ; In the formula, is the second stability factor of the second blood glucose data subsequence, U is the number of abnormal blood glucose data sequences corresponding to the second blood glucose data subsequence, u is the sequence number of the abnormal blood glucose data sequence corresponding to the second blood glucose data subsequence, is the time ratio of the rising phase and the falling phase of the u-th abnormal blood sugar data sequence, It is the average value of the time ratio between the rising phase and the falling phase of all abnormal blood glucose data series.

7. A method for dynamic monitoring of blood sugar in diabetes according to claim 1, It is characterized in that The step of determining the wearer's blood sugar control degree within the time range corresponding to the pending blood sugar data subsequence according to the pending blood sugar data subsequence and the stability index of each historical pending blood sugar data subsequence includes: Calculate the average value of the stability index of the pending blood sugar data subsequence and each historical pending blood sugar data subsequence, and then calculate the standard deviation of the stability index of the pending blood sugar data subsequence and each historical pending blood sugar data subsequence; The ratio of the average value of the stability index to the standard deviation is used as the blood sugar control factor, and the blood sugar control factor is normalized to obtain the wearer's blood sugar control degree within the corresponding time range of the pending blood sugar data subsequence.

8. A method for dynamic monitoring of blood sugar in diabetes according to claim 1, It is characterized in that The method of determining the revised K value of each blood sugar data in the initial blood sugar data sequence according to the pre-obtained initial K value of the blood sugar data, each blood sugar data in the initial blood sugar data sequence, and the degree of blood sugar control at the time point of each blood sugar data within its corresponding time range comprises: For any blood sugar data in the initial blood sugar data sequence, a correction coefficient of the blood sugar data is determined according to the blood sugar data, a second preset number of blood sugar data closest to the blood sugar data, and the degree of blood sugar control at the time point of the blood sugar data within the time range to which it belongs; The product of the correction coefficient of the blood sugar data and the initial K value is calculated as the corrected K value of the blood sugar data.

9. A method for dynamic monitoring of blood sugar in diabetes according to claim 8, It is characterized in that Determining the correction coefficient of the blood sugar data according to the blood sugar data, a second preset number of blood sugar data closest to the blood sugar data, and the degree of blood sugar control at the time point of the blood sugar data within the time range to which it belongs, comprises: The blood glucose data and a second preset number of blood glucose data closest to the blood glucose data are formed into an array, and the first-order difference absolute values ​​corresponding to the array are determined; the average values ​​of all the first-order difference absolute values ​​are normalized in inverse proportion, and the product of the normalized average value and the degree of blood glucose control at the time point of the blood glucose data within the time range to which it belongs is used as the correction coefficient of the blood glucose data.

10. A diabetes blood sugar dynamic monitoring system, It is characterized in that It comprises a processor and a memory, wherein the processor is used to process instructions stored in the memory to implement a diabetic blood glucose dynamic monitoring method as described in any one of claims 1-9.