Blood glucose trend prediction method for continuous glucose monitoring and related devices
By filtering and differentially calculating the blood glucose monitor signal, the system predicts future blood glucose trends and generates preventive intervention information, solving the problem of insufficient blood glucose trend prediction in traditional technologies. It achieves keen identification and accurate prediction of blood glucose fluctuations and provides intuitive visualization analysis.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BIOLAND TECH (SHENZHEN) CO LTD
- Filing Date
- 2026-03-13
- Publication Date
- 2026-06-12
Smart Images
Figure CN122201573A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of clinical technology, and in particular to a method and related equipment for predicting blood glucose trends through continuous blood glucose monitoring. Background Technology
[0002] In the field of diabetes management and blood glucose monitoring, real-time monitoring of blood glucose dynamics and early prediction of trends are crucial for avoiding acute complications such as hypoglycemia and hyperglycemia. Currently, in clinical and home settings, traditional finger-prick blood glucose meters, while providing immediate blood glucose values, suffer from low sampling frequency and an inability to reflect continuous blood glucose fluctuations. This is especially problematic in cases of sudden spikes or drops in blood glucose at night or delayed postprandial blood glucose peaks, where a single test cannot capture the complete change, preventing users and healthcare professionals from promptly detecting potential blood glucose risks. While existing continuous glucose monitoring devices can achieve high-frequency data collection, they largely remain at the level of presenting historical and current blood glucose values, lacking the ability to effectively predict blood glucose trends in the short term, thus failing to meet users' needs for proactive preventative measures such as dietary adjustments, exercise interventions, or medication adjustments. Summary of the Invention
[0003] The main technical problem addressed in this application is to provide a method and related equipment for predicting blood glucose trends in continuous blood glucose monitoring, thereby solving the technical problem that traditional technologies lack the ability to effectively predict blood glucose trends in the short term.
[0004] To address the aforementioned technical problems, this application employs a method for predicting blood glucose trends based on continuous blood glucose monitoring, comprising the following steps: The raw blood glucose signals continuously collected by the blood glucose monitor are filtered and denoised to obtain a clean blood glucose sequence. First-order and second-order difference calculations are performed on the clean blood glucose sequences at adjacent time points to obtain the rate of change and acceleration of blood glucose change. Based on the blood glucose change rate and blood glucose change acceleration, predict the user's blood glucose trend within a future preset time window, and plot a blood glucose curve based on the blood glucose trend; Based on the blood glucose curve, the threshold boundary is cross-determined to obtain the threshold trigger time point and the corresponding blood glucose risk level identifier, and prevention and intervention information is generated based on the blood glucose risk level identifier and the blood glucose curve.
[0005] Furthermore, the filtering and denoising process performed on the raw blood glucose signals continuously acquired by the blood glucose monitor to obtain a clean blood glucose sequence includes: The raw blood glucose signal continuously collected by the blood glucose monitor is sampled at a preset sampling time interval to obtain a blood glucose sampling sequence, and the blood glucose sampling sequence is processed by moving average to obtain a smooth blood glucose sequence. The smoothed blood glucose sequence is subjected to outlier detection and removal to obtain an outlier-removed sequence, and baseline drift correction is performed based on the outlier-removed sequence to obtain the clean blood glucose sequence.
[0006] Furthermore, the step of performing first-order and second-order difference calculations on the clean blood glucose sequences at adjacent time points to obtain the rate of blood glucose change and the acceleration of blood glucose change includes: The difference between the blood glucose values at two adjacent time points in the clean blood glucose sequence is calculated to obtain a blood glucose difference sequence. The blood glucose difference sequence is then divided by the corresponding time interval to obtain the rate of blood glucose change. Dividing the rate of change of blood glucose by the corresponding time interval yields the acceleration of the change in blood glucose.
[0007] Furthermore, the prediction of the user's blood glucose trend within a future preset time window based on the blood glucose change rate and blood glucose change acceleration includes: The blood glucose change rate and blood glucose change acceleration are numerically combined and arranged to obtain a blood glucose dynamic parameter matrix. The blood glucose dynamic parameter matrix is then divided into time steps according to a future preset time window to obtain a predicted time node sequence. Based on the numerical parameters in the blood glucose dynamic parameter matrix, the blood glucose values of the predicted time node sequence are calculated point by point to obtain the future blood glucose value sequence. The blood glucose values of each time node in the future blood glucose value sequence are then checked for continuity to obtain the checked blood glucose sequence. The verification blood glucose sequence is marked with trend direction according to time order to obtain blood glucose trend identifier sequence. The blood glucose trend is obtained by combining and mapping the blood glucose trend identifier sequence with the verification blood glucose sequence.
[0008] Furthermore, the step of plotting a blood glucose curve based on the blood glucose trend includes: The blood glucose values and time nodes in the blood glucose trend are mapped and transformed to obtain a set of blood glucose coordinate points. The blood glucose coordinate points are then connected and plotted according to the time series to obtain a preliminary blood glucose trend curve segment. Based on the extreme points of the rate of change of blood glucose in the preliminary blood glucose trend curve segment, feature marking processing is performed to obtain blood glucose fluctuation feature points. The blood glucose fluctuation feature points are then graphically superimposed and combined with the preliminary blood glucose trend curve segment to obtain the blood glucose curve.
[0009] Furthermore, the step of cross-judging the threshold boundary based on the blood glucose curve to obtain the threshold trigger time point and the corresponding blood glucose risk level identifier includes: The blood glucose coordinate point values in the blood glucose curve are compared with the preset threshold boundary to obtain the threshold cross-judgment result. The lower cross-judgment point and the upper cross-judgment point in the threshold cross-judgment result are extracted to obtain the threshold trigger time point. Based on the blood glucose value corresponding to the threshold trigger time point, the deviation of blood glucose from the threshold boundary is calculated, and the blood glucose deviation is combined with the blood glucose change rate to divide the interval into grades, thereby obtaining the blood glucose risk level identifier.
[0010] Furthermore, the generation of preventive intervention information based on the blood glucose risk level identifier and the blood glucose curve includes: Based on the blood glucose risk level identifier, a time interval mapping is performed to obtain the risk warning period, and based on the risk warning period, the blood glucose fluctuation characteristics of the blood glucose curve are extracted to obtain the blood glucose fluctuation parameters. Based on the blood glucose fluctuation parameters, a preset intervention measure library is matched and screened to obtain a candidate intervention measure set. The intervention measures in the candidate intervention measure set are then sorted according to the blood glucose risk level identifier to obtain the prevention intervention information.
[0011] The present invention also provides a blood glucose trend prediction system for continuous blood glucose monitoring, comprising: The calculation module is used to filter and denoise the raw blood glucose signals continuously collected by the blood glucose monitor to obtain a clean blood glucose sequence, and to perform first-order and second-order difference calculations on the clean blood glucose sequences at adjacent time points to obtain the rate of change and acceleration of blood glucose change. The prediction module is used to predict the user's blood glucose trend within a future preset time window based on the blood glucose change rate and blood glucose change acceleration, and to draw a blood glucose curve based on the blood glucose trend. The determination module is used to perform cross-determination of the threshold boundary based on the blood glucose curve, obtain the threshold trigger time point and the corresponding blood glucose risk level identifier, and generate prevention and intervention information based on the blood glucose risk level identifier and the blood glucose curve.
[0012] The above scheme filters and denoises the raw blood glucose signals continuously collected by the blood glucose monitor to obtain a clean blood glucose sequence. First-order and second-order difference calculations are then performed on the clean blood glucose sequences at adjacent time points to obtain the rate and acceleration of blood glucose change. Based on these rates and accelerations, the scheme predicts the user's blood glucose trend within a preset future time window and plots a blood glucose curve. The scheme then cross-determines threshold boundaries based on the blood glucose curve to obtain threshold trigger times and corresponding blood glucose risk level indicators. Based on these risk level indicators and the blood glucose curve, preventative intervention information is generated. This solves the technical problem of traditional technologies lacking the ability to effectively predict blood glucose trends in the short term. It enables more sensitive identification of potential turning points in blood glucose fluctuations—for example, when the rate of blood glucose change is flat but the acceleration has shown positive or negative changes, it can detect an impending rise or fall in blood glucose, thus more accurately predicting the blood glucose trend within a preset future time window. Simultaneously, the blood glucose curve plotted based on the prediction results transforms abstract numerical changes into intuitive visual trends, facilitating quick understanding of the technical effects of future blood glucose dynamics by users or medical personnel. Attached Figure Description
[0013] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0014] Figure 1 This is a schematic diagram of the steps of a continuous blood glucose monitoring method for predicting blood glucose trends in one embodiment of the present invention; Figure 2 This is a structural block diagram of a blood glucose trend prediction system for continuous blood glucose monitoring according to an embodiment of the present invention; Figure 3 This is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.
[0015] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation
[0016] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.
[0017] Specifically, the blood glucose trend prediction method for continuous blood glucose monitoring in this embodiment includes the following steps: like Figure 1 As shown, Figure 1 This invention provides a method for predicting blood glucose trends through continuous blood glucose monitoring, comprising the following steps: Step S1: Filter and denoise the raw blood glucose signal continuously collected by the blood glucose monitor to obtain a clean blood glucose sequence. Then, perform first-order and second-order difference calculations on the clean blood glucose sequences at adjacent time points to obtain the rate of change and acceleration of blood glucose change.
[0018] Specifically, when processing the raw blood glucose signal collected by the blood glucose monitor, filtering and noise reduction operations are first performed to address any interference that may exist in the signal. For example, when the user's arm moves slightly during the monitoring process, causing a brief fluctuation in the signal, or when changes in ambient temperature affect the sensor output, appropriate filtering algorithms (such as Kalman filtering or moving average filtering) are used to suppress these irrelevant interferences, gradually removing the noise components from the raw signal, and finally obtaining a clean blood glucose sequence that can truly reflect the blood glucose level. After acquiring the clean blood glucose sequence, data from adjacent time points are selected for calculation. The first-order difference calculation involves subtracting the clean blood glucose value from the previous time point from the value at the next time point, then dividing by the interval between the two time points. For example, if the interval between two adjacent monitoring time points is 5 minutes, and the blood glucose value at the previous time point is 5.2 mmol / L and at the next time point is 5.5 mmol / L, the first-order difference can be used to calculate the rate of change of blood glucose within those 5 minutes. The second-order difference calculation, based on the obtained rates of change of blood glucose between adjacent time points, subtracts the rate of change of the previous time point from the rate of change of the next time period, and then divides by the time interval to obtain the acceleration of blood glucose change. For example, the rate of change for the first 5 minutes is 0.06 mmol / L. The rate was 0.03 mmol / (L) for the first 5 minutes and 0.03 mmol / (L) for the last 5 minutes. The acceleration of blood glucose change during this period can be obtained by using the second-order difference (min).
[0019] Step S2: Based on the blood glucose change rate and blood glucose change acceleration, predict the user's blood glucose trend within a future preset time window, and draw a blood glucose curve based on the blood glucose trend.
[0020] Specifically, when predicting a user's blood glucose trend within a preset time window based on the blood glucose change rate and acceleration, the previously calculated data from both methods will be integrated and analyzed first. For example, if the current blood glucose change rate is 0.04 mmol / (L) (min), the acceleration change is -0.01 mmol / (L) The system calculates the blood glucose level based on the changes in these two parameters (min²) and a preset time window of 30 minutes. It then predicts that blood glucose levels will initially rise slowly at the current rate over the next 30 minutes, then gradually slow down due to negative acceleration, and may even slightly decline at the end of the window. This process is used to deduce the predicted blood glucose value for each time point within the window, forming a complete blood glucose trend prediction. After trend prediction, the predicted blood glucose values for each point within the preset time window are sequentially marked on a coordinate system with time as the horizontal axis and predicted blood glucose value as the vertical axis. These marked points are then connected in an orderly manner using a smooth curve. Finally, a blood glucose curve is plotted based on the predicted blood glucose trend, visually representing future blood glucose changes. For example, if the predicted blood glucose level rises from 5.3 mmol / L to 5.8 mmol / L and then falls back to 5.6 mmol / L within 30 minutes, the curve will show a trend of first rising and then gradually declining.
[0021] Step S3: Based on the blood glucose curve, cross-determine the threshold boundary to obtain the threshold trigger time point and the corresponding blood glucose risk level identifier, and generate prevention and intervention information based on the blood glucose risk level identifier and the blood glucose curve.
[0022] Specifically, when performing cross-judgment of threshold boundaries based on the blood glucose curve, the upper and lower thresholds of the normal blood glucose range are preset first (for example, 3.9 mmol / L is usually set as the hypoglycemia threshold and 7.8 mmol / L as the hyperglycemia threshold). Then, the trend of the plotted blood glucose curve within the preset time window is observed. When the predicted value of the curve at a certain moment touches or crosses the preset threshold boundary, that moment is recorded as the threshold trigger time point. At the same time, the corresponding blood glucose risk level is marked according to the type of threshold crossed. For example, if the blood glucose curve drops from the normal range to 3.8 mmol / L in the next 15 minutes and crosses the hypoglycemia threshold, the 15-minute moment is set as the threshold trigger time point and marked with the label "hypoglycemia risk (mild)". After determining the threshold trigger time and blood glucose risk level, preventive intervention information will be generated by combining these two pieces of information with the overall trend of the blood glucose curve. For example, in the case of a mild hypoglycemia risk that will occur after 15 minutes, the specific intervention content will be generated by referring to whether the curve continues to decline afterward, such as "It is recommended to supplement 15g of carbohydrates (such as half a cup of juice) within the next 10 minutes and pay attention to subsequent blood glucose changes".
[0023] In a specific embodiment, the step of filtering and denoising the raw blood glucose signals continuously acquired by the blood glucose monitor to obtain a clean blood glucose sequence includes: The raw blood glucose signal continuously collected by the blood glucose monitor is sampled at a preset sampling time interval to obtain a blood glucose sampling sequence, and the blood glucose sampling sequence is processed by moving average to obtain a smooth blood glucose sequence. The smoothed blood glucose sequence is subjected to outlier detection and removal to obtain an outlier-removed sequence, and baseline drift correction is performed based on the outlier-removed sequence to obtain the clean blood glucose sequence.
[0024] Specifically, when processing the raw blood glucose signal collected by the blood glucose monitor, the raw signal is first sampled at a preset sampling time interval (e.g., 1 minute / time or 5 minutes / time, adjusted according to monitoring needs). This converts the continuous signal into a discrete blood glucose sampling sequence. For example, blood glucose values are recorded every 5 minutes, resulting in a sequence of data such as 6.1 mmol / L, 6.3 mmol / L, and 6.2 mmol / L. Next, a moving average is applied to this blood glucose sampling sequence. For example, the average of the values of three adjacent sampling points is used as the smoothing value at the midpoint. If a certain sampling sequence is 5.9, 6.5, and 6.1, the smoothing value at the midpoint is calculated to be (5.9 + 6.5 + 6.1) / 3 = 6.17. The smoothed blood glucose sequence is obtained by calculating point by point in this way. Then, outlier detection and removal are performed on the smoothed blood glucose sequence. For example, when an abnormal data such as 12.0 mmol / L (far exceeding the adjacent normal value) suddenly appears in the sequence, it will be identified as an outlier and removed by statistical methods, resulting in an outlier removal sequence. Finally, baseline drift correction is performed based on the sequence. For example, if the sequence as a whole has a slow upward drift trend, the algorithm will eliminate the influence of this trend, and finally obtain a clean blood glucose sequence.
[0025] In a specific embodiment, the step of performing first-order and second-order difference calculations on clean blood glucose sequences at adjacent time points to obtain the rate of blood glucose change and the acceleration of blood glucose change includes: The difference between the blood glucose values at two adjacent time points in the clean blood glucose sequence is calculated to obtain a blood glucose difference sequence. The blood glucose difference sequence is then divided by the corresponding time interval to obtain the rate of blood glucose change. Dividing the rate of change of blood glucose by the corresponding time interval yields the acceleration of the change in blood glucose.
[0026] Specifically, when processing a clean blood glucose sequence, the first step is to calculate the difference between blood glucose values at two adjacent time points. For example, if the blood glucose value at one time point is 5.4 mmol / L and the value at the next time point is 5.7 mmol / L, the difference is 0.3 mmol / L. This process of arranging all such differences between adjacent time points sequentially forms a blood glucose difference sequence. Then, each difference in this sequence is divided by the time interval between the two adjacent time points. If the interval between the two time points is 5 minutes, then 0.3 mmol / L divided by 5 minutes yields 0.06 mmol / L. By calculating the differences one by one (min), the rate of blood glucose change can be obtained. Then, for the obtained rate of blood glucose change, the difference between the rate values of two adjacent time points is calculated. For example, if the rate in the previous time period is 0.06 mmol / (L) The concentration of 0.03 mmol / (L) was measured in the first min and the second min was measured in the second min. The difference was -0.03 mmol / (L) min. This generates a rate difference sequence; then, each difference in the sequence is divided by the corresponding time interval (still taking 5 minutes as an example), resulting in -0.03 mmol / (L). Dividing the result by 5 minutes (min) yields -0.006 mmol / (L) The acceleration of blood glucose change was obtained by calculating min².
[0027] In a specific embodiment, predicting the user's blood glucose trend within a future preset time window based on the blood glucose change rate and blood glucose change acceleration includes: The blood glucose change rate and blood glucose change acceleration are numerically combined and arranged to obtain a blood glucose dynamic parameter matrix. The blood glucose dynamic parameter matrix is then divided into time steps according to a future preset time window to obtain a predicted time node sequence. Based on the numerical parameters in the blood glucose dynamic parameter matrix, the blood glucose values of the predicted time node sequence are calculated point by point to obtain the future blood glucose value sequence. The blood glucose values of each time node in the future blood glucose value sequence are then checked for continuity to obtain the checked blood glucose sequence. The verification blood glucose sequence is marked with trend direction according to time order to obtain blood glucose trend identifier sequence. The blood glucose trend is obtained by combining and mapping the blood glucose trend identifier sequence with the verification blood glucose sequence.
[0028] Specifically, when predicting blood glucose trends based on the aforementioned rate of change and acceleration of blood glucose change, the previously obtained data for both are first arranged in chronological order. For example, the rate of change every 5 minutes within a certain hour (e.g., 0.05, 0.04, 0.03, 0.02, 0.01, 0, -0.01, -0.02, -0.03, -0.04, -0.05, -0.06 mmol / (L)) min) and changes in acceleration (e.g., -0.002, -0.002, -0.002, -0.002, -0.002, -0.002, -0.002, -0.002, -0.002, -0.002, -0.002, -0.002, -0.002, -0.002, -0.002, -0.002 mmol / (L) The min²)) are combined one-to-one to form a dynamic blood glucose parameter matrix containing three columns of data: time, rate, and acceleration. Then, the time step is divided according to the future preset time window (e.g., set to 30 minutes). If the step is set to 5 minutes, the 30 minutes are divided into 6 time nodes, resulting in the predicted time node sequence from the current time to 5 minutes, 10 minutes, 15 minutes, 20 minutes, 25 minutes, and 30 minutes later. Then, based on the numerical parameters in the blood glucose dynamic parameter matrix, the blood glucose values of the predicted time node sequence are calculated point by point. Taking the current blood glucose value of 5.5 mmol / L as an example, the blood glucose value at the 5-minute node is 5.5 + 0.05 × 5 = 5.75 mmol / L, and the blood glucose value at the 10-minute node is 5.75 + 0.04 × 5 = 5.95 mmol / L. The values of all nodes are calculated in this way to form the future blood glucose value sequence. Then, the continuity of the sequence is checked, such as checking whether the difference between blood glucose values of adjacent nodes is within a reasonable range (e.g., not exceeding 0.5 mmol / L). If the calculated value of a node suddenly jumps by 1.2 mmol / L, the parameters and calculation process are rechecked and corrected to obtain the verified blood glucose sequence. Finally, the calibration blood glucose sequence is marked with trend direction according to time sequence. If the blood glucose value of adjacent nodes continues to rise, it is marked as "rise", if it continues to fall, it is marked as "fall", and if it is basically stable, it is marked as "flat", forming a blood glucose trend label sequence such as "rise, rise, flat, fall, fall, fall". This sequence is then combined and mapped with the specific values in the calibration blood glucose sequence (such as 5.75, 5.95, 5.96, 5.86, 5.76, 5.66 mmol / L). For example, the correspondence such as "5 minutes (rise, 5.75 mmol / L), 10 minutes (rise, 5.95 mmol / L)" is integrated to finally obtain the complete blood glucose trend.
[0029] In a specific embodiment, the step of plotting a blood glucose curve based on the blood glucose trend includes: The blood glucose values and time nodes in the blood glucose trend are mapped and transformed to obtain a set of blood glucose coordinate points. The blood glucose coordinate points are then connected and plotted according to the time series to obtain a preliminary blood glucose trend curve segment. Based on the extreme points of the rate of change of blood glucose in the preliminary blood glucose trend curve segment, feature marking processing is performed to obtain blood glucose fluctuation feature points. The blood glucose fluctuation feature points are then graphically superimposed and combined with the preliminary blood glucose trend curve segment to obtain the blood glucose curve.
[0030] Specifically, when plotting a blood glucose curve based on the blood glucose trend, the first step is to perform a coordinate axis mapping transformation on the blood glucose values and corresponding time points included in the blood glucose trend. Typically, a Cartesian coordinate system is established with the horizontal axis representing time and the vertical axis representing blood glucose value. For example, 5 minutes, 10 minutes, and 15 minutes in the predicted time point sequence are mapped to the coordinates 5, 10, and 15 on the horizontal axis, respectively. Similarly, 5.75 mmol / L, 5.95 mmol / L, and 5.96 mmol / L in the verification blood glucose sequence are mapped to the coordinates 5.75, 5.95, and 5.96 on the vertical axis, respectively. Coordinate points are created by combining each time point with its corresponding blood glucose value into coordinate points such as (5, 5.75), (10, 5.95), and (15, 5.96). All these coordinate points are then aggregated to form a set of blood glucose coordinate points. Next, these coordinate points are connected sequentially in chronological order using smooth lines, for example, from (5, 5.75) to (10, 5.95), then to (15, 5.96), and so on, until all coordinate points are connected. This process is called line plotting, ultimately forming a preliminary blood glucose trend curve segment. Afterwards, the extreme points of the blood glucose rate of change in the preliminary blood glucose trend curve segment will be analyzed. For example, in the curve segment, the rate of blood glucose rise from 5 minutes to 10 minutes is 0.04 mmol / (L). The rate of increase slowed to 0.002 mmol / (L) after 10 to 15 minutes. The rate of decline changed from -0.02 mmol / (L) after 15 to 20 minutes. If we consider the time interval (min), then the point (15, 5.96) corresponding to 15 minutes is the turning point between the rise and fall, which is the extreme point where the rate of blood glucose change changes from positive to negative. The process of marking this point with a special symbol (such as a red dot) is called feature marking. After finding all similar points in this way, we obtain the blood glucose fluctuation feature points. Finally, these marked blood glucose fluctuation feature points are superimposed on the corresponding positions of the preliminary blood glucose trend curve segment. For example, the red dot is precisely placed at (15, 5.96) of the curve segment, so that the feature points and the curve segment are completely integrated. After such graphic superposition and combination, we finally obtain a blood glucose curve that can clearly show the blood glucose change trend and key fluctuation nodes.
[0031] In a specific embodiment, the step of cross-judging the threshold boundary based on the blood glucose curve to obtain the threshold trigger time point and the corresponding blood glucose risk level identifier includes: The blood glucose coordinate point values in the blood glucose curve are compared with the preset threshold boundary to obtain the threshold cross-judgment result. The lower cross-judgment point and the upper cross-judgment point in the threshold cross-judgment result are extracted to obtain the threshold trigger time point. Based on the blood glucose value corresponding to the threshold trigger time point, the deviation of blood glucose from the threshold boundary is calculated, and the blood glucose deviation is combined with the blood glucose change rate to divide the interval into grades, thereby obtaining the blood glucose risk level identifier.
[0032] Specifically, when performing cross-judgment of threshold boundaries based on the blood glucose curve, a preset threshold boundary is first determined. Typically, a low blood glucose threshold (e.g., 3.9 mmol / L) and a high blood glucose threshold (e.g., 7.8 mmol / L) are set as the judgment criteria. Then, the blood glucose coordinate points in the blood glucose curve are extracted one by one and compared with these two threshold boundaries. For example, the blood glucose value of 3.8 mmol / L corresponding to a certain coordinate point (20, 3.8) is lower than the low blood glucose threshold of 3.9 mmol / L, and the blood glucose value of 8.0 mmol / L corresponding to (25, 8.0) is higher than the high blood glucose threshold of 7.8 mmol / L. These results that meet the cross-judgment conditions are classified as threshold cross-judgment results. Then, the lower crossing point (i.e., the point where the blood glucose value drops from above the threshold to below the threshold, such as (20, 3.8)) and the upper crossing point (i.e., the point where the blood glucose value rises from below the threshold to above the threshold, such as (25, 8.0)) are selected from these results, and the time nodes of 20 minutes and 25 minutes corresponding to these two types of points are extracted to obtain the threshold trigger time point.
[0033] Next, the numerical difference will be calculated based on the deviation of the blood glucose value at the threshold trigger time point from the threshold boundary. For example, the difference between 3.8 mmol / L at (20, 3.8) and the low blood glucose threshold of 3.9 mmol / L is -0.1 mmol / L, and the difference between 8.0 mmol / L at (25, 8.0) and the high blood glucose threshold of 7.8 mmol / L is 0.2 mmol / L. These differences represent the blood glucose deviation amplitude. Then, the blood glucose change rate at the corresponding time point will be combined to divide the data into graded intervals. Assuming that the absolute value of the deviation amplitude is ≤0.3 mmol / L and the absolute value of the rate is ≤0.02 mmol / (L)... (min) indicates a mild risk; the absolute value of the deviation is >0.3 mmol / L or the absolute value of the rate is >0.02 mmol / (L). "min) is considered a moderate risk", if the rate corresponding to (20, 3.8) is -0.01 mmol / (L) The rate corresponding to (25, 8.0) was 0.03 mmol / (L) min). If (20, 3.8) corresponds to a risk level of "mild hypoglycemia risk", and (25, 8.0) corresponds to a risk level of "moderate hyperglycemia risk", these classification results are the blood glucose risk level indicators.
[0034] In a specific embodiment, the step of generating preventive intervention information based on the blood glucose risk level identifier and the blood glucose curve includes: Based on the blood glucose risk level identifier, a time interval mapping is performed to obtain the risk warning period, and based on the risk warning period, the blood glucose fluctuation characteristics of the blood glucose curve are extracted to obtain the blood glucose fluctuation parameters. Based on the blood glucose fluctuation parameters, a preset intervention measure library is matched and screened to obtain a candidate intervention measure set. The intervention measures in the candidate intervention measure set are then sorted according to the blood glucose risk level identifier to obtain the prevention intervention information.
[0035] Specifically, when generating preventive intervention information based on the blood glucose risk level identifier and the blood glucose curve, the process first maps time intervals according to the type of blood glucose risk level identifier and the corresponding threshold trigger time point. For example, if a threshold trigger time point is 20 minutes later, the corresponding blood glucose risk level identifier is "mild hypoglycemia risk". Usually, 5 minutes before and after this time point are defined as the risk warning period, i.e., 15-25 minutes. If the risk level is "moderate hyperglycemia risk" and the trigger time point is 25 minutes later, the warning period may be set to 22-30 minutes. The risk warning period is obtained through this correspondence. Then, based on this risk warning period, the blood glucose fluctuation within this period is extracted from the blood glucose curve. For example, if the blood glucose drops from 4.2 mmol / L to 3.8 mmol / L within 15-25 minutes and then slowly rises back to 3.9 mmol / L, the corresponding minimum blood glucose value, the magnitude of the drop, and the rate of recovery are recorded. These data are then summarized to obtain the blood glucose fluctuation parameters.
[0036] Subsequently, based on the obtained blood glucose fluctuation parameters, a matching and screening process is performed in a pre-set intervention library. This library typically stores intervention plans corresponding to different risk types and fluctuation characteristics. For example, for the case of "mild hypoglycemia risk with slow blood glucose decline," the library may contain plans such as "supplementing 10g of carbohydrates (e.g., one slice of whole-wheat bread)" or "reducing exercise." Based on the extracted blood glucose fluctuation parameters (e.g., mild hypoglycemia, decrease of 0.4 mmol / L, slow rate of decline), these plans that meet the criteria are selected to form a candidate intervention set. Finally, the intervention measures in the candidate intervention set are sorted according to the urgency and time sequence corresponding to the blood glucose risk level. For example, for mild hypoglycemia risk of 15-25 minutes, the first step is to "supplement 10g of carbohydrates at 15 minutes," then to "monitor blood glucose at 20 minutes," and finally to "adjust diet if blood glucose does not rise at 25 minutes." Through this sorting, complete prevention and intervention information is finally obtained.
[0037] Please see Figure 2 , Figure 2 This is a schematic diagram of the framework of an embodiment of the continuous blood glucose monitoring blood glucose trend prediction system of this application. Figure 2 As shown, the continuous blood glucose monitoring blood glucose trend prediction system includes a calculation module 1, which is used to filter and denoise the raw blood glucose signals continuously collected by the blood glucose monitor to obtain a clean blood glucose sequence, and to perform first-order and second-order difference calculations on the clean blood glucose sequences at adjacent time points to obtain the blood glucose change rate and blood glucose change acceleration; a prediction module 2, which is used to predict the user's blood glucose trend within a future preset time window based on the blood glucose change rate and blood glucose change acceleration, and to draw a blood glucose curve based on the blood glucose trend; and a judgment module 3, which is used to perform cross-judgment on the threshold boundary based on the blood glucose curve to obtain the threshold trigger time point and the corresponding blood glucose risk level identifier, and to generate prevention and intervention information based on the blood glucose risk level identifier and the blood glucose curve.
[0038] Reference Figure 3 This invention also provides a computer device whose internal structure can be as follows: Figure 3As shown, the computer device includes a processor, memory, display screen, input device, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores the data corresponding to this embodiment. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements the above-described method.
[0039] Those skilled in the art will understand that Figure 3 The structures shown are merely block diagrams of some structures related to the present invention and do not constitute a limitation on the computer devices on which the present invention is applied.
[0040] An embodiment of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method. It is understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.
[0041] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the present invention and embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.
[0042] In some embodiments, the functions or modules of the apparatus provided in this disclosure can be used to perform the methods described in the above method embodiments. The specific implementation can be referred to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.
[0043] The description of the various embodiments above tends to emphasize the differences between the various embodiments. The similarities or similarities between them can be referred to, and for the sake of brevity, they will not be repeated here.
[0044] In the several embodiments provided in this application, it should be understood that the disclosed methods and apparatus can be implemented in other ways. For example, the apparatus implementations described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection of devices or units may be electrical, mechanical, or other forms.
[0045] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.
[0046] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0047] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods of various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0048] If the technical solution of this application involves personal information, the product using this technical solution has clearly informed the user of the personal information processing rules and obtained the user's voluntary consent before processing the personal information. If the technical solution of this application involves sensitive personal information, the product using this technical solution has obtained the user's separate consent before processing the sensitive personal information, and also meets the requirement of "express consent". For example, at personal information collection devices such as cameras, clear and prominent signs are set up to inform users that they have entered the scope of personal information collection and that personal information will be collected. If an individual voluntarily enters the collection scope, it is deemed that they have agreed to the collection of their personal information; or on the personal information processing device, with clear signs / information informing users of the personal information processing rules, authorization is obtained from the individual through pop-up information or by asking the individual to upload their personal information; wherein, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the types of personal information processed.
Claims
1. A method for predicting blood glucose trends based on continuous blood glucose monitoring, characterized in that, Includes the following steps: The raw blood glucose signals continuously collected by the blood glucose monitor are filtered and denoised to obtain a clean blood glucose sequence. First-order and second-order difference calculations are performed on the clean blood glucose sequences at adjacent time points to obtain the rate of change and acceleration of blood glucose change. Based on the blood glucose change rate and blood glucose change acceleration, predict the user's blood glucose trend within a future preset time window, and plot a blood glucose curve based on the blood glucose trend; Based on the blood glucose curve, the threshold boundary is cross-determined to obtain the threshold trigger time point and the corresponding blood glucose risk level identifier, and prevention and intervention information is generated based on the blood glucose risk level identifier and the blood glucose curve.
2. The method for predicting blood glucose trends based on continuous blood glucose monitoring according to claim 1, characterized in that, The process of filtering and denoising the raw blood glucose signals continuously acquired by the blood glucose monitor to obtain a clean blood glucose sequence includes: The raw blood glucose signal continuously collected by the blood glucose monitor is sampled at a preset sampling time interval to obtain a blood glucose sampling sequence, and the blood glucose sampling sequence is processed by moving average to obtain a smooth blood glucose sequence. The smoothed blood glucose sequence is subjected to outlier detection and removal to obtain an outlier-removed sequence, and baseline drift correction is performed based on the outlier-removed sequence to obtain the clean blood glucose sequence.
3. The method for predicting blood glucose trends based on continuous blood glucose monitoring according to claim 1, characterized in that, The calculation of first-order and second-order differences on clean blood glucose sequences at adjacent time points to obtain the rate of blood glucose change and the acceleration of blood glucose change includes: The difference between the blood glucose values at two adjacent time points in the clean blood glucose sequence is calculated to obtain a blood glucose difference sequence. The blood glucose difference sequence is then divided by the corresponding time interval to obtain the rate of blood glucose change. Dividing the rate of change of blood glucose by the corresponding time interval yields the acceleration of the change in blood glucose.
4. The method for predicting blood glucose trends based on continuous blood glucose monitoring according to claim 1, characterized in that, The method of predicting a user's blood glucose trend within a future preset time window based on the blood glucose change rate and blood glucose change acceleration includes: The blood glucose change rate and blood glucose change acceleration are numerically combined and arranged to obtain a blood glucose dynamic parameter matrix. The blood glucose dynamic parameter matrix is then divided into time steps according to a future preset time window to obtain a predicted time node sequence. Based on the numerical parameters in the blood glucose dynamic parameter matrix, the blood glucose values of the predicted time node sequence are calculated point by point to obtain the future blood glucose value sequence. The blood glucose values of each time node in the future blood glucose value sequence are then checked for continuity to obtain the checked blood glucose sequence. The verification blood glucose sequence is marked with trend direction according to time order to obtain blood glucose trend identifier sequence. The blood glucose trend is obtained by combining and mapping the blood glucose trend identifier sequence with the verification blood glucose sequence.
5. The method for predicting blood glucose trends based on continuous blood glucose monitoring according to claim 1, characterized in that, The process of plotting a blood glucose curve based on the blood glucose trend includes: The blood glucose values and time nodes in the blood glucose trend are mapped and transformed to obtain a set of blood glucose coordinate points. The blood glucose coordinate points are then connected and plotted according to the time series to obtain a preliminary blood glucose trend curve segment. Based on the extreme points of the rate of change of blood glucose in the preliminary blood glucose trend curve segment, feature marking processing is performed to obtain blood glucose fluctuation feature points. The blood glucose fluctuation feature points are then graphically superimposed and combined with the preliminary blood glucose trend curve segment to obtain the blood glucose curve.
6. The method for predicting blood glucose trends based on continuous blood glucose monitoring according to claim 1, characterized in that, The step of cross-judging the threshold boundary based on the blood glucose curve to obtain the threshold trigger time point and the corresponding blood glucose risk level identifier includes: The blood glucose coordinate point values in the blood glucose curve are compared with the preset threshold boundary to obtain the threshold cross-judgment result. The lower cross-judgment point and the upper cross-judgment point in the threshold cross-judgment result are extracted to obtain the threshold trigger time point. Based on the blood glucose value corresponding to the threshold trigger time point, the deviation of blood glucose from the threshold boundary is calculated, and the blood glucose deviation is combined with the blood glucose change rate to divide the interval into grades, thereby obtaining the blood glucose risk level identifier.
7. The method for predicting blood glucose trends based on continuous blood glucose monitoring according to claim 1, characterized in that, The generation of preventive intervention information based on the blood glucose risk level identifier and the blood glucose curve includes: Based on the blood glucose risk level identifier, a time interval mapping is performed to obtain the risk warning period, and based on the risk warning period, the blood glucose fluctuation characteristics of the blood glucose curve are extracted to obtain the blood glucose fluctuation parameters. Based on the blood glucose fluctuation parameters, a preset intervention measure library is matched and screened to obtain a candidate intervention measure set. The intervention measures in the candidate intervention measure set are then sorted according to the blood glucose risk level identifier to obtain the prevention intervention information.
8. A blood glucose trend prediction system for continuous blood glucose monitoring, characterized in that, include: The calculation module is used to filter and denoise the raw blood glucose signals continuously collected by the blood glucose monitor to obtain a clean blood glucose sequence, and to perform first-order and second-order difference calculations on the clean blood glucose sequences at adjacent time points to obtain the rate of change and acceleration of blood glucose change. The prediction module is used to predict the user's blood glucose trend within a future preset time window based on the blood glucose change rate and blood glucose change acceleration, and to draw a blood glucose curve based on the blood glucose trend. The determination module is used to perform cross-determination of the threshold boundary based on the blood glucose curve, obtain the threshold trigger time point and the corresponding blood glucose risk level identifier, and generate prevention and intervention information based on the blood glucose risk level identifier and the blood glucose curve.
9. A computer device, characterized in that, The method includes a memory and a processor coupled to each other, wherein the memory stores program instructions and the processor executes the program instructions to implement the blood glucose trend prediction method for continuous blood glucose monitoring as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, The system stores program instructions that can be executed by a processor, the program instructions being used to implement the blood glucose trend prediction method for continuous blood glucose monitoring as described in any one of claims 1 to 7.