A method and apparatus for adaptive compensation of continuous analyte monitoring

Abstract

This application relates to an adaptive compensation method and apparatus for continuous analyte monitoring. It obtains and determines an adaptive adjustment coefficient negatively correlated with the historical changes in tissue fluid and blood analyte concentrations. It then obtains and determines the blood analyte concentration monitoring value based on the real-time tissue fluid analyte concentration, the adaptive adjustment coefficient, and a preset conversion model. In this application, the stability of the sensor signal is quantified by real-time evaluation of the historical changes in tissue fluid and blood analyte concentrations, and the adjustment coefficient is adaptively determined to dynamically adjust the influence of the hysteresis compensation term on the final monitoring value. This allows the hysteresis compensation to be fully utilized to improve accuracy when the signal is stable, and automatically weakens the compensation to suppress noise amplification when the sensor experiences abnormal fluctuations. This significantly improves the output stability, anti-interference capability, and overall reliability of the monitoring system in complex usage scenarios.

Description

Technical Field

[0001] This application belongs to the field of medical device technology, and in particular relates to an adaptive compensation method and apparatus for continuous analyte monitoring. Background Technology

[0002] Continuous analyte monitoring systems (such as continuous glucose monitoring systems) continuously measure the concentration of analytes (such as glucose) in tissue fluid through subcutaneous sensors and use a preset conversion model to convert the tissue fluid analyte concentration into the corresponding blood analyte concentration value and output it to the user. One of the core aspects of this conversion process is to compensate for the physiological lag between tissue fluid and blood.

[0003] Currently widely used hysteresis compensation methods are usually based on preset fixed compensation parameters. However, such fixed parameter compensation strategies have significant drawbacks: when the sensor experiences short-term abnormal fluctuations due to compression, temperature changes, or unstable operating conditions, these abnormal values ​​are amplified by the fixed compensation mechanism, causing non-physiological abrupt changes in the final output blood analyte concentration value. This directly leads to distorted monitoring results, causing unnecessary interference or even misleading to users, and may trigger false alarms, seriously affecting the reliability, stability, and user experience of the continuous monitoring system. Summary of the Invention

[0004] In view of this, this application provides an adaptive compensation method and apparatus for continuous analyte monitoring, which aims to significantly improve the output stability, anti-interference ability and overall reliability of the monitoring system under complex usage scenarios.

[0005] This application provides an adaptive compensation method for continuous analyte monitoring, the method comprising:

[0006] Obtain the shift changes in historical tissue fluid analyte concentrations and historical blood analyte concentrations;

[0007] An adaptive adjustment coefficient is determined based on the shift changes in the historical tissue fluid analyte concentration and the shift changes in the historical blood analyte concentration, wherein the adaptive adjustment coefficient is negatively correlated with the shift change.

[0008] Obtain real-time concentrations of analytes in tissue fluid;

[0009] The blood analyte concentration monitoring value is determined based on the real-time tissue fluid analyte concentration, the adaptive adjustment coefficient, and the preset conversion model. The preset conversion model is used to obtain the blood analyte concentration based on the tissue fluid analyte concentration and the hysteresis compensation term, and the adaptive adjustment coefficient is used to adjust the degree of influence of the hysteresis compensation term on the blood analyte concentration monitoring value.

[0010] Optionally, determining the adaptive adjustment coefficient based on the shift changes in the historical tissue fluid analyte concentration and the shift changes in the historical blood analyte concentration includes:

[0011] Based on the historical changes in the concentration of analytes in tissue fluid and the historical changes in the concentration of analytes in blood, a historical analyte concentration fluctuation index is constructed.

[0012] The adaptive adjustment coefficient is determined based on the historical analyte concentration fluctuation index.

[0013] Optionally, the historical analyte concentration fluctuation index is a weighted combination of the shift changes in the historical tissue fluid analyte concentration and the shift changes in the historical blood analyte concentration.

[0014] Optionally, the historical analyte concentration fluctuation index is:

[0015] S(t) = ,

[0016] Wherein, S(t) is the historical analyte concentration fluctuation index at time t;

[0017] , which represents the shift in the concentration of analytes in historical tissue fluids;

[0018] This represents the shift in historical blood analyte concentrations;

[0019] n is the preset historical step size; i is the historical time index; , These are weighting coefficients; This is a preset power parameter; The concentration of the tissue fluid analyte at a historical time (ti); The concentration of the tissue fluid analyte at a historical time (ti-1); The concentration of the blood analyte at a historical moment (ti); The concentration of the blood analyte at a historical moment (ti-1); and These are the baseline changes in analyte concentrations in tissue fluid and blood, respectively, and are the average values ​​of analyte concentration changes calculated using a sliding window.

[0020] Optionally, determining the adaptive adjustment coefficient based on the historical analyte concentration fluctuation index includes:

[0021] The adaptive adjustment coefficient is determined based on the historical analyte concentration fluctuation index and the preset compensation coefficient function.

[0022] Optionally, the preset compensation coefficient function is a monotonically decreasing function.

[0023] Optionally, the monotonically decreasing function is an exponentially decreasing function, or a monotonically decreasing function with a range of [0,1] obtained by transforming the Sigmoid function or the Tanh function.

[0024] Optionally, the preset conversion model is:

[0025] ,

[0026] in, t represents the blood analyte concentration monitoring value at time t; z represents the current tissue fluid analyte concentration coefficient. Let be the concentration of the analyte in the tissue fluid at time t; This is the adaptive adjustment coefficient; This is the lag compensation term at time t.

[0027] Optionally, the lag compensation term is:

[0028] ,or ,

[0029] Where, h(t) and is the decay function; b is the model parameter.

[0030] This application also provides an adaptive compensation device for continuous analyte monitoring, the device comprising:

[0031] The first acquisition module is used to acquire the changes in the concentration of historical tissue fluid analytes and the changes in the concentration of historical blood analytes.

[0032] The first determining module is used to determine an adaptive adjustment coefficient based on the offset change in the historical tissue fluid analyte concentration and the offset change in the historical blood analyte concentration, wherein the adaptive adjustment coefficient is negatively correlated with the offset change.

[0033] The second acquisition module is used to acquire the real-time concentration of analytes in tissue fluid.

[0034] The second determining module is used to determine the blood analyte concentration monitoring value based on the real-time tissue fluid analyte concentration, the adaptive adjustment coefficient, and the preset conversion model. The preset conversion model is used to obtain the blood analyte concentration based on the tissue fluid analyte concentration and the hysteresis compensation term, and the adaptive adjustment coefficient is used to adjust the degree of influence of the hysteresis compensation term on the blood analyte concentration monitoring value.

[0035] Compared with existing technologies, this application provides an adaptive compensation method and apparatus for continuous analyte monitoring. This method acquires historical shifts in tissue fluid analyte concentration and historical shifts in blood analyte concentration. Based on these shifts, an adaptive adjustment coefficient is determined, where the adaptive adjustment coefficient is negatively correlated with the shift change. Real-time tissue fluid analyte concentration is then acquired. Based on the real-time tissue fluid analyte concentration, the adaptive adjustment coefficient, and a preset conversion model, the blood analyte concentration monitoring value is determined. The preset conversion model is used to determine the blood analyte concentration monitoring value based on the relationship between the tissue fluid analyte concentration and the hysteresis loop. The post-compensation term yields the blood analyte concentration, and an adaptive adjustment coefficient is used to adjust the influence of the hysteresis compensation term on the blood analyte concentration monitoring value. In this application, the stability of the sensor signal is quantified by evaluating the deviation changes of historical tissue fluid and blood analyte concentrations in real time, and the adjustment coefficient is adaptively determined to dynamically adjust the influence of the hysteresis compensation term on the final monitoring value. This allows the hysteresis compensation to be fully utilized to improve accuracy when the signal is stable, and the compensation to be automatically weakened to suppress noise amplification when the sensor experiences abnormal fluctuations. This significantly improves the output stability, anti-interference ability, and overall reliability of the monitoring system in complex usage scenarios. Attached Figure Description

[0036] 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.

[0037] Figure 1 This is a schematic flowchart of an adaptive compensation method for continuous analyte monitoring disclosed in an embodiment of this application;

[0038] Figure 2 This is a schematic diagram of the structure of an adaptive compensation device for continuous analyte monitoring disclosed in an embodiment of this application. Detailed Implementation

[0039] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0040] It should be noted that when a component is referred to as being "fixed to" or "set on" another component, it can be directly on or indirectly set on the other component; when a component is referred to as being "connected to" another component, it can be directly connected to or indirectly connected to the other component.

[0041] It should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0042] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "a plurality of" or "several" means two or more, unless otherwise explicitly specified.

[0043] It should be noted that the structures, proportions, sizes, etc., shown in the accompanying drawings of this specification are only for the purpose of assisting those skilled in the art in understanding and reading the content disclosed in the specification, and are not intended to limit the conditions under which this application can be implemented. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportions, or adjustments to the size should still fall within the scope of the technical content disclosed in this application, provided that they do not affect the effects and purposes that this application can produce.

[0044] like Figure 1 As shown in the embodiments of this application, an adaptive compensation method for continuous analyte monitoring is provided, the method comprising:

[0045] S1. Obtain the shift changes in historical tissue fluid analyte concentrations and historical blood analyte concentrations;

[0046] In this embodiment, the deviation of historical tissue fluid analyte concentration can be obtained by calculating the difference between the change in tissue fluid analyte concentration and the baseline change in tissue fluid analyte concentration, which is used to characterize the short-term fluctuation of tissue fluid analyte concentration. Similarly, the deviation of historical blood analyte concentration can be obtained by calculating the difference between the change in blood analyte concentration and the baseline change in blood analyte concentration, which is also used to characterize the short-term fluctuation of blood analyte concentration.

[0047] The changes in historical tissue fluid analyte concentrations can be obtained by calculating the differences in tissue fluid analyte concentrations at adjacent or specified historical time intervals, which characterizes the rate or magnitude of change in tissue fluid analyte concentrations. Similarly, the changes in historical blood analyte concentrations can be obtained by calculating the differences in blood analyte concentrations at adjacent or specified historical time intervals, which characterizes the rate or magnitude of change in blood analyte concentrations. Specifically, the changes can be first-order differences, rates of change, or values ​​after mathematical transformations (such as taking absolute values ​​or squaring). The baseline changes in tissue fluid analyte concentrations can be the first-order differences, rates of change, or values ​​after mathematical transformations of tissue fluid analyte concentrations within a dynamic time window, and the baseline changes in blood analyte concentrations can also be the first-order differences, rates of change, or values ​​after mathematical transformations of blood analyte concentrations within a dynamic time window. By obtaining a series of offset changes at historical time intervals, a data foundation can be provided for subsequent evaluation of signal stability and determination of adaptive adjustment coefficients.

[0048] S2. Determine the adaptive adjustment coefficient based on the historical shift changes in tissue fluid analyte concentration and the historical shift changes in blood analyte concentration, wherein the adaptive adjustment coefficient is negatively correlated with the shift change.

[0049] In this embodiment, the adaptive adjustment coefficient is negatively correlated with both the historical changes in tissue fluid analyte concentration and the historical changes in blood analyte concentration. The adaptive adjustment coefficient is used to dynamically adjust the intensity of hysteresis compensation. Its basic principle is that the difference between the historical changes in tissue fluid analyte concentration and the baseline changes, and the difference between the historical changes in blood analyte concentration and the baseline changes, directly reflect the recent stability of the sensor signal. A smaller shift indicates a more stable signal and more reliable sensor operation; in this case, hysteresis compensation should be enhanced to ensure accuracy. Conversely, a larger shift indicates a significant difference between the signal and the baseline, potentially indicating abnormal fluctuations or sensor interference; in this case, hysteresis compensation should be weakened to prevent noise amplification. Therefore, by establishing a negative correlation between the adaptive adjustment coefficient and historical shift changes, the adaptive adjustment coefficient used for subsequent intelligent adjustment of compensation intensity can be determined.

[0050] S3. Obtain the real-time concentration of analytes in tissue fluid;

[0051] In this embodiment, the real-time tissue fluid analyte concentration is the signal value directly measured by the analyte sensor implanted under the skin at the current sampling time. Before being input into subsequent calculations, this signal can undergo necessary preprocessing, such as basic filtering or unit conversion, to obtain a concentration value that reflects the current tissue fluid analyte level. This step provides the necessary real-time input for subsequent calculation of the final monitoring value based on real-time data and adaptive adjustment coefficient.

[0052] S4. Determine the blood analyte concentration monitoring value based on the real-time tissue fluid analyte concentration, the adaptive adjustment coefficient, and the preset conversion model. The preset conversion model is used to obtain the blood analyte concentration based on the tissue fluid analyte concentration and the hysteresis compensation term, and the adaptive adjustment coefficient is used to adjust the degree of influence of the hysteresis compensation term on the blood analyte concentration monitoring value.

[0053] In this embodiment, a preset conversion model linearly combines the real-time tissue fluid analyte concentration with a hysteresis compensation term to calculate the blood analyte concentration monitoring value. The hysteresis compensation term can be calculated from the real-time tissue fluid analyte concentration and historical tissue fluid analyte concentration, depending on the specific implementation (e.g., the hysteresis compensation term is the first-order difference of the tissue fluid analyte concentration), or from historical tissue fluid analyte concentration and historical blood analyte concentration (e.g., the hysteresis compensation term is a cumulative term of historical blood analyte concentration weighted by the initial tissue fluid analyte concentration and attenuation). The adaptive adjustment coefficient determined in step S2 directly participates in the calculation as the multiplicative coefficient of the hysteresis compensation term, thereby dynamically adjusting the degree of influence of the hysteresis compensation term on the final output result. Combining the aforementioned steps, this method, through an adaptive adjustment mechanism based on historical signal stability, can effectively suppress output distortion caused by short-term abnormal fluctuations of the sensor while ensuring the accuracy of normal physiological hysteresis compensation, thus significantly improving the output stability, anti-interference ability, and overall reliability of the continuous monitoring system in complex application scenarios.

[0054] Compared with existing technologies, this application provides an adaptive compensation method and apparatus for continuous analyte monitoring. This method acquires historical shifts in tissue fluid analyte concentration and historical shifts in blood analyte concentration. Based on these shifts, an adaptive adjustment coefficient is determined, where the adaptive adjustment coefficient is negatively correlated with the shift change. Real-time tissue fluid analyte concentration is then acquired. Based on the real-time tissue fluid analyte concentration, the adaptive adjustment coefficient, and a preset conversion model, the blood analyte concentration monitoring value is determined. The preset conversion model is used to determine the blood analyte concentration monitoring value based on the relationship between the tissue fluid analyte concentration and the hysteresis loop. The post-compensation term yields the blood analyte concentration, and an adaptive adjustment coefficient is used to adjust the influence of the hysteresis compensation term on the blood analyte concentration monitoring value. In this application, the stability of the sensor signal is quantified by evaluating the deviation changes of historical tissue fluid and blood analyte concentrations in real time, and the adjustment coefficient is adaptively determined to dynamically adjust the influence of the hysteresis compensation term on the final monitoring value. This allows the hysteresis compensation to be fully utilized to improve accuracy when the signal is stable, and the compensation to be automatically weakened to suppress noise amplification when the sensor experiences abnormal fluctuations. This significantly improves the output stability, anti-interference ability, and overall reliability of the monitoring system in complex usage scenarios.

[0055] As one implementation method, in this embodiment of the application, an adaptive adjustment coefficient is determined based on the shift changes in historical tissue fluid analyte concentrations and historical blood analyte concentrations, including:

[0056] Based on the historical shifts in analyte concentrations in tissue fluid and blood, a historical analyte concentration fluctuation index is constructed.

[0057] In this embodiment, fusion construction refers to the comprehensive processing of two types of historical offset change information to form a single scalar index that can quantify the overall fluctuation of the historical signal. For example, the fluctuation index can be the weighted sum, square sum, or other statistical quantity of the two types of offset changes within a specified historical time window. By constructing this comprehensive index, the recent working status of the sensor can be evaluated more stably and comprehensively, laying the foundation for the precise adjustment of subsequent coefficients.

[0058] The adaptive adjustment coefficient was determined based on historical analyte concentration fluctuation indicators.

[0059] In this embodiment, the adaptive adjustment coefficient can be determined based on the fluctuation index by calculating a preset function or querying a preset mapping table. The mapping relationship or function is configured such that when the fluctuation index is small, a larger coefficient value (approaching 1) is output to enhance hysteresis compensation; when the fluctuation index is large, a smaller coefficient value (approaching 0) is output to suppress hysteresis compensation. In this way, intelligent and automatic matching between the adjustment coefficient and the signal stability is achieved.

[0060] As one implementation method, in this embodiment of the application, the historical analyte concentration fluctuation index is a weighted combination of the shift change in historical tissue fluid analyte concentration and the shift change in historical blood analyte concentration.

[0061] In this embodiment, weighted combination refers to the process of assigning corresponding weight coefficients to the analyte concentration shifts from different sources (tissue fluid and blood) and at different historical times, and then merging them into a single value. This method allows for flexible adjustment of the contribution of various shift changes to the final fluctuation index by adjusting the values ​​of the weight coefficients. For example, higher weights can be assigned to shift changes from more recent times to improve sensitivity to recent signal changes; or different weights can be assigned to blood concentration shift changes than to tissue fluid concentration shift changes to reflect their different importance in physiological delay assessment. Constructing a fluctuation index through weighted combination enables a more refined and targeted quantification of historical signal fluctuation characteristics, providing a more reliable decision-making basis for subsequent adaptive adjustments.

[0062] As one implementation method, in this embodiment of the application, the historical analyte concentration fluctuation index is:

[0063] S(t) = ,

[0064] Wherein, S(t) is the historical analyte concentration fluctuation index at time t;

[0065] , which represents the shift in the concentration of analytes in historical tissue fluids;

[0066] This represents the shift in historical blood analyte concentrations;

[0067] n is the preset historical step size; i is the historical time index; , These are weighting coefficients; This is a preset power parameter; The concentration of the tissue fluid analyte at a historical time (ti); The concentration of the tissue fluid analyte at a historical time (ti-1); The concentration of the blood analyte at a historical moment (ti); The concentration of the blood analyte at a historical moment (ti-1); and These are the baseline changes in analyte concentrations in tissue fluid and blood, respectively, and are the average values ​​of analyte concentration changes calculated using a sliding window.

[0068] In this embodiment, the preset historical step size n determines the range of historical data involved in the calculation, and the value range of the historical time index i is from 1 to n.

[0069] The baseline change in tissue fluid analyte concentration is obtained by: within a sliding window of a preset time length, calculating a series of changes in tissue fluid analyte concentration based on the tissue fluid analyte concentration values ​​at each historical moment within the window; wherein, the change can be the first difference between concentrations at adjacent moments, or the concentration difference between moments with a specified step interval; subsequently, the series of changes in tissue fluid analyte concentration are statistically processed, and the processing result is used as the baseline change in tissue fluid analyte concentration; wherein, the statistical processing includes, but is not limited to, calculating the arithmetic mean, weighted average, or median.

[0070] The baseline change in blood analyte concentration is obtained by: within a sliding window of a preset time length, calculating a series of changes in blood analyte concentration based on the blood analyte concentration values ​​at each historical moment within the window; wherein, the change can be the first difference between concentrations at adjacent moments, or the concentration difference between moments with a specified step interval; subsequently, the series of changes in blood analyte concentration are statistically processed, and the processing result is used as the baseline change in blood analyte concentration; wherein, the statistical processing includes, but is not limited to, calculating the arithmetic mean, weighted average, or median.

[0071] This formula constructs a composite index that can sensitively reflect the intensity of historical signal fluctuations through weighted and exponential operations. Its design principle lies in: using weighting coefficients... , Setting a value that decreases over time makes the fluctuation index more sensitive to recent signal changes. By setting the power parameters r and j to positive even numbers (such as 2 or 4), the response weight of the index to large abnormal fluctuations can be significantly increased. Thus, the fluctuation index can more accurately and stably characterize the overall working status and signal quality of the sensor within the recent historical window, thereby providing key input for the high-reliability calculation of the subsequent adaptive adjustment coefficient.

[0072] As one implementation method, in this embodiment of the application, determining the adaptive adjustment coefficient based on historical analyte concentration fluctuation indicators includes:

[0073] The adaptive adjustment coefficient is determined based on historical analyte concentration fluctuation indicators and a preset compensation coefficient function.

[0074] In this embodiment, a preset compensation coefficient function defines a deterministic mapping relationship between historical analyte concentration fluctuation indicators and adaptive adjustment coefficients. This function is constructed as a monotonically decreasing function with respect to its input (i.e., the fluctuation indicator). The technical design logic is as follows: when the input fluctuation indicator value is small, it indicates that the historical signal is stable, and the function outputs a larger adaptive adjustment coefficient value (approaching 1) to enhance the contribution of the lag compensation term in the subsequent conversion model, thereby improving the steady-state monitoring accuracy. When the input fluctuation indicator value is large, it indicates that there are significant abnormal fluctuations in the historical signal, and the function outputs a smaller adaptive adjustment coefficient value (approaching 0) to weaken the contribution of the lag compensation term in the subsequent conversion model, thereby preferentially suppressing noise amplification and ensuring output stability. Through the calculation of this preset function, the adaptive adjustment coefficient achieves an adaptive and automatic response to signal quality.

[0075] As one implementation method, in this embodiment of the application, the preset compensation coefficient function is a monotonically decreasing function.

[0076] In this embodiment, a monotonically decreasing function refers to a mathematical function whose function value decreases strictly or not strictly as the input variable increases. Specifically configuring the compensation coefficient function as such a function is the key mathematical guarantee for realizing the core relationship of "negative correlation between the adaptive adjustment coefficient and the historical analyte concentration fluctuation index." Its technical significance lies in establishing a stable and predictable control rule for the system: when the fluctuation index increases (signal instability), the function output (i.e., the adaptive adjustment coefficient) will inevitably decrease, thereby automatically reducing the hysteresis compensation intensity; conversely, when the fluctuation index decreases (signal stability), the function output will inevitably increase, thereby automatically increasing the hysteresis compensation intensity. This deterministic, inverse correspondence establishes a robust causal logic between compensation behavior and signal quality, avoiding randomness or contradictions in the adjustment process and ensuring the consistency and reliability of the adaptive compensation system's response.

[0077] As one implementation method, in this embodiment of the application, the monotonically decreasing function is an exponentially decreasing function, or a monotonically decreasing function with a range of [0,1] obtained by transforming the Sigmoid function or the Tanh function.

[0078] In this embodiment, these specific function forms are selected to optimize the adaptive dynamic response curve with different mathematical properties.

[0079] Exponential decay functions (e.g., g(x) = (k>0) has the characteristics of smooth output as input increases and rapid convergence to zero, and can produce a sensitive suppression response to abnormal fluctuations; the Sigmoid function and the Tanh function are not monotonically decreasing themselves, but can be transformed by simple linear transformations (such as g(x)=1 sigmoid(kx) or g(x)=(1 The expression tanh(kx) / 2 can construct a monotonically decreasing function with a range of [0,1]. Such functions change gradually when the input value is in the middle range, providing a smooth transition region, while tending to saturate when the input value is extremely large or small, enhancing the stability of the system. By selecting and configuring different specific functions (such as adjusting the parameter k), the dynamic behavior of adaptive compensation can be finely tuned according to the actual application scenario (such as sensitivity requirements for abnormal response, signal noise level), thereby achieving the best balance between response speed and stability.

[0080] As one implementation method, in this embodiment of the application, the preset conversion model is:

[0081] ,

[0082] in, Let be the blood analyte concentration at time t; and z be the current tissue fluid analyte concentration coefficient. Let be the concentration of the analyte in the tissue fluid at time t; For adaptive adjustment coefficient This is the lag compensation term at time t.

[0083] In this embodiment, the current tissue fluid analyte concentration coefficient is a constant. The model integrates real-time measurements with historical compensation information in a linear combination to calculate the final blood analyte concentration monitoring value Gp(t), where the first term... Represented by the current tissue fluid concentration Instantaneous estimated components generated by fixed-scale z-mapping; second term The dynamic adjustment component used to compensate for physiological lag consists of a lag compensation term f(t) composed of historical data and an adaptive adjustment coefficient determined in real time.

[0084] The result is obtained by multiplying q(t). The adaptive adjustment coefficient q(t) acts as a dynamic weight in this model. Through the mechanism established in step S2, the contribution of the lag compensation term f(t) to the final output Gp(t) is adjusted in real time according to the stability of the historical signal. The model has a clear structure. By introducing a dynamically adjustable coefficient q(t), it achieves intelligent and adaptive fine control of the lag compensation intensity while maintaining the stability of the basic conversion framework.

[0085] As one implementation method, in this embodiment of the application, the lag compensation term is:

[0086] ,or ,

[0087] Where, h(t) and is the decay function; b is the model parameter.

[0088] In this embodiment, the two forms of hysteresis compensation terms correspond to different levels of complexity in physiological dynamics modeling and implementation. The first form... The first difference of the concentration of the analyte in the tissue fluid is given by, where, Let be the concentration of the analyte in the tissue fluid at time t. Let be the concentration of the analyte in the tissue fluid at time t-1. The first form of hysteresis compensation directly and efficiently captures the instantaneous trend of concentration change, suitable for scenarios sensitive to computational resources or requiring rapid response; the second form, based on convolution operations, models the physiological hysteresis process in greater detail: where, This represents the initial concentration of the analyte in the tissue fluid. The term characterizes the attenuation of the effect of the initial tissue fluid analyte concentration, while the summation term... This is achieved by analyzing historical blood analyte concentrations (either historical estimates or obtained through other means). A decay-weighted summation is performed to reflect the cumulative delayed effect of historical blood analyte levels on the current tissue fluid concentration; the decay function... The second form determines the decay rate of historical influence through parameters. and decay function It offers more adjustable dimensions, enabling more accurate fitting of complex physiological dynamics, and is suitable for scenarios with extremely high monitoring accuracy requirements. The system can select or switch different compensation terms according to actual needs.

[0089] Specifically, ,in, , This refers to the uptake and metabolic breakdown of analytes by subcutaneous tissue. P1 can also be a constant, for example, obtained through experimental calibration or system optimization, to simplify the model implementation, representing the forward transport rate of the analyte through the capillary.

[0090] like Figure 2 As shown, this application also provides an adaptive compensation device for continuous analyte monitoring, the device comprising:

[0091] The first acquisition module 21 is used to acquire the offset change of historical tissue fluid analyte concentration and the offset change of historical blood analyte concentration.

[0092] The first determining module 22 is used to determine an adaptive adjustment coefficient based on the offset change of historical tissue fluid analyte concentration and the offset change of historical blood analyte concentration, wherein the adaptive adjustment coefficient is negatively correlated with the offset change.

[0093] The second acquisition module 23 is used to acquire the real-time concentration of tissue fluid analytes;

[0094] The second determining module 24 is used to determine the blood analyte concentration monitoring value based on the real-time tissue fluid analyte concentration, the adaptive adjustment coefficient, and the preset conversion model. The preset conversion model is used to obtain the blood analyte concentration based on the tissue fluid analyte concentration and the hysteresis compensation term, and the adaptive adjustment coefficient is used to adjust the degree of influence of the hysteresis compensation term on the blood analyte concentration monitoring value.

[0095] The embodiments in this specification are described in a progressive manner, with each embodiment focusing on the related aspects.

[0096] For any differences between the embodiments, or for the same or similar parts between the embodiments, please refer to each other.

[0097] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An adaptive compensation method for continuous analyte monitoring, characterized in that, The method includes: The offset changes in historical tissue fluid analyte concentrations and historical blood analyte concentrations are obtained, wherein the offset change in historical tissue fluid analyte concentrations is the difference between the change in tissue fluid analyte concentration and the baseline change in tissue fluid analyte concentration, and the offset change in historical blood analyte concentrations is the difference between the change in blood analyte concentration and the baseline change in blood analyte concentration. An adaptive adjustment coefficient is determined based on the shift changes in the historical tissue fluid analyte concentration and the shift changes in the historical blood analyte concentration, wherein the adaptive adjustment coefficient is negatively correlated with the shift change. Obtain real-time concentrations of analytes in tissue fluid; The blood analyte concentration monitoring value is determined based on the real-time tissue fluid analyte concentration, the adaptive adjustment coefficient, and the preset conversion model. The preset conversion model is used to obtain the blood analyte concentration based on the tissue fluid analyte concentration and the hysteresis compensation term, and the adaptive adjustment coefficient is used to adjust the degree of influence of the hysteresis compensation term on the blood analyte concentration monitoring value.

2. The method according to claim 1, characterized in that, The step of determining the adaptive adjustment coefficient based on the shift changes in the historical tissue fluid analyte concentration and the shift changes in the historical blood analyte concentration includes: Based on the historical changes in the concentration of analytes in tissue fluid and the historical changes in the concentration of analytes in blood, a historical analyte concentration fluctuation index is constructed. The adaptive adjustment coefficient is determined based on the historical analyte concentration fluctuation index.

3. The method according to claim 2, characterized in that, The historical analyte concentration fluctuation index is a weighted combination of the shift changes in the historical tissue fluid analyte concentration and the shift changes in the historical blood analyte concentration.

4. The method according to claim 3, characterized in that, The historical analyte concentration fluctuation index is as follows: S(t)= , Wherein, S(t) is the historical analyte concentration fluctuation index at time t; , which represents the shift in the concentration of analytes in historical tissue fluids; This represents the shift in historical blood analyte concentrations; n is the preset historical step size; i is the historical time index; , These are weighting coefficients; This is a preset power parameter; The concentration of the tissue fluid analyte at a historical time (ti); The concentration of the tissue fluid analyte at a historical time (ti-1); The concentration of the blood analyte at a historical moment (ti); The concentration of the blood analyte at a historical moment (ti-1); and These are the baseline changes in analyte concentrations in tissue fluid and blood, respectively, and are the average values ​​of analyte concentration changes calculated using a sliding window.

5. The method according to claim 2, characterized in that, The step of determining the adaptive adjustment coefficient based on the historical analyte concentration fluctuation index includes: The adaptive adjustment coefficient is determined based on the historical analyte concentration fluctuation index and the preset compensation coefficient function.

6. The method according to claim 5, characterized in that, The preset compensation coefficient function is a monotonically decreasing function.

7. The method according to claim 6, characterized in that, The monotonically decreasing function is an exponentially decreasing function, or a monotonically decreasing function with a range of [0,1] obtained by transforming the Sigmoid function or the Tanh function.

8. The method according to claim 1, characterized in that, The preset conversion model is: ， in, Let be the blood analyte concentration at time t; z is the current concentration coefficient. Let be the concentration of the analyte in the tissue fluid at time t; This is the adaptive adjustment coefficient; This is the lag compensation term at time t.

9. The method according to claim 8, characterized in that, The delayed compensation item is: ,or , Where, h(t) and is the decay function; b is the model parameter.

10. An adaptive compensation device for continuous analyte monitoring, characterized in that, The device includes: The first acquisition module is used to acquire the offset change of historical tissue fluid analyte concentration and the offset change of historical blood analyte concentration, wherein the offset change of historical tissue fluid analyte concentration is the difference between the change in tissue fluid analyte concentration and the baseline change in tissue fluid analyte concentration, and the offset change of historical blood analyte concentration is the difference between the change in blood analyte concentration and the baseline change in blood analyte concentration. The first determining module is used to determine an adaptive adjustment coefficient based on the offset change in the historical tissue fluid analyte concentration and the offset change in the historical blood analyte concentration, wherein the adaptive adjustment coefficient is negatively correlated with the offset change. The second acquisition module is used to acquire the real-time concentration of analytes in tissue fluid. The second determining module is used to determine the blood analyte concentration monitoring value based on the real-time tissue fluid analyte concentration, the adaptive adjustment coefficient, and the preset conversion model. The preset conversion model is used to obtain the blood analyte concentration based on the tissue fluid analyte concentration and the hysteresis compensation term, and the adaptive adjustment coefficient is used to adjust the degree of influence of the hysteresis compensation term on the blood analyte concentration monitoring value.

Images