AKI patient blood glucose dynamic monitoring system based on clustering processing

Abstract

The invention discloses an AKI patient blood glucose dynamic monitoring system based on clustering processing. The AKI patient blood glucose dynamic monitoring system comprises a data acquisition module, an initialization module, a blood glucose sample clustering processing module and a blood glucose dynamic monitoring module. The invention belongs to the field of data processing, and particularly relates to an AKI patient blood glucose dynamic monitoring system based on clustering processing, according to the scheme, an adaptive forgetting rate is introduced to construct a time decay weighting matrix, the sensitivity to severe fluctuation is improved, and slow response caused by excessive smoothness is avoided; the time weight and intra-cluster covariance information are fused by adopting a forgetting mahalanobis distance, so that noise interference is reduced; the dynamic blood glucose monitoring effect is further improved; through non-correlation component filtration, the blood glucose is not dragged by historical rigid components when suddenly decreased or increased; through elastic credibility calibration, a secondary forgetting factor is introduced to perform exponential attenuation on a historical effective blood glucose sample number, in order to inhibit micro fluctuation misjudgment in a stationary period, a stationary detection factor is introduced, blood glucose mutation is quickly responded, and the blood glucose dynamic monitoring accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and specifically refers to a blood glucose dynamic monitoring system for AKI patients based on clustering processing. Background Art

[0002] The blood glucose dynamic monitoring system for AKI patients is a real-time analysis and early warning system for the blood glucose status of acute kidney injury patients. Its core goal is to quickly and accurately identify potential blood glucose abnormalities through dynamic modeling of the blood glucose changes of patients. However, generally, the blood glucose dynamic monitoring system for AKI patients lacks temporal sensitivity to the dynamic changes of blood glucose, fails to distinguish abnormal or marginal blood glucose samples, and thus leads to poor blood glucose dynamic monitoring effects; generally, the blood glucose dynamic monitoring system for AKI patients cannot focus on effective information, is easily misled by outdated or noisy centers, and delays early warning; rare states are submerged, and it is impossible to balance sensitivity to mutations and suppression of fluctuations during stable periods, thus leading to poor accuracy of blood glucose dynamic monitoring. Summary of the Invention

[0003] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides a blood glucose dynamic monitoring system for AKI patients based on clustering processing. Aiming at the problem that the general blood glucose dynamic monitoring system for AKI patients lacks temporal sensitivity to the dynamic changes of blood glucose, fails to distinguish abnormal or marginal blood glucose samples, and thus leads to poor blood glucose dynamic monitoring effects, this solution introduces an adaptive forgetting rate to construct a time-decaying weighted matrix, improves the sensitivity to sharp fluctuations, timely captures and focuses on the latest blood glucose trend, and avoids slow response caused by excessive smoothing; uses the forgetting Mahalanobis distance to fuse time weights and within-cluster covariance information, more accurately quantifies the probability that each new blood glucose sample belongs to each state cluster, and reduces noise interference; thereby improving the blood glucose dynamic monitoring effect; aiming at the problem that the general blood glucose dynamic monitoring system for AKI patients cannot focus on effective information, is easily misled by outdated or noisy centers, and delays early warning; rare states are submerged, and it is impossible to balance sensitivity to mutations and suppression of fluctuations during stable periods, thus leading to poor accuracy of blood glucose dynamic monitoring, this solution filters out irrelevant components, and when the blood glucose drops or rises suddenly, the system quickly completes the judgment without being dragged down by historical rigid components; through elastic confidence calibration, even in rare cases of severe hypoglycemia or acute hyperglycemia, a minimum alarm probability can be maintained, introducing a secondary forgetting factor to exponentially decay the number of historical effective blood glucose samples, and introducing a stable detection factor to suppress misjudgment of small fluctuations during the stable period, quickly respond to blood glucose mutations, and improve the accuracy of blood glucose dynamic monitoring.

[0004] The technical solution adopted by the present invention is as follows: The blood glucose dynamic monitoring system for AKI patients based on clustering processing provided by the present invention includes a data acquisition module, an initialization module, a blood glucose sample clustering processing module, and a blood glucose dynamic monitoring module;

[0005] The data acquisition module collects the blood glucose measurement sequences of AKI patients to form blood glucose samples;

[0006] The initialization module offline trains a Gaussian mixture model based on historical continuous blood glucose data;

[0007] The blood glucose sample clustering processing module calculates the trust degree by introducing the time-decaying weighted Mahalanobis distance, and completes the online clustering of new blood glucose samples by filtering out irrelevant components, elastic calibration and combining incremental parameter update with a forgetting factor;

[0008] The blood glucose dynamic monitoring module realizes blood glucose dynamic monitoring for the clustering results of new blood glucose samples.

[0009] Furthermore, the data acquisition module continuously collects the blood glucose measurement sequences of AKI patients , and takes the nearest L measurement values at the current moment to form a blood glucose sample ; where is the time stamp of the k-th blood glucose measurement; T is the transpose operation; , and are the blood glucose values of the k-th, (i - L + 1)-th and i-th measurements respectively.

[0010] Furthermore, the initialization module collects the historical continuous blood glucose data of AKI patients and offline trains a Gaussian mixture model with C components; expressed as: ; where is the probability density function of the Gaussian mixture model for any blood glucose sample vector x; j is the component index; is the initial weight of the j-th component; is the multivariate Gaussian distribution density function corresponding to the j-th component; is the covariance matrix.

[0011] Furthermore, the blood glucose sample clustering processing module applies the time-decaying weighted Mahalanobis distance to calculate the trust degree for new blood glucose samples, eliminates irrelevant components and performs trust degree calibration, and then uses incremental parameter update with a forgetting factor to cluster new blood glucose samples and assign the clusters to which the blood glucose samples belong; specifically including the following content:

[0012] Calculate the original trust degree; for the newly arrived blood glucose sample , construct a time-sensitive weighted matrix , and introduce an adaptive forgetting rate, expressed as: , ; where , and are forgetting weights; is the adaptive forgetting rate; and is the adjustment parameter; calculate the blood glucose sample the forgetting Mahalanobis distance from each cluster center , expressed as: ; where is the mean vector of the j-th cluster at the t-th iteration; is the covariance matrix of the j-th cluster at the t-th iteration; then calculate the original trust degree , expressed as: ; ; where is the mixing coefficient; and are the trust strengths of the blood glucose sample for the j-th cluster and the k-th cluster respectively;

[0013] Uncorrelated component filtering; calculate the displacement ; perform uncorrelated component filtering, expressed as: ; where is the mean vector of the j-th cluster at the (t - 1)-th iteration; and are the forgetting Mahalanobis distance and the displacement for the -th cluster respectively; is the set after uncorrelated component filtering;

[0014] Elastic trust degree calibration; to ensure that each blood glucose sample and the component trust degree have a lower bound, calculate the minimum trust degree threshold, and the formula used is: ; ; where is the lowest trust degree threshold; is the weight strength; compress the uncorrelated component trust degree to the minimum, and uniformly amplify the remaining cluster trust degrees, and the formula used is: ; where is the calibrated trust degree; is the blood glucose sample 's original trust degree for the g-th cluster;

[0015] Incremental parameter update; introduce a forgetting factor to perform exponential decay on the number of historical valid blood glucose samples, and introduce a steady state detection factor during the mean vector update, and finally perform incremental parameter update, and the formula used is: ; ; ; ; ; where and are the decayed weights of the j-th cluster at the (t - 1)-th iteration and the t-th iteration respectively; is the number of samples of the j-th cluster at the (t - 1)-th iteration; is the mixing coefficient of the j-th cluster in the (t + 1)-th iteration; l is the cluster index; is the weighted value after attenuation of the l-th cluster in the t-th iteration; and are the mean vector and covariance matrix of the j-th cluster in the (t + 1)-th iteration respectively; is the quadratic forgetting factor; and are the mean vectors of the j-th cluster and the l-th cluster in the (t - 1)-th iteration respectively; is the exponential parameter

[0016] Allocation; according to the calibrated confidence, the blood glucose is allocated to the cluster with the maximum confidence; when the maximum number of iterations is reached, the iteration is completed; the clustering of new blood glucose samples ends.

[0017] Furthermore, after the calculation by the blood glucose sample clustering processing module, for the newly arrived blood glucose samples, the blood glucose status is obtained based on the cluster to which they belong; and the blood glucose dynamic monitoring is carried out based on the current blood glucose status of the blood glucose samples.

[0018] The beneficial effects achieved by the present invention using the above scheme are as follows:

[0019] (1) Aiming at the problem that the blood glucose dynamic monitoring system for general AKI patients lacks the temporal sensitivity to the dynamic changes of blood glucose, fails to distinguish abnormal or marginal blood glucose samples, and thus leads to poor blood glucose dynamic monitoring effect, this scheme introduces an adaptive forgetting rate to construct a time decay weighted matrix, improves the sensitivity to violent fluctuations, timely captures and focuses on the latest blood glucose trend, and avoids the response slowness caused by excessive smoothing; uses the forgetting Mahalanobis distance to fuse the time weight and the within-cluster covariance information, more accurately quantifies the probability that each new blood glucose sample belongs to each state cluster, and reduces noise interference; thereby improving the blood glucose dynamic monitoring effect.

[0020] (2) Aiming at the problem that the blood glucose dynamic monitoring system for general AKI patients cannot focus on effective information, is easily misled by outdated or noisy centers, delays early warning; rare states are submerged, and it is impossible to balance mutation sensitivity and stable suppression, thus leading to poor blood glucose dynamic monitoring accuracy, this scheme filters out non-correlated components. When the blood glucose drops or rises suddenly, the system quickly makes a judgment and is not dragged down by historical rigid components; through elastic confidence calibration, even in rare severe hypoglycemia or acute hyperglycemia states, a minimum alarm probability can be maintained. A quadratic forgetting factor is introduced to exponentially decay the number of historical effective blood glucose samples. To suppress misjudgments of small fluctuations during the stable period, a stable detection factor is introduced to quickly respond to blood glucose mutations and improve the accuracy of blood glucose dynamic monitoring. Description of the Drawings

[0021] Figure 1Schematic diagram of the blood glucose dynamic monitoring system for AKI patients based on clustering processing provided by the present invention;

[0022] Figure 2 Schematic diagram of the process of the blood glucose sample clustering processing module.

[0023] The accompanying drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention, and do not constitute a limitation to the present invention. Detailed implementation manners

[0024] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0025] In the description of the present invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present invention.

[0026] Example 1, referring to Figure 1 , the blood glucose dynamic monitoring system for AKI patients based on clustering processing provided by the present invention includes a data acquisition module, an initialization module, a blood glucose sample clustering processing module, and a blood glucose dynamic monitoring module;

[0027] The data acquisition module collects the blood glucose measurement sequence of AKI patients to form blood glucose samples; and sends the data to the initialization module;

[0028] The initialization module offline trains a Gaussian mixture model based on historical continuous blood glucose data; and sends the data to the blood glucose sample clustering processing module;

[0029] The blood glucose sample clustering processing module introduces a time decay weighted Mahalanobis distance to calculate the trust degree, and completes the online clustering of new blood glucose samples by filtering non-related components, elastic calibration, and combining incremental parameter updates with a forgetting factor; and sends the data to the blood glucose dynamic monitoring module;

[0030] The blood glucose dynamic monitoring module realizes blood glucose dynamic monitoring on the clustering results of new blood glucose samples.

[0031] Example 2, referring to Figure 1, this embodiment is based on the above embodiment, and the data acquisition module continuously acquires the blood glucose measurement sequence of AKI patients , at the current moment, the most recent L measurement values are taken to form a blood glucose sample ; where is the timestamp of the k-th blood glucose measurement; T is the transpose operation; , and are the blood glucose values of the k-th, (i - L + 1)-th, and i-th measurements respectively.

[0032] Embodiment Three, refer to Figure 1 , this embodiment is based on the above embodiment, and the initialization module collects the continuous blood glucose data of historical AKI patients and offline trains a Gaussian mixture model with C components; expressed as: ; where is the probability density function of the Gaussian mixture model for any blood glucose sample vector x; j is the component index; is the initial weight of the j-th component; is the multivariate Gaussian distribution density function corresponding to the j-th component; is the covariance matrix; the clinical cluster number C = 4, corresponding to severe hypoglycemia, mild hypoglycemia, normal, and hyperglycemia; initialize using the positions of the blood glucose clinical boundaries in the artificially selected feature space; and initialize using the overall blood glucose sample covariance; after training is completed, the four components respectively correspond to different Gaussian distributions, and each distribution represents a specific blood glucose state cluster.

[0033] Embodiment Four, refer to Figure 1 and Figure 2 , this embodiment is based on the above embodiment, and the blood glucose sample clustering processing module includes calculating the original trust degree; for the newly arrived blood glucose sample , construct a time-sensitive weighted matrix , and introduce an adaptive forgetting rate, which is adaptively adjusted according to the speed of blood glucose change and is more sensitive to rapid blood glucose changes to avoid response lag caused by over-smoothing, expressed as: , ; strengthen recent data so that the distance can better reflect the true difference between the current blood glucose trend and the state centers. Among them, , and are the forgetting weights; is the adaptive forgetting rate; and are the adjustment parameters; calculate the forgetting Mahalanobis distance between the blood glucose sample , expressed as: ; where is the mean vector of the j-th cluster at the t-th iteration; is the covariance matrix of the j-th cluster at the t-th iteration; furthermore, the original trust degree is calculated , which is expressed as: ; ; making the contribution of recent blood glucose fluctuations to the current state judgment greater; among them, is the mixing coefficient; and are respectively the trust strengths of the blood glucose sample for the j-th cluster and the k-th cluster.

[0034] By performing the above operations, aiming at the problem that the blood glucose dynamic monitoring system for general AKI patients lacks the temporal sensitivity to the dynamic changes of blood glucose, differentiating abnormal or marginal blood glucose samples, and thus resulting in poor blood glucose dynamic monitoring effect, this solution introduces an adaptive forgetting rate to construct a time decay weighted matrix, improves the sensitivity to drastic fluctuations, timely captures and focuses on the latest blood glucose trend, and avoids the response slowness caused by over-smoothing; uses the forgetting Mahalanobis distance to fuse the time weight and the within-cluster covariance information, more accurately quantifies the probability that each new blood glucose sample belongs to each state cluster, and reduces noise interference; thereby improving the blood glucose dynamic monitoring effect.

[0035] Example 5, refer to Figure 1 and Figure 2 , this example is based on the above example. The blood glucose sample clustering processing module applies the time decay weighted Mahalanobis distance to calculate the trust degree for new blood glucose samples, eliminates irrelevant components and performs trust degree calibration, and then uses incremental parameter update with a forgetting factor to cluster new blood glucose samples and assign the clusters to which the blood glucose samples belong; it also includes the following content:

[0036] Irrelevant component filtering; calculating the displacement ; performing irrelevant component filtering. If the component is always farther compared to the current nearest component of the blood glucose sample, it can be regarded as irrelevant, and the trust degree has little impact on this update, which is expressed as: ; among them, is the mean vector of the j-th cluster at the (t - 1)-th iteration; and are respectively the forgetting Mahalanobis distance and the displacement for the -th cluster; is the set after irrelevant component filtering; for continuous blood glucose monitoring, most irrelevant centers are eliminated, significantly reducing the subsequent calculation amount; using component displacement to avoid misjudgment caused by center update;

[0037] Elastic trust degree calibration; to ensure that each blood glucose sample and the component trust degree have a lower bound, calculate the minimum trust degree threshold, and the formula used is: ; ; among them, is the lowest confidence threshold; is the weight strength; suppress the confidence of non-relevant components to the minimum and uniformly amplify the confidence of the remaining clusters. The formula used is: ; where is the calibrated confidence; is the blood glucose sample is the original confidence in the g-th cluster; retain the lowest response ability;

[0038] Incremental parameter update; use the adjusted confidence to more accurately reflect the confidence of the current blood glucose sample in each state. Especially when the blood glucose suddenly rises / drops, it can quickly push the components to the true aggregation area. Introduce a forgetting factor to exponentially decay the number of effective historical blood glucose samples to avoid hysteresis caused by excessive historical accumulation. It is sensitive to blood glucose mutations in AKI patients and has a smoothing effect on the stable stage; introduce a stable state detection factor when updating the mean vector , to avoid misjudging the fluctuations in the stable stage as mutation states; finally, perform incremental parameter update. The formula used is: ; ; ; ; ; where and are the decayed weights of the j-th cluster in the (t - 1)-th iteration and the t-th iteration respectively; is the number of samples in the j-th cluster in the (t - 1)-th iteration; is the mixing coefficient of the j-th cluster in the (t + 1)-th iteration; l is the cluster index; is the decayed weight of the l-th cluster in the t-th iteration; and are the mean vector and covariance matrix of the j-th cluster in the (t + 1)-th iteration respectively; is the secondary forgetting factor; and are the mean vectors of the j-th cluster and the l-th cluster in the (t - 1)-th iteration respectively; is the exponential parameter; it can capture new blood glucose dynamic patterns faster and reduce the risk of misjudging mutation states as normal states;

[0039] Allocation; according to the calibrated confidence, allocate the blood glucose to the cluster with the highest confidence. When the maximum number of iterations is reached, the iteration is completed; the clustering of new blood glucose samples ends.

[0040] By performing the above operations, for the blood glucose dynamic monitoring system of general AKI patients, it is unable to focus on effective information, is easily misled by outdated or noisy centers, and delays early warnings; rare states are submerged, and it is impossible to balance mutation sensitivity and stable inhibition, thus resulting in poor accuracy of blood glucose dynamic monitoring. In this solution, through non-correlated component filtering, when the blood glucose drops or rises suddenly, the system quickly makes a judgment without being dragged down by historical rigid components; through elastic confidence calibration, even in rare cases of severe hypoglycemia or acute hyperglycemia, a minimum probability can be maintained. A secondary forgetting factor is introduced to perform exponential decay on the number of historical effective blood glucose samples. To suppress misjudgment of minor fluctuations during the stable period, a stable detection factor is introduced to quickly respond to blood glucose mutations and improve the accuracy of blood glucose dynamic monitoring.

[0041] Example 6, refer to Figure 1 , based on the above example, the blood glucose dynamic monitoring module, after the calculation of the blood glucose sample clustering processing module, for newly arrived blood glucose samples, obtains the blood glucose state based on the cluster to which they belong; performs blood glucose dynamic monitoring based on the current blood glucose state of the blood glucose samples; for the state of severe hypoglycemia, triggers a high-priority alarm system to notify medical staff; for the state of mild hypoglycemia, issues a relatively lower-level alarm to remind patients and medical staff to pay attention to the low blood glucose situation and increase the monitoring frequency; for the normal state, performs blood glucose monitoring at the normal monitoring frequency; for the state of hyperglycemia, issues an alarm to inform patients and medical staff of the high blood glucose situation.

[0042] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention.

[0043] The above describes the present invention and its implementation manners. This description is not restrictive, and what is shown in the drawings is only one of the implementation manners of the present invention. The actual structure is not limited thereto. Generally speaking, if those of ordinary skill in the art are inspired by it and, without departing from the purpose of the present invention, creatively design a structural manner and an embodiment similar to this technical solution without creative efforts, they shall fall within the protection scope of the present invention.

Claims

1. AKI patient blood glucose dynamic monitoring system based on clustering processing, characterized in that: The system includes a data acquisition module, an initialization module, a blood glucose sample clustering processing module, and a blood glucose dynamic monitoring module; The data acquisition module collects the blood glucose measurement sequences of AKI patients to form blood glucose samples; The initialization module offline trains a Gaussian mixture model based on historical continuous blood glucose data; The blood glucose sample clustering processing module calculates the confidence by introducing the time decay weighted Mahalanobis distance, and completes the online clustering of new blood glucose samples by filtering non-related components, elastic calibration, and combining incremental parameter update with a forgetting factor; The blood glucose dynamic monitoring module realizes blood glucose dynamic monitoring for the clustering results of new blood glucose samples.

2. The AKI patient blood glucose dynamic monitoring system based on clustering processing according to claim 1, wherein: The data acquisition module continuously acquires the blood glucose measurement sequence of AKI patients , and at the current moment, the most recent L measurement values are taken to form a blood glucose sample ; where is the timestamp of the k-th blood glucose measurement; T is the transpose operation; , and are the blood glucose values of the k-th, (i - L + 1)-th, and i-th measurements respectively.

3. The AKI patient blood glucose dynamic monitoring system based on clustering processing according to claim 2, wherein: The initialization module collects the continuous blood glucose data of historical AKI patients and offline trains a Gaussian mixture model with C components, which is expressed as: ; where is the probability density function of the Gaussian mixture model for any blood glucose sample vector x; j is the component index; is the initial weight of the j-th component; is the multivariate Gaussian distribution density function corresponding to the j-th component; is the covariance matrix.

4. The AKI patient blood glucose dynamic monitoring system based on clustering processing according to claim 3, wherein: The blood glucose sample clustering processing module applies the time decay weighted Mahalanobis distance to calculate the confidence for new blood glucose samples, eliminates irrelevant components and performs confidence calibration, and then uses incremental parameter update with a forgetting factor to cluster new blood glucose samples and assign the clusters to which the blood glucose samples belong; specifically, it includes the following contents: Calculate the original trust degree; for the newly arrived blood glucose samples , construct a time-sensitive weighted matrix , and introduce an adaptive forgetting rate, expressed as: , ; where , and are forgetting weights; is the adaptive forgetting rate; and are adjustment parameters; calculate the forgetting Mahalanobis distance between the blood glucose sample and the cluster centers, expressed as: ; where is the mean vector of the j-th cluster at the t-th iteration; is the covariance matrix of the j-th cluster at the t-th iteration; and then calculate the original trust degree , expressed as: ; ; where is the mixing coefficient; and are the trust strengths of the blood glucose sample for the j-th cluster and the k-th cluster respectively; Non-related component filtering; Elastic confidence calibration; Incremental parameter update; Allocation: According to the calibrated confidence, the blood glucose is allocated to the cluster with the highest confidence; when the maximum number of iterations is reached, the iteration is completed; the clustering of new blood glucose samples ends.

5. The AKI patient blood glucose dynamic monitoring system based on clustering processing according to claim 4, characterized in that: The non-correlated component filtering calculates the displacement ; perform non-correlated component filtering, expressed as: ; where is the mean vector of the j-th cluster at the (t-1)-th iteration; and are the forgetting Mahalanobis distance and the displacement of the -th cluster, respectively; is the set after non-correlated component filtering.

6. The AKI patient blood glucose dynamic monitoring system based on clustering processing according to claim 5, characterized in that: The elastic trust calibration is to ensure that each blood glucose sample has a lower bound of component trust, and calculate the minimum trust threshold. The formula used is: ; ; where is the lowest trust threshold; is the weight intensity; Press the trust of non - relevant components to the minimum and uniformly amplify the trust of the remaining clusters. The formula used is: ; where is the calibrated trust; is the blood glucose sample 's original trust in the g - th cluster.

7. The AKI patient blood glucose dynamic monitoring system based on clustering processing according to claim 6, characterized in that: The incremental parameter update is to introduce a forgetting factor to perform exponential decay on the number of historical valid blood glucose samples, and introduce a steady state detection factor when updating the mean vector. , and finally perform incremental parameter update. The formula used is: ; ; ; ; ; where, and are the decayed weights of the j-th cluster in the (t - 1)-th iteration and the t-th iteration respectively; is the number of samples in the j-th cluster in the (t - 1)-th iteration; is the mixing coefficient of the j-th cluster in the (t + 1)-th iteration; l is the cluster index; is the decayed weight of the l-th cluster in the t-th iteration; and are the mean vector and covariance matrix of the j-th cluster in the (t + 1)-th iteration respectively; is the secondary forgetting factor; is the exponential parameter; and are the mean vectors of the j-th cluster and the l-th cluster in the (t - 1)-th iteration respectively.

8. The AKI patient blood glucose dynamic monitoring system based on clustering processing according to claim 7, characterized in that: The blood glucose dynamic monitoring module, after the calculation of the blood glucose sample clustering processing module, for newly arrived blood glucose samples, obtains the blood glucose status based on the belonging cluster; performs blood glucose dynamic monitoring based on the current blood glucose status of the blood glucose samples.

Images