Patient multi-parameter coupled intravenous insulin intelligent control method and system

Abstract

The invention relates to the technical field of medical health management, and particularly discloses a patient multi-parameter coupled venous insulin intelligent control method and a patient multi-parameter coupled venous insulin intelligent control system. The method realizes accurate regulation and control of the intravenous insulin infusion rate and ensures that the blood glucose level of a patient is maintained in a safe target interval, and specifically comprises the following steps: evaluating in-vivo energy conversion efficiency and night body sugar regulation efficiency, constructing a comprehensive feature vector by using the parameters, and inputting the comprehensive feature vector into a pre-trained random forest model; the blood glucose fluctuation range of a patient in the current and future preset time periods is predicted in combination with a blood glucose kinetic model, and an optimal insulin infusion rate adjustment strategy is calculated through an optimization algorithm based on a prediction result and a set target blood glucose interval, so that the insulin infusion rate is dynamically adjusted, and it is ensured that the blood glucose level is stably maintained in the target interval.

Description

Technical Field

[0001] The present invention relates to the field of medical health management technology, and in particular to a method and system for intelligent control of intravenous insulin with multi-parameter coupling for patients. Background Art

[0002] Blood glucose monitoring and insulin therapy mainly rely on patients to self-monitor their blood glucose levels and manually adjust insulin dosages according to doctor's advice. However, this method is not only time-consuming and error-prone, but may also lead to excessive blood glucose fluctuations and increase the risk of hypoglycemia or hyperglycemia. Therefore, developing a technology that can automatically and accurately adjust the intravenous insulin infusion rate has become an important research direction in the field of diabetes management. With the development of machine learning, data-driven models and biomedical engineering technology, it has become possible to use multi-parameter coupling methods to optimize insulin infusion strategies, providing a new solution for personalized and intelligent diabetes management.

[0003] The existing technology has the following problems:

[0004] Traditional fixed or semi-automatic adjustment strategies are often based on limited physiological parameters and lack sufficient consideration of individual differences, resulting in low prediction accuracy and difficulty in meeting the personalized needs of different patients. Existing methods usually fail to fully utilize the complex interactions between the patient's various physiological parameters, such as the relationship between changes in chromium concentration and energy conversion efficiency, and the body's unique blood sugar regulation mechanism during deep sleep, which limits the effectiveness of blood sugar prediction and control. Most current systems lack the ability to self-learn and dynamically adjust, and cannot adapt to changes in the patient's lifestyle or health status in real time, thereby affecting the effectiveness of long-term management. Due to the lack of an effective closed-loop control system, the existing technology is difficult to ensure real-time optimization of the insulin infusion rate, increasing the risk of hypoglycemia or hyperglycemia events. These problems highlight the limitations of the existing technology in diabetes management and also point out the key improvements targeted by the present invention. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and system for intelligent control of intravenous insulin with multi-parameter coupling for patients, so as to solve the problems in the above background.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] The method for intelligent intravenous insulin control based on multi-parameter coupling of a patient includes the following steps:

[0008] S1: Obtain and monitor the concentration level of chromium in the patient's blood as a first specific physiological parameter, and evaluate the energy conversion efficiency in the patient's body based on the changing trend of the chromium concentration;

[0009] S2: Obtain and monitor the proportion of deep sleep time during the patient's sleep cycle as a second specific physiological parameter. Based on the body's blood sugar regulation mechanism during deep sleep, the patient's nighttime body sugar regulation efficiency is evaluated according to the proportion of deep sleep time.

[0010] S3: Comprehensively analyzing the real-time changes of the first specific physiological parameter and the second specific physiological parameter, constructing a dynamic model of the patient's blood glucose state, and predicting the patient's current blood glucose fluctuation range;

[0011] S4: Automatically adjust the intravenous insulin infusion rate based on the predicted patient's current blood sugar fluctuation range to ensure that the patient's blood sugar level is maintained within the predetermined target range.

[0012] As a further solution of the present invention: the evaluation of the energy conversion efficiency in the patient's body according to the changing trend of the chromium concentration specifically includes:

[0013] During the monitoring period, the concentration level of chromium in the patient's blood is obtained in real time according to the time series. The characteristic value of the chromium concentration change is calculated based on the change amplitude of the chromium concentration in the patient's blood, and it is determined whether the characteristic value of the chromium concentration change is greater than or equal to the preset threshold. If so, the energy conversion efficiency in the patient's body is abnormal; if not, the energy conversion efficiency in the patient's body is normal.

[0014] As a further solution of the present invention: the process of obtaining the characteristic value of the chromium element concentration change is:

[0015] For the chromium concentration data obtained in real time during the monitoring period, Morlet wavelet was used to convert the chromium concentration time series data into wavelet coefficients at different scales to extract the local characteristics of the chromium concentration changing with time.

[0016] Calculate each scale The energy density under the condition of chromium is determined, the key scale range is determined according to the actual data analysis results, and the average energy density within the selected key scale range is calculated as the characteristic value of the chromium element concentration change.

[0017] As a further embodiment of the present invention, the evaluation of the patient's nighttime sugar regulation efficiency specifically includes:

[0018] During the monitoring period, the proportion of deep sleep time in the patient's sleep cycle is obtained in real time. Based on the blood sugar regulation mechanism of the patient's body during deep sleep, the sugar regulation characteristic value is calculated to determine whether the patient's sugar regulation characteristic value is greater than or equal to the preset threshold. If so, the patient's body sugar regulation efficiency at night is normal; if not, the patient's body sugar regulation efficiency at night is abnormal.

[0019] As a further solution of the present invention: the process of obtaining the sugar regulation characteristic value is:

[0020] During the monitoring period, the patient's sleep state time series and corresponding blood glucose level time series are obtained in real time, where the sleep states include wakefulness, light sleep, and deep sleep. Based on the sleep state, the blood glucose level data corresponding to the deep sleep stage is screened out. Fast Fourier transform is applied to the blood glucose level time series during deep sleep to convert the blood glucose level time series into frequency domain data, thereby analyzing the frequency characteristics of blood glucose levels changing over time. A low-frequency interval is preset, and the proportion of energy in the low-frequency interval to the total energy is calculated, and the low-frequency energy proportion is recorded as the sugar regulation characteristic value.

[0021] As a further solution of the present invention: the construction of a dynamic model of the patient's blood glucose status specifically includes:

[0022] The patient's chromium concentration change characteristic value and sugar regulation characteristic value are constructed into a comprehensive characteristic vector as the input of the dynamic model of the patient's blood sugar status. The dynamic model of the patient's blood sugar status is trained with minimizing the error between the predicted patient's current blood sugar fluctuation range and the actual patient's current blood sugar fluctuation range as the training goal. Based on the trained dynamic model of the patient's blood sugar status, the patient's current blood sugar fluctuation range is output. The dynamic model of the patient's blood sugar status is a random forest model.

[0023] As a further solution of the present invention: the predicting of the patient's current blood sugar fluctuation range specifically includes:

[0024] The random forest model was trained using historical data. The dataset was divided into a training set and a test set. The training set was used to train the random forest model. Multiple decision trees were constructed and their prediction results were summarized. The model performance was evaluated on the test set, and the model parameters were adjusted. A comprehensive feature vector was constructed by collecting the characteristic values of the chromium concentration changes and the sugar regulation characteristic values of the patients to be predicted. The comprehensive feature vector was input into the trained random forest model. Based on the model output results, the predicted value of the patient's current blood sugar fluctuation range was obtained.

[0025] As a further embodiment of the present invention, the automatic adjustment of the intravenous insulin infusion rate specifically includes:

[0026] A dynamic regulation system based on a model predictive control algorithm is constructed. The system obtains the patient's chromium concentration change characteristic value, sugar regulation characteristic value, and current blood glucose level in real time as input, uses a pre-trained random forest model to predict the patient's short-term blood glucose fluctuation range, and combines it with a blood glucose kinetic model to predict the blood glucose change trend within a preset time period in the future. Based on the prediction results and the set target blood glucose range, the optimal insulin infusion rate adjustment strategy is calculated to enable the future blood glucose level to reach the target range while satisfying all constraints. The constraints include: a safe range for the insulin infusion rate and a safety threshold for the blood glucose level. At each time step, only the insulin infusion rate adjustment amount at the current moment in the calculated strategy is executed.

[0027] As a further solution of the present invention: the method of predicting the blood glucose change trend within a preset time period in the future by combining the blood glucose kinetic model specifically includes:

[0028] A blood glucose kinetics model was established and applied to predict blood glucose change trends within a preset time period in the future. The model was based on the patient's chromium concentration change characteristic values, sugar regulation characteristic values, and current blood glucose levels, and was dynamically adjusted in combination with the real-time updated insulin infusion rate. By inputting the patient's chromium concentration change characteristic values, sugar regulation characteristic values, and current blood glucose levels into the blood glucose kinetics model, the model equation was solved using a numerical simulation method to accurately predict changes in blood glucose levels at various time points in the future under different insulin infusion strategies.

[0029] The patient multi-parameter coupled intravenous insulin intelligent control system includes:

[0030] A data acquisition module, the data acquisition module is used to collect the concentration level of chromium in the patient's blood, the proportion of deep sleep time in the patient's sleep cycle, and the blood sugar level;

[0031] an energy conversion efficiency evaluation module, wherein the energy conversion efficiency evaluation module obtains and monitors the concentration level of chromium in the patient's blood as a first specific physiological parameter, and evaluates the energy conversion efficiency in the patient's body according to a change trend of the chromium concentration;

[0032] a nighttime glucose regulation effectiveness assessment module, which obtains and monitors the proportion of deep sleep time during a patient's sleep cycle as a second specific physiological parameter, and assesses the patient's nighttime glucose regulation effectiveness based on the body's blood glucose regulation mechanism during deep sleep and the proportion of deep sleep time;

[0033] a patient blood glucose fluctuation range prediction module, which constructs a dynamic model of the patient's blood glucose state by comprehensively analyzing the real-time changes of the first specific physiological parameter and the second specific physiological parameter, and predicts the patient's current blood glucose fluctuation range;

[0034] The insulin infusion rate automatic adjustment module automatically adjusts the intravenous insulin infusion rate based on the predicted patient's current blood sugar fluctuation range to ensure that the patient's blood sugar level is maintained within a predetermined target range.

[0035] Beneficial effects of the present invention:

[0036] (1) The present invention integrates multiple specific physiological parameters of patients, such as the chromium concentration level in the blood and the proportion of deep sleep time in the sleep cycle, and uses these parameters to construct a comprehensive feature vector as input. Combined with the pre-trained random forest model and blood glucose dynamics model, it achieves accurate prediction of the patient's current and future blood glucose fluctuation range. This method not only deeply considers individual differences, but also evaluates the energy conversion efficiency in the body by analyzing the trend of changes in chromium concentration and the nighttime glucose regulation efficiency based on the body's unique blood glucose regulation mechanism during deep sleep, thereby reflecting the unique physiological characteristics of each patient. It also uses a dynamic adjustment mechanism to optimize the intravenous insulin infusion rate in real time to ensure that the blood glucose level is stably maintained within a safe target range. Compared with traditional fixed or semi-automatic adjustment strategies, the present invention significantly improves the accuracy and safety of blood glucose management and effectively prevents the occurrence of hypoglycemia and hyperglycemia events. In addition, the system's built-in data-driven model optimization mechanism allows for continuous self-adjustment and optimization of model parameters based on newly collected patient data, further improving long-term prediction accuracy and individual adaptability, providing diabetic patients with a more personalized, accurate and efficient treatment plan, and greatly enhancing the safety and effectiveness of diabetes management. This innovative approach not only represents a major advancement in diabetes management technology, but also brings patients a higher quality of life.

[0037] (2) The present invention introduces an advanced data-driven model optimization mechanism, which enables the system to self-adjust and optimize model parameters based on real-time updated patient data, thereby ensuring that as time and the patient's physiological state change, the system can not only more accurately capture and reflect the individual's unique physiological response pattern, but also flexibly adapt to the influence of external factors such as lifestyle changes or health status fluctuations. This adaptive ability not only improves the accuracy of short-term blood glucose prediction, but also provides a solid foundation for long-term management, ensuring that efficient and stable blood glucose control can be achieved even in complex and changing situations. Specifically, by continuously monitoring and analyzing the characteristic values of chromium concentration changes and sugar regulation characteristic values, combined with the dynamic regulation of random forest models and blood glucose dynamics models, the present invention can automatically optimize the insulin infusion rate within each time step, effectively preventing the occurrence of hypoglycemia and hyperglycemia events. In addition, the system's self-learning function further enhances the effectiveness of its long-term management. By continuously iterating and optimizing the algorithm, the personalization and response speed of the treatment plan are significantly improved, providing a highly reliable and sustainable blood glucose management solution for diabetic patients, greatly promoting the improvement of patients' health management quality and quality of life. This feature is particularly suitable for chronic disease management scenarios that require long-term monitoring and fine-tuning, marking a major advancement in diabetes treatment technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The present invention will be further described below with reference to the accompanying drawings.

[0039] Figure 1 It is a flowchart of the patient multi-parameter coupled intravenous insulin intelligent control method of the present invention;

[0040] Figure 2 It is a flow chart of the patient multi-parameter coupled intravenous insulin intelligent control system of the present invention. DETAILED DESCRIPTION

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0042] See also Figure 1 As shown, the present invention is a patient multi-parameter coupled intravenous insulin intelligent control method, comprising the following steps:

[0043] S1: Obtain and monitor the concentration level of chromium in the patient's blood as a first specific physiological parameter, and evaluate the energy conversion efficiency in the patient's body based on the changing trend of the chromium concentration;

[0044] S2: Obtain and monitor the proportion of deep sleep time during the patient's sleep cycle as a second specific physiological parameter. Based on the body's blood sugar regulation mechanism during deep sleep, the patient's nighttime body sugar regulation efficiency is evaluated according to the proportion of deep sleep time.

[0045] S3: Comprehensively analyzing the real-time changes of the first specific physiological parameter and the second specific physiological parameter, constructing a dynamic model of the patient's blood glucose state, and predicting the patient's current blood glucose fluctuation range;

[0046] S4: Automatically adjust the intravenous insulin infusion rate based on the predicted patient's current blood sugar fluctuation range to ensure that the patient's blood sugar level is maintained within the predetermined target range.

[0047] In S1, the concentration level of chromium in the patient's blood is obtained and monitored as a first specific physiological parameter. Based on the changing trend of the chromium concentration, the energy conversion efficiency in the patient's body is evaluated, specifically including:

[0048] During the monitoring period, the concentration level of chromium in the patient's blood is obtained in real time according to a time series. Based on the variation range of the chromium concentration in the patient's blood, a characteristic value of the chromium concentration variation is calculated. It is determined whether the characteristic value of the chromium concentration variation is greater than or equal to a preset threshold. If so, the energy conversion efficiency in the patient's body is abnormal; if not, the energy conversion efficiency in the patient's body is normal.

[0049] The process of obtaining the characteristic value of the chromium element concentration change is as follows:

[0050] For the chromium concentration data obtained in real time according to the time series during the monitoring period, Morlet wavelet is selected to convert the chromium concentration time series data into wavelet coefficients at different scales. ;in, represents the scale parameter, Represents the time position parameter to extract the local characteristics of chromium concentration changing with time;

[0051] Calculate each scale The energy density under , the calculation expression is: ;in, Representation scale The energy density under Indicates the total number of time position parameters, and determines the key scale range based on actual data analysis results The energy density within the key scale range can reflect the main change mode and amplitude of the chromium concentration. The average value of the energy density within the selected key scale range is calculated as the characteristic value of the chromium concentration change.

[0052] It should be noted that the characteristic value of the change in chromium concentration reflects the abnormal degree of energy conversion efficiency in the patient's body, and the larger the value of the characteristic value of the change in chromium concentration, the higher the abnormal degree of energy conversion efficiency in the corresponding patient's body.

[0053] In S2, the proportion of deep sleep time in the patient's sleep cycle is obtained and monitored as a second specific physiological parameter. Based on the body's blood sugar regulation mechanism during deep sleep, the patient's nighttime body sugar regulation efficiency is evaluated according to the deep sleep time proportion, specifically including:

[0054] During the monitoring period, the proportion of deep sleep time in the patient's sleep cycle is obtained in real time. Based on the patient's blood sugar regulation mechanism during deep sleep, the sugar regulation characteristic value is calculated to determine whether the patient's sugar regulation characteristic value is greater than or equal to a preset threshold. If so, the patient's nighttime body sugar regulation efficiency is normal; if not, the patient's nighttime body sugar regulation efficiency is abnormal.

[0055] The process of obtaining the sugar regulation characteristic value is as follows:

[0056] During the monitoring period, the patient's sleep state time series and corresponding blood glucose level time series are obtained in real time. The sleep states include wakefulness, light sleep, and deep sleep. Based on the sleep state, the blood glucose level data corresponding to the deep sleep stage is screened out. Fast Fourier transform is applied to the blood glucose level time series during deep sleep to convert the blood glucose level time series into frequency domain data, thereby analyzing the frequency characteristics of blood glucose level changes over time. A low-frequency interval is preset, and the proportion of energy in the low-frequency interval to the total energy is calculated. The calculation expression is: ;in, Represents the frequency domain data converted from the blood glucose level time series, represents a frequency variable, represents the low-frequency range, represents the proportion of low-frequency energy, and the proportion of low-frequency energy is recorded as the characteristic value of sugar regulation;

[0057] It should be noted that the sugar regulation characteristic value reflects the abnormal degree of the patient's body's sugar regulation efficiency at night, and the larger the sugar regulation characteristic value, the lower the abnormal degree of the patient's body's sugar regulation efficiency at night.

[0058] In S3, a comprehensive analysis is performed on the real-time changes of the first specific physiological parameter and the second specific physiological parameter to construct a dynamic model of the patient's blood glucose state and predict the patient's current blood glucose fluctuation range, specifically including:

[0059] The patient's chromium concentration change characteristic value and sugar regulation characteristic value are constructed into a comprehensive characteristic vector as the input of the dynamic model of the patient's blood sugar status. The dynamic model of the patient's blood sugar status is trained with minimizing the error between the predicted patient's current blood sugar fluctuation range and the actual patient's current blood sugar fluctuation range as the training goal. Based on the trained dynamic model of the patient's blood sugar status, the patient's current blood sugar fluctuation range is output. The dynamic model of the patient's blood sugar status is a random forest model.

[0060] The predicted patient's current blood sugar fluctuation range specifically includes:

[0061] Use historical data to train the random forest model. Divide the dataset into a training set and a test set. Use the training set to train the random forest model. Improve prediction accuracy by constructing multiple decision trees and aggregating their prediction results. Evaluate model performance on the test set. Adjust model parameters to optimize model performance. Collect a comprehensive feature vector constructed from the chromium concentration change characteristic values and sugar regulation characteristic values of the patient to be predicted. Input the comprehensive feature vector into the trained random forest model. Based on the model output results, obtain the predicted value of the patient's current blood sugar fluctuation range.

[0062] In S4, the intravenous insulin infusion rate is automatically adjusted based on the patient's predicted current blood sugar fluctuation range to ensure that the patient's blood sugar level is maintained within the predetermined target range, including:

[0063] A dynamic regulation system based on a model predictive control algorithm is constructed. The system obtains the patient's chromium concentration change characteristic value, sugar regulation characteristic value and current blood glucose level in real time as input, uses a pre-trained random forest model to predict the patient's short-term blood glucose fluctuation range, and combines it with a blood glucose kinetic model to predict the blood glucose change trend within a preset time period in the future. Based on the prediction results and the set target blood glucose range, an optimization algorithm is used to calculate the optimal insulin infusion rate adjustment strategy that enables the future blood glucose level to reach the target range while satisfying all constraints. The constraints include: a safe range for the insulin infusion rate and a safety threshold for the blood glucose level. At each time step, only the current insulin infusion rate adjustment amount in the calculated strategy is executed, and this process is continuously repeated to achieve precise dynamic regulation of the patient's blood glucose level, thereby ensuring that the patient's blood glucose is maintained within a predetermined safe range.

[0064] The method of combining the blood glucose dynamics model to predict the blood glucose change trend within a preset time period in the future specifically includes:

[0065] A blood glucose dynamics model is established and applied to predict the trend of blood glucose changes within a preset time period in the future. The model is based on patient-specific parameters such as the characteristic value of chromium concentration changes, the characteristic value of sugar regulation and the current blood glucose level, and is dynamically adjusted in combination with the real-time updated insulin infusion rate; by inputting these parameters into the blood glucose dynamics model, the model equation is solved using a numerical simulation method to accurately predict the changes in blood glucose levels at various time points in the future under different insulin infusion strategies. This process particularly emphasizes the importance of individualized parameters, so that the model can more accurately reflect the patient's unique physiological responses, thereby achieving precise regulation of the insulin infusion rate, effectively avoiding the occurrence of hypoglycemia and hyperglycemia events, and ensuring that blood glucose levels are stably maintained within the target range. In addition, the data-driven model optimization mechanism adopted by the present invention can continuously self-adjust and optimize model parameters according to new patient data, further improving the prediction accuracy and treatment effect, and providing a safer and more effective management solution for diabetic patients.

[0066] See also Figure 2 As shown, the patient multi-parameter coupled intravenous insulin intelligent control system includes:

[0067] A data acquisition module, the data acquisition module is used to collect the concentration level of chromium in the patient's blood, the proportion of deep sleep time in the patient's sleep cycle, and the blood sugar level;

[0068] an energy conversion efficiency evaluation module, wherein the energy conversion efficiency evaluation module obtains and monitors the concentration level of chromium in the patient's blood as a first specific physiological parameter, and evaluates the energy conversion efficiency in the patient's body according to a change trend of the chromium concentration;

[0069] a nighttime glucose regulation effectiveness assessment module, which obtains and monitors the proportion of deep sleep time during a patient's sleep cycle as a second specific physiological parameter, and assesses the patient's nighttime glucose regulation effectiveness based on the body's blood glucose regulation mechanism during deep sleep and the proportion of deep sleep time;

[0070] a patient blood glucose fluctuation range prediction module, which constructs a dynamic model of the patient's blood glucose state by comprehensively analyzing the real-time changes of the first specific physiological parameter and the second specific physiological parameter, and predicts the patient's current blood glucose fluctuation range;

[0071] The insulin infusion rate automatic adjustment module automatically adjusts the intravenous insulin infusion rate based on the predicted patient's current blood sugar fluctuation range to ensure that the patient's blood sugar level is maintained within a predetermined target range.

[0072] The present invention operates as follows: The data acquisition module acquires and monitors the patient's blood chromium concentration and the proportion of deep sleep duration during the sleep cycle in real time as the first and second specific physiological parameters. The energy conversion efficiency assessment module evaluates the body's energy conversion efficiency based on the changing trend of chromium concentration. The nighttime glucose regulation effectiveness assessment module utilizes the body's unique blood glucose regulation mechanism during deep sleep to assess the body's nighttime glucose regulation effectiveness based on the proportion of deep sleep duration. Next, the patient's blood glucose fluctuation range prediction module constructs a comprehensive feature vector based on the real-time changes of these two specific physiological parameters and inputs it into a pre-trained random forest model to construct a dynamic model of the patient's blood glucose status and predict the current blood glucose fluctuation range. Finally, the automatic insulin infusion rate adjustment module utilizes a dynamic adjustment system based on a model predictive control algorithm, combined with a blood glucose dynamics model to predict blood glucose trends within a preset time period. An optimization algorithm is then used to calculate the optimal insulin infusion rate adjustment strategy, executing only the insulin infusion rate adjustment for the current time step. This process is continuously repeated to achieve precise dynamic regulation of the patient's blood glucose level. The system places particular emphasis on the importance of individualized parameters, enabling the model to more accurately reflect the patient's unique physiological responses, effectively preventing hypoglycemia and hyperglycemia, providing more personalized and precise treatment options for diabetic patients, and significantly improving the safety and effectiveness of diabetes management. Furthermore, the system possesses self-learning and optimization capabilities, continuously adjusting and optimizing model parameters based on new patient data to further enhance long-term prediction accuracy and individual adaptability.

[0073] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0074] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0075] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0076] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0077] The above is a detailed description of an embodiment of the present invention. However, the content described is only a preferred embodiment of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A multi-parameter coupled intravenous insulin intelligent control method for patients, characterized by: The following steps are involved: S1: Obtain and monitor the concentration level of chromium in the patient's blood as a first specific physiological parameter, and evaluate the energy conversion efficiency in the patient's body based on the changing trend of the chromium concentration; S2: Obtain and monitor the proportion of deep sleep time during the patient's sleep cycle as a second specific physiological parameter. Based on the body's blood sugar regulation mechanism during deep sleep, the patient's nighttime body sugar regulation efficiency is evaluated according to the proportion of deep sleep time. S3: Comprehensively analyzing the real-time changes of the first specific physiological parameter and the second specific physiological parameter, constructing a dynamic model of the patient's blood glucose state, and predicting the patient's current blood glucose fluctuation range; S4: Automatically adjust the intravenous insulin infusion rate based on the predicted patient's current blood sugar fluctuation range to ensure that the patient's blood sugar level is maintained within the predetermined target range.

2. The method for intelligent intravenous insulin control based on multi-parameter coupling of patients according to claim 1, characterized in that: The method of evaluating the energy conversion efficiency in the patient's body based on the changing trend of chromium concentration specifically includes: During the monitoring period, the concentration level of chromium in the patient's blood is obtained in real time according to the time series. The characteristic value of the chromium concentration change is calculated based on the change amplitude of the chromium concentration in the patient's blood, and it is determined whether the characteristic value of the chromium concentration change is greater than or equal to the preset threshold. If so, the energy conversion efficiency in the patient's body is abnormal; if not, the energy conversion efficiency in the patient's body is normal.

3. The method for intelligent intravenous insulin control based on multi-parameter coupling of patients according to claim 2, characterized in that: The process of obtaining the characteristic value of the chromium element concentration change is as follows: For the chromium concentration data obtained in real time during the monitoring period, Morlet wavelet was used to convert the chromium concentration time series data into wavelet coefficients at different scales to extract the local characteristics of the chromium concentration changing with time. Calculate each scale The energy density under the condition of chromium is determined, the key scale range is determined according to the actual data analysis results, and the average energy density within the selected key scale range is calculated as the characteristic value of the chromium element concentration change.

4. The method for intelligent intravenous insulin control based on multi-parameter coupling of patients according to claim 1, characterized in that: The assessment of the patient's nighttime body sugar regulation efficiency specifically includes: During the monitoring period, the proportion of deep sleep time in the patient's sleep cycle is obtained in real time. Based on the blood sugar regulation mechanism of the patient's body during deep sleep, the sugar regulation characteristic value is calculated to determine whether the patient's sugar regulation characteristic value is greater than or equal to the preset threshold. If so, the patient's body sugar regulation efficiency at night is normal; if not, the patient's body sugar regulation efficiency at night is abnormal.

5. The method for intelligent intravenous insulin control based on multi-parameter coupling of patients according to claim 4, characterized in that: The process of obtaining the sugar regulation characteristic value is as follows: During the monitoring period, the patient's sleep state time series and corresponding blood glucose level time series are obtained in real time, wherein the sleep state includes wakefulness, light sleep and deep sleep; Based on the sleep state, the blood glucose level data corresponding to the deep sleep stage is filtered out; the fast Fourier transform is applied to the blood glucose level time series during deep sleep, and the blood glucose level time series is converted into frequency domain data, so as to analyze the frequency characteristics of blood glucose level changes over time; a low-frequency interval is preset, and the proportion of energy in the low-frequency interval to the total energy is calculated, and the low-frequency energy proportion is recorded as the sugar regulation characteristic value.

6. The method for intelligent intravenous insulin control based on multi-parameter coupling of patients according to claim 1, characterized in that: The construction of a dynamic model of the patient's blood glucose status specifically includes: The patient's chromium concentration change characteristic value and sugar regulation characteristic value are constructed into a comprehensive characteristic vector as the input of the dynamic model of the patient's blood sugar status. The dynamic model of the patient's blood sugar status is trained with minimizing the error between the predicted patient's current blood sugar fluctuation range and the actual patient's current blood sugar fluctuation range as the training goal. Based on the trained dynamic model of the patient's blood sugar status, the patient's current blood sugar fluctuation range is output. The dynamic model of the patient's blood sugar status is a random forest model.

7. The method for intelligent intravenous insulin control based on multi-parameter coupling of patients according to claim 1, characterized in that: The predicted patient's current blood sugar fluctuation range specifically includes: The random forest model was trained using historical data. The dataset was divided into a training set and a test set. The training set was used to train the random forest model. Multiple decision trees were constructed and their prediction results were summarized. The model performance was evaluated on the test set, and the model parameters were adjusted. A comprehensive feature vector was constructed by collecting the characteristic values of the chromium concentration changes and the sugar regulation characteristic values of the patients to be predicted. The comprehensive feature vector was input into the trained random forest model. Based on the model output results, the predicted value of the patient's current blood sugar fluctuation range was obtained.

8. The method for intelligent intravenous insulin control based on multi-parameter coupling of patients according to claim 1, characterized in that: The automatic adjustment of the intravenous insulin infusion rate specifically includes: A dynamic regulation system based on a model predictive control algorithm is constructed. The system obtains the patient's chromium concentration change characteristic value, sugar regulation characteristic value, and current blood glucose level in real time as input, uses a pre-trained random forest model to predict the patient's short-term blood glucose fluctuation range, and combines it with a blood glucose kinetic model to predict the blood glucose change trend within a preset time period in the future. Based on the prediction results and the set target blood glucose range, the optimal insulin infusion rate adjustment strategy is calculated to enable the future blood glucose level to reach the target range while satisfying all constraints. The constraints include: a safe range for the insulin infusion rate and a safety threshold for the blood glucose level. At each time step, only the insulin infusion rate adjustment amount at the current moment in the calculated strategy is executed.

9. The method for intelligent intravenous insulin control based on multi-parameter coupling of patients according to claim 8, characterized in that: The method of combining the blood glucose dynamics model to predict the blood glucose change trend within a preset time period in the future specifically includes: A blood glucose kinetics model was established and applied to predict blood glucose change trends within a preset time period in the future. The model was based on the patient's chromium concentration change characteristic values, sugar regulation characteristic values, and current blood glucose levels, and was dynamically adjusted in combination with the real-time updated insulin infusion rate. By inputting the patient's chromium concentration change characteristic values, sugar regulation characteristic values, and current blood glucose levels into the blood glucose kinetics model, the model equation was solved using a numerical simulation method to accurately predict changes in blood glucose levels at various time points in the future under different insulin infusion strategies.

10. Patient multi-parameter coupled intravenous insulin intelligent control system, characterized by: The method for intelligent intravenous insulin control with multi-parameter coupling for a patient as claimed in any one of claims 1 to 9 comprises: A data acquisition module, the data acquisition module is used to collect the concentration level of chromium in the patient's blood, the proportion of deep sleep time in the patient's sleep cycle, and the blood sugar level; an energy conversion efficiency evaluation module, wherein the energy conversion efficiency evaluation module obtains and monitors the concentration level of chromium in the patient's blood as a first specific physiological parameter, and evaluates the energy conversion efficiency in the patient's body according to a change trend of the chromium concentration; a nighttime glucose regulation effectiveness assessment module, which obtains and monitors the proportion of deep sleep time during a patient's sleep cycle as a second specific physiological parameter, and assesses the patient's nighttime glucose regulation effectiveness based on the body's blood glucose regulation mechanism during deep sleep and the proportion of deep sleep time; a patient blood glucose fluctuation range prediction module, which constructs a dynamic model of the patient's blood glucose state by comprehensively analyzing the real-time changes of the first specific physiological parameter and the second specific physiological parameter, and predicts the patient's current blood glucose fluctuation range; The insulin infusion rate automatic adjustment module automatically adjusts the intravenous insulin infusion rate based on the predicted patient's current blood sugar fluctuation range to ensure that the patient's blood sugar level is maintained within a predetermined target range.

Images