Blood glucose real-time monitoring and hypoglycemia early warning system in hemodialysis

Abstract

The invention relates to the technical field of blood glucose monitoring, and discloses a blood glucose real-time monitoring and hypoglycemia early warning system in hemodialysis, which comprises dialysis blood real-time data: acquiring the change of blood glucose of a dialysis patient in real time through a blood glucose monitoring device of a blood glucose monitor connected with a hemodialysis machine; patient body data: acquiring basic data of a patient body, and acquiring data of weight, age, height and related blood glucose; acquiring and recording medicine data and disease history of the patient, and acquiring and recording past medicine and disease history of the patient; the patient data transceiving device is used for compressing and transmitting the acquired data through a data receiving device outside the hemodialysis machine; and dialysis data is stored, processed and uploaded information is stored, attention fatigue caused by real-time monitoring of blood glucose of a patient by medical staff is reduced, and dialysis blood glucose prediction accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of blood sugar monitoring, and in particular to a real-time blood sugar monitoring and hypoglycemia early warning system during hemodialysis. Background Art

[0002] The kidney is one of the important organs in our body. It is responsible for filtering waste and excess fluid from the blood and regulating electrolyte and acid-base balance in the body. However, when kidney function is impaired, these wastes and fluids will accumulate in the body. At this time, artificial dialysis devices are needed to remove waste, salts and excess water from the blood to help maintain balance in the body and ensure normal functioning of the body.

[0003] During dialysis, especially during hemodialysis, some patients may experience hypoglycemia. Changes in blood sugar during dialysis may be due to changes in waste removal and metabolic rate. Increased blood sugar during dialysis may be related to spontaneous eating during dialysis, stress response in the body, release of adrenaline, or use of glucose in the dialysate. Patients with advanced chronic kidney disease on dialysis, especially those who receive long-term hemodialysis or peritoneal dialysis, often experience hypoglycemia. The definition of hypoglycemia is that the blood sugar concentration is lower than the normal level, among which hypoglycemia is usually defined as a blood sugar concentration lower than 2.8mmol / L. The incidence of hypoglycemia is relatively high in dialysis patients, including dialysis treatment itself can cause hypoglycemia, the use of insulin or oral hypoglycemic drugs and malnutrition or loss of appetite, etc. Although this situation is relatively rare, since dialysis patients are originally accompanied by multiple symptoms, some mild hypoglycemia symptoms may be ignored or confused with other dialysis-related symptoms, which makes hypoglycemia difficult to detect and deal with in time. Therefore, a real-time monitoring system for blood sugar and hypoglycemia warning during hemodialysis is proposed. Summary of the invention

[0004] The purpose of the present invention is to reduce the attention fatigue caused by medical staff's real-time monitoring of patients' blood sugar, improve the accuracy of blood sugar monitoring during dialysis, and reduce or avoid the impact of hypoglycemia on patients during dialysis. A real-time blood sugar monitoring and hypoglycemia early warning system during hemodialysis is proposed.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] Real-time data of dialysis blood, through the blood glucose monitoring device connected to the hemodialysis machine, the blood glucose changes of dialysis patients can be obtained in real time;

[0007] The patient's physical data is obtained by obtaining the patient's basic physical data, including weight, age, height and blood sugar data;

[0008] Patient medication data and medical history: learn and record the patient's past medication and medical history;

[0009] The patient data transceiver compresses and transmits the acquired data through a data receiving device outside the hemodialysis machine;

[0010] Dialysis data storage, the process of storing processed and uploaded information and saving data in computer systems or other electronic devices;

[0011] The data processing module processes and transmits various data to monitor and process the patient's blood sugar in real time, and trains the time series model for the acquired information;

[0012] Blood sugar prediction data output: predict the patient's blood sugar during dialysis through a time series model and process the predicted data;

[0013] The real-time dialysis blood data is monitored in real time by a real-time blood glucose monitoring device externally connected to the dialysis machine, and the processed information is transmitted to the patient data transceiver. The patient's body data is transmitted to the patient data transceiver after the patient's basic body data is acquired. The patient's medication data and medical history are transmitted to the patient's data transceiver after the patient's past medical history and medication history are acquired. The patient data transceiver receives and decodes the signal through a real-time receiver, and then converts it into understandable data or information through a wirelessly transmitted signal and transmits it to a data processing module. The data processing module processes the incoming data and transmits the data to the blood glucose prediction data output. The data processing module packages and compresses the incoming data and transmits the data to the dialysis data storage for storage.

[0014] The above technical solution further includes:

[0015] The real-time data of dialyzed blood can provide more accurate treatment and monitoring for patients through the application of an external real-time blood glucose monitor on the dialysis machine, and the data is transmitted to the patient data transceiver. The patient's body data is obtained by collecting the patient's body temperature, weight, height and respiratory rate data, and the data is transmitted to the patient's data transceiver. The patient's medication data and medical history are obtained by understanding the patient's past medication history and the patient's medical history, and then the data is transmitted to the patient's data transceiver.

[0016] The patient data transceiver preprocesses the data, and packages and compresses the preprocessed data and transmits it to the patient data transceiver. After collecting and observing the patient data, the patient data transceiver transmits the data to data stabilization, and the data stabilization performs a differential operation on the data to eliminate or weaken the influence of the trend and seasonality of the data, and transmits the data to an autoregressive moving average model, and the autoregressive moving average model observes the linear relationship between the current observation value and the random error term, and transmits the data to parameter estimation, and the parameter estimation uses the maximum likelihood estimation method to estimate the parameters in the time series model, and transmits the data to model fitting and diagnosis, and the model fitting and diagnosis uses the estimated parameters to fit the time series model and diagnose the model, and transmits the data to prediction and evaluation, and the prediction and evaluation uses the fitted time series model to predict and determine the accuracy and stability of the model, and transmits the data to model adjustment and change, and the model adjustment and change adjusts and improves the model according to the prediction results and evaluation indicators.

[0017] The autoregressive moving average model is a time series analysis model used to describe trends and cyclical changes in time series data. The ARMA model combines the characteristics of the autoregressive model and the moving average model, where the autoregressive model represents the linear relationship between the current observation value and the past observation value, while the moving average model represents the linear relationship between the current observation value and the random error term. The ARMA model is usually expressed as a mathematical equation:

[0018] [X t =c+φ1X t-1 +φ2X t-2 +…+φ p X t-p +θ1ε t-1 +θ2ε t-2 +…+θ q ε t-q +ε t ], where: (X_t) represents the observed value of the time series at time (t); (c) is a constant term; (φ1, φ2,…, φ p ) are the coefficients of the autoregressive model, which represent the linear relationship between the current observation and the past (p) observations; (θ1,θ2,…,θ q ) is the coefficient of the moving average model, which represents the linear relationship between the current observation and the past (q) random error terms, (ε t ,ε t-1 ,...,ε t-q ) is a white noise error term. The estimation of the parameters of the ARMA model is usually performed using maximum likelihood estimation or other statistical methods. After the parameters are estimated, the ARMA model can be used to predict the blood glucose values ​​of the future time series.

[0019] The model fitting and diagnosis are commonly used concepts in statistics and machine learning, which are used to evaluate the degree of fit of the established model to the data and the effectiveness of the model. They involve testing whether the distribution of the residuals meets the assumptions, whether the model has multicollinearity or heteroscedasticity, and other issues. The fitting of the linear regression model can be expressed as:

[0020] [Y=β0+β1X1+β2X2+...+β p X p +ε], where (Y) is the dependent variable, (X1,X2,...,X p ) are independent variables, (β0,β1,β2,...,β p ) is the model coefficient, (ε) is the error term. When we want to predict a person's blood sugar (Y) by height (X_1) and weight (X_2), we can use a linear regression model to fit it and estimate the coefficients by the least squares method to obtain the best fitting linear relationship.

[0021] The blood glucose prediction data output predicts the patient's blood glucose changes through a time series model, and compares the predicted blood glucose data with the safe blood glucose range. When it is lower or higher than the safe blood glucose range, the patient's blood glucose prediction information alarm is triggered. The patient's blood glucose prediction information alarm reminds and warns medical staff to intervene in the patient's blood glucose in time through the patient's blood glucose alarm.

[0022] The blood sugar prediction data is output by comparing the blood sugar prediction data with the normal range, and the result of calculating the blood sugar concentration in the nutrient solution that needs to be regulated is output through the data processing module, and the result is transmitted to the medical staff for confirmation and regulation. After the medical staff confirms the regulation, the adjustment instruction is transmitted to the automatic regulation of glucose concentration. After receiving the instruction, the automatic regulation of glucose concentration regulates blood sugar. After confirming the regulation, the medical staff can manually modify the value after receiving the information, and transmit the instruction to the manual regulation of glucose concentration. The manual regulation of glucose concentration executes the blood sugar instruction modified by the medical staff.

[0023] The present invention has the following beneficial effects:

[0024] 1. In the present invention, the real-time blood glucose monitoring and hypoglycemia early warning system during hemodialysis is proposed. The real-time data of dialysis blood, the patient's physical data, the patient's drug data and the key data in the medical history are learned through a time series model, and then deployed through multiple trained models to predict the patient's blood glucose in real time during the dialysis process.

[0025] 2. In the present invention, the real-time blood sugar monitoring and hypoglycemia early warning system during hemodialysis is proposed. By comparing the predicted blood sugar with the normal blood sugar range, the patient's blood sugar prediction information alarm is issued for excessively high and low blood sugar. At the same time, the glucose concentration is automatically adjusted to facilitate timely and rapid regulation of the patient's blood sugar during dialysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a system block diagram of a real-time blood sugar monitoring and hypoglycemia early warning system during hemodialysis proposed by the present invention;

[0027] Figure 2 This is a partial block diagram of a real-time blood sugar monitoring and hypoglycemia early warning system during hemodialysis proposed by the present invention.

[0028] In the figure: 1. Real-time data of dialysis blood; 2. Patient physical data; 3. Patient drug data and medical history; 4. Patient data transceiver; 5. Dialysis data storage; 6. Data processing module; 7. Blood glucose prediction data output; 8. Patient blood glucose prediction information alarm; 9. Medical staff confirms and controls; 10. Manual control of glucose concentration; 11. Automatic control of glucose concentration; 12. Patient data collection and observation; 13. Data stabilization; 14. Autoregressive moving average model; 15. Parameter estimation; 16. Model fitting and diagnosis; 17. Prediction and evaluation; 18. Model adjustment and change. DETAILED DESCRIPTION

[0029] Embodiment 1

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0031] like Figure 1-2As shown, a real-time blood glucose monitoring and hypoglycemia early warning system in hemodialysis proposed by the present invention comprises: real-time dialysis blood data 1, which is used to obtain the changes of blood glucose of dialysis patients in real time through the blood glucose monitoring device connected to the hemodialysis machine; patient body data 2, which is used to obtain the basic data of the patient's body, weight, age, height and data related to blood glucose; patient drug data and medical history 3, which is used to learn and record the patient's past drug and medical history; patient data transceiver 4, which is used to compress and transmit the acquired data through the data receiving device outside the hemodialysis machine; dialysis data storage 5, which is used to store the processed and uploaded information and save the data in the computer system or other electronic devices; data processing module 6, which is used to grasp and process the patient's blood glucose in real time through the processing and transmission of various data, and to train the time series model of the acquired information; blood glucose prediction data output 7;

[0032] The blood sugar of the patient during dialysis is predicted through the time series model, and the predicted data is processed; the real-time dialysis blood data 1 is monitored in real time by the real-time blood sugar monitoring device connected to the dialysis machine, and the processed information is transmitted to the patient data transceiver 4; the patient's body data 2 is transmitted to the patient data transceiver 4 after the basic body data of the patient is obtained; the patient's drug data and medical history 3 is transmitted to the patient data transceiver 4 after the past medical history and drug history of the patient are obtained; the patient data transceiver 4 receives and decodes the signal through the real-time receiver, and then transmits the signal through wireless transmission. Convert it into understandable data or information and transmit it to the data processing module 6. After processing the incoming data, the data processing module 6 transmits the data to the blood glucose prediction data output 7. After packaging and compressing the incoming data, the data processing module 6 transmits the data to the dialysis data storage 5 for storage. The dialysis blood real-time data 1 can provide more accurate treatment and monitoring for patients through the application of an external real-time blood glucose monitor on the dialysis machine, and transmit the data to the patient data transceiver 4. The patient's body data 2 is obtained by obtaining the patient's body temperature, weight, height and respiratory rate data, and the data is transmitted to the patient data transceiver 4.

[0033] Patient drug data and disease history 3 is obtained through understanding of the patient's past drug history and the patient's disease history, and then the data is transmitted to the patient data transceiver 4. The data processing module 6 compresses the processed information and transmits the data to the dialysis data storage 5, and manages and accesses a large amount of information in a safe and efficient manner, and safely and effectively stores and accesses important data at any time. The blood glucose prediction data output 7 predicts the patient's blood glucose changes through a time series model, and compares the predicted blood glucose data with the safe blood glucose range. When it is lower or higher, the patient's blood glucose prediction information alarm 8 is triggered. The patient's blood glucose prediction information alarm 8 reminds and warns the medical staff to intervene in the patient's blood glucose in time through the patient's blood glucose alarm. The blood glucose prediction data output 7 compares the blood glucose prediction data with the normal range;

[0034] The result of calculating the blood sugar concentration in the nutrient solution that needs to be regulated is output through the data processing module 6, and the result is transmitted to the medical staff for confirmation and regulation 9. After the medical staff confirms the regulation 9, the adjustment instruction is transmitted to the automatic glucose concentration control 11. After receiving the instruction, the automatic glucose concentration control 11 regulates the blood sugar. After receiving the information, the medical staff confirms the regulation 9 and can manually modify the value and transmit the instruction to the manual glucose concentration control 10. The manual glucose concentration control 10 executes the blood sugar instruction modified by the medical staff.

[0035] The dialysis blood real-time data 1 is used to monitor the patient's blood sugar in real time through the real-time blood sugar monitoring device connected to the dialysis machine, and the processed information is transmitted to the patient data transceiver 4. The patient's body data 2 is transmitted to the patient's data transceiver 4 after the patient's basic body data is learned. The patient's drug data and medical history 3 are transmitted to the patient's data transceiver 4 after the patient's past medical history and drug history are obtained. The patient's data transceiver 4 receives and decodes the signal through a real-time receiver, and then converts it into understandable data or information through a wirelessly transmitted signal and transmits it to the data processing module 6. After the data processing module 6 processes the incoming data, the signal real-time receiver is used to receive the physiological parameter data transmitted by the medical device and transmit the data to the blood sugar prediction data output 7. The data processing module 6 packages and compresses the incoming data and transmits the data to the dialysis data storage 5 for storage, which can manage and access a large amount of medical information in a safer and more efficient manner.

[0036] Embodiment 2

[0037] like Figure 1-2As shown, in the embodiment of the present invention, the patient data transceiver 4 pre-processes the data, and packs and compresses the pre-processed data to transmit to the patient data transceiver 4. After collecting and observing the patient data 12, the patient data transceiver 4 transmits the data to the data stabilization 13. The data stabilization 13 performs a differential operation on the data to eliminate or weaken the influence of the trend and seasonality of the data, and transmits the data to the autoregressive moving average model 14. The autoregressive moving average model 14 observes the linear relationship between the current observation value and the random error term, and transmits the data to the parameter estimation 15. The parameter estimation 15 uses the maximum likelihood estimation method to estimate the parameters in the time series model, and transmits the data to the model fitting and diagnosis 16. The model fitting and diagnosis 16 uses the estimated parameters;

[0038] Fit the time series model, diagnose the model, and pass the data to prediction and evaluation 17. Prediction and evaluation 17 predicts by using the fitted time series model, and determines the accuracy and stability of the model, and passes the data to model adjustment and change 18. Model adjustment and change 18 adjusts and improves the model according to the prediction results and evaluation indicators. The autoregressive moving average model 14 is a model of time series analysis, which is used to describe the trend and cyclical changes in time series data. The ARMA model combines the characteristics of the autoregressive model and the moving average model, where the autoregressive model represents the linear relationship between the current observation value and the past observation value;

[0039] The moving average model represents the linear relationship between the current observation and the random error term. The ARMA model is usually expressed as a mathematical equation:

[0040] [X t =c+φ1X t-1 +φ2X t-2 +...+φ p X t-p +θ1ε t-1 +θ2ε t-2 +...+θ q ε t-q +ε t ], where: (X_t) represents the observed value of the time series at time (t); (c) is a constant term; (φ1, φ2, ..., φ p ) are the coefficients of the autoregressive model, which represent the linear relationship between the current observation and the past (p) observations; (θ1,θ2,...,θ q ) is the coefficient of the moving average model, which represents the linear relationship between the current observation and the past (q) random error terms, (ε t ,ε t-1 ,...,ε t-q) is a white noise error term. The estimation of the parameters of the ARMA model is usually performed using maximum likelihood estimation or other statistical methods;

[0041] After the parameters are estimated, the ARMA model can be used to predict the blood glucose values ​​of the future time series. Model fitting and diagnosis16 are commonly used concepts in statistics and machine learning. They are used to evaluate the degree of fit of the established model to the data and the effectiveness of the model. They involve testing whether the distribution of the residuals meets the assumptions, whether the model has multicollinearity or heteroscedasticity, and other issues. The fitting of the linear regression model can be expressed as:

[0042] [Y=β0+β1X1+β2X2+...+β p X p +ε], where (Y) is the dependent variable, (X1,X2,...,X p ) are independent variables, (β0,β1,β2,...,β p ) is the model coefficient, (ε) is the error term. When we want to predict a person's blood sugar (Y) by height (X_1) and weight (X_2), we can use a linear regression model to fit it and estimate the coefficients by the least squares method to obtain the best fitting linear relationship.

[0043] Patient data collection and observation 12 passes the data to data stabilization 13. Data stabilization 13 performs a difference operation on the data to eliminate or weaken the influence of trends and seasonality of the data, and passes the data to the autoregressive moving average model 14. The autoregressive moving average model 14 observes the linear relationship between the current observation value and the random error term. The ARMA model combines the characteristics of the autoregressive model and the moving average model, and passes the data to parameter estimation 15. Parameter estimation 15 uses the maximum likelihood estimation method to estimate the parameters in the time series model, and passes the data to model fitting and diagnosis 16. Model fitting and diagnosis 16 uses the estimated parameters to evaluate the degree of fit of the established model to the data and the effectiveness of the model, involving testing whether the distribution of the residuals meets the assumptions, whether the model has multicollinearity or heteroscedasticity, etc., fitting of the linear regression model, and passing the data to prediction and evaluation 17. Prediction and evaluation 17 uses the fitted time series model to make predictions and determine the accuracy and stability of the model, and passes the data to model adjustment and change 18. Model adjustment and change 18 adjusts and improves the model according to the prediction results and evaluation indicators.

[0044] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A real-time blood sugar monitoring and hypoglycemia early warning system during hemodialysis, characterized in that: include: Real-time data of dialysis blood (1), through the blood glucose monitoring device connected to the hemodialysis machine, the blood glucose changes of the dialysis patient are obtained in real time; Patient physical data (2), by obtaining the patient's basic physical data, weight, age, height and blood sugar data; Patient medication data and medical history (3): Obtain and record the patient’s past medication and medical history; The patient data transceiver (4) compresses and transmits the acquired data through a data receiving device outside the hemodialysis machine; Dialysis data storage (5), the process of storing processed and uploaded information and saving data in a computer system or other electronic device; The data processing module (6) processes and transmits various data to grasp and process the patient's blood sugar in real time, and trains the time series model for the acquired information; Blood glucose prediction data output (7), blood glucose prediction of patients during dialysis is performed using a time series model, and the predicted data is processed; The dialysis blood real-time data (1) monitors the patient's blood sugar in real time through a real-time blood sugar monitoring device external to the dialysis machine, and transmits the processed information to the patient data transceiver (4). The patient's body data (2) is obtained by obtaining the patient's basic body data and then transmits the information to the patient's data transceiver (4). The patient's medication data and medical history (3) is obtained by obtaining the patient's past medical history and medication history and then transmits the information to the patient's data transceiver (4). The patient's data transceiver (4) receives and decodes the signal through a real-time receiver and then converts it into understandable data or information through a wirelessly transmitted signal and transmits it to a data processing module (6). The data processing module (6) processes the incoming data and then transmits the data to a blood sugar prediction data output (7). The data processing module (6) packages and compresses the incoming data and then transmits the data to the dialysis data storage (5) for storage.

2. A real-time blood sugar monitoring and hypoglycemia early warning system during hemodialysis according to claim 1, characterized in that: The real-time dialysis blood data (1) can provide more accurate treatment and monitoring for patients by using an external real-time blood glucose monitor on the dialysis machine, and transmit the data to the patient data transceiver (4). The patient's body data (2) is obtained by collecting the patient's BMI, body temperature, blood pressure and respiratory rate data, and the data is transmitted to the patient's data transceiver (4). The patient's medication data and medical history (3) is obtained by understanding the patient's past medication history and the patient's medical history, and then the data is transmitted to the patient's data transceiver (4).

3. A real-time blood sugar monitoring and hypoglycemia early warning system during hemodialysis according to claim 1, characterized in that: The patient data transceiver (4) pre-processes the data, and packs and compresses the pre-processed data and transmits it to the patient data transceiver (4). The patient data transceiver (4) collects and observes the patient data (12), and then transmits the data to the data stabilization (13). The data stabilization (13) performs a differential operation on the data to eliminate or weaken the influence of the trend and seasonality of the data, and transmits the data to the autoregressive moving average model (14). The autoregressive moving average model (14) observes the linear relationship between the current observation value and the random error term, and transmits the data to the parameter estimation (15). ), the parameter estimation (15) estimates the parameters in the time series model using the maximum likelihood estimation method, and passes the data to the model fitting and diagnosis (16), the model fitting and diagnosis (16) uses the estimated parameters to fit the time series model, and diagnoses the model, and passes the data to the prediction and evaluation (17), the prediction and evaluation (17) predicts by using the fitted time series model and determines the accuracy and stability of the model, and passes the data to the model adjustment and change (18), the model adjustment and change (18) adjusts and improves the model according to the prediction results and evaluation indicators.

4. A real-time blood sugar monitoring and hypoglycemia early warning system during hemodialysis according to claim 3, characterized in that: The autoregressive moving average model (14) is a model of time series analysis, which is used to describe the trend and periodic changes in time series data. The ARMA model combines the characteristics of the autoregressive model and the moving average model, wherein the autoregressive model represents the linear relationship between the current observation value and the past observation value, while the moving average model represents the linear relationship between the current observation value and the random error term. The ARMA model is usually expressed as a mathematical equation: [X t =c+φ1X t-1 +φ2X t-2 +…+φ p X t-p +θ1ε t-1 +θ2ε t-2 +…+θ q ε t-q +ε t ], where: (X_t) represents the observed value of the time series at time (t); (c) is a constant term; (φ1, φ2,…, φ p ) are the coefficients of the autoregressive model, which represent the linear relationship between the current observation and the past (p) observations; (θ1,θ2,...,θ q ) is the coefficient of the moving average model, which represents the linear relationship between the current observation and the past (q) random error terms, (ε t ,ε t-1 ,...,ε t-q ) is a white noise error term. The estimation of the parameters of the ARMA model is usually performed using maximum likelihood estimation or other statistical methods. After the parameters are estimated, the ARMA model can be used to predict the blood glucose values ​​of the future time series.

5. A real-time blood sugar monitoring and hypoglycemia early warning system during hemodialysis according to claim 3, characterized in that: The model fitting and diagnosis (16) is a commonly used concept in statistics and machine learning, which is used to evaluate the degree of fit of the established model to the data and the effectiveness of the model. It involves testing whether the distribution of residuals meets the assumptions, whether the model has multicollinearity or heteroscedasticity, etc. The fitting of the linear regression model can be expressed as: [Y=β0+β1X1+β2X2+...+β p X p +ε], where (Y) is the dependent variable, (X1,X2,...,X p ) are independent variables, (β0,β1,β2,...,β p ) is the model coefficient, (ε) is the error term. When we want to predict a person's blood sugar (Y) by height (X_1) and weight (X_2), we can use a linear regression model to fit it and estimate the coefficients by the least squares method to obtain the best fitting linear relationship.

6. A real-time blood sugar monitoring and hypoglycemia early warning system during hemodialysis according to claim 1, characterized in that: The data processing module (6) compresses the processed information and transmits the data to the dialysis data storage (5), and manages and accesses a large amount of information in a safe and efficient manner, and safely and effectively stores and accesses important data at any time.

7. A real-time blood sugar monitoring and hypoglycemia early warning system during hemodialysis according to claim 1, characterized in that: The blood glucose prediction data output (7) predicts the patient's blood glucose changes through a time series model, and compares the predicted blood glucose data with the safe blood glucose range. When it is lower or higher than the safe blood glucose range, the patient's blood glucose prediction information alarm (8) is triggered. The patient's blood glucose prediction information alarm (8) reminds and warns medical staff to intervene in the patient's blood glucose in a timely manner through the patient's blood glucose alarm.

8. A real-time blood sugar monitoring and hypoglycemia early warning system during hemodialysis according to claim 7, characterized in that: The blood sugar prediction data output (7) is obtained by comparing the blood sugar prediction data with the normal range, and the result of calculating the blood sugar concentration in the nutrient solution to be regulated is output through the data processing module (6), and the result is transmitted to the medical staff for confirmation and regulation (9). After the medical staff confirms the regulation (9), the adjustment instruction is transmitted to the automatic glucose concentration regulation (11). After receiving the instruction, the automatic glucose concentration regulation (11) regulates the blood sugar. After receiving the information, the medical staff confirms the regulation (9), and can manually modify the value, and transmit the instruction to the manual glucose concentration regulation (10). The manual glucose concentration regulation (10) executes the blood sugar instruction modified by the medical staff.

Images