A blood glucose monitoring and control method and system based on medical robots

Abstract

The present invention provides a blood glucose monitoring and control method and system based on a medical robot, wherein the method includes: real-time monitoring of the acquired blood glucose value, determining whether the blood glucose value falls within a first preset range; if not, generating a first alarm signal; if so, obtaining a continuous blood glucose value sequence within a set time window, and constructing a blood glucose change curve; traversing the blood glucose change curve, calculating the blood glucose change rate and blood glucose change rate corresponding to each time node in the time window, and constructing a change rate sequence and a change rate sequence; calculating the blood glucose predicted value after the current time node based on the change rate sequence, the change rate sequence, the maximum peak-to-valley difference of the blood glucose change curve, and the blood glucose value at the current time node; when it is determined that the blood glucose predicted value does not fall within the first preset range, generating a second alarm signal and generating a control list for the blood glucose predicted value. The blood glucose value prediction process of the method of the present invention is simpler, does not require a large amount of computing power resources, and improves the real-time performance of the user's blood glucose monitoring and control.

Description

Technical Field

[0001] The present invention generally relates to the technical field of blood sugar prediction, and in particular to a blood sugar monitoring and control method based on a medical robot. Background Art

[0002] In recent years, with the advancement of medical technology, blood glucose monitoring and regulation have gradually become a core research area in diabetes management. Traditional blood glucose monitoring systems rely on continuous glucose monitoring devices to obtain real-time data, combined with manual analysis or simple threshold alarm mechanisms for early warning. However, existing technologies still have significant limitations in practical applications.

[0003] To improve prediction accuracy, existing technologies often use prediction models with complex algorithms such as deep learning to predict blood sugar. However, this prediction method requires high-performance computing resources, resulting in high system power consumption and large response delays, which cannot meet real-time warning needs. Summary of the Invention

[0004] In view of the above-mentioned defects or deficiencies in the prior art, it is desired to provide a blood glucose monitoring and control method and system based on a medical robot to solve the above-mentioned problems.

[0005] A first aspect of the present invention provides a blood glucose monitoring and control method based on a medical robot, comprising:

[0006] monitoring the obtained blood glucose value in real time, and determining whether the blood glucose value falls within a first preset range;

[0007] If not, a first alarm signal is generated;

[0008] If so, obtaining a continuous blood glucose value sequence within a set time window, and constructing a blood glucose change curve based on the continuous blood glucose value sequence;

[0009] Traversing the blood glucose change curve, calculating the change rate and change rate corresponding to each time node in the time window, and constructing a change rate sequence and a change rate sequence;

[0010] Calculating a predicted blood glucose value after a current time node based on the change rate sequence, the change speed sequence, the maximum peak-to-valley difference of the blood glucose change curve, and the blood glucose value at the current time node;

[0011] When it is determined that the predicted blood glucose value does not fall within the first preset range, a second alarm signal is generated and a control list is generated for the predicted blood glucose value; the control list is used to indicate how to regulate blood glucose.

[0012] According to the technical solution provided by the present invention, the step of obtaining a continuous blood glucose value sequence within a set time window and constructing a blood glucose change curve based on the continuous blood glucose value sequence includes:

[0013] Obtaining all blood glucose values within the set time window, wherein the set time window includes a plurality of time nodes, the plurality of time nodes are evenly spaced, and each time node corresponds to a blood glucose value;

[0014] A continuous blood glucose value sequence is constructed based on all blood glucose values within the set time window, and a blood glucose change curve is constructed based on the continuous blood glucose value sequence; the continuous blood glucose value sequence includes multiple blood glucose values and a time node corresponding to each blood glucose value.

[0015] Inputting all blood glucose values, blood glucose change rates and blood glucose change speeds within the set time window into a long-term prediction model, and outputting a long-term trend parameter of blood glucose, wherein the long-term trend parameter is used to characterize the influence of basal metabolism on the predicted blood glucose value;

[0016] Calculating a blood glucose parameter set based on all blood glucose values within the set time window, the blood glucose parameter set including the blood glucose mean, the blood glucose fluctuation standard deviation, the maximum peak-to-valley difference of the blood glucose change curve, and the linear regression slope of the blood glucose change curve;

[0017] Inputting the blood glucose parameter set into a short-term prediction model and outputting a short-term fluctuation parameter of blood glucose, wherein the short-term fluctuation parameter is used to characterize the influence of external intervention on the blood glucose prediction value;

[0018] The blood glucose prediction value is obtained by calculation based on the long-term trend parameter, the short-term fluctuation parameter and the current blood glucose value.

[0019] According to the technical solution provided by the present invention, the long-term prediction model is:

[0020]

[0021] in, represents the long-term trend parameter; Indicates the current moment; Indicates the time interval between the time node corresponding to the blood glucose prediction value and the current moment; Indicates the Blood glucose value at each time point; Indicates the Blood glucose change rate at each time point; Indicates the The rate of change of blood glucose at each time point; represents the global scaling factor; represents the attenuation coefficient, represents the attenuation weight; Indicates the sampling time interval; express The importance coefficient of express The importance coefficient of

[0022] The short-term prediction model is:

[0023]

[0024] in, represents the short-term volatility parameter; Indicates mean blood sugar level; represents the variance of blood glucose fluctuation; Indicates the maximum peak-to-valley difference; and Respectively and Baseline value for standardization and ; It represents the linear regression slope of the blood glucose change curve; 、 、 and Respectively 、 、 and The importance coefficient of .

[0025] According to the technical solution provided by the present invention, after traversing the blood glucose change curve, calculating the blood glucose change rate and blood glucose change rate corresponding to each time node in the time window, and constructing the change rate sequence and change rate sequence, the method further includes:

[0026] traversing the change rate sequence and outputting a data guidance sequence when determining that the blood glucose change rate in the change rate sequence is greater than a first preset rate; the data guidance sequence includes a plurality of prompt messages related to blood glucose changes, and is used to instruct the user to input external data, the external data including at least exercise intensity, dietary carbohydrate content, and hypoglycemic drug dosage;

[0027] According to the external data, the weight database is called to obtain 、 、 、 、 、 Correction value of

[0028] according to 、 、 、 、 、 The blood glucose prediction value is corrected by the correction value.

[0029] According to the technical solution provided by the present invention, the method further includes:

[0030] Calculate a first absolute difference; the first absolute difference is the absolute value of the difference between the blood glucose value corresponding to the same time node and the blood glucose predicted value;

[0031] determining that the external data has been acquired within a first set time period before the time node, and comparing the first absolute difference with a first threshold;

[0032] When it is determined that the first absolute difference is greater than the first threshold, a third alarm signal is generated; the third alarm signal is used to instruct the user to provide a manually measured blood glucose value and update the blood glucose value at the time node.

[0033] According to the technical solution provided by the present invention, when it is determined that the predicted blood glucose value does not fall within the first preset range, a second alarm signal is generated and a control list is generated for the predicted blood glucose value, including:

[0034] Obtaining a healthy blood sugar preset value, where the healthy blood sugar preset value is the user's optimal blood sugar value;

[0035] According to the blood glucose health preset value and the blood glucose predicted value, the control database corresponding to the user identity is called to obtain the corresponding control list; the control database includes the physiological data and behavioral data updated by the user within a third set time period before the current time node, as well as the control lists corresponding to the physiological data, behavioral data and different blood glucose predicted values.

[0036] According to the technical solution provided by the present invention, after calculating the blood glucose predicted value after the current time node based on the blood glucose change rate, the blood glucose change velocity, the maximum peak-to-valley difference of the blood glucose change curve, and the blood glucose value at the current time node, the method further includes:

[0037] When it is determined that the predicted blood glucose value belongs to the first preset range, an update display signal is generated, where the update display signal is used to instruct the predicted blood glucose value to be displayed.

[0038] According to the technical solution provided by the present invention, the method further includes:

[0039] Generate a regular reminder instruction based on the control list, and the regular reminder instruction is used to remind the user to execute according to the control list.

[0040] A second aspect of the present invention provides a blood glucose monitoring and control system based on a medical robot, which is used to perform the above-mentioned blood glucose monitoring and control method based on a medical robot. The system comprises:

[0041] a first processing module configured to monitor the acquired blood glucose value in real time and determine whether the blood glucose value falls within a first preset range; if not, generate a first alarm signal; if so, acquire a continuous sequence of blood glucose values within a set time window and construct a blood glucose change curve based on the continuous sequence of blood glucose values;

[0042] a data calculation module configured to traverse the blood glucose change curve, calculate the blood glucose change rate and blood glucose change rate corresponding to each time node in the time window, and construct a change rate sequence and a change rate sequence;

[0043] a blood glucose prediction module configured to calculate a predicted blood glucose value after a current time node based on the change rate sequence, the change rate sequence, the maximum peak-to-valley difference of the blood glucose change curve, and the blood glucose value at the current time node;

[0044] The second processing module is configured to generate a second alarm signal and generate a control list for the blood glucose prediction value when it is determined that the blood glucose prediction value does not fall within the first preset range; the control list is used to indicate how to adjust blood glucose.

[0045] Compared with the prior art, the present invention has the following advantages: on the one hand, by monitoring the acquired blood glucose value in real time and determining whether it meets the first preset range, a first alarm signal is generated when it does not meet the range, so that the user's blood glucose can be monitored in real time, and an alarm is issued when the blood glucose value deviates from the safe range, prompting the user to take timely action; on the other hand, by constructing a blood glucose change curve based on a continuous blood glucose value sequence within a set time window when the blood glucose value falls within the first preset range, and calculating the blood glucose change rate and blood glucose change rate corresponding to each time point within the set time window, and then predicting the blood glucose value at a certain moment in the future based on conventional and easily accessible parameters such as the blood glucose change rate, blood glucose change rate, maximum peak-to-valley value, and blood glucose value, the prediction process is simpler, the output result is faster, and the required computing power is smaller. The embedded system inside the medical robot can realize real-time monitoring and prediction of blood glucose, thereby improving the real-time nature of blood glucose prediction; it can also issue an early warning based on the predicted blood glucose value and feedback a targeted control list, making it convenient for the user to control blood glucose by themselves. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Other features, objects and advantages of the present invention will become more apparent upon reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0047] Figure 1 A flowchart of the steps of the blood glucose monitoring and control method based on a medical robot provided in Example 1;

[0048] Figure 2 This is a schematic diagram of the structure of the blood glucose monitoring and control system based on the medical robot provided in Example 4.

[0049] Figure numbers: 10, first processing module; 20, data calculation module; 30, blood glucose prediction module; 40, second processing module. DETAILED DESCRIPTION

[0050] The present invention will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are intended only to illustrate the relevant invention and are not intended to limit the invention. It should also be noted that, for ease of description, only portions relevant to the invention are shown in the accompanying drawings.

[0051] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments of the present invention can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0052] Example 1

[0053] Please refer to Figure 1 This embodiment provides a blood glucose monitoring and control method based on a medical robot, comprising:

[0054] S100: Monitor the acquired blood glucose level in real time and determine whether the blood glucose level falls within a first preset range.

[0055] Specifically, the method provided in this embodiment is implemented using a medical robot, which can be used in either hospital or home environments. The robot integrates a blood glucose monitor, a smart medicine cabinet, and a touchscreen display for human-machine interaction. The robot communicates with a mobile terminal (e.g., a user's phone) via Bluetooth or wireless networks. The robot can establish connections with multiple users, identifying them through identification or other verification information. The robot then performs targeted monitoring and control for each user, ensuring that monitoring and control by different users do not interfere with each other.

[0056] In this embodiment, blood glucose is tested using a continuous blood glucose monitoring device. This device can be a micro-sensor implanted under the user's skin or a non-invasive blood glucose sensor worn by the user. The continuous blood glucose monitoring device communicates with the medical robot via Bluetooth or a wireless network. The continuous blood glucose monitoring device samples the user's blood glucose at set intervals and outputs the collected blood glucose value to the medical robot via Bluetooth or a wireless network. In this embodiment, the sampling interval of the continuous blood glucose monitoring device is 5 minutes.

[0057] Specifically, in step S100, the medical robot obtains the blood glucose value detected by the continuous blood glucose monitoring device in real time, and records the blood glucose value and the time node of the collection after obtaining the blood glucose value, ensuring that the record corresponds to the identity of the user to avoid crosstalk with the data of other users; at the same time, after each blood glucose value is obtained, it is determined whether the blood glucose value meets the first preset range. The first preset range is the normal fluctuation range of the user's blood glucose value, including a maximum blood glucose threshold and a minimum blood glucose threshold. The purpose is to determine whether the user's current blood glucose is abnormal.

[0058] S200: If not, generate a first alarm signal.

[0059] Specifically, in step S200, after determining whether the blood glucose level meets the first preset range, if the blood glucose level is greater than the maximum blood glucose threshold or less than the minimum blood glucose threshold, it indicates that the user's blood glucose level is abnormal and requires timely human intervention. The medical robot then generates a first alarm signal. The first alarm signal acts on the touch screen of the medical robot, displaying an abnormal blood glucose level indicator and emitting a sound reminder. It also acts on the user's mobile phone, causing the phone to emit a sound reminder. The first alarm signal also includes detailed information on whether the blood glucose level is greater than the maximum blood glucose threshold or less than the minimum blood glucose threshold. When the blood glucose level is greater than the maximum blood glucose threshold, the first alarm signal also activates the medical robot's smart medicine box, allowing the user to lower their blood glucose through medication.

[0060] S300: If yes, obtain a continuous blood glucose value sequence within a set time window, and construct a blood glucose change curve according to the continuous blood glucose value sequence.

[0061] Specifically, in step S300, after determining whether the blood glucose value meets the first preset range, if the blood glucose value is less than or equal to the maximum blood glucose threshold and greater than or equal to the minimum blood glucose threshold, it means that the user's blood glucose at this time is within the normal range. Then the medical robot begins to predict the user's blood glucose value at a certain moment in the future to prevent blood glucose abnormalities.

[0062] Step S300 specifically includes:

[0063] S301: Acquire all blood glucose values within the set time window, where the set time window includes multiple time nodes, which are evenly spaced and each time node corresponds to a blood glucose value.

[0064] Specifically, in step S301, the medical robot obtains the blood glucose value within the set time window. The set time window includes the current time node and multiple consecutive time nodes before the current time node. In this embodiment, the set time window is 50 minutes. Since each time node is 5 minutes apart, the set time window includes a total of 10 time nodes, that is, a total of 10 blood glucose values including the blood glucose value corresponding to the current time node are obtained.

[0065] S302: Constructing a continuous blood glucose value sequence based on all blood glucose values within the set time window, and constructing a blood glucose change curve based on the continuous blood glucose value sequence; the continuous blood glucose value sequence includes multiple blood glucose values and a time node corresponding to each blood glucose value.

[0066] Specifically, in step S302, after obtaining the blood glucose values corresponding to the 10 time nodes, a continuous blood glucose value sequence is obtained based on the 10 time nodes and the blood glucose values corresponding to each time node. Then, a blood glucose change curve is constructed based on the obtained continuous blood glucose value sequence. Here, taking a user with high blood glucose as an example, a continuous blood glucose value sequence of the user is listed as shown in Table 1 below:

[0067] Table 1

[0068]

[0069] in, The blood sugar value at the time point can be regarded as the basic blood sugar value when fasting. to The blood sugar value at the time point can be regarded as the blood sugar rising stage after the meal, with a peak value of At the time point, blood sugar reaches its peak. to The gradual decrease in blood sugar at this time point may be related to delayed insulin secretion or mild exercise.

[0070] S400: traverse the blood glucose change curve, calculate the blood glucose change rate and blood glucose change rate corresponding to each time node in the time window, and construct a change rate sequence and a change rate sequence.

[0071] Specifically, in step S400, after establishing the blood sugar change curve, the medical robot calculates the blood sugar change rate corresponding to each time node on the blood sugar change curve. The blood sugar change rate is used to represent the rising or falling trend of blood sugar. The calculation formula is as follows:

[0072] Formula (1)

[0073] because For the first time point, there is no prodromal data, so the blood glucose change rate is Start the calculation, and after calculating all the blood glucose change rates within the set time window, you can get the blood glucose change rate sequence. Taking the continuous blood glucose value sequence shown in Table 1 as an example, you can calculate the blood glucose change rate sequence shown in Table 2 below:

[0074] Table 2

[0075]

[0076] in, to The blood sugar change rate gradually increases at the time point, which can reflect the accelerated rise of blood sugar after the meal. The blood sugar change rate at the time node reaches the maximum value, At the time point, the blood sugar change rate turns negative. to The absolute value of the blood sugar change rate at the time node decreases, and blood sugar tends to be stable.

[0077] After establishing the blood sugar change rate sequence, the medical robot continues to calculate the blood sugar change rate corresponding to each time node in the blood sugar change rate sequence. The blood sugar change rate is used to represent the speed of blood sugar rise or fall. Its calculation formula is as follows:

[0078] Formula (2)

[0079] because As the first calculation point, there is no precursor data, so the blood glucose change rate is Start the calculation, and after calculating all the blood sugar change rates within the set time window, you can get the blood sugar change rate sequence. Taking the blood sugar change rate sequence shown in Table 2 as an example, you can calculate the blood sugar change rate sequence shown in Table 3 below:

[0080] Table 3

[0081]

[0082] in, to The blood sugar change rate at the time node is positive, indicating that the blood sugar rises faster. The blood sugar change rate reaches its peak at the time point. The blood sugar change rate at the time point dropped sharply to 0.004, The time node suddenly changed to -0.048, indicating that blood sugar changed from rising to falling rapidly. to The blood sugar change rate at a time node tends to 0, which can reflect that the blood sugar decreases at a uniform rate.

[0083] Through the rate of change sequence and the rate of change sequence, we can clearly obtain the rising or falling trend of blood sugar in a period of time before the current time node and the corresponding rising rate and falling rate, thereby accurately presenting the user's recent blood sugar changes.

[0084] S500: Calculating a predicted blood glucose value after the current time node based on the change rate sequence, the change speed sequence, the maximum peak-to-valley difference of the blood glucose change curve, and the blood glucose value at the current time node.

[0085] Specifically, in step S500, the medical robot predicts blood sugar based on the previously constructed change rate sequence, change speed sequence, maximum peak-to-valley difference, and visible parameters such as blood sugar value. On the one hand, this simplifies the prediction process, reduces computing resource usage, and ensures the real-time nature of blood sugar prediction. On the other hand, it enables the data used for prediction to evenly reflect the parameters of the user's recent blood sugar condition, thereby improving the accuracy of blood sugar prediction.

[0086] Step S500 specifically includes:

[0087] Step S501: Input all blood glucose values, blood glucose change rates and blood glucose change rates within the set time window into the long-term prediction model, and output long-term trend parameters of blood glucose, which are used to characterize the impact of basal metabolism on blood glucose prediction values.

[0088] Specifically, in step S501, the blood glucose values, blood glucose change rates, and blood glucose change rates corresponding to all time nodes within the set time window are retrieved. A preset long-term prediction model is then retrieved, and all retrieved blood glucose values, blood glucose change rates, and blood glucose change rates are output to the long-term prediction model. The long-term prediction model then outputs long-term trend parameters, which can be used to characterize the impact of human basal metabolism on blood glucose. The long-term prediction model is represented by the following formula (3):

[0089] Formula (3)

[0090] in, represents the long-term trend parameter; Indicates the current moment; Indicates the time interval between the time node corresponding to the blood glucose prediction value and the current moment, It can be adjusted according to needs, such as If you select half an hour, you can calculate the predicted blood sugar value half an hour later; Indicates the Blood glucose value at each time point; Indicates the Blood glucose change rate at each time point; Indicates the The rate of change of blood glucose at each time point; Represents the global scaling factor, which is used to adjust the numerical amplitude of the entire trend item and control the overall contribution of historical blood glucose data to the current trend. Setting a larger value (such as close to 1) indicates that the blood sugar trend is more dependent on recent measured data, which is suitable for users with large basal metabolic fluctuations (such as patients in the acute stage of diabetes). Setting a smaller value (such as close to 0.1) indicates that the blood sugar trend is more dependent on the smooth continuation of the historical trend, which is suitable for users with stable basal metabolism (such as patients with chronic diseases with good blood sugar control); for patients with low insulin sensitivity, It can be set to 0.8 to enable the model to quickly capture the trend of sudden increase / decrease in blood sugar. For healthy people or diabetic patients with stable blood sugar, It can be set to 0.3 to avoid false alarms caused by accidental fluctuations; Represents the decay coefficient, which is used to control the speed at which the weight of historical data decays over time. The larger it is, the faster the weight of the long-term data decays, and the more the model focuses on recent blood sugar changes. The smaller it is, the more weight is retained for long-term data, and the model relies more on long-term trends; represents the decay weight. In this embodiment, the weight value corresponding to the recent time node is higher;

[0091] Indicates the sampling time interval, used for and Standardization; Indicates that Standardized to consistent; Indicates that Standardized to consistent; express The importance coefficient of express The importance coefficient of .

[0092] S502: Calculate a blood glucose parameter set based on all blood glucose values within the set time window, wherein the blood glucose parameter set includes a blood glucose mean value, a blood glucose fluctuation standard deviation, a maximum peak-to-valley difference of the blood glucose change curve, and a linear regression slope of the blood glucose change curve.

[0093] Specifically, in step S502, the mean blood glucose value is first calculated based on the blood glucose values corresponding to all time nodes within the set time window. The mean blood glucose value is calculated as shown in the following formula (IV):

[0094] Formula (IV);

[0095] in, Indicates mean blood sugar level.

[0096] After calculating the mean blood sugar value, the blood sugar fluctuation variance is calculated based on the mean blood sugar value. The calculation of the blood sugar fluctuation variance is expressed as the following formula (5):

[0097] Formula (5);

[0098] in, represents the variance of blood sugar fluctuation;

[0099] The corresponding blood sugar standard deviation can be obtained:

[0100] Formula (6).

[0101] Then, the maximum peak-to-valley difference is calculated based on the maximum and minimum blood glucose values within the set time window. The maximum peak-to-valley difference is used to guide the limitation of the blood glucose variation range.

[0102] Finally, a linear regression is performed on all blood glucose values within the set time window. The slope of the fitted line is the linear regression slope, which is used to measure the overall trend of blood glucose over the entire period, balance parameters such as blood glucose mean and blood glucose variance, and avoid misjudgment of a single indicator. The linear regression slope is calculated as shown in the following formula (VII):

[0103] Formula (7)

[0104] in, Represents the time mean.

[0105] S503: Input the blood glucose parameter set into a short-term prediction model, and output a short-term fluctuation parameter of blood glucose, wherein the short-term fluctuation parameter is used to characterize the impact of external intervention on the blood glucose prediction value.

[0106] Specifically, in step S503, a preset short-term prediction model is retrieved, and then the calculated blood glucose mean value, blood glucose fluctuation standard deviation, maximum peak-to-valley difference of the blood glucose change curve, and linear regression slope of the blood glucose change curve are input into the short-term prediction model. The short-term prediction model outputs a short-term fluctuation parameter, which can be used to characterize the impact of external intervention on the predicted blood glucose value of the human body. The short-term prediction model is represented by the following formula (six):

[0107] Formula (8)

[0108] in, represents the short-term volatility parameter; Indicates the maximum peak-to-valley difference; and The baseline values of mean blood glucose and standard deviation of blood glucose are used for standardization and ; express Normalized by time and baseline values; 、 、 and Respectively 、 、 and The importance coefficient of

[0109] Further, and The method is as follows: the user's fasting blood sugar baseline value is collected every day, and after collecting the set number of days, the average of the collected blood sugar baseline values is obtained. , correspondingly, can be achieved by Calculated .

[0110] S504: Calculate and obtain the blood glucose prediction value based on the long-term trend parameter, the short-term fluctuation parameter and the current blood glucose value.

[0111] Specifically, in step S504, the blood glucose value corresponding to the current moment in the continuous blood glucose value sequence is first obtained, and then the blood glucose prediction value is calculated based on the current blood glucose value, the long-term trend parameter, and the short-term fluctuation parameter. The calculation formula of the blood glucose prediction value is shown in the following formula (7):

[0112] Formula (9)

[0113] in, Indicates the time from the current moment predicted blood glucose values after the time interval; Indicates the current blood sugar value; Indicates that Standardized to and consistent;

[0114] So far, according to The blood sugar prediction value at the corresponding moment in the future can be calculated by selecting .

[0115] For the setting of formula (VII), you can also introduce it when you need to adjust the proportion of each component globally. 、 and The weights can be adjusted according to the user's exercise intensity, dietary carbohydrate content, insulin injection amount and other parameters, or a fixed ratio can be determined based on clinical verification.

[0116] This embodiment collects the user's recent blood sugar data and predicts blood sugar in combination with the long-term trend and short-term fluctuations of blood sugar, so as to ensure the accuracy of blood sugar prediction and the regular and easy-to-obtain parameters used for prediction, making the prediction process simpler and thus achievable through the embedded system built into the medical robot; at the same time, corresponding importance coefficients are assigned to long-term trends and short-term fluctuations, and different weights can be assigned to long-term trends and short-term fluctuations, so that the blood sugar prediction method can be personalized for different users, thereby further improving the accuracy of blood sugar prediction.

[0117] S600: When it is determined that the predicted blood glucose value does not fall within the first preset range, a second alarm signal is generated and a control list is generated for the predicted blood glucose value; the control list is used to indicate how to regulate blood glucose.

[0118] Specifically, step S600 specifically includes: the medical robot obtains the time from the current time node After the blood glucose prediction value is obtained, the user continues to determine whether the blood glucose prediction value meets the first preset range. If the blood glucose prediction value is greater than the maximum blood glucose threshold or less than the minimum blood glucose threshold, it indicates that the user is After a certain period of time, blood sugar becomes abnormal, which requires the user's attention. Then, the medical robot generates a second alarm signal. On the one hand, the second alarm signal acts on the touch screen of the medical robot, and displays an indication that blood sugar is about to become abnormal on the touch screen, and displays the time point when blood sugar is about to become abnormal, and issues a voice prompt to remind the user to pay attention to his or her own blood sugar; the second alarm signal acts on the user's mobile phone, and displays an indication that blood sugar is about to become abnormal on the mobile phone, and displays the time point when blood sugar is about to become abnormal, and issues a voice prompt at the same time. When it is judged that the predicted blood sugar value meets the first preset range, it indicates that If the user's blood sugar remains within the normal range after a certain period of time, it is only necessary to continue to obtain and update the blood sugar value collected by the continuous blood sugar monitoring device and calculate and update the blood sugar prediction value.

[0119] Specifically, after generating the second alarm signal, the medical robot will also generate a targeted control list based on the predicted blood sugar value. The control list includes whether sugar supplementation is needed, whether hypoglycemic drugs are needed, whether exercise is needed, detailed exercise types and exercise time, diet adjustments, etc. After generating the control list, the medical robot will send the control list to the touch screen and the user's mobile phone for the user to view.

[0120] Furthermore, step S600 includes:

[0121] S610: Obtain a healthy blood sugar preset value, where the healthy blood sugar preset value is the user's optimal blood sugar value.

[0122] Specifically, in step S610, when the medical robot determines that the blood glucose prediction value does not fall within the first preset range, it obtains a blood glucose health preset value. The blood glucose health preset value is determined based on the user's physical condition and can be the average blood glucose value within the time period set by the user.

[0123] S620: Call the control database corresponding to the user identity according to the blood glucose health preset value and the blood glucose predicted value to obtain the corresponding control list; the control database includes the physiological data and behavioral data updated by the user before the current time node, and the control lists corresponding to the physiological data, behavioral data and different blood glucose predicted values.

[0124] Specifically, in step S620, a control database is preset in advance based on the user's blood sugar health preset value and physiological data and behavioral data. The control database stores a control list when the blood sugar health preset value corresponds to different blood sugar prediction values, and the control list is adjusted under the influence of physiological data and behavioral data; wherein, physiological data includes the user's recent sleep quality, mental stress and other physiological factors, and behavioral data includes external intervention factors such as the user's recent frequent exercise and eating habits; physiological data and behavioral data are updated by the user, and the control list in the control database is adjusted as the blood sugar health preset value and physiological data and behavioral data are updated. When it is determined that the blood sugar prediction value does not meet the first preset range, the control database is retrieved according to the user's blood sugar health preset value and blood sugar prediction value to obtain a personalized control list.

[0125] Furthermore, after step S500, the method further includes:

[0126] When it is determined that the predicted blood glucose value belongs to the first preset range, an update display signal is generated, where the update display signal is used to instruct the predicted blood glucose value to be displayed.

[0127] Specifically, after the blood glucose prediction value is calculated, when it is determined that the blood glucose prediction value falls within the first preset range, that is, the user After a period of time, blood sugar will not become abnormal. The medical robot will update the blood sugar prediction value on the touch screen and the user's mobile phone. In addition, the medical robot will also update the real-time detected blood sugar value on the touch screen and the user's mobile phone, so that the user can view the current blood sugar value and future blood sugar prediction value in real time.

[0128] Furthermore, after step S620, the method further includes:

[0129] Generate a regular reminder instruction based on the control list, and the regular reminder instruction is used to remind the user to execute according to the control list.

[0130] Specifically, the control list instructs the user to perform corresponding control measures at a specified time, such as consuming a certain amount of carbohydrates or doing some exercise in five minutes. When the medical robot generates the control list, it also sets an alarm to remind the user to follow the control list.

[0131] Example 2

[0132] Based on the above-mentioned embodiment 1, this embodiment provides another blood glucose monitoring and control method based on a medical robot. The same contents as those in embodiment 1 are not repeated here. The difference lies in that:

[0133] After step S400, the following steps are further included:

[0134] S410: Traversing the change rate sequence, and when determining that the blood glucose change rate in the change rate sequence is greater than a first preset rate, outputting a data guidance sequence; the data guidance sequence includes a plurality of prompt messages related to blood glucose changes, for instructing the user to input external data, the external data including at least exercise intensity, dietary carbohydrate content, and hypoglycemic drug dosage.

[0135] Specifically, in step S410, after obtaining the change rate sequence, each blood glucose change rate in the change rate sequence is compared with the first preset rate one by one. When the blood glucose change rate is greater than the first preset rate, it indicates that the blood glucose is rising or falling rapidly. At this time, the change in blood glucose exceeds the influence of the human body's basal metabolism, which may be caused by other external factors, such as exercise intensity, dietary carbohydrate content and insulin injection amount. Therefore, the method provided in Example 1 can no longer meet the prediction requirements and the prediction method needs to be adjusted. Therefore, the medical robot outputs a data guidance sequence to the user's mobile phone or touch screen. The data guidance sequence causes multiple questions to be displayed one by one on the user's mobile phone and touch screen. The questions include whether the user has exercised and the type and time of exercise, whether the user has consumed carbohydrates and the type and amount of carbohydrates consumed, whether the user uses hypoglycemic drugs and the type and dosage of hypoglycemic drugs, etc.; the questions displayed on the mobile phone and touch screen are used to guide the user to supplement external data such as his or her own physiological characteristics and behavioral characteristics to improve the accuracy of the blood glucose prediction value.

[0136] S420: Call the weight database according to the external data to obtain 、 、 、 、 、 Correction value.

[0137] Specifically, in step S420, a weight database is preset in advance based on physiological characteristics and behavioral characteristics, and the weight database includes multiple exercise types, multiple exercise duration ranges corresponding to each exercise type, multiple carbohydrate types corresponding to each exercise duration range, multiple dietary amount ranges corresponding to each carbohydrate type, multiple hypoglycemic drug types corresponding to each dietary amount range, multiple hypoglycemic drug dosage ranges corresponding to each hypoglycemic drug type, and a set of hypoglycemic drug dosage ranges corresponding to each hypoglycemic drug dosage range. 、 、 、 、 、 Here we only list the parameters included in physiological characteristics and behavioral characteristics. In practice, more types of physiological characteristics and behavioral characteristics can be added to make the correction value more accurate.

[0138] Further explanation: if used to increase exercise intensity, Increase, so that the model pays more attention to the rapid changes in blood sugar caused by exercise; if the user adjusts the dosage of hypoglycemic drugs, Increased to reflect the effect of the drug on the acceleration of blood sugar changes; if the user consumes high carbohydrates, and Increase to highlight the impact of diet on blood sugar fluctuations; if the user records specific medications, Adjustments are made to reflect the long-term effects of the medication on blood sugar trends. For example, if the user consumes 50g of carbohydrates, From 0.3 to 0.5, From 0.2 to 0.3 to reflect the increased blood sugar fluctuations caused by high carbohydrates; From 0.4 to 0.3, the dominant factor in short-term fluctuations has become diet rather than basal metabolism.

[0139] S430: 、 、 、 、 、 The blood glucose prediction value is corrected by the correction value.

[0140] Specifically, in step S430, after obtaining a set of 、 、 、 、 、 After the correction value is obtained, the calculation results of the above formulas (3) and (8) are corrected according to the correction value to make the long-term prediction parameters and short-term fluctuation parameters more accurate, thereby making the final blood glucose prediction value more accurate.

[0141] The method provided in this embodiment deeply integrates the physiological and behavioral characteristics of the user with the prediction method of Example 1, thereby improving the personalized accuracy of blood sugar prediction without affecting blood sugar prediction, so that the prediction results can better represent the user's actual physical condition.

[0142] Example 3

[0143] Based on the above-mentioned embodiment 2, this embodiment provides another blood glucose monitoring and control method based on a medical robot. The same contents as those in embodiment 2 are not repeated here. The difference lies in that:

[0144] The method further comprises:

[0145] Calculate a first absolute difference; the first absolute difference is the absolute value of the difference between the blood glucose value corresponding to the same time node and the blood glucose predicted value;

[0146] determining that the external data has been acquired within a first set time period before the time node, and comparing the first absolute difference with a first threshold;

[0147] When it is determined that the first absolute difference is greater than the first threshold, a third alarm signal is generated; the third alarm signal is used to instruct the user to provide a manually measured blood glucose value and update the blood glucose value at the time node.

[0148] Specifically, since the medical robot will obtain the current blood sugar value in real time and calculate the subsequent blood sugar prediction value, there will be a real-time blood sugar value and a blood sugar prediction value calculated at the previous node at each time node. By comparing the blood sugar value and the blood sugar prediction value at the corresponding time node, it is possible to check whether the blood sugar value and / or the blood sugar prediction value are incorrect, thereby facilitating the verification of the method of this embodiment. The specific method is:

[0149] First, calculate the absolute value of the difference between the blood glucose value corresponding to the time node and the blood glucose predicted value, that is, the first absolute difference. The first absolute difference is used to reflect the degree to which the blood glucose predicted value deviates from the blood glucose value. If the degree of deviation is large, there are two possibilities. One possibility is that there is a problem with the prediction algorithm, resulting in a deviation in the blood glucose predicted value. The other possibility is that there is a malfunction in the continuous blood glucose detection device, resulting in inaccurate blood glucose detection. The degree to which the blood glucose predicted value deviates from the blood glucose value is judged by comparing the first absolute difference with the first threshold. When the first absolute difference is greater than the first threshold, it is determined that the blood glucose value and / or the blood glucose predicted value are erroneous. When the first absolute difference is less than or equal to the first threshold, it is determined that the blood glucose value and the blood glucose predicted value are both normal. In order to identify the two possible errors, before comparing the first absolute difference with the first threshold, it is also determined whether external data input by the user is received within the first set time period before the time node. If no external data is received, the first absolute difference is greater than the first threshold, which may be due to the lack of external data. 、 、 、 、 、 These parameters are corrected, which leads to inaccurate blood glucose prediction values calculated by the algorithm. At this time, the medical robot reminds the user again to input external data to update the algorithm. If external data is received, 、 、 、 、 、 The update of these parameters reduces the probability of errors in the blood glucose prediction value. If the first absolute difference is greater than the first threshold, it is likely that the continuous blood glucose monitoring device has malfunctioned, resulting in inaccurate blood glucose values. At this time, the medical robot generates a third alarm signal. The third alarm signal is displayed on the touch screen and the user's mobile phone, and the built-in smart medicine box is activated, prompting the user to manually measure blood glucose using the fingertip blood collection device in the smart medicine box, and prompting the user to upload the manual measurement results to the medical robot via the touch screen or mobile phone. This method can be used to check whether the continuous blood glucose monitoring device is faulty and problems that arise during wear, avoiding inaccurate blood glucose predictions later due to errors in the blood glucose value obtained by the medical robot itself. After the manually measured blood glucose value is uploaded to the medical robot, the medical robot updates the blood glucose value at that time point.

[0150] Example 4

[0151] Please refer to Figure 2 This embodiment provides a blood glucose monitoring and control system based on a medical robot, which is used to execute the blood glucose monitoring and control method based on a medical robot as described in Example 1. The system includes:

[0152] A first processing module 10 is configured to monitor the obtained blood glucose value in real time and determine whether the blood glucose value falls within a first preset range; if not, generate a first alarm signal; if so, obtain a continuous blood glucose value sequence within a set time window and construct a blood glucose change curve based on the continuous blood glucose value sequence;

[0153] A data calculation module 20 is configured to traverse the blood glucose change curve, calculate the blood glucose change rate and blood glucose change rate corresponding to each time node in the time window, and construct a change rate sequence and a change rate sequence;

[0154] a blood glucose prediction module 30 configured to calculate a predicted blood glucose value after a current time node based on the change rate sequence, the change rate sequence, the maximum peak-to-valley difference of the blood glucose change curve, and the blood glucose value at the current time node;

[0155] The second processing module 40 is configured to generate a second alarm signal and generate a control list for the blood glucose prediction value when it is determined that the blood glucose prediction value does not fall within the first preset range; the control list is used to indicate how to adjust blood glucose.

[0156] The above description is merely a preferred embodiment of the present invention and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of the invention is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, but also encompasses other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the inventive concept. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in the present invention.

Claims

1. A blood glucose monitoring and control method based on a medical robot, characterized in that: include: monitoring the obtained blood glucose value in real time, and determining whether the blood glucose value falls within a first preset range; If not, a first alarm signal is generated; If so, obtaining a continuous blood glucose value sequence within a set time window, and constructing a blood glucose change curve based on the continuous blood glucose value sequence; Traversing the blood glucose change curve, calculating the blood glucose change rate and blood glucose change rate corresponding to each time node in the time window, and constructing a change rate sequence and a change rate sequence; Calculating a predicted blood glucose value after a current time node based on the change rate sequence, the change speed sequence, the maximum peak-to-valley difference of the blood glucose change curve, and the blood glucose value at the current time node; When it is determined that the predicted blood sugar value does not fall within the first preset range, a second alarm signal is generated and a control list is generated for the predicted blood sugar value; the control list is used to indicate how to adjust blood sugar; The calculating of the blood glucose prediction value after the current time node based on the change rate sequence, the change rate sequence, the maximum peak-to-valley difference of the blood glucose change curve, and the blood glucose value at the current time node includes: Inputting all blood glucose values, blood glucose change rates and blood glucose change speeds within the set time window into a long-term prediction model, and outputting a long-term trend parameter of blood glucose, wherein the long-term trend parameter is used to characterize the influence of basal metabolism on the predicted blood glucose value; Calculating a blood glucose parameter set based on all blood glucose values within the set time window, the blood glucose parameter set including the blood glucose mean, the blood glucose fluctuation standard deviation, the maximum peak-to-valley difference of the blood glucose change curve, and the linear regression slope of the blood glucose change curve; Inputting the blood glucose parameter set into a short-term prediction model and outputting a short-term fluctuation parameter of blood glucose, wherein the short-term fluctuation parameter is used to characterize the influence of external intervention on the blood glucose prediction value; Calculating the blood glucose prediction value based on the long-term trend parameter, the short-term fluctuation parameter and the current blood glucose value; The long-term prediction model is: in, represents the long-term trend parameter; Indicates the current moment; Indicates the time interval between the time node corresponding to the blood glucose prediction value and the current moment; Indicates the Blood glucose value at each time point; Indicates the Blood glucose change rate at each time point; Indicates the The rate of change of blood glucose at each time point; represents the global scaling factor; represents the attenuation coefficient, represents the attenuation weight; Indicates the sampling time interval; express The importance coefficient of express The importance coefficient of The short-term prediction model is: in, represents the short-term volatility parameter; Indicates mean blood sugar level; represents the variance of blood glucose fluctuation; Indicates the maximum peak-to-valley difference; and Respectively and Baseline value for standardization and ; It represents the linear regression slope of the blood glucose change curve; 、 、 and Respectively 、 、 and The importance coefficient of .

2. The blood glucose monitoring and control method based on a medical robot according to claim 1, characterized in that: The step of obtaining a continuous blood glucose value sequence within a set time window and constructing a blood glucose change curve according to the continuous blood glucose value sequence includes: Obtaining all blood glucose values within the set time window, wherein the set time window includes a plurality of time nodes, the plurality of time nodes are evenly spaced, and each time node corresponds to a blood glucose value; A continuous blood glucose value sequence is constructed based on all blood glucose values within the set time window, and a blood glucose change curve is constructed based on the continuous blood glucose value sequence; the continuous blood glucose value sequence includes multiple blood glucose values and a time node corresponding to each blood glucose value.

3. The blood glucose monitoring and control method based on a medical robot according to claim 2, characterized in that: After traversing the blood glucose change curve, calculating the blood glucose change rate and blood glucose change rate corresponding to each time node in the time window, and constructing the change rate sequence and change rate sequence, the method further includes: traversing the change rate sequence and outputting a data guidance sequence when determining that the blood glucose change rate in the change rate sequence is greater than a first preset rate; the data guidance sequence includes a plurality of prompt messages related to blood glucose changes, and is used to instruct the user to input external data, the external data including at least exercise intensity, dietary carbohydrate content, and hypoglycemic drug dosage; According to the external data, the weight database is called to obtain 、 、 、 、 、 Correction value of according to 、 、 、 、 、 The blood glucose prediction value is corrected by the correction value.

4. The blood glucose monitoring and control method based on a medical robot according to claim 3, characterized in that: The method further comprises: Calculate a first absolute difference; the first absolute difference is the absolute value of the difference between the blood glucose value corresponding to the same time node and the blood glucose predicted value; determining that the external data has been acquired within a first set time period before the time node, and comparing the first absolute difference with a first threshold; When it is determined that the first absolute difference is greater than the first threshold, a third alarm signal is generated; the third alarm signal is used to instruct the user to provide a manually measured blood glucose value and update the blood glucose value at the time node.

5. The blood glucose monitoring and control method based on a medical robot according to claim 4, characterized in that: When it is determined that the predicted blood glucose value does not fall within the first preset range, generating a second alarm signal and generating a control list for the predicted blood glucose value, including: Obtaining a healthy blood sugar preset value, where the healthy blood sugar preset value is the user's optimal blood sugar value; According to the blood glucose health preset value and the blood glucose predicted value, the control database corresponding to the user identity is called to obtain the corresponding control list; the control database includes the physiological data and behavioral data updated by the user within a third set time period before the current time node, as well as the control lists corresponding to the physiological data, behavioral data and different blood glucose predicted values.

6. The blood glucose monitoring and control method based on a medical robot according to claim 5, characterized in that: After calculating the blood glucose predicted value after the current time node according to the blood glucose change rate, the blood glucose change velocity, the maximum peak-to-valley difference of the blood glucose change curve, and the blood glucose value at the current time node, the method further includes: When it is determined that the predicted blood glucose value belongs to the first preset range, an update display signal is generated, where the update display signal is used to instruct the predicted blood glucose value to be displayed.

7. The blood glucose monitoring and control method based on a medical robot according to claim 6, characterized in that: The method further comprises: Generate a regular reminder instruction based on the control list, and the regular reminder instruction is used to remind the user to execute according to the control list.

8. A blood glucose monitoring and control system based on a medical robot, characterized in that: A system for executing the blood glucose monitoring and control method based on a medical robot according to any one of claims 1 to 7, the system comprising: A first processing module (10), the first processing module (10) is configured to monitor the acquired blood glucose value in real time, determine whether the blood glucose value falls within a first preset range; if not, generate a first alarm signal; if so, acquire a continuous blood glucose value sequence within a set time window, and construct a blood glucose change curve based on the continuous blood glucose value sequence; A data calculation module (20), the data calculation module (20) is configured to traverse the blood glucose change curve, calculate the blood glucose change rate and blood glucose change rate corresponding to each time node in the time window, and construct a change rate sequence and a change rate sequence; A blood glucose prediction module (30), configured to calculate a blood glucose prediction value after a current time node based on the change rate sequence, the change rate sequence, the maximum peak-to-valley difference of the blood glucose change curve, and the blood glucose value at the current time node; The second processing module (40) is configured to generate a second alarm signal and generate a control list for the blood glucose prediction value when it is determined that the blood glucose prediction value does not fall within the first preset range; the control list is used to indicate how to adjust blood glucose.

Images