WO2022083124A1

WO2022083124A1 - Personalized diabetes health management system and device, and storage medium

Info

Publication number: WO2022083124A1
Application number: PCT/CN2021/097230
Authority: WO
Inventors: 赵婷婷; 孙行智; 廖希洋; 赵惟; 徐卓扬
Original assignee: 平安科技（深圳）有限公司
Priority date: 2020-10-22
Filing date: 2021-05-31
Publication date: 2022-04-28
Also published as: CN112102953A; CN112102953B

Abstract

Provided are a personalized diabetes health management system and device, and a storage medium. The system comprises: a data collection unit used for acquiring the mode in which a patient is in and collecting blood glucose data of time-series changes of the patient in said mode, the mode comprising sleep, diet, exercise and leisure modes; a blood glucose prediction unit used for inputting the blood glucose data into a blood glucose prediction model to predict and output a predicted blood glucose value in a future period of time, wherein the blood glucose prediction model is obtained by training by means of using the patient's own data in the mode; and a comparison unit used for comparing the predicted blood glucose value with a preset abnormal blood glucose threshold, wherein if the predicted blood glucose value exceeds the abnormal blood glucose threshold, an early warning reminder is issued. Personalized diabetes health management can be achieved, the management is more comprehensive, and blood glucose warning information is more accurate.

Description

Personalized diabetes health management system, device and storage medium

This application claims the priority of the Chinese patent application with the application number 202011138904.2 and the invention titled "Personalized Diabetes Health Management System, Device and Storage Medium" filed with the China Patent Office on October 22, 2020, the entire contents of which are incorporated by reference in this application.

technical field

The present application relates to the field of artificial intelligence technology and the field of digital medicine, and in particular to a personalized diabetes health management system, equipment and storage medium.

Background technique

Diabetes is a long-term chronic disease that requires long-term adherence to treatment and management. Patients' daily behavior and self-management ability are one of the keys to diabetes control. Therefore, in addition to drug treatment, diabetes control requires individualized lifestyle interventions, including diet, exercise, and self-monitoring of blood sugar.

Among them, for diet: diabetic patients need to control their diet to maintain a standard body weight, correct metabolic disorders that have occurred, and reduce the burden on islet beta cells. The total daily intake of nutrients has a certain standard to help diabetic patients calculate and count the intake of various substances. Intakes help in the establishment of a healthy eating plan. For exercise: A personalized exercise program can promote diabetes patients to adhere to exercise therapy and improve their physical condition. For self-monitoring of blood sugar: patients may experience hypoglycemia during treatment. Hypoglycemia can lead to discomfort and even life-threatening, and is also the main obstacle to reaching the blood sugar target. Greatly reduces the risk of hypoglycemia in patients.

The inventor realizes that most of the existing diabetes management systems only have a blood glucose monitoring module, which lacks various lifestyle management. At the same time, the existing diabetes management systems lack individualization and cannot meet the needs of various types of diabetes patients. For example, Chinese patent CN110680340A discloses an all-weather blood glucose monitoring system for diabetes health management, which collects blood glucose data by setting a blood glucose acquisition module including a graphene flexible glucose sensor attached to the wrist, and then directly determines through the early warning module Whether the collected data falls into the preset alarm threshold, and then give an early warning, the early warning of this patent is the current early warning. Although it also discloses a diet monitoring module and a food blood sugar calculation module, it does not involve other aspects of management, and blood sugar early warning The accuracy of the results also needs to be improved.

SUMMARY OF THE INVENTION

Based on the above problems, the purpose of this application is to provide a personalized diabetes health management system, equipment and storage medium to solve the problems in the prior art that the diabetes health management system lacks individualization, management is not comprehensive, and blood glucose warning information is not accurate enough.

The above purpose is achieved through the following technical solutions:

According to one aspect of the present application, the present application provides a personalized diabetes health management system, comprising:

a data acquisition unit, configured to acquire the mode in which the patient is located, and to acquire the blood glucose data of the patient with time series changes in the mode, wherein the mode includes: a sleep mode, a diet mode, an exercise mode and a leisure mode;

A blood sugar prediction unit, configured to input the blood sugar data into a blood sugar prediction model for prediction, and output a blood sugar prediction value in a future period of time through the blood sugar prediction model, wherein the blood sugar prediction model is performed in the mode by adopting The patient's own data is trained;

A comparison unit, configured to compare the predicted blood glucose value with a preset blood glucose abnormality threshold, and issue an early warning reminder if the blood glucose predicted value exceeds the blood glucose abnormality threshold.

According to another aspect of the present application, an electronic device provided by the present application includes:

at least one processor; and,

a memory communicatively coupled to the at least one processor; wherein,

The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to execute the following elements and / or the function of the module:

According to yet another aspect of the present application, the present application further provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, realizes the following units and/or modules in the personalized diabetes health management system Features:

Compared with the prior art, the personalized diabetes health management system, device and storage medium of the present application are obtained based on technologies such as predictive analysis of artificial intelligence, health management and risk assessment of digital medicine. The application collects and monitors blood glucose data in four different modes, and inputs the blood glucose data in each mode into the blood glucose prediction model trained in this mode using the patient's own blood glucose data for prediction, and finally outputs the future time period. The blood sugar prediction value is compared with the preset threshold value, and an early warning signal is issued according to the comparison result, which makes the blood sugar management more comprehensive, and the blood sugar early warning result is more accurate. need.

Description of drawings

Other objects and results of the present application will be more apparent and readily understood by reference to the following description in conjunction with the accompanying drawings, and as the present application is more fully understood. In the attached picture:

1 is a schematic diagram of the use of the personalized diabetes health management system according to an embodiment of the present application;

FIG. 2 is a schematic diagram of the operation process framework of the personalized diabetes health management system according to an embodiment of the present application;

3 is a schematic diagram of a sliding window in a data processing module according to an embodiment of the present application;

4 is a schematic diagram of the framework structure of an embodiment of the present application using LSTM to build a blood glucose prediction model;

5 is a schematic diagram of a layer structure of a blood glucose prediction model constructed using LSTM according to an embodiment of the present application;

6 is a schematic structural diagram of a personalized diabetes health management system according to an embodiment of the present application;

FIG. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application:

The present application provides a personalized diabetes health management system, including: a data acquisition unit, a blood glucose prediction unit, and a comparison unit. The system realizes personalized diabetes health management, and the management is more comprehensive, and the blood sugar warning information is more accurate, which can promote patients to develop good living habits, thereby improving their health status.

In the present application, the data acquisition unit is used to acquire the mode in which the patient is located, and to acquire the blood glucose data of the patient that changes in time in the mode. The data acquisition unit may, for example, collect blood glucose data of time series in each mode by wearing a blood glucose meter by a patient. The mode may reflect the current activities performed by the patient, and the mode may include sleep mode, eating mode, exercise mode, leisure mode, and the like. In order to detect blood sugar reasonably, this application provides blood sugar intervention in multiple modes, covering the diet, exercise, sleep and rest modes of diabetic patients. Monitoring blood sugar in each mode can make blood sugar alarm information more accurate, so that health management and prediction results are more comprehensive and accurate, that is, providing a more comprehensive diabetes management model for diabetic patients, which is helpful to improve the self-management ability of patients. Foster healthy lifestyles that ultimately improve their clinical outcomes, health status, and quality of life. It should be noted that the mode described in this application is not limited to this, for example, the mode of medication may also be included.

The blood glucose prediction unit is used for inputting the blood glucose data into a blood glucose prediction model for prediction, and the blood glucose prediction model outputs a blood glucose prediction value in a future period of time, wherein the blood glucose prediction model is performed in the mode by using the patient It is obtained by training its own data, and a personalized prediction model is trained according to the personalized data, which fully reflects the personalized blood sugar management method, so as to realize the personalized diabetes health management system. The future period of time is a preset future period of time, such as 0.5 to 4 hours in the future. The blood glucose data input by the blood glucose prediction model is the blood glucose data of the time series change from time T to T+t, and the output blood glucose prediction value is the average value of the blood glucose prediction sequence from T+t to T+t+Δt, where Δt is the average value of the blood glucose prediction sequence. The specific value can be determined according to the patient's living habits and the characteristics of blood sugar changes, for example, it can be 3 to 4 hours. The blood glucose prediction model of the present application is obtained by training in four modes, the blood glucose prediction model is more robust, the prediction effect is more accurate, and the blood glucose management is more comprehensive; in addition, the blood glucose prediction model in each mode of the present application uses the corresponding mode It is obtained by training the patient's own blood sugar data, realizing personalized health management, and the prediction results are more accurate, meeting the needs of all kinds of diabetic patients.

The comparison unit is used for comparing the predicted blood glucose value with a preset blood glucose abnormality threshold, and if the blood glucose predicted value exceeds the blood glucose abnormality threshold, an early warning reminder is issued. The threshold for abnormal blood glucose varies with different activity modes, and it can also vary from person to person. In the comparison unit of this application, the threshold for abnormal blood glucose is preset through the threshold setting module. In addition, if the patient needs to customize the threshold, it can also pass the threshold Modify the settings module. The abnormal blood glucose threshold includes: the low blood sugar threshold G _low in any mode, the highest blood glucose threshold G _{high_relax} in the leisure mode, the highest blood glucose threshold G _{high_meal} in the eating mode, etc., and the highest blood glucose threshold in the eating mode is higher than The highest threshold of blood glucose in leisure mode, ie G _{high_meal} >G _{high_relax} . For example, set the blood sugar prediction value as y _predict , if y _predict <G _low , a low blood sugar warning is issued, where G _low is the low blood sugar threshold in any mode; in leisure mode, if y _predict > G _{high_relax} , a low blood sugar warning is issued. High reminder; in diet mode, if y _predict >G _{high_meal} , a high blood sugar reminder will be sent. Similarly, the lowest blood glucose threshold and the highest blood glucose threshold in the sleep mode and in the exercise mode can be preset through the threshold setting unit according to the patient's own conditions. The personalized model and personalized early warning threshold of the present application can meet the daily blood sugar management of diabetic patients, and avoid the risk of serious hypoglycemia events or hyperglycemia.

The personalized diabetes health management system in the embodiment of the present application will be described in detail below with reference to FIG. 1 and FIG. 2 . FIG. 1 schematically shows a schematic diagram of the use of the personalized diabetes health management system of the embodiment, and FIG. 2 schematically shows the operation process framework of the personalized diabetes health management system of the embodiment. Figure 1 includes: a mobile smart terminal (smart phone), a Bluetooth electronic scale, a continuous blood glucose meter and a smart bracelet connected through a network. Wherein, using the mobile intelligent terminal can complete the input of the patient's basic information and the calling of the blood glucose prediction model, and at the same time, the function setting of the system can be completed, and it is also used to send out alarm information. The smart bracelet can receive alarm information, and at the same time, it can record the movement data of the patient in the movement mode. The continuous blood glucose meter is used to dynamically collect real-time blood glucose data of a patient and upload it to a mobile intelligent terminal for data processing, model building, prediction and other operations. The bluetooth electronic scale with bluetooth function can be used to collect the patient's diet data and upload it to the mobile smart terminal. The blood sugar of the patient will fluctuate greatly within a day. For example, the blood sugar after meals may exceed 10mmol/L, and when hypoglycemia occurs, the blood sugar can be as low as 3.9mmol/L. The personalized diabetes health management system of this application can be used in different The blood sugar is monitored separately in the mode, and then predicted by the blood sugar prediction model in the corresponding mode, so that the blood sugar alarm prediction information is more accurate.

In an optional embodiment, the system may further include an input and call unit for inputting basic patient information, and calling a blood glucose prediction model trained by using the patient's own blood glucose data according to the patient's basic information. The basic information may include: patient name, gender, date of birth, body mass index BMI, and the like. As shown in Figure 2, the patient logs into the personalized diabetes health management system through a mobile smart terminal, and basic information needs to be entered when using it for the first time; then select the corresponding mode to train the blood glucose prediction model according to the patient's daily activities or the time period in between, or Directly enter the health management state to predict blood sugar through the blood sugar prediction model, and finally make an alarm reminder through mobile smart terminals or smart bracelets after the comparison. The present application further ensures that the blood glucose prediction model corresponds to the patient one-to-one by calling the blood glucose prediction model belonging to the individual patient (that is, the model obtained by using the patient's own data) according to the basic information of the patient, so that the subsequent blood glucose prediction results are more personalized. meet the needs of different patients.

In this application, the personalized diabetes health management system further includes: a model building unit for building blood glucose prediction models in different modes by using the LSTM algorithm. Among them, in the model building process, the model building process in each mode is the same, and the difference is the parameters for building the model and the data of the model (the data are different because they are collected under different moduli). In order to provide a personalized management plan, the present application proposes to use the patient's own blood glucose data to train the blood glucose prediction model, and use different prediction models in different modes. Added personalization so that blood glucose predictions not only behave differently on an individual basis, but also in different modes. LSTM (longshort-term memory) is very suitable for the prediction of time series series, that is, data that can process changes in the sequence. Figure 4 schematically shows the framework structure of LSTM. As shown in Figure 4, it is controlled by a forget gate z ^f , an input gate z ⁱ , and an output gate ^zo , selectively forget the input from the previous node, and selectively forget the input of this stage. Memory, add these two parts to get ct that is transmitted to the next state, the output stage will decide which information will be regarded as the output of the current state, controlled by zo, get ^h ^t ^. The final output y ^t is obtained by transforming h ^t . Among them, the expressions of c _t , h _t , and y _t are respectively: c ^t =z ^f ⊙c ^t-1 +z ⁱ ⊙z; y ^t =σ(W′h ^t ); h ^t =z ^o ⊙tanh (c ^t ). Fig. ₅ schematically shows the layer structure of LSTM, as shown in Fig. ₅ , including input layer (X1, X2... _Xt ), LSTM layer (A1, A2...At; h1, h2...ht ), and the output layer (y ₁ , y ₂ , ... y _t ); the input is the processed time series blood glucose data, and the output is the blood glucose data of the next period.

In an optional embodiment, the model building unit may include: a data processing module, a model training module, a model verification module, and the like.

Wherein, the data processing module is used to process the blood sugar data in a sliding window manner; wherein, the blood sugar data is the blood sugar data of the patient's own time series changes in each mode collected by the data acquisition unit. FIG. 3 schematically shows the structure of the sliding window, and the sliding window method is adopted, which can increase the training data. The length of the sliding window T represents the length of each time step s, which is the length of the input X. Further, the data processing module is also used for dividing the blood glucose data into two parts, one part is used for the model training module to train the blood glucose prediction model, and the other part is used for the model verification module to verify the blood glucose prediction model. For example, blood glucose data can be split into 80% for training and 20% for validation.

The model training module uses the processed blood glucose data of the patient to train the blood glucose prediction model; wherein, different blood glucose prediction models are obtained by training the blood glucose data in different modes. That is to say, in order to train a personalized blood glucose prediction model, the patient needs to use it for a period of time in advance, such as a month, and use the data acquisition unit to collect the blood glucose data of the patient's blood sugar changes in different modes for model training, so as to fully reflect the personalized In the same way, during the use process, the training of the blood glucose prediction models in different modes uses the blood glucose data of the patient in the corresponding mode. As for the mode, the present application can switch the mode according to the patient's own activities, and can also switch the mode according to the time. That is to say, the mode can be determined according to the activity performed by the patient when the blood glucose data is collected. For example, the patient can switch the mode according to the current activity, and combine the mode with the blood glucose data collected in this mode. Upload, so as to label the blood sugar data later, so as to improve the prediction accuracy of the model; of course, the mode can also be determined according to the time period when the blood sugar data is collected. For example, the blood sugar data can be determined according to the time period. The mode is set to the default state, that is, when the user does not switch the mode selection, the system switches the mode according to the time, for example, 00:00:00-06:00:00 belongs to the sleep mode.

The model verification module uses the processed blood glucose data of the patient to verify the blood glucose prediction model. If the error between the predicted value and the actual value is within the threshold, the blood glucose prediction model is constructed. That is, the error between the actual value and the predicted value of the blood glucose data in the next period is tested on 20% of the data. Specifically, the verification is carried out according to 20% of the blood glucose data. If the error is within the threshold (allowed range), the model verification is passed, then The blood glucose prediction model trained in the corresponding mode can be obtained. If the error is not within the threshold, it is necessary to continue training and verify the blood glucose prediction model until the model is verified and the trained model is obtained.

In an optional embodiment, the data collection unit further includes: a data labeling module. When the model is built, the data needs to be labeled during the data collection process, that is, the mode label and the abnormal occurrence label are required, so that the subsequent model training can be carried out. In this application, the data labeling module is used to label the mode in which the blood glucose data is located to obtain the mode label, and at the same time, it is used to label the abnormal situation of the blood glucose data to obtain the abnormality occurrence label, and the mode label and the abnormality label are obtained. The occurrence label is uploaded to the model building unit to construct and train the blood glucose prediction model. Further, the mode in which the blood glucose data is located may be marked according to the activities performed by the patient when the blood glucose data is collected, or according to the time period in which the blood glucose data is collected. In addition, if the user does not select the mode, the time period may be used by default to determine the mode of the blood glucose data. Among them, the mode label includes exercise mode label, sleep mode label, diet mode label, and leisure mode label. For example, the blood glucose data collected when the patient is exercising is labeled as the exercise mode label, and 00:00:00-06:00: The blood glucose data obtained in the 00 period is marked as the sleep mode label. The abnormality label may include: low blood sugar label, high blood sugar label, etc. For example, when low blood sugar or high blood sugar occurs, the user can return to the abnormal label through the smart device (for example, click the abnormal report button on the smart bracelet), In this way, the blood glucose data has the mode label and the abnormal occurrence label.

In an optional embodiment, the personalized diabetes health management system may further include: a calorie reminder unit, configured to acquire the type of food in the eating mode, calculate the calorie of the food, and determine whether the calorie of the food exceeds a preset calorie threshold, If it exceeds, a reminder of unreasonable diet will be issued. Specifically, when the patient eats, he or she can choose the type of food, including staple food, vegetables, fruit, meat, etc., after taking the weight with a Bluetooth electronic scale, calculate the calories of the food to remind the patient whether the diet is reasonable.

In another optional embodiment, the calorie reminder unit is configured to determine whether the calorie difference exceeds a preset calorie threshold, and if it exceeds, a reminder of unreasonable diet is issued. The calorie difference refers to the difference between the food calorie in the eating mode and the calorie consumed in the exercise mode.

FIG. 6 schematically shows the structure of the personalized diabetes health management system according to the embodiment of the present application. As shown in FIG. 6 , in the personalized diabetes health management system of the present application, when the model is constructed: the patient information is entered through the entry calling unit; The acquisition unit collects the blood glucose data of the same patient in different modes, and uses the data labeling module for labeling; and then enters the model building unit, uses the data processing module to process the data, and inputs the processed data to the model training module and model verification module. Carry out model training and verification, and finally obtain a trained blood glucose prediction model. When the model is applied: call the model matching the patient through the input calling unit, collect the blood glucose data of the patient in a certain mode through the data acquisition unit, input the blood glucose data into the blood glucose prediction unit, and pass the blood glucose prediction model (with the patient and the mode) model) to predict blood sugar, output the predicted value of blood sugar in a period of time in the future, and then compare it with the threshold value through the comparison unit to make an early warning reminder.

In the present application, after inputting the collected blood glucose data in a certain mode into the blood glucose prediction model trained by the patient's own blood glucose data in this mode for prediction, the blood glucose prediction value in the future time period can be output and based on the difference between the blood glucose prediction value and the threshold value The comparison results are used to issue early warning signals, which makes blood sugar management more comprehensive, and the blood sugar early warning results are more accurate. Personalized blood sugar management can be realized, which can meet the needs of all kinds of diabetic patients.

FIG. 7 is a schematic structural diagram of an electronic device implementing a management program in the present application. As shown in FIG. 7 , the electronic device 1 may include a processor 10, a memory 11 and a bus, and may also include a computer program stored in the memory 11 and executed on the processor 10, such as a management program 12 .

Wherein, the memory 11 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (for example: SD or DX memory, etc.), magnetic memory, magnetic disk, CD etc. The memory 11 may be an internal storage unit of the electronic device 1 in some embodiments, such as a mobile hard disk of the electronic device 1 . In other embodiments, the memory 11 may also be an external storage device of the electronic device 1, such as a pluggable mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital) equipped on the electronic device 1. , SD) card, flash memory card (Flash Card), etc. Further, the memory 11 may also include both an internal storage unit of the electronic device 1 and an external storage device. The memory 11 can not only be used to store application software installed in the electronic device 1 and various types of data, such as codes of management programs, etc., but also can be used to temporarily store data that has been output or will be output.

In some embodiments, the processor 10 may be composed of integrated circuits, for example, may be composed of a single packaged integrated circuit, or may be composed of multiple integrated circuits packaged with the same function or different functions, including one or more integrated circuits. Central Processing Unit (CPU), microprocessor, digital processing chip, graphics processor and combination of various control chips, etc. The processor 10 is the control core (Control Unit) of the electronic device, and uses various interfaces and lines to connect the various components of the entire electronic device, by running or executing programs or modules (such as management and control) stored in the memory 11. programs, etc.), and call data stored in the memory 11 to execute various functions of the electronic device 1 and process data.

The bus may be a peripheral component interconnect (PCI for short) bus or an extended industry standard architecture (Extended industry standard architecture, EISA for short) bus or the like. The bus can be divided into address bus, data bus, control bus and so on. The bus is configured to implement connection communication between the memory 11 and at least one processor 10 and the like.

FIG. 7 only shows an electronic device with components, and those skilled in the art can understand that it does not constitute a limitation on the electronic device 1, and may include fewer or more components than the one shown, or a combination of certain components may be included. some components, or a different arrangement of components.

For example, although not shown, the electronic device 1 may also include a power supply (such as a battery) for powering the various components, preferably, the power supply may be logically connected to the at least one processor 10 through a power management device, so that the power management The device implements functions such as charge management, discharge management, and power consumption management. The power source may also include one or more DC or AC power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, and any other components. The electronic device 1 may further include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

Further, the electronic device 1 may also include a network interface, optionally, the network interface may include a wired interface and/or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is usually used in the electronic device 1 Establish a communication connection with other electronic devices.

Optionally, the electronic device 1 may further include a user interface, and the user interface may be a display (Display), an input unit (eg, a keyboard (Keyboard)), optionally, the user interface may also be a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode, organic light-emitting diode) touch device, and the like. The display may also be appropriately called a display screen or a display unit, which is used for displaying information processed in the electronic device 1 and for displaying a visualized user interface.

It should be understood that the embodiments are only used for illustration, and are not limited by this structure in the scope of the patent application.

The management program 12 stored in the memory 11 in the electronic device 1 is a combination of multiple instructions, and when running in the processor 10, the functions of each unit and/or module can be implemented, for example: for obtaining a patient A data acquisition unit that collects the blood glucose data of the patient in the mode and the time series changes in the mode; used to input the blood glucose data into the blood glucose prediction model for prediction, and the blood glucose prediction model outputs the blood glucose prediction value for a period of time in the future The blood glucose prediction unit; a comparison unit for comparing the blood glucose prediction value with a preset blood glucose abnormality threshold, and issuing an early warning if the blood glucose prediction value exceeds the blood glucose abnormality threshold.

Specifically, for the specific implementation method of the above-mentioned instruction by the processor 10, reference may be made to the relevant description in the corresponding embodiment of FIG. 6, and details are not described herein.

Further, if the modules/units integrated by the electronic device 1 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium, and the computer-readable storage medium Can be non-volatile or volatile. Wherein, the computer-readable storage medium can be any tangible medium that contains or stores programs or instructions, on which a computer program that can be executed is stored. When the computer program is executed by the processor, the stored program instructs the relevant hardware through the stored program. The functions of each unit/module of the personalized diabetes health management system of the present application are realized. For example, the stored program instructs the relevant hardware to obtain the mode the patient is in and collect the blood glucose data of the patient's time-series changes in the mode, input the blood glucose data into the blood glucose prediction model for prediction, and the blood glucose prediction model outputs the future The predicted blood glucose value within a period of time, compares the predicted blood glucose value with a preset blood glucose abnormality threshold, and issues a warning reminder if the blood glucose predicted value exceeds the blood glucose abnormality threshold. The computer-readable medium, for example, may include: any entity or device capable of carrying the computer program code, recording medium, U disk, removable hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory) Memory).

In the several embodiments provided in this application, it should be understood that the disclosed devices and apparatuses may be implemented in other manners. For example, the apparatus embodiments described above are only illustrative. For example, the division of the modules is only a logical function division, and there may be other division manners in actual implementation.

The modules described as separate components may or may not be physically separated, and components shown as modules may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment.

In addition, each functional module in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware, or can be implemented in the form of hardware plus software function modules.

It will be apparent to those skilled in the art that the present application is not limited to the details of the above-described exemplary embodiments, but that the present application can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application.

Accordingly, the embodiments are to be regarded in all respects as illustrative and not restrictive, and the scope of the application is to be defined by the appended claims rather than the foregoing description, which is therefore intended to fall within the scope of the claims. All changes within the meaning and scope of the equivalents of , are included in this application. Any reference signs in the claims shall not be construed as limiting the involved claim.

Furthermore, it is clear that the word "comprising" does not exclude other units or steps and the singular does not exclude the plural. Several units or means recited in the system claims can also be realized by one unit or means by means of software or hardware. Second-class terms are used to denote names and do not denote any particular order.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application rather than limitations. Although the present application has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present application can be Modifications or equivalent substitutions can be made without departing from the spirit and scope of the technical solutions of the present application.

Claims

A personalized diabetes health management system, wherein the system includes:

a data acquisition unit, configured to acquire the mode in which the patient is located, and to acquire the blood glucose data of the patient with time series changes in the mode, wherein the mode includes: a sleep mode, a diet mode, an exercise mode and a leisure mode;

A blood sugar prediction unit, configured to input the blood sugar data into a blood sugar prediction model for prediction, and output a blood sugar prediction value in a future period of time through the blood sugar prediction model, wherein the blood sugar prediction model is performed in the mode by adopting The patient's own data is trained;

A comparison unit, configured to compare the predicted blood glucose value with a preset blood glucose abnormality threshold, and issue an early warning reminder if the blood glucose predicted value exceeds the blood glucose abnormality threshold.
The personalized diabetes health management system according to claim 1, wherein the system further comprises: a model building unit for building blood glucose prediction models in different modes by using the LSTM algorithm.
The personalized diabetes health management system according to claim 2, wherein the model building unit includes:

a data processing module for processing blood glucose data in a sliding window manner, wherein the blood glucose data is the blood glucose data of the patient's own time-series changes in each mode collected by the data acquisition unit;

The model training module uses the processed blood glucose data of the patient to train the blood glucose prediction model, wherein different blood glucose prediction models are obtained by training the blood glucose data in different modes;

The model verification module uses the processed patient's own blood glucose data to verify the blood glucose prediction model. If the error between the predicted value and the actual value is within the threshold, the blood glucose prediction model is constructed.
The personalized diabetes health management system according to claim 3, wherein the data processing module is further configured to divide the blood glucose data into two parts, one part is used for the model training module to train the blood glucose prediction model, and the other part is used for training the blood glucose prediction model. The model validation module validates the blood glucose prediction model.
The personalized diabetes health management system according to claim 1, wherein the data acquisition unit further comprises:

The data labeling module is used to label the mode of the blood glucose data to obtain the mode label, and to label the abnormal situation of the blood glucose data to obtain the abnormal occurrence label, and upload the mode label and the abnormal occurrence label to the model A construction unit is used to construct a blood glucose prediction model; wherein, according to the activities performed by the patient when the blood glucose data is collected, or the time period in which the blood glucose data is collected, the mode of the blood glucose data is marked.
The personalized diabetes health management system according to claim 1, wherein the system further comprises: an input and call unit for inputting the basic information of the patient, and calling and training the blood glucose data of the patient according to the basic information of the patient. blood glucose prediction model.
The personalized diabetes health management system according to claim 1, wherein the input blood sugar data of the blood sugar prediction model is blood sugar data of time series change from time T to T+t, and the output blood sugar prediction value is from T+t to T+t. The mean of the blood glucose prediction series for T+t+Δt.
The personalized diabetes health management system according to claim 1, wherein the abnormal blood sugar threshold includes: a low blood sugar warning threshold in any mode, a maximum blood sugar threshold in a leisure mode, and a maximum blood sugar threshold in a diet mode, Wherein, the highest blood glucose threshold in the eating mode is higher than the highest blood glucose threshold in the leisure mode.
An electronic device, wherein the electronic device comprises:

at least one processor; and,

a memory communicatively coupled to the at least one processor; wherein,

The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to execute the following elements in the personalized diabetes health management system function:

a data acquisition unit, configured to acquire the mode in which the patient is located, and to acquire the blood glucose data of the patient with time series changes in the mode, wherein the mode includes: a sleep mode, a diet mode, an exercise mode and a leisure mode;

A blood sugar prediction unit, configured to input the blood sugar data into a blood sugar prediction model for prediction, and output a blood sugar prediction value in a future period of time through the blood sugar prediction model, wherein the blood sugar prediction model is performed in the mode by adopting The patient's own data is trained;

A comparison unit, configured to compare the predicted blood glucose value with a preset blood glucose abnormality threshold, and issue an early warning reminder if the blood glucose predicted value exceeds the blood glucose abnormality threshold.
The electronic device of claim 9, wherein the at least one processor is further capable of performing the functions of the following units in the personalized diabetes health management system:

The model building unit is used to build blood glucose prediction models in different modes by using the LSTM algorithm.
The electronic device according to claim 10, wherein, when executing the function of the model building unit, it also executes the functions of the following modules:

a data processing module for processing blood glucose data in a sliding window manner, wherein the blood glucose data is the blood glucose data of the patient's own time-series changes in each mode collected by the data acquisition unit;

The model training module uses the processed blood glucose data of the patient to train the blood glucose prediction model, wherein different blood glucose prediction models are obtained by training the blood glucose data in different modes;

The model verification module uses the processed patient's own blood glucose data to verify the blood glucose prediction model. If the error between the predicted value and the actual value is within the threshold, the blood glucose prediction model is constructed.
The electronic device according to claim 11, wherein when performing the function of the data processing module, the blood glucose data is further divided into two parts, one part is used for the model training module to train the blood glucose prediction model, and the other part is used for model verification The module validates the blood glucose prediction model.
The electronic device according to claim 9, wherein, when the function of the data acquisition unit is performed, the function of the following modules is also performed:

The data labeling module is used to label the mode of the blood glucose data to obtain the mode label, and to label the abnormal situation of the blood glucose data to obtain the abnormal occurrence label, and upload the mode label and the abnormal occurrence label to the model. A construction unit is used to construct a blood glucose prediction model; wherein, according to the activities performed by the patient when the blood glucose data is collected, or the time period in which the blood glucose data is collected, the mode of the blood glucose data is marked.
The electronic device of claim 9, wherein the at least one processor is further capable of performing the functions of the following units in the personalized diabetes health management system:

The input calling unit is used for inputting the basic information of the patient, and calling the blood glucose prediction model obtained by training with the blood glucose data of the patient according to the basic information of the patient.
The electronic device according to claim 9, wherein the input blood sugar data of the blood sugar prediction model is blood sugar data of time series change from time T to T+t, and the output blood sugar prediction value is T+t to T+t+ The mean of the blood glucose prediction series for Δt.
The electronic device according to claim 9, wherein the abnormal blood glucose threshold includes: a low blood sugar warning threshold in any mode, a maximum blood glucose threshold in a leisure mode, and a maximum blood glucose threshold in a diet mode, wherein the The maximum blood glucose threshold in the eating mode was higher than that in the leisure mode.
A computer-readable storage medium storing a computer program, wherein, when the computer program is executed by a processor, the functions of the following units in a personalized diabetes health management system are realized:

a data acquisition unit, configured to acquire the mode in which the patient is located, and to acquire the blood glucose data of the patient with time series changes in the mode, wherein the mode includes: a sleep mode, a diet mode, an exercise mode and a leisure mode;

A blood sugar prediction unit, configured to input the blood sugar data into a blood sugar prediction model for prediction, and output a blood sugar prediction value in a future period of time through the blood sugar prediction model, wherein the blood sugar prediction model is performed in the mode by using The patient's own data is trained;

A comparison unit, configured to compare the predicted blood glucose value with a preset blood glucose abnormality threshold, and issue an early warning reminder if the blood glucose predicted value exceeds the blood glucose abnormality threshold.
The computer-readable storage medium according to claim 17, wherein, when the computer program is executed by the processor, it further implements the functions of the following units in the personalized diabetes health management system:

The model building unit is used to build blood glucose prediction models in different modes by using the LSTM algorithm.
The computer-readable storage medium according to claim 17, wherein, when the computer program is executed by the processor, the functions of the following modules in the data acquisition unit are further implemented:

The data labeling module is used to label the mode of the blood glucose data to obtain the mode label, and to label the abnormal situation of the blood glucose data to obtain the abnormal occurrence label, and upload the mode label and the abnormal occurrence label to the model A construction unit is used to construct a blood glucose prediction model; wherein, according to the activities performed by the patient when the blood glucose data is collected, or the time period in which the blood glucose data is collected, the mode of the blood glucose data is marked.
The computer-readable storage medium according to claim 17, wherein, when the computer program is executed by the processor, it further implements the functions of the following units in the personalized diabetes health management system:

The input calling unit is used for inputting the basic information of the patient, and calling the blood glucose prediction model obtained by training with the blood glucose data of the patient according to the basic information of the patient.