CN115662616A

CN115662616A - Critical patient intelligent blood sugar management system based on CGM and management method thereof

Info

Publication number: CN115662616A
Application number: CN202211301659.1A
Authority: CN
Inventors: 郭劲宏; 褚正康
Original assignee: Chongqing Lianxin Zhikang Biotechnology Co ltd
Current assignee: Chongqing Lianxin Zhikang Biotechnology Co ltd
Priority date: 2022-10-24
Filing date: 2022-10-24
Publication date: 2023-01-31

Abstract

The invention discloses an intelligent blood sugar management system for critical patients based on CGM, which comprises: the medical record statistical database is used for collecting medical record information of critical patients with hyperglycemia to generate a medical record statistical table; the intelligent analysis module carries out deep learning training through a medical record statistical table, analyzes the hyperglycemia morbidity information data of the critical patient to be treated before treatment within preset days, establishes a treatment history table of the critical patient to be treated, and generates first treatment scheme data; the retraining module is used for continuously analyzing the treatment history table of the critical patient in the treatment period based on the intelligent analysis module and continuously generating at least one round of subsequent treatment scheme data; the scheme of the invention effectively reduces the incidence of hyperglycemia of critically ill patients, improves the overall body function, and has important significance for exploring the pathology of the incidence of hyperglycemia.

Description

Critical patient intelligent blood sugar management system based on CGM and management method thereof

Technical Field

The invention relates to the technical field of blood sugar management, in particular to an intelligent blood sugar management system for critically ill patients based on CGM and a management method thereof.

Background

The incidence of hyperglycemia in critically ill patients during hospitalization is as high as 80.00%, wherein the incidence of hyperglycemia (> 10 mmol/L) per day in average critically ill patients is even as high as 32.20%, and hyperglycemia causes disorders in the water, electrolyte and acid-base balance of patients, damages cardiac muscle, brain and liver tissues, increases the infection risk of patients, and leads to increased complication incidence and mortality, prolonged hospitalization time and increased need for post-hospital care. The management of blood sugar for critically ill patients has become an important part of the management of critically ill patients. Effective blood glucose management requires the assistance of accurate blood glucose monitoring and normative scientific insulin infusion protocols.

The regulation and control are mainly carried out according to a paper-edition blood sugar infusion scheme after bedside rapid blood sugar detection in the traditional clinical practice, so as to maintain the blood sugar level of critically ill patients within a proper range. Currently, advanced technologies (for example, patent application No. CN 202011353708.7) adopt a precise algorithm control to achieve precise regulation of insulin, so as to achieve strict control of blood sugar of critical patients. Whether according to a paper edition or a high-precision algorithm, the aim is to quickly reduce the blood sugar of a patient and control the blood sugar within a reasonable range. The applicant finds, through a large amount of clinical data, that by adopting the above two methods, the frequency of daily hyperglycemia of a patient after insulin and nutrient infusion is not affected, and even the daily incidence of hyperglycemia tends to increase as the condition worsens.

The applicant discovers that the insulin injection amount, the insulin injection frequency and the disease frequency of a large number of patients have subtle mutual influence, the daily hyperglycemia incidence of the patients can be reduced by 3.64% -4.98% through slightly high frequency and low insulin injection amount in the period without disease, how to reasonably control the insulin injection frequency and the insulin injection amount, the purpose of reducing the hyperglycemia incidence of the patients is the problem which is continuously solved by the industry at present, and the method has important significance for researching the pathology of the hyperglycemia incidence.

Disclosure of Invention

Aiming at the problem of reasonably controlling the insulin injection frequency and the insulin injection amount, the invention provides the CGM-based intelligent blood sugar management system for the critically ill patients, which realizes the continuous detection of the blood sugar of the critically ill patients, generates a targeted insulin injection scheme and reduces the incidence of hyperglycemia of the critically ill patients.

Critical patient intelligent blood sugar management system based on CGM includes:

the medical record statistical database is used for collecting medical record information of critical patients with hyperglycemia to generate a medical record statistical table;

the intelligent analysis module carries out deep learning training through a medical record statistical table, analyzes the hyperglycemia morbidity information data of the critical patient to be treated before treatment within preset days, establishes a treatment history table of the critical patient to be treated, and generates first treatment scheme data;

and the retraining module is used for continuously analyzing the treatment history table of the critical patient in the treatment period based on the intelligent analysis module and continuously generating at least one round of subsequent treatment scheme data.

The system collects a large number of hyperglycemia morbidity medical records of critically ill patients in a hospital period in history through the medical record statistical database, meanwhile, after the system is applied, the medical records of critically ill patients after treatment through the system are also taken as the hyperglycemia morbidity medical records and are included in the medical record statistical database, the hyperglycemia morbidity mode is presented as truly as possible to the greatest extent through continuously increasing the number of the medical records, deep learning training is carried out on a medical record statistical table through the intelligent analysis module, then initial treatment is carried out on new critically ill patients, the retraining module carries out analysis according to the treatment medical record of the patients in the initial treatment period of the initial treatment to generate initial subsequent treatment scheme data, and the latest subsequent treatment scheme data are continuously generated according to the treatment medical record data collected in the treatment period.

The intelligent portable terminal is used for acquiring the patient condition development prediction model after the adjustment module receives the hyperglycemia morbidity information data continuously collected by the detected patient, the continuously updated patient basic information and the prediction logic adjustment; the intelligent portable terminal is also used for acquiring picture information of the patient when eating, judging the type of food of the patient when eating according to the picture information, generating a first prediction result of blood sugar according to the type of the food, then acquiring sound information of the eating process to judge eating sequence information of the user, correcting prediction logic for the first prediction result according to the eating sequence information, obtaining a blood sugar prediction result of the patient after eating this time according to the corrected prediction logic, and sending the eating blood sugar prediction result to the intelligent analysis module; the intelligent analysis module is also used for generating first treatment scheme data according to the blood glucose prediction result after eating.

In the scheme, the picture of food intake is acquired through the intelligent portable terminal, and the type of the food intake of the patient can be known. The type of food has a large influence on the change of blood sugar, so that the change range of blood sugar can be well predicted by combining the type of food. But in addition to the kind of food having a greater influence on blood glucose, the sequence of eating may also have a greater influence on the outcome of blood glucose. For example, in the common combination of carbohydrate (staple food) + dish, the intake timing of carbohydrate (staple food) will have a great influence on blood sugar. The mode of taking dishes firstly and then taking staple food can effectively reduce the peak value of blood sugar in the absorption process. According to the scheme, the feeding order can be effectively judged through the collection of feeding sounds, so that a more accurate blood sugar prediction result is obtained. The influence of the eating condition of the patient on the blood sugar can be predicted, the prediction result is more accurate, and the data of the first treatment scheme can be more fit with the actual condition of the patient.

Further, the medical record information includes: the incidence of hyperglycemia, a blood sugar detection value and an insulin injection amount within a preset number of days are preset;

the pre-treatment hyperglycemia onset information data comprises: the number of times of hyperglycemia, blood glucose detection value and insulin injection amount of a critical patient to be treated every day.

Furthermore, the treatment history table is used for recording the hyperglycemia morbidity information data of the critically ill patient to be treated before treatment, recording the first treatment history data and continuously generating at least one round of subsequent treatment history data.

Further, the first treatment protocol data includes: the first treatment injection frequency and the first treatment injection amount of insulin;

the follow-up treatment protocol data comprises: the injection frequency of the subsequent treatment of the insulin and the injection quantity of the subsequent treatment of the insulin.

Furthermore, the retraining module acquires corresponding treatment case history data through the first treatment scheme data and the subsequent treatment scheme data,

wherein, the treatment medical record data comprises: the daily actual insulin infusion frequency, the actual insulin infusion amount, the incidence frequency of hyperglycemia and the blood sugar level of critically ill patients during the treatment period.

Further, the continuous analysis of the treatment history of the critically ill patient in the treatment cycle by the retraining module based on the intelligent analysis module comprises:

the retraining module analyzes the treatment history table updated each time to generate subsequent treatment scheme data;

the retraining module collects the first treatment medical record data of the critical patient in the treatment period, continuously generates at least one round of subsequent treatment medical record data, and continuously updates the treatment medical record.

Further, the retraining module generates first-round subsequent treatment plan data according to treatment medical record data corresponding to the first-time treatment plan data.

The management method of the CGM-based critical patient intelligent blood sugar management system comprises the following steps:

s1, acquiring medical record information of critical patients with hyperglycemia to generate a medical record statistical table;

s2, analyzing the pre-treatment hyperglycemia morbidity information data of the critical patient to be treated within preset days, establishing a treatment history table of the critical patient to be treated, and generating first treatment scheme data;

and S3, analyzing the treatment history table of the critically ill patient in the treatment period, and continuously generating at least one round of subsequent treatment scheme data.

Further, the step S3 includes:

s301, analyzing the updated treatment history table every time to generate subsequent treatment scheme data;

s302, collecting first treatment medical record data of a critical patient in a treatment period, continuously generating at least one round of subsequent treatment medical record data, and continuously updating a treatment medical record table.

Compared with the prior art, the invention has the advantages and beneficial effects that: by adopting the scheme of the invention, the critically ill patients are continuously monitored, different critically ill patients are treated by adjusting the injection frequency and the injection amount of the insulin in a targeted manner, the incidence of hyperglycemia of the critically ill patients is effectively reduced, the overall body function is improved, and the method has important significance for exploring the pathology of the incidence of hyperglycemia.

Drawings

FIG. 1 is a block diagram of the system of the present invention;

FIG. 2 is a flow chart of the method of the present invention;

FIG. 3 is a flowchart of step S3 of the method of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings by way of specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

Example 1

the intelligent analysis module performs deep learning training through a medical record statistical table, analyzes the hyperglycemia morbidity information data of the critical patient to be treated before treatment within preset days, establishes a treatment history table of the critical patient to be treated, and generates first treatment scheme data;

Specifically, the medical record information includes: the incidence of hyperglycemia, the blood sugar detection value and the insulin injection amount within the preset days are preset.

The medical record information is used for counting the medical records of the hyperglycemia during the hospitalization period of a large number of critical patients in history, the critical patients do not use the scheme of the system, and initial analysis logic is established through the disease analysis of the critical patients in history.

Specifically, the information data of hyperglycemia before treatment includes: the incidence of hyperglycemia, blood sugar detection value and insulin injection amount of the critically ill patient to be treated every day.

Pre-treatment hyperglycemic incidence information data was taken as a statistical history of the disease manifestations once at the time of hyperglycemic incidence for each critically ill patient.

Specifically, the treatment history table is used for recording hyperglycemia morbidity information data of the critically ill patient to be treated before treatment, recording first treatment history data and continuously generating at least one round of subsequent treatment history data.

The treatment schedule is used for counting the critical patients who receive the treatment scheme generated by the system, and a specific treatment scheme is generated according to the occurrence condition of hyperglycemia of each critical patient based on the performance of treatment effect during treatment.

Specifically, the first treatment protocol data includes: the first treatment injection frequency of insulin and the first treatment injection amount of insulin;

follow-up treatment protocol data included: the injection frequency of the insulin for the subsequent treatment and the injection amount of the insulin for the subsequent treatment.

Specifically, the retraining module acquires corresponding treatment case history data through the first treatment scheme data and the subsequent treatment scheme data,

The retraining module acquires corresponding treatment case history data through the first treatment scheme data and the subsequent treatment scheme data and records the treatment case history data into a treatment case history table, the incidence of the critical patient under the treatment scheme generated by the system is recorded, and the subsequent treatment scheme data generated by the subsequent retraining module is also generated based on the continuously updated treatment case history table.

Specifically, the analysis of the treatment history table of the critically ill patient in the treatment period by the retraining module based on the intelligent analysis module comprises the following steps:

the retraining module collects the first treatment medical record data of the critical patient in the treatment period, continuously generates at least one round of subsequent treatment medical record data and continuously updates the treatment medical record table.

The retraining module analyzes the treatment history table of the critically ill patient during treatment through deep learning training of the intelligent analysis module, updates treatment scheme data according to the updating frequency of the treatment history table to reach the end of each treatment period, and can generate the latest treatment scheme data according to the previous treatment period and all treatment history data in the whole treatment period.

Specifically, the retraining module generates first-round subsequent treatment plan data according to treatment medical record data corresponding to the first-time treatment plan data.

and S3, analyzing the treatment history table of the critical patient in the treatment period, and continuously generating at least one round of subsequent treatment scheme data.

Specifically, step S3 includes:

In practical application, after a critical patient receives treatment, blood sugar values of the critical patient after injection need to be continuously detected, different critical patients have certain difference to the performance of the critical patient after insulin injection, physical conditions of the critical patient are further known through the blood sugar value performance of the critical patient, and the conditions about insulin receiving influence in practical application include:

plasma glucose concentration is the most important factor affecting insulin secretion. Insulin release is a biphasic response following oral or intravenous glucose administration. The early rapid phase, the insulin in the portal blood plasma reaches the maximum value within 2 minutes and then rapidly decreases; slow phase, plasma insulin level gradually rising after 10 minutes, lasting more than 1 hour. The early fast phase shows that glucose promotes the release of stored insulin, the delayed slow phase shows insulin synthesis and proinsulin conversion.

After eating a protein-rich diet, the blood amino acid concentration increases and insulin secretion increases. Arginine, lysine, leucine and phenylalanine all have strong effects of stimulating insulin secretion.

Increasing gastrointestinal hormone after meal intake can promote insulin secretion, such as gastrin, secretin, pepstatin, and vasoactive intestinal peptide.

Promoting insulin secretion when the autonomic nervous function state vagus nerve is excited; when sympathetic nerve is excited, insulin secretion is inhibited.

It should be noted that, in practical application, the dose of insulin required by a critical patient is often larger than that of a normal patient, and when the critical patient faces the critical patient, the problem of insulin tolerance of the critical patient needs to be particularly noticed, so that acute tolerance is often caused by stress states such as concurrent infection, trauma, operation, emotional agitation and the like. When insulin resistance in blood is increased or ketoacidosis is caused, glucose uptake and utilization are hindered by the presence of large amounts of free fatty acids and ketone bodies in blood. In the case of acute tolerance, insulin doses are increased by thousands of units over a short period of time. The cause of chronic tolerance is complex (it means that more than 200U of insulin is needed daily and there is no complication). It may be that anti-insulin receptor antibodies (AIRA) are produced in vivo, for which immunosuppressors may be used to control the symptoms and restore normal insulin sensitivity to the patient; it may also be a change in the number of insulin receptors, such as a decrease in the number of insulin receptors on the target cell membrane in hyperinsulinemia; it may also be a malfunction of the glucose transport system on the target cell membrane. In this case, it is usually effective to use insulin from other animals or high-purity insulin, and to adjust the dosage appropriately.

The adjustment of the insulin infusion dosage for the critically ill patient during the treatment process needs to be remarked in a special form in the medical record of treatment so as to provide more reference for the subsequent treatment of the critically ill patient.

Subcutaneously injected insulin is stored subcutaneously. As subcutaneous storage increases, fluctuations in absorption increase and the net absorption decreases. This can have a significant impact on patients who require multiple bolus injections per day due to insulin resistance. One reason for the smooth glycemic control of continuous subcutaneous infusion of insulin is that it uses only normal or fast acting insulin (the latter being preferred), with small subcutaneous storage, because the drug is stored in a syringe or other container outside the body and, in addition, syringe injections have inherent variability, including differences in the location, angle, depth, and underlying blood flow of each injection, which can be reduced by having each catheter fixed in the same location during continuous subcutaneous infusion.

Among these, insulin toxicity is also a concern for critically ill patients who, due to the high frequency and overall high dose of injected insulin, have a much higher likelihood of insulin being severe than normal patients.

Example 2

Compared with the embodiment 1, the intelligent portable terminal is different in that the intelligent portable terminal further comprises an intelligent portable terminal and a weight acquisition device which is in signal connection with the intelligent portable terminal, wherein the intelligent portable terminal is used for acquiring the patient condition development prediction model which is obtained by the adjustment module after the adjustment module receives the hyperglycemia morbidity information data which is continuously collected by the detected patient, the continuously updated patient basic information and the prediction logic are adjusted; the weight acquisition device is used for acquiring weight information of current food and current weighing time information; the intelligent portable terminal is also used for acquiring picture information of the patient during eating, judging the type of food of the patient during eating according to the picture information, generating a first prediction result of blood sugar according to the type of the food, then acquiring sound information of the eating process to judge eating sequence information of the user, correcting prediction logic on the first prediction result according to the eating sequence information (namely according to the influence coefficient of the stored eating sequence on the blood sugar), obtaining a blood sugar prediction result of the patient after eating this time according to the corrected prediction logic, and sending the blood sugar prediction result after eating to the intelligent analysis module; the intelligent analysis module is also used for generating first treatment scheme data according to the blood glucose prediction result after eating.

When the intelligent portable terminal is used specifically, the intelligent portable terminal is a smart phone of a patient, pictures of food to be eaten are taken through the smart phone, then the pictures are uploaded to a network, and the types of the food in the pictures are identified through a trained neural network model (the intelligent portable terminal is mature technology and is not repeated, functions similar to photographing and identifying objects are provided, and the lower section of the intelligent portable terminal is specifically exemplified) so as to be used as a basis (of course, the characteristics of the patient are also combined), and a first prediction result is obtained.

In the actual use process, the situation that the accuracy rate of identifying the food types through the pictures is low is also considered. However, in combination with the actual dietary notes of the user (mostly diabetics), the cooked forms of green vegetables, corn and the like are easy to identify. The food which has a large influence on the blood sugar of the user is the food with high content of starch and sugar. The recognition range is relatively limited, and can be roughly divided into: meat, eggs, rice, noodles, vegetables and the like, so that the overall identification accuracy and the use experience can be ensured.

Then, the eating sequence of the user is judged according to the collected sound information of the eating process, and the collection of the technology can be directly realized through an audio collection module of a Bluetooth headset (which is carried out on the premise of obtaining the permission of the user, and can be realized through the permission of a microphone permitted by the user in other embodiments) which is in signal communication with the smart phone. Collecting such sound information can be used as a basis for determining the order of eating. The weight collecting device is a Bluetooth scale connected with the smart phone through signals.

The specific means can still refer to the neural network model to match the timbre in the sound information with the sound characteristics of different foods during eating, so that the food types during eating can be identified. And then the prediction logic is corrected for the first prediction result according to the feeding sequence information, and the blood sugar prediction result of the patient after the current feeding is obtained according to the corrected prediction logic. And presented to the patient via the display screen. Of course, the information can be sent to and developed by the disease condition analysis module to provide basis for the basic model of disease condition development in the disease condition analysis module.

In the embodiment, the influence of the eating condition of the patient on the blood sugar can be predicted, the prediction result is more accurate, and the data of the first treatment scheme can be matched with the actual condition of the patient.

Example 3

Compared with the embodiment 1, the difference is that the intelligent portable terminal is also used for acquiring picture information of the patient when the patient eats, judging the type of food of the patient when the patient eats according to the picture information, generating an input box to acquire a correction opinion of the type of the food by the user, generating an input box of eating sequence information of the user according to the change speed of the weight information of the weight acquisition device in the eating process, correcting the prediction logic of the first prediction result according to the eating sequence information, and obtaining a blood glucose prediction result of the patient after the patient eats according to the corrected prediction logic.

When the intelligent food recognition method is used specifically, if food recognition is carried out only by taking pictures through the intelligent mobile phone, generally, only specific food can be recognized, modeling needs to be carried out on all food to be recognized, cooking methods are different, and a certain deviation exists in the recognition success rate in the stage of the existing algorithm. Such food products are difficult to identify accurately if they are deliberately made into watermelon-like bread. After the picture is taken, food which is wanted to be ingested is placed on the weight acquisition device to be weighed, the food on the weight acquisition device is taken out for eating during eating, weight information on the weight acquisition device can be changed at the moment, and the ingestion speed of the food of a user can be identified through the change speed. Namely, the user can eat the food for multiple times after weighing the food for one time; or all the ingredients can be eaten after being weighed every time. The weight acquisition device acquires the reduction amount at each time and also acquires the time and the time interval for generating the reduction amount at each time, so that the food intake speed of the user can be accurately acquired.

Go to judge that there is certain discernment degree of difficulty in the food classification through bluetooth headset in last embodiment 2, the leading cause lies in, what bluetooth headset can gather is that the sound information of the information of chewing and bone conduction that transmits in the air, because of the problem of eating habit, the sound that some personnel sent in the eating process is less and also there is environmental disturbance, the kind information of gathering may be comparatively difficult. Therefore, in the embodiment, the correct opinion input by the user is used for performing more accurate recognition, so that the recognition accuracy is ensured.

And most importantly, food needs to be taken in a matched manner, so that the weight acquisition device and the method for filling food contents by self are used for realizing the food filling method, and the technical realization difficulty is reduced.

In addition to the intake amount, the intake speed of food also affects the change of blood sugar, so in this embodiment, the intake speed is represented by collecting the change speed of the weight information of the weight collecting device during eating, and the change of blood sugar can be predicted more accurately.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a," "8230," "8230," or "comprising" does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

The foregoing is a more detailed description of the present invention that is presented in conjunction with specific embodiments, and the practice of the invention is not to be considered limited to those descriptions. For those skilled in the art to which the invention pertains, numerous and varied simplifications or substitutions may be made without departing from the spirit of the invention, which should be construed as falling within the scope of the invention.

Claims

1. CGM-based critical patient intelligent blood sugar management system is characterized by comprising:

2. The CGM-based intelligent blood sugar management system for critical patients according to claim 1, further comprising an intelligent portable terminal, wherein the intelligent portable terminal is used for acquiring the disease state development prediction model after the adjustment module receives the data of hyperglycemia onset information, the continuously updated patient basic information and the prediction logic adjustment, which are continuously collected by the detected patient; the intelligent portable terminal is also used for acquiring picture information of the patient when eating, judging the type of food of the patient when eating according to the picture information, generating a first prediction result of blood sugar according to the type of the food, then acquiring sound information of the eating process to judge eating sequence information of the user, correcting prediction logic for the first prediction result according to the eating sequence information, obtaining a blood sugar prediction result of the patient after eating this time according to the corrected prediction logic, and sending the eating blood sugar prediction result to the intelligent analysis module; the intelligent analysis module is also used for generating first treatment scheme data according to the blood glucose prediction result after eating.

3. The CGM-based critical patient intelligent blood glucose management system according to claim 1, wherein the medical record information comprises: the incidence of hyperglycemia, a blood glucose detection value and an insulin injection amount within a preset day range are preset;

the pre-treatment hyperglycemia onset information data comprises: the incidence of hyperglycemia, blood sugar detection value and insulin injection amount of the critically ill patient to be treated every day.

4. The CGM-based intelligent blood glucose management system for critically ill patients according to claim 1, wherein the treatment history table is used for recording information data of hyperglycemia onset before treatment of critically ill patients to be treated, recording first treatment history data and continuously generating at least one round of subsequent treatment history data.

5. The CGM-based critically ill patient intelligent blood glucose management system according to claim 1, wherein the first treatment protocol data comprises: the first treatment injection frequency of insulin and the first treatment injection amount of insulin;

6. The CGM-based critically ill patient intelligent blood glucose management system according to claim 4, wherein the retraining module obtains corresponding treatment case history data through first treatment protocol data and subsequent treatment protocol data,

7. The CGM-based critical patient intelligent blood glucose management system of claim 6, wherein the retraining module based on intelligent analysis module continuously analyzing the treatment history of critical patients during the treatment cycle comprises:

8. The CGM-based critically ill patient intelligent blood glucose management system according to claim 7, wherein the retraining module generates first round of subsequent treatment protocol data according to treatment medical record data corresponding to the first treatment protocol data.

9. The management method of CGM-based critically ill patient intelligent blood glucose management system according to claims 1-8, characterized by comprising the steps of:

10. The CGM-based critical patient intelligent blood glucose management method according to claim 9, wherein the step S3 comprises: