CN113270204A

CN113270204A - Method for predicting initial dose of insulin pump

Info

Publication number: CN113270204A
Application number: CN202110624360.9A
Authority: CN
Inventors: 荣曦
Original assignee: Individual
Current assignee: Individual
Priority date: 2021-06-04
Filing date: 2021-06-04
Publication date: 2021-08-17

Abstract

The invention provides a method for predicting an initial dose of an insulin pump, which belongs to the crossing field of diabetes treatment technology and computer technology, and is based on the clinical data and machine learning technology of a large number of diabetics receiving the insulin pump to treat, after data cleaning, data transformation and data correction, a reasonable training set and a testing set are established, supervised machine learning is carried out by utilizing an error back propagation neural network algorithm to establish prediction models under different application scenes, the initial dose of the insulin pump can be accurately predicted through the prediction models, the accuracy of the initial dose can reach more than 90 percent, the technical problems that in the existing medical technology, the professional threshold of the insulin pump is high, the initial dose of the insulin pump is calculated only by personal experience of a clinician, the dose is difficult to accurately estimate are solved, and the technical problems that the blood sugar of the diabetics can be finely adjusted, The popularization of the insulin pump for strengthening the blood sugar reduction treatment makes a beneficial contribution.

Description

Method for predicting initial dose of insulin pump

Technical Field

The invention belongs to the cross field of diabetes treatment technology and computer technology, and particularly relates to a method for predicting initial dose of an insulin pump.

Background

As a big country with diabetes, the incidence of diabetes is on the rise year by year. According to investigation, 1 diabetic patient exists in every 11 adults in China on average, and the disease seriously harms the health and life of people. The insulin pump is an insulin input device controlled by a microcomputer, and simulates physiological insulin secretion mode to the maximum extent in a mode of continuous subcutaneous insulin infusion, so that an insulin treatment method for better controlling blood sugar is achieved.

The insulin pump enters the Chinese market for more than 20 years, and according to survey, the number of long-term users of the insulin pump currently exceeds 4 thousands in China, wherein 44% of the users are type 1 diabetes, 54% of the users are type 2 diabetes, and the rest 2% of the users are diabetes caused by other reasons. Over 3000 hospitals now develop insulin pump therapy, and it is speculated that over a million patients receiving short-term insulin pump therapy. However, the technical threshold for installing and using insulin pumps is high, especially when the insulin pump is used for the first time, and the dosage of insulin required by the patient needs to be estimated.

According to the recommendation of 'Chinese insulin pump treatment guideline' 2014 (physicians 'division of endocrinologists of China physician's Association of endocrinology, endocrinology division of China medical society, diabetes and diabetes division of China medical society, Chinese insulin pump treatment guideline (2014 edition) selected (upper) [ J ]. diabetes clinic, 2014,8(8):353 and 359.), the daily insulin dosage calculation is determined according to the diabetes type, the blood sugar level and the weight condition of a patient. The initial recommended dose is as follows: calculation of insulin doses for patients not treated with insulin are set according to the insulin doses for the different diabetes types:

type 1 diabetes mellitus: total weight (U) ═ body weight (kg) × (0.4 to 0.5) per day

Type 2 diabetes: total weight (U) ═ body weight (kg) × (0.5 to 1.0) per day

During the use process, personalized dose adjustment is carried out according to the blood sugar monitoring level. A patient who has received insulin therapy can be calculated from the amount of insulin used before the insulin pump therapy. The method can be specifically determined according to the blood sugar control condition of a patient, personalized dose adjustment is carried out according to the blood sugar monitoring level in the using process, and the conversion of the insulin dose when a hypoglycemic therapist changes to an insulin pump for therapy is accepted:

1) if the blood sugar control condition before using the insulin pump is good and no hypoglycemia exists, the recommended dose during insulin pump treatment is the total amount of insulin before using the insulin pump, x (75-80%);

2) if hypoglycemia frequently occurs in the blood sugar control condition before the insulin pump is used, the recommended dose during insulin pump treatment is the total amount of insulin before the pump is used, which is x (70-75%);

3) if the blood sugar control condition before using the insulin pump is hyperglycemia, little or no hypoglycemia occurs, the recommended dose during insulin pump treatment is the total amount of insulin before using the pump x (95-100%);

it follows that the initial dosage of insulin pump for diabetic patients is not formulated with an accurate calculation formula and is estimated by the experience of the physician. The formulation of an accurate insulin dose is crucial to treatment, and the accurate insulin dose not only can quickly relieve the toxic effect of continuous hyperglycemia of a patient, enhance the confidence of the patient in overcoming diseases, but also can avoid hypoglycemia. However, in clinical work, even an endocrine professional doctor can hardly give accurate insulin dosage every time, the insulin dosage is too much or too little, the risk of hypoglycemia and blood sugar fluctuation is increased, the difficulty of blood sugar regulation is increased, the compliance of a patient is reduced, and resistance is brought to follow-up doctor-patient cooperation. This severely limits the use of insulin pumps and the therapeutic advantages that can be realized.

Disclosure of Invention

In view of this, the present invention provides a method for predicting an initial dose of an insulin pump, which has an accuracy of more than 90%, for the technical problem that the initial dose of the insulin pump of a diabetic patient cannot be accurately estimated.

In order to realize the purpose of the invention, the invention adopts the following technical scheme:

a method for predicting an initial dose of an insulin pump is characterized by comprising the following steps:

1) acquiring original data: collecting a large amount of physical information, disease information, diabetes treatment information and biochemical inspection information of a diabetic patient who has been treated by an insulin pump previously;

2) processing raw data: performing data cleaning, data transformation and data correction on the acquired original data to construct an effective database;

3) training and testing set construction: and (4) enabling the effective database to be as follows: 4 to 8:2, dividing the ratio into a training set and a test set;

4) BP neural network training: taking the training sample of the training set as a training object of the BP neural network, taking the test sample of the test set as a verification object of the model, and carrying out supervised machine learning training;

5) constructing a prediction model: adjusting the fitting degree of the model, and constructing a prediction model based on the recognition training result meeting the accuracy requirement;

6) prediction of insulin dose: and inputting the body information, disease information, diabetes treatment information and biochemical inspection information of the new diabetes patient into the prediction model, and acquiring the initial recommended dose of the insulin pump as the output of the prediction model.

Preferably, the physical information includes: one or more of gender, age, height, weight; the condition information includes: one or more of diabetes typing, course of diabetes, complications of diabetes; the diabetes treatment information includes: one or more of previous insulin use cases and previous non-insulin hypoglycemic drug use cases; the biochemical test information includes: one or more of random blood glucose before insulin pump treatment, fasting blood glucose the next day after insulin pump treatment, glycated hemoglobin, fasting and postprandial C-peptide levels.

Preferably, the previous insulin hypoglycemic drug conditions include: the brand of a factory which uses the insulin in the past, the classification, the usage and the dosage according to the action time of the medicine, the brand of the insulin which is used in the past comprises one or more of quick-acting, short-acting, middle-acting, long-acting, super-long-acting and premixing, the usage comprises the daily usage times, and the dosage comprises the international unit value of the insulin which is used each time; previous non-insulin hypoglycemic drug conditions include: previous biguanide drug use and dosage, previous alpha glucosidase inhibitor use and dosage, previous sulfonylurea drug use and dosage, previous sodium-glucose cotransporter 2 inhibitor use and dosage, previous dipeptidyl peptidase IV inhibitor use and dosage, previous glinide drug use and dosage, previous thiazolidinedione drug use and dosage, and previous glucagon-like peptide-1 receptor agonist use and dosage.

Preferably, the diabetic complication condition comprises ketoacidosis, lactic acidosis, hyperglycemic hyperosmolar syndrome, diabetes-related infection, hypoglycemia, atherosclerotic cardiovascular disease, hypertension, diabetic foot, diabetic retinopathy, diabetic peripheral neuropathy, diabetic autonomic neuropathy, and diabetic nephropathy condition.

Preferably, the prediction model comprises an untreated drug model and a treated drug model, wherein: the input variables of the non-drug treatment model comprise body information, disease information, diabetes treatment information and biochemical inspection information, wherein the diabetes treatment information is a default value, and the output of the non-drug treatment model is an initial recommended dose of the insulin pump; the input variables of the used drug therapy model comprise body information, disease information, diabetes therapy information and biochemical test information, wherein the diabetes therapy information comprises the use condition of the used insulin and the use condition of other non-insulin hypoglycemic drugs, and the output of the used drug therapy model is the initial recommended dose of the insulin pump.

Preferably, the neural network recognition training is BP artificial neural network training, and specifically includes the following steps: setting the number N of layers and the number M of nodes of hidden neurons in the multilayer perceptron, wherein: n is more than or equal to 2 and less than or equal to 3, M is more than or equal to 5 and less than or equal to 35, a training stopping rule of the multilayer perceptron is set to be that the error cannot be further reduced, a training sample of a training set is used as a training object of the multilayer perceptron, a test sample of a test set is used as a verification object of the multilayer perceptron, an artificial neural network is used for identifying and training and judging whether the accuracy rate meets the requirement, when the accuracy rate at least meets 90%, a prediction model is built based on an identification training result meeting the accuracy rate requirement, and when the accuracy rate does not meet 90%, the parameters of the multilayer perceptron are adjusted again and retrained.

Preferably, the parameter adjustment method of the multilayer sensor includes the following steps:

1) setting the initial value of the number N of layers of hidden neurons in the multilayer perceptron as a minimum value and setting the initial value of the number M of nodes as a minimum value;

2) performing supervised machine learning training based on the number N of layers and the number M of nodes, and judging whether the accuracy rate meets the requirement;

3) when the accuracy meets the requirement, stopping parameter adjustment;

4) when the accuracy rate does not meet the requirement, increasing the node number M, and judging whether the node number M exceeds the maximum value:

when the node number M does not exceed the maximum value, performing the step 2);

and when the node number M exceeds the maximum value, increasing the layer number N, setting the node number M as the minimum value, and performing the step 2).

Preferably, the data cleaning in the raw data processing process includes the following steps: checking and correcting the original data, and discarding incomplete data when missing values exist in the original data; when the abnormal value exists in the original data, the abnormal value is corrected by referring to the original medical record.

Preferably, the data transformation in the process of processing the original data specifically includes the following steps: other non-insulin hypoglycemic agents including different types of oral hypoglycemic agents and glucagon-like peptide-1 receptor agonist injections are normalized and converted into equivalent insulin doses.

Preferably, the data rectification in the raw data processing process includes the following steps:

1) type 2 diabetic patients are evaluated according to the Chinese guide for preventing and treating type 2 diabetes for the disease information of diabetic patients, and different levels of blood sugar control targets are formulated according to different clinical conditions of patients, wherein the control targets comprise general control, strict control and loose control, and the following steps are included:

the general control is suitable for most non-pregnant adult type 2 diabetes patients, namely the fasting blood sugar of the diabetes patients is controlled to be 4.4-7.0 mmol/l, and the glycosylated hemoglobin HbA1c is less than 7%;

the strict control is suitable for type 2 diabetes with short course of disease, long life expectancy, no complication and no ASCVD, and the fasting blood sugar of a diabetic patient is controlled to be 4.4-6.0 mmol/l, and the glycosylated hemoglobin HbA1c is less than 6.5%;

the loose control is suitable for patients with serious hypoglycemia history, short expected life and obvious microvascular and macrovascular complications, and is characterized in that the fasting blood sugar of the diabetic is controlled to be 5.6-8.0 mmol/l, and the glycated hemoglobin HbA1c is less than 8%;

similarly, type 1 diabetic and gestational diabetic patients may determine individualized glycemic control goals based on the age of the patient and the risk of hypoglycemia, with reference to corresponding clinical guidelines.

2) Based on the actual insulin dosage of the first day of insulin pump treatment of the diabetic patient in the blood sugar control target and the second sky abdominal blood sugar value after insulin pump treatment in the database, correcting the original insulin pump basic dosage of the diabetic patient in the database, wherein:

when the blood sugar control target is "general control", the correction method is as follows:

if the second sky abdominal blood sugar value after the insulin pump treatment is less than or equal to 3.9, the basic amount of the original insulin pump of the diabetic patient in the database is reduced by 4.8U;

if the second sky abdominal blood sugar value is less than or equal to 4.4 after the insulin pump treatment is more than 3.9, the original insulin pump basic amount of the diabetic patient in the database is reduced by 2.4U;

if the second sky abdominal blood sugar value is less than or equal to 7.0 after the insulin pump treatment is more than 4.4, the original insulin pump basic amount of the diabetic patient in the database is kept unchanged;

if the second sky abdominal blood sugar value after the insulin pump treatment is more than 7.0 and less than or equal to 8.0, the original insulin pump basic amount of the diabetic patient in the database is increased by 2.4U;

if the second sky abdominal blood sugar value after the insulin pump treatment is more than 8.0, the basic amount of the original insulin pump of the diabetic patient in the database is increased by 4.8U;

when the blood sugar control target is "strict control", the correction method is as follows:

if the second sky abdominal blood sugar value is less than or equal to 6.0 after 4.4 < insulin pump treatment, the original insulin pump basic amount of the diabetic patient in the database is kept unchanged;

if the second sky abdominal blood sugar value is less than or equal to 8.0 after the insulin pump treatment is more than 6.0, the original insulin pump basic amount of the diabetic patient in the database is increased by 2.4U;

when the blood sugar control target is "loose control", the correction method is as follows:

if the second sky abdominal blood sugar value is less than or equal to 5.6 after the insulin pump treatment is more than 3.9, the original insulin pump basic amount of the diabetic patient in the database is reduced by 2.4U;

if the second sky abdominal blood sugar value after the insulin pump treatment is more than 5.6 and less than or equal to 8.0, the original insulin pump basic amount of the diabetic patient in the database is kept unchanged;

if the second sky abdominal blood sugar value after the insulin pump treatment is more than 8.0 and less than or equal to 9.0, the original insulin pump basic amount of the diabetic patient in the database is increased by 2.4U;

if the second sky abdominal blood sugar value after the insulin pump treatment is more than 9.0, the basic amount of the original insulin pump of the diabetic patient in the database is increased by 4.8U;

specifically, when the patient is gestational diabetes, the correction method comprises the following steps:

if the second sky abdominal blood sugar value after the insulin pump treatment is less than or equal to 2.8, the basic amount of the original insulin pump of the diabetic patient in the database is reduced by 4.8U;

if the second sky abdominal blood sugar value after 2.8 < insulin pump treatment is less than 4.0, the original insulin pump basic amount of the diabetic patient in the database is reduced by 2.4U;

if the second sky abdominal blood sugar value is less than or equal to 5.3 after the insulin pump treatment is more than or equal to 4.0, keeping the original insulin pump basic amount of the diabetic patient in the database unchanged;

if the second sky abdominal blood sugar value after the insulin pump treatment is more than 5.3 and less than or equal to 7.0, the original insulin pump basic amount of the diabetic patient in the database is increased by 2.4U;

if the second sky abdominal blood sugar value after the insulin pump treatment is more than 7.0, the basic amount of the original insulin pump of the diabetic patient in the database is increased by 4.8U;

preferably, the insulin pump initial recommended dose comprises a basal insulin amount and a prandial insulin amount, the basal insulin amount and the prandial insulin amount being in a ratio of 1:1 and both employing a mean subcutaneous infusion rate, wherein: the insulin basic quantity adopts a multi-section structure, and the multi-section structure comprises one of a three-section type, a four-section type or a five-section type.

The method for predicting the initial dose of the insulin pump has the following beneficial effects:

the method for predicting the initial dose of the insulin pump is based on a large amount of clinical data and machine learning technology for receiving the insulin pump to treat diabetes, establishes a reasonable training set and a reasonable testing set after data cleaning, data transformation and data correction, and performs supervised machine learning by using a BP neural network algorithm to establish prediction models in different application scenes, so that the initial dose of the insulin pump can be accurately predicted through the prediction models. The dosage treatment can enable the blood sugar of a patient to quickly reach a control target, reduce the times of repeatedly adjusting insulin due to inaccurate initial dosage, reduce the risk of hypoglycemia of the patient, and ensure the prediction accuracy to be more than 90%. The technical problems that in the prior medical technology, the professional threshold for using the insulin pump is high, the initial dose of the insulin pump is calculated only by the personal experience of a clinician, and the insulin pump is difficult to accurately measure are solved, and the method makes a beneficial contribution to the fine adjustment of the blood sugar of a diabetic patient and the popularization of the insulin pump strengthened blood sugar reduction treatment.

Drawings

FIG. 1 is a schematic overall flow diagram of the present invention;

FIG. 2 is a schematic diagram of raw data processing in the present invention;

FIG. 3 is a schematic diagram of the artificial neural network recognition training of the present invention;

FIG. 4 is a schematic diagram of a model constructed according to the present invention;

FIG. 5 is a schematic flow chart of the present invention;

FIG. 6 is a graph showing the relationship between the basal amount of insulin and the amount of insulin at meal time in the present invention.

Detailed Description

In the description of the present invention, it is to be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience of description and simplicity of description, and do not indicate or imply that the referenced devices or elements must have a particular orientation, be constructed and operated in a particular orientation, and thus, are not to be construed as limiting the present invention. The terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present invention, "a plurality" means two or more unless otherwise specified.

Example one

As shown in fig. 1, the method for predicting the initial dose of the insulin pump comprises the following steps:

2) processing raw data: and performing data cleaning, data transformation and data rectification on the acquired original data to construct an effective database, as shown in fig. 2.

Specifically, the data cleaning includes checking and correcting the original data, and when missing values exist in the original data, the incomplete data are discarded; when the abnormal value exists in the original data, the abnormal value is corrected by referring to the original medical record.

Specifically, the data transformation comprises the step of carrying out normalization processing on other non-insulin hypoglycemic medicines to transform the other non-insulin hypoglycemic medicines into equivalent insulin dosage, wherein the other non-insulin hypoglycemic medicines comprise different types of oral hypoglycemic medicines and glucagon-like peptide-1 receptor stimulant injections.

Specifically, the data correction comprises the evaluation of disease information of the diabetic patient, a blood sugar control target is formulated according to clinical conditions of different patients, and meanwhile, the original insulin pump basic quantity of the diabetic patient in the database is corrected based on the blood sugar control target, the actual insulin dosage of the diabetic patient in the first day of insulin pump treatment in the database and the second sky abdominal blood sugar value after insulin pump treatment.

specifically, 6:4 to 8:2 is a ratio range, and 6:4, 7:3, 8:2, etc. can be selected, such as: the proportion of training and validation samples was 70% to 30%.

4) BP neural network training: taking the training sample of the training set as a training object of the BP neural network, taking the test sample of the test set as a verification object of the model, and carrying out supervised machine learning training; the BP neural networks are all referred to as error back propagation neural networks.

6) prediction of insulin dose: inputting body information, disease information, diabetes treatment information and biochemical inspection information of a new diabetic into a prediction model, and obtaining an initial recommended dose of an insulin pump as the output of the prediction model, wherein: the initial recommended dose of the insulin pump refers to the daily insulin dosage and the subcutaneous infusion rate, including the insulin basal amount and the insulin meal time amount, which are set in the device before the insulin pump is connected to the body of a diabetic who needs to be treated by the insulin pump for a short time or a long time.

Specifically, after the prediction model training finishes, run into the new diabetes mellitus patient who needs to use the insulin pump, only need input relevant information, can acquire the initial suggestion dosage of insulin pump, and clinician can combine patient actual conditions according to artificial intelligence's suggestion, and nimble increase and decrease insulin dosage formulates the initial dosage of insulin pump more accurately.

It should be noted that, the user needs to update the original database periodically, and the more data in the database, the higher the prediction accuracy. Secondly, the patients in storage are selected as much as possible to have general representativeness, and special patients with complicated storage conditions (such as diabetic patients who have undergone major operations, enter an intensive care unit and have various acute and chronic complications) are avoided, so that the data in the database have general representativeness.

Particularly, for diabetes patients with critical illness, postoperative patients, pregnant patients, old people of advanced age, teenagers and children, and the like, which are complicated or have special age groups, an independent database is established to group a certain number of cases meeting conditions, and an independent prediction model is established to carry out neural network training and prediction.

Finally, because type 2 is the most common type of diabetes in diabetes typing, gestational diabetes is the second type of diabetes, and type 1 diabetes and special type diabetes are relatively rare, a certain number of cases of various types of diabetes in a database must be guaranteed in the training of the model so as to guarantee the generalization capability of the training model in practical prediction.

Example two

In this embodiment, the original data includes body information, disease information, diabetes treatment information, and biochemical examination information of a diabetic patient who has been treated by an insulin pump, and the information is obtained by way of inquiry, physical examination, and laboratory blood drawing examination, and the units are international standard units, where:

1) the physical information includes: one or more of gender, age, height, and weight.

2) The disease condition information includes: one or more of diabetes typing, course of diabetes, complications of diabetes;

in particular, the diabetes classification includes type 1 diabetes, type 2 diabetes, gestational diabetes and other special type diabetes, and the diabetes course includes the age limit.

It should be noted that the diabetes complication information includes whether or not ketoacidosis, lactic acidosis, hyperglycemic hyperosmolar syndrome, diabetes-related infection, hypoglycemia, atherosclerotic cardiovascular disease (ASCVD), hypertension, diabetic foot, diabetic retinopathy, diabetic peripheral neuropathy, diabetic autonomic neuropathy, and diabetic nephropathy is present.

3) The diabetes treatment information includes: one or more of previous insulin use cases and previous non-insulin hypoglycemic drug use cases;

the previous insulin use cases include: the brand, the brand type, the brand usage and the brand dosage of the past insulin use factory, the brand type of the past insulin use factory comprises one or more of quick acting, short acting, medium acting, long acting, super long acting and premixing, the usage comprises the times of daily use, specifically comprises three short times and one long time, one time before sleep, premixing for multiple times a day, short acting for multiple times before meal and the like, and the dosage is an international unit.

It should be noted that the conditions of the non-insulin hypoglycemic drugs in the past include: previous biguanide drug use and dosage, previous alpha glucosidase inhibitor use and dosage, previous sulfonylurea drug use and dosage, previous sodium-glucose cotransporter 2 (SGLT-2) inhibitor use and dosage, previous dipeptidyl peptidase IV (DPPIV) inhibitor use and dosage, previous glinide drug use and dosage, previous thiazolidinedione drug use and dosage, and previous glucagon-like peptide-1 (glucagon-like peptide-1, GLP-1) receptor agonist use and dosage.

4) The biochemical test information includes: one or more of random blood glucose before insulin pump treatment, fasting blood glucose the next day after insulin pump treatment, glycated hemoglobin, fasting and postprandial C-peptide levels. Specifically, the fasting blood glucose in the fasting blood glucose on the second day after the insulin pump treatment refers to the blood glucose measured before breakfast in the next morning after fasting for more than 8-12 hours every night.

It should be noted that the actual values of the continuous variable and the classification variable are summarized into the original data, and the data is updated regularly, so that the number of patients in the database is enriched continuously.

EXAMPLE III

In this embodiment, the data transformation in the raw data processing includes normalizing other non-insulin hypoglycemic drugs to an equivalent insulin dosage, and the other non-insulin hypoglycemic drugs include different types of oral hypoglycemic drugs and GLP-1 receptor agonist injections. The normalization processing aims at normalizing the diabetes treatment information into a specific and quantized equivalent insulin quantity, then determining the theoretical dosage of the insulin pump according to the value of the equivalent insulin quantity, and inputting the theoretical dosage into a training model.

The diabetes mellitus has a plurality of treatment medicines except insulin, the hypoglycemic medicines with the same mechanism have different types and specifications, and the hypoglycemic medicines with the same type are subjected to equivalent normalization when entering a model. For example: acarbose and voglibose belong to the same alpha-glucosidase inhibitor, and 50 mg of acarbose is equivalent to 0.2 mg of voglibose in the normalization according to the principle of one tablet for one tablet. For example, most patients use acarbose in 50 mg tablets, whereas occasionally some patients use voglibose in 0.2 mg tablets, the input variable being entered as 50 mg of acarbose, a commonly used drug. The details are shown in the following table.

Example four

The actual situation is that statistical analysis is carried out on collected original data of insulin pump treatment for diabetes in the trimethyl hospital, the clinical effect of the insulin pump after initial dose treatment actually formulated by endocrine professional doctors is taken as a research object, and the clinical effect is divided into 'control excellent', 'control in' and 'control poor' according to the deviation of the actual fasting blood glucose of a patient from a blood glucose control target after the patient receives the dose treatment. Wherein, the fasting blood glucose in the blood glucose control target interval is defined as 'control optimal'; a deviation of blood glucose from the control target interval, but less than 2mmol/l, is defined as "in control", and an occurrence of hypoglycemia or a deviation from the upper limit of the control target by 2mmol/l is defined as "poor control". The results show that the blood glucose results in the original data are 21.3% excellent, 32.5% excellent and 46.2% poor. The results show that even the endocrine professional in the third hospital, given the patient's initial insulin dose empirically, only a few patients achieve a second sky fasting glucose optimum. Therefore, the initial dose of the insulin pump actually prescribed by the endocrine professional in the raw data is mostly not the originally optimal dose for that patient. The physical, pathological, diabetic and biochemical information of each patient are different, and theoretically, each patient corresponds to an independent optimal dosage. The optimal dosage is a necessary condition for supervising learning (super learning) in model training and improving the prediction accuracy of the model. The existing basic amount of the original insulin pump in the original data needs to be corrected to be used as a sample for machine learning, the step is the key for improving the prediction accuracy, the evaluation needs to be carried out according to the disease information of the diabetic patient, the blood sugar control target of the patient is determined according to the clinical conditions of different patients, and the basic amount of the original insulin pump in the database is corrected.

The existing theoretical dose formulation process is divided into two specific cases:

1) calculation of insulin doses for patients not treated with insulin are set according to the insulin doses for the different diabetes types:

type 1 diabetes mellitus: an insulin pump theoretical dose (U) is weight (kg) × (0.4-0.5);

type 2 diabetes: the theoretical dosage (U) of insulin pump is weight (kg) × (0.5-1.0).

2) A patient who has received medication is pumped with the theoretical pump insulin dose (U) (total amount of insulin used throughout the day + equivalent dose of non-insulin hypoglycemic agent) multiplied by a corresponding factor as described in the background.

The actual clinical endocrinologist formulates the initial process of insulin pump: the dose is set according to the theory, and the actually formulated insulin dose is obtained by combining the increase and decrease of clinical experience.

Namely: the clinician sets the insulin pump basic amount as the theoretical dose plus the clinical experience increment and decrement

Specifically, the type 2 diabetic can evaluate the disease information of the diabetic according to the Chinese guide for the prevention and treatment of type 2 diabetes, and the blood sugar control target is formulated according to the clinical conditions of different patients, wherein the control target comprises general control, strict control and loose control. Wherein the general control requires that the glycosylated hemoglobin HbA1c is less than 7%, the strict control requires that the glycosylated hemoglobin HbA1c is less than 6.5%, and the loose control requires that the glycosylated hemoglobin HbA1c is less than 8%. Similarly, type 1 diabetic patients and gestational diabetic patients define individualized glycemic control goals according to the corresponding clinical guidelines.

The principle and process of insulin pump basal volume correction is as follows:

the initial insulin dose has the following logical relationship with the next day of fasting glucose: initial insulin dose (basal amount) appropriate → fasting glycemia is excellent the next day; the initial insulin dose (basal amount) is slightly larger or smaller → the second day of fasting plasma glucose, the initial insulin dose (basal amount) is excessive or insufficient → the second day of fasting plasma glucose is poor. Therefore, we need to correct the data of the original insulin pump basic amount in fasting blood sugar and poor patients in the database before training.

Based on the actual insulin dosage of the first day of insulin pump treatment of the diabetic patient in the blood sugar control target and the second sky abdominal blood sugar value after insulin pump treatment in the database, correcting the original insulin pump basic dosage of the diabetic patient in the database, wherein:

in addition, when the patient is gestational diabetes, the correction method is as follows:

it should be noted that, by correcting, the original insulin initial dose in the database is corrected up and down according to the actual blood sugar deviation from the control target. The scheme provides a logic scheme for large-data batch processing, but the blood sugar control target cannot be generalized and is individualized due to the diversity and complexity of diabetes diseases. In order to further improve the accuracy, the diabetes specialist can perform manual review and re-correction, the senior professional endocrine specialist can look over the original insulin pump basic amount, the insulin amount during meal, the blood sugar before meal and at night, the treated fasting blood sugar and the specific medical history retrospectively, and the clinical experience is combined to perform manual adjustment and education, so that the insulin pump basic amount in the database is just like the guidance of the specialist, and the optimal dosage of the patient is approached as much as possible after optimization.

EXAMPLE five

In this embodiment, in the training of the BP neural network, a training sample of a training set is used as a training object of the BP neural network, a test sample of a test set is used as a verification object of a model, and supervised machine learning training is performed, as shown in fig. 3, the method specifically includes the following steps:

setting the number N of layers and the number M of nodes of hidden neurons in the multilayer perceptron, wherein: n is more than or equal to 2 and less than or equal to 3, M is more than or equal to 5 and less than or equal to 35, a training stopping rule of the multilayer perceptron is set to be that the error cannot be further reduced, a training sample of a training set is used as a training object of the multilayer perceptron, a test sample of a test set is used as a verification object of the multilayer perceptron, an artificial neural network is used for identifying and training and judging whether the accuracy rate meets the requirement, when the accuracy rate at least meets 90%, a prediction model is built based on an identification training result meeting the accuracy rate requirement, and when the accuracy rate does not meet 90%, the parameters of the multilayer perceptron are adjusted again and retrained.

It should be noted that the parameter adjustment method of the multilayer perceptron includes the following steps:

3) when the accuracy meets the requirement, stopping parameter adjustment;

Specifically, the parameter adjustment method can traverse all parameters within the range until the accuracy rate meets 90% or more.

In this embodiment, the number N of layers of hidden neurons in the multilayer perceptron is set to 2, the number M of nodes is set to 30, and the training stopping rule of the multilayer perceptron is set to be such that the error cannot be further reduced, and the accuracy of the training stopping rule can reach 93.5%.

EXAMPLE six

The construction of the prediction model comprises the steps of adjusting the fitting degree of the model and constructing the prediction model based on the recognition training result meeting the accuracy requirement.

In this embodiment, the prediction model includes an untreated drug model and a treated drug model, as shown in fig. 4, where:

the input variables of the non-drug treatment model comprise body information, disease information, diabetes treatment information and biochemical inspection information, wherein the diabetes treatment information is a default value, and the output of the non-drug treatment model is an initial recommended dose of the insulin pump;

the input variables of the used drug therapy model comprise physical information, disease information, diabetes therapy information and biochemical test information, wherein the diabetes therapy information comprises the use condition of the previous insulin and the use condition of other non-insulin hypoglycemic drugs, and the output of the used drug therapy model is the initial recommended dose of the insulin pump.

It should be noted that the required data can be obtained through on-site inquiry, physical examination and laboratory blood drawing examination, and the initial recommended dosage of the insulin pump can be obtained through model calculation, as shown in fig. 5.

EXAMPLE seven

The insulin pump initial recommended dose includes a basal amount of insulin and a prandial amount of insulin, the general ratio of the basal amount of insulin to the prandial amount of insulin being 1:1, and both employing a mean subcutaneous infusion rate, wherein: the insulin basic amount adopts a multi-section structure, and the multi-section structure comprises one of a three-section structure, a four-section structure or a multi-section structure. Therefore, after acquiring the insulin basal amount, the insulin meal time amount can be matched quickly, and the proportional relation between the insulin basal amount and the insulin meal time amount is shown in fig. 6.

The invention selects the insulin pump basic quantity as the machine learning object based on the following theoretical facts:

1. the suitability of the insulin pump basal amount and the control of the fasting blood sugar have certain logical relations. The control range of fasting blood glucose in the blood glucose control target in the 'guide' has strict limitation (4.4-7.0 mmol/l), and the requirement on postprandial blood glucose (< 10mmol/l) is relatively wide.

2. For patients with poor glycemic control, i.e., patients requiring short or long term treatment with an insulin pump, fasting glucose "contributes" more to the elevation of glycated hemoglobin than postprandial glucose.

3. Based on the clinical experience of high water rising and low water falling, if the fasting blood sugar can be reduced to a normal level, the postprandial blood sugar is easy to control, if the fasting blood sugar is very high, the postprandial blood sugar can reach the standard difficultly, only if the fasting blood sugar and the postprandial blood sugar are well controlled, the glycosylated hemoglobin can reach the standard, and the control of the fasting blood sugar is the primary consideration of blood sugar reduction treatment.

4. Insulin pump commonly used insulin aspart has half-life period of about 90 minutes, and after the insulin dose is acted during meal, the effect disappears quickly, which is not enough to influence the next day of fasting blood glucose. While the insulin pump basal volume is a continuous micro pump, which can reach a pharmacological steady state and affect the fasting glucose the next day for the patient.

Therefore, the invention adopts the strategy of firstly determining the basic amount by artificial intelligence and then matching the meal time according to the formula.

It should be noted that the usage of the insulin pump is relatively flexible, the basic amount means that the insulin pump will pump a small amount of insulin into the body every hour when the patient does not eat food in daily life to simulate the normal insulin secretion mode, each time period can have different dosage, and the maximum 24 hours can be different every hour, namely divided into 24 segments. For adjustment of treatment, 3 or 4 segments are used in clinical work, and the principle is as follows: generally, 8am to 22 pm are the major active hours of a person's day, with the most insulin required; the time from 22 to 3 a.m. is in the sleep period, the insulin requirement is minimum, the time from 3 a.m. to 8 a.night is dawn, the physiological function is gradually recovered, and the insulin requirement is gradually increased. Similarly, some hospitals can also adopt 4 sections and multiple sections in one day, and the principle is that the hourly dosage of the basic amount of each section cannot be greatly different, and the total amount of the basic amount of insulin and the meal time of insulin are close to each other.

Taking a three-stage example, all insulin basal amounts are fixed three-stage time-stage: "8 am-10 pm", "10 pm-3 am", "3 am-8 am". For example, if the base amount of an insulin pump for one patient is 10.5U, the three-stage distribution is "8 am-10pm 0.5U/h", "10 pm-3am 0.3U/h", and "3 am-8am 0.4U/h", as shown in the following table.

It should be noted that the table is only a common example, and the basic amount of each adjustment is increased by 0.1U every hour all day and 24 hours a day, so that the difference between two adjacent basic amounts is 2.4U, and similarly, the basic amount can also be increased by 0.2U every hour, so that 4.8U is increased in 24 hours, and each hospital can have its own experience summary at present, but the general principle is not changed.

Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting an initial dose of an insulin pump is characterized by comprising the following steps:

5) constructing a prediction model: adjusting the fitting degree of the model, and constructing a prediction model based on the training result meeting the accuracy requirement;

2. The method of claim 1, wherein said method comprises:

the body information includes: one or more of gender, age, height, weight;

the condition information includes: one or more of diabetes typing, course of diabetes, complications of diabetes;

the diabetes treatment information includes: one or more of previous insulin use cases and previous non-insulin hypoglycemic drug use cases;

the biochemical test information includes: one or more of random blood glucose before insulin pump treatment, fasting blood glucose the next day after insulin pump treatment, glycated hemoglobin, fasting and postprandial C-peptide levels.

3. The method of claim 2, wherein said method comprises:

the conditions of the previous insulin hypoglycemic drugs comprise: the brand of a factory using the insulin in the past, the classification, the usage and the dosage according to the action time of the medicament, the type of the insulin used in the past comprises one or more of quick-acting, short-acting, intermediate-acting, long-acting, super-long-acting and premixing, the usage comprises the daily usage times, and the dosage comprises the international unit value of the insulin used each time;

the previous non-insulin hypoglycemic drug conditions include: previous biguanide drug use and dosage, previous alpha glucosidase inhibitor use and dosage, previous sulfonylurea drug use and dosage, previous sodium-glucose cotransporter 2 inhibitor use and dosage, previous dipeptidyl peptidase IV inhibitor use and dosage, previous glinide drug use and dosage, previous thiazolidinedione drug use and dosage, and previous glucagon-like peptide-1 receptor agonist use and dosage.

4. The method of claim 2, wherein said diabetic complication comprises ketoacidosis, lactic acidosis, hyperglycemic hyperosmolar syndrome, diabetes related infections, hypoglycemia, atherosclerotic cardiovascular disease, hypertension, diabetic foot disease, diabetic retinopathy, diabetic peripheral neuropathy, diabetic autonomic neuropathy, and diabetic nephropathy.

5. The method of claim 2, wherein said predictive models comprise an untreated model and a treated model, wherein:

the input variables of the used drug therapy model comprise body information, disease information, diabetes therapy information and biochemical test information, wherein the diabetes therapy information comprises the use condition of the used insulin and the use condition of other non-insulin hypoglycemic drugs, and the output of the used drug therapy model is the initial recommended dose of the insulin pump.

6. The method for predicting initial dose of insulin pump according to claim 1, wherein the neural network training is based on BP neural network algorithm training, comprising the following steps:

setting the number N of layers and the number M of nodes of hidden neurons in the multilayer perceptron, wherein: n is more than or equal to 2 and less than or equal to 3, M is more than or equal to 5 and less than or equal to 35, a training stopping rule of the multilayer perceptron is set to be that the error cannot be further reduced, a training sample of a training set is used as a training object of the multilayer perceptron, a test sample of a test set is used as a verification object of the multilayer perceptron, a BP neural network is used for training and judging whether the accuracy rate meets the requirement or not, when the accuracy rate at least meets 90%, a prediction model is built based on the recognition training result meeting the accuracy rate requirement, and when the accuracy rate does not meet 90%, the parameters of the multilayer perceptron are adjusted again and retrained.

7. The method for predicting initial dose of insulin pump according to claim 6, wherein the method for adjusting parameters of the multi-layer sensor comprises the following steps:

3) when the accuracy meets the requirement, stopping parameter adjustment;

8. The method of claim 3, wherein said method comprises:

the data cleaning in the original data processing process comprises the following steps:

checking and correcting the original data, and discarding incomplete data when missing values exist in the original data; when the original data has an abnormal value, correcting the abnormal value by referring to the original medical record;

the data transformation in the process of processing the original data specifically comprises the following steps:

other non-insulin hypoglycemic agents including different types of oral hypoglycemic agents and glucagon-like peptide-1 receptor agonist injections are normalized and converted into equivalent insulin doses.

9. The method of claim 1, wherein said data correction during raw data processing comprises the steps of:

the general control is to control the fasting blood sugar of the diabetic patient to be 4.4-7.0 mmol/l, and the glycosylated hemoglobin HbA1c is less than 7%;

the strict control is to control the fasting blood sugar of the diabetic patient to be 4.4-6.0 mmol/l, and the glycosylated hemoglobin HbA1c is less than 6.5%;

the loose control is to control the fasting blood sugar of the diabetic patient to be 5.6-8.0 mmol/l, and the glycosylated hemoglobin HbA1c is less than 8%; 2) based on the actual insulin dosage of the first day of insulin pump treatment of the diabetic patient in the blood sugar control target and the second sky abdominal blood sugar value after insulin pump treatment in the database, correcting the original insulin pump basic dosage of the diabetic patient in the database, wherein:

if the second sky abdominal blood glucose value after the insulin pump treatment is more than 9.0, the original insulin pump basic amount of the diabetic patient in the database is increased by 4.8U.

10. The method of claim 9, wherein the insulin pump initial recommended dose comprises a basal insulin amount and a prandial insulin amount, wherein the basal insulin amount and the prandial insulin amount are in a ratio of 1:1 and both adopt a mean subcutaneous infusion rate, and wherein: the insulin basic quantity adopts a multi-section structure, and the multi-section structure comprises one of a three-section structure, a four-section structure or a multi-section structure.