Automatic dose recommendation method for subcutaneous injection insulin system based on big data

Abstract

The invention relates to a hypodermic injection insulin system automatic dose recommendation method based on a big data AI algorithm, and the method comprises the following steps: (1) adding a current patient to a hypodermic injection insulin system for information setting; (2) calculating the initial daily total dose TDD of insulin needing to be injected by the patient at present; (3) calculating a correction large-dose coefficient CF of the insulin to be injected; (4) collecting the blood glucose value of the patient in the current state, and further calculating an adjustment factor AF; and (5) comparing the currently collected blood glucose value of the patient with a preset threshold value to prevent the occurrence of hypoglycemia. By adopting the automatic dose recommendation method for the subcutaneous injection insulin system based on the big data AI algorithm, the probability of occurrence of hypoglycemia can be effectively reduced by 97% or above, so that the time of reaching the blood glucose standard of a patient is saved by 34% or above, and the method has certain practicability in the aspects of reducing the hospitalization time of the patient and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of big data processing, and particularly to the technical field of adjuvant therapy. Specifically, it refers to a method for automatically recommending doses of a subcutaneous insulin injection system based on big data. Background Art

[0002] When a patient needs to receive subcutaneous insulin injection, the common practice is as follows:

[0003] 1. The doctor evaluates the patient's blood glucose condition and calculates the medication advice based on parameters such as the patient's weight.

[0004] 2. To prevent hypoglycemia, the initial dose is generally conservative.

[0005] 3. Nurses manually record the patient's blood glucose, and it is difficult for one person to manage multiple patients well.

[0006] Based on the above practices, the prior art has the following defects:

[0007] (1) Dose adjustment is not timely;

[0008] (2) Control of hyperglycemia and hypoglycemia is not ideal;

[0009] (3) It takes a relatively long time to reach the target control range;

[0010] (4) Medical resources are relatively tense, and it is difficult for one doctor to timely pay attention to the insulin responses of multiple patients;

[0011] (5) When the patient has hypoglycemia, the patient's needs cannot be responded to in a timely manner;

[0012] (6) For patients across clinics who need consultation by endocrinology experts, they often need to make an appointment with a doctor in advance, which is time-consuming, laborious, and has low efficiency;

[0013] (7) Calculating the pre-meal large dose and correction large dose requires a professional doctor.

[0014] Therefore, in view of the defects in the prior art, it is necessary to propose a treatment method that can be based on personalized treatment parameters set by the patient himself. Summary of the Invention

[0015] The purpose of the present invention is to overcome the above-mentioned defects of the prior art and provide a method for automatically recommending doses of a subcutaneous insulin injection system based on a big data AI algorithm.

[0016] To achieve the above purpose, the method for automatically recommending doses of a subcutaneous insulin injection system based on a big data AI algorithm of the present invention is as follows:

[0017] The method for automatic dose recommendation of a subcutaneous insulin injection system based on big data AI algorithms, the main features of which are that the method includes the following steps:

[0018] (1) Add the current patient to the subcutaneous insulin injection system for information setting;

[0019] (2) Calculate the initial total daily dose TDD of insulin to be injected by the current patient;

[0020] (3) Calculate the correction bolus factor CF of the insulin to be injected;

[0021] (4) Collect the blood glucose value of the patient in the current state and further calculate the adjustment factor AF;

[0022] (5) Compare the currently collected blood glucose value of the patient with a preset threshold, and calculate the currently recommended actual dose according to the insulin treatment method. If the system model believes that insulin injection should be stopped and relevant treatment suggestions are given to prevent hypoglycemia.

[0023] Preferably, the step (1) specifically includes the following steps:

[0024] (1.1) Add the patient to the subcutaneous insulin injection system and select the insulin treatment method as subcutaneous injection;

[0025] (1.2) Set the patient's own basic information, including date of birth, hemoglobin A1c, height, and weight;

[0026] (1.3) Set the initial personalized medical order information, including insulin type, whether there is diabetes, target blood glucose control range, type of pre-meal bolus, type of basal insulin, daily basal insulin administration frequency and administration time, and the amount of carbohydrates per meal.

[0027] Preferably, the insulin administration method is basal / bolus + correction bolus.

[0028] Preferably, the step (2) is specifically:

[0029] If calculating the initial total daily dose TDD according to body weight, specifically:

[0030] TDD = body weight (kg) × x%

[0031] Among them, the basal total amount (initial basal amount = basal total amount / daily administration frequency) = TDD × y%,

[0032] The total amount of pre-meal bolus (initial dose per meal = total pre-meal dose / 3) = TDD - basal total amount

[0033] x% is the percentage of the total dose, and y% is the percentage of the basal amount in the total dose.

[0034] Preferably, the correction large dose coefficient CF is specifically calculated in the following manner:

[0035] CF = 1700 / TDD.

[0036] Preferably, the step (4) is specifically as follows:

[0037] Compare the currently collected blood glucose with the target blood glucose. If the current blood glucose is within the target range, the adjustment factor AF is set to 1. If it is higher than the target range, the adjustment factor AF is set to 1.2. If it is lower than the target range, the adjustment factor AF is set to 0.8.

[0038] Preferably, the step (5) specifically includes the following steps:

[0039] (5.1) If the current blood glucose is lower than the set hypoglycemic threshold, it is recommended to stop insulin injection, and it is recommended to re-check the patient's blood glucose after 15 minutes. At the same time, glucose injection is recommended to prevent hypoglycemia.

[0040] (5.2) If it is the time to inject the basal dose set by the system, it is recommended to directly give the basal injection dose.

[0041] (5.3) If it is the blood glucose monitoring time point, it is recommended to collect the patient's current blood glucose currentBG, with the blood glucose unit mmol / l.

[0042] (5.4) If the currentBG collected in step (5.3) is the pre-breakfast blood glucose, the currently recommended insulin dose = pre-meal large dose + correction large dose, where the correction large dose = (current blood glucose currentBG - median of the target blood glucose) × 18 / CF.

[0043] (5.5) If the current blood glucose currentBG collected in step (5.3) is the pre-lunch blood glucose, the currently recommended insulin dose = pre-meal large dose + correction large dose, where the correction large dose = (current blood glucose currentBG - median of the target blood glucose) × 18 / CF), and at the same time calculate the pre-breakfast large dose for the next day = the pre-breakfast large dose of the current day × AF.

[0044] (5.6) If the current blood glucose currentBG collected in step (5.3) is the pre-dinner blood glucose, the currently recommended insulin dose = pre-meal large dose + correction large dose, where the correction large dose = (current blood glucose currentBG - median of the target blood glucose) × 18 / CF), and at the same time calculate the pre-lunch large dose for the next day = the pre-lunch large dose of the current day × AF.

[0045] (5.7) If the current blood glucose (currentBG) collected in step (5.3) is the bedtime blood glucose, then the currently recommended insulin dose = correction bolus dose, where the correction bolus dose = (currentBG - median of the target blood glucose) × 18 / CF), and at the same time calculate the dinner pre-meal bolus dose for the next day = the dinner pre-meal bolus dose of the current day × AF;

[0046] (5.8) If the current blood glucose (currentBG) collected in step (5.3) is the midnight blood glucose, then the currently recommended insulin dose = correction bolus dose, where the correction bolus dose = (currentBG - median of the target blood glucose) × 18 / CF);

[0047] (5.9) Calculate the recommended basal insulin, where the basal insulin for the next day = AF × the amount of the previous basal insulin, and calculate the adjustment factor AF based on the lowest blood glucose value at breakfast or midnight;

[0048] (5.10) Loop through the above steps until the dosage standard is met.

[0049] With the method of automatic dose recommendation of the subcutaneous insulin injection system based on big data of the present invention, when a patient needs subcutaneous insulin treatment, the system automatically calculates and recommends an appropriate insulin dose according to the personalized treatment parameters set by the patient himself and the AI big data algorithm model, which can enable the patient's blood glucose level to recover within the control range relatively quickly. At the same time, the probability of hypoglycemia is reduced by more than 97%, the time for the patient's blood glucose to reach the standard is saved by more than 34%, and the patient's hospital stay is reduced. Description of the Drawings

[0050] Figure 1 is the conventional insulin treatment process.

[0051] Figure 2 is the first insulin automatic dose recommendation process of the present invention.

[0052] Figure 3 is the second insulin automatic dose recommendation process of the present invention. Detailed Embodiments

[0053] In order to more clearly describe the technical content of the present invention, the following will be further described in combination with specific embodiments.

[0054] Before elaborating on the embodiments according to the present invention, it should be noted that, hereinafter, the terms "comprising", "including" or any other variations are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or elements inherent to such process, method, article or device.

[0055] Please refer to Figure 2 and Figure 3 shown, the method for automatic dose recommendation of a subcutaneous insulin injection system based on a big data AI algorithm, wherein the method comprises the following steps:

[0056] (1) Add the current patient to the subcutaneous insulin injection system for information setting;

[0057] (2) Calculate the initial total daily dose TDD of insulin to be injected for the current patient;

[0058] (3) Calculate the correction bolus factor CF of the insulin to be injected;

[0059] (4) Collect the blood glucose value of the patient in the current state and further calculate the adjustment factor AF;

[0060] (5) Compare the currently collected blood glucose value of the patient with a preset threshold, and calculate the currently recommended actual dose according to the insulin treatment method. If the system model determines that insulin injection should be stopped, relevant treatment suggestions are given to prevent hypoglycemia.

[0061] As a preferred embodiment of the present invention, the step (1) specifically comprises the following steps:

[0062] (1.1) Add the patient to the subcutaneous insulin injection system and select the insulin treatment method as subcutaneous injection;

[0063] (1.2) Set the patient's own basic information, including date of birth, hemoglobin A1c, height, and weight;

[0064] (1.3) Set the initialized personalized medical advice information, including insulin type, whether there is diabetes, target blood glucose control range, type of pre-meal bolus, type of basal insulin, daily basal insulin administration frequency and administration time, and the amount of carbohydrates per meal.

[0065] As a preferred embodiment of the present invention, the insulin administration method is suitable for the basal / pre-meal bolus + correction bolus mode.

[0066] As a preferred embodiment of the present invention, the step (2) is specifically:

[0067] If the initial total daily dose TDD is calculated according to body weight, specifically:

[0068] TDD = body weight (kg) × x%

[0069] Among them, the basal total amount (initial basal amount = basal total amount / number of daily administrations) = TDD × y%,

[0070] The total amount of pre-meal large dose (initial dose per meal = total pre-meal dose / 3) = TDD - basal total amount

[0071] x% is the percentage of the total dose, and y% is the percentage of the basal amount in the total dose.

[0072] As a preferred embodiment of the present invention, the correction large dose coefficient CF is specifically calculated in the following manner:

[0073] CF = 1700 / TDD.

[0074] As a preferred embodiment of the present invention, the step (4) is specifically:

[0075] Compare the currently collected blood glucose with the target blood glucose. If the current blood glucose is within the target range, the adjustment factor AF is set to 1. If it is higher than the target range, the adjustment factor AF is set to 1.2. If it is lower than the target range, the adjustment factor AF is set to 0.8.

[0076] As a preferred embodiment of the present invention, the step (5) specifically includes the following steps:

[0077] (5.1) If the current blood glucose is lower than the set hypoglycemic threshold, it is recommended to stop insulin injection, and it is recommended to re-check the patient's blood glucose after 15 minutes. At the same time, glucose injection is recommended to prevent hypoglycemia;

[0078] (5.2) If it is time to inject the basal dose set by the system, it is recommended to directly give the basal injection dose;

[0079] (5.3) If it is time for blood glucose monitoring, it is recommended to collect the patient's current blood glucose currentBG, with the blood glucose unit mmol / l;

[0080] (5.4) If the currentBG collected in step (5.3) is the pre-breakfast blood glucose, the currently recommended insulin dose = pre-meal large dose + correction large dose, where the correction large dose = (current blood glucose currentBG - median of the target blood glucose) × 18 / CF;

[0081] (5.5) If the current blood glucose (currentBG) collected in step (5.3) is the pre-lunch blood glucose, then the currently recommended insulin dose = pre-meal bolus dose + correction bolus dose, where the correction bolus dose = (currentBG - median of target blood glucose) × 18 / CF). At the same time, calculate the pre-breakfast bolus dose for the next day = the pre-breakfast bolus dose of the current day × AF;

[0082] (5.6) If the current blood glucose (currentBG) collected in step (5.3) is the pre-dinner blood glucose, then the currently recommended insulin dose = pre-meal bolus dose + correction bolus dose, where the correction bolus dose = (currentBG - median of target blood glucose) × 18 / CF). At the same time, calculate the pre-lunch bolus dose for the next day = the pre-lunch bolus dose of the current day × AF;

[0083] (5.7) If the current blood glucose (currentBG) collected in step (5.3) is the bedtime blood glucose, then the currently recommended insulin dose = correction bolus dose, where the correction bolus dose = (currentBG - median of target blood glucose) × 18 / CF). At the same time, calculate the pre-dinner bolus dose for the next day = the pre-dinner bolus dose of the current day × AF;

[0084] (5.8) If the current blood glucose (currentBG) collected in step (5.3) is the midnight blood glucose, then the currently recommended insulin dose = correction bolus dose, where the correction bolus dose = (currentBG - median of target blood glucose) × 18 / CF);

[0085] (5.9) Calculate the recommended basal insulin, where the basal insulin for the next day = AF × the amount of the previous basal insulin. Calculate the adjustment factor AF based on the lowest blood glucose value at breakfast or midnight;

[0086] (5.10) Loop through the above steps until the dosage standard is met.

[0087] In practical applications, the method for automatic dose recommendation of the subcutaneous insulin injection system based on the big data AI algorithm of this technical solution is specifically operated as follows:

[0088] 1. Add the patient to the system, select the insulin treatment method as subcutaneous injection, and set the patient's own basic information: date of birth, A1c, height, weight. At the same time, set the initialized personalized doctor's order information including: type of insulin (basal / pre-meal bolus dose + correction), whether having diabetes, target blood glucose control range (targetLowBG ~ targetHighBG), type of pre-meal bolus dose (including pre-meal administration time point), type of basal insulin, as well as the daily basal insulin administration frequency and administration time, and the amount of carbohydrates per meal;

[0089] 2. Calculation of the initial total daily dose (TDD): The basis for the calculation (weight and customization), the percentage x% of the total dose, the percentage y% of the basal amount in the total dose, the pre-meal large dose is (1 - y%). If calculating the initial TDD based on weight = weight (kg) × x%, the total basal amount (the initial basal amount = total basal amount / number of daily administrations) = TDD × y%, and the total pre-meal large dose (the initial dose per meal = total pre-meal dose / 3) = TDD - total basal amount. The system recommends the first relevant dose as the initial dose (basal insulin amount and pre-meal large dose);

[0090] 3. Calculation of TDD: On the first day, it is equal to the initial total dose. For subsequent days, TDD = yesterday's basal amount + yesterday's pre-meal amount + yesterday's correction amount);

[0091] 4. CF (Correction Factor) = 1700 / TDD;

[0092] 5. Collect the current blood glucose (currentBG), with the unit of mmol / l;

[0093] 6. Calculation of the AF adjustment factor. Compare the current blood glucose (currentBG) with the target blood glucose. If it is within the target range, it is 1. If it is higher than the target range, it is recommended to be 1.2. If it is lower than the target range, it is recommended to be 0.8;

[0094] 7. If the current blood glucose (currentBG) is lower than the set hypoglycemia threshold, it is recommended to stop insulin injection, re-check the patient's blood glucose after 15 minutes, and at the same time recommend injecting glucose to prevent hypoglycemia;

[0095] 8. Recommendation of the subcutaneous insulin injection dose;

[0096] (8.1) If it is time to inject the basal amount set by the system, it is recommended to directly give the basal amount;

[0097] (8.2) If it is time for blood glucose monitoring, it is recommended to collect the current blood glucose (currentBG), with the unit of mmol / l;

[0098] (8.3) If the currentBG collected in 8.2 is the blood glucose before breakfast, the currently recommended insulin dose = pre-meal large dose + correction large dose (correction large dose = (current blood glucose (currentBG) - median of the target blood glucose) × 18 / CF);

[0099] (8.4) If the current blood glucose (currentBG) collected in 8.2 is the blood glucose before lunch, the currently recommended insulin dose

[0100] = Pre - meal large dose + Correction large dose (Correction large dose = (Current blood glucose currentBG - Median of target blood glucose) × 18 / CF), and at the same time calculate the pre - meal large dose for breakfast on the next day = Current day's pre - meal large dose for breakfast × AF;

[0101] (8.5) If the current blood glucose currentBG collected at 8.2 is the pre - dinner blood glucose, the currently recommended insulin dose

[0102] = Pre - meal large dose + Correction large dose (Correction large dose = (Current blood glucose currentBG - Median of target blood glucose) × 18 / CF), and at the same time calculate the pre - meal large dose for lunch on the next day = Current day's pre - meal large dose for lunch × AF;

[0103] (8.6) If the current blood glucose currentBG collected at 8.2 is the bedtime blood glucose, the currently recommended insulin dose = Correction large dose (Correction large dose = (Current blood glucose currentBG - Median of target blood glucose) × 18 / CF), and at the same time calculate the

[0104] pre - meal large dose for dinner on the next day = Current day's pre - meal large dose for dinner × AF;

[0105] (8.7) If the current blood glucose currentBG collected at 8.2 is the midnight blood glucose, the currently recommended insulin dose =

[0106] Correction large dose (Correction large dose = (Current blood glucose currentBG - Median of target blood glucose) × 18 / CF);

[0107] (8.8) Calculate the recommended basal insulin. The basal insulin for the next day = AF × The amount of the previous basal insulin. The AF calculation rule (Calculate AF based on the lower value of the blood glucose at breakfast and midnight);

[0108] 9. Loop and execute from 8.3 to 8.8.

[0109] Any process or method description shown in the flowchart or described in other ways herein can be understood as representing a module, segment, or part of code including one or more executable instructions for implementing a specific logical function or process. And the scope of the preferred embodiments of the present invention includes additional implementations, where the functions can be executed not in the order shown or discussed, including in a substantially simultaneous manner according to the involved functions or in the reverse order, which should be understood by those skilled in the technical field to which the embodiments of the present invention belong.

[0110] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution device.

[0111] Those of ordinary skill in the art can understand that all or part of the steps carried by the method of implementing the above embodiments can be completed by instructing relevant hardware through a program. The program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

[0112] The above-mentioned storage medium can be a read-only memory, a magnetic disk, an optical disk, etc.

[0113] In the description of this specification, the descriptions with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "embodiment", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0114] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

[0115] Since the conventional treatment method for diabetes is insulin injection, the adjustment of insulin infusion dose depends to a large extent on the subjective judgment of doctors. Due to the individual differences in patients' responses to insulin, and as the disease progresses, patients' insulin requirements change. The traditional insulin infusion mode is difficult to meet the dynamically changing clinical needs. Therefore, it is of great significance to customize an individualized, precise, and dynamic insulin regulation strategy for patients. However, when injecting with a fixed amount, and when the condition changes, that is, when the blood glucose concentration in the patient's body changes after treatment, doctors need to re-evaluate the patient's condition and then adjust the drug dosage for the patient. Patients cannot obtain the required insulin dose in real time and precisely, which results in a longer time for patients to achieve blood glucose control, and there may also be hypoglycemia or hyperglycemia events during this period.

[0116] With the method of automatic dose recommendation of the subcutaneous insulin injection system based on big data of the present invention, when a patient needs subcutaneous insulin treatment, the system automatically calculates and recommends an appropriate insulin dose according to the personalized treatment parameters set by the patient himself and the AI big data algorithm, which can enable the patient's blood glucose level to recover within the control range relatively quickly, while greatly reducing the probability of hypoglycemia by more than 97%, saving more than 34% of the time for the patient to reach the blood glucose standard, and at the same time reducing the hospital stay.

[0117] In this specification, the present invention has been described with reference to its specific embodiments. However, it is obvious that various modifications and variations can still be made without departing from the spirit and scope of the present invention. Therefore, the specification and the drawings should be regarded as illustrative rather than restrictive.

Claims

1. A method for automatic dose recommendation of a subcutaneous insulin injection system based on big data AI algorithms, characterized in that, The described method includes the following steps: (1) Add the current patient to the subcutaneous insulin injection system for information setting; (2) Calculate the initial total daily dose TDD of insulin to be injected for the current patient; (3) Calculate the correction bolus factor CF of the insulin to be injected; (4) Collect the blood glucose value of the patient in the current state and further calculate the adjustment factor AF; (5) Compare the currently collected blood glucose value of the patient with a preset threshold, and calculate the currently recommended actual dose according to the insulin treatment method. If the system model deems that insulin injection should be stopped, relevant treatment suggestions are given to prevent hypoglycemia.

2. The method for automatic dose recommendation of a subcutaneous insulin injection system based on a big data AI algorithm according to claim 1, characterized in that, The specific steps of step (1) include the following steps: (1.1) Add the patient to the subcutaneous insulin injection system and select the insulin treatment method as subcutaneous injection; (1.2) Set the patient's own basic information, including date of birth, hemoglobin A1c, height, and weight; (1.3) Set the initialized personalized medical order information, including insulin type, whether there is diabetes, target blood glucose control range, type of pre-meal bolus, type of basal insulin, number of daily basal insulin administrations and administration times, and the amount of carbohydrates per meal.

3. The method for automatic dose recommendation of the subcutaneous insulin injection system based on the big data AI algorithm according to claim 2, wherein, The insulin administration plan is a combination of basal / pre-meal bolus + correction bolus.

4. The method for automatic dose recommendation of a subcutaneous insulin injection system based on a big data AI algorithm according to claim 1, characterized in that, The specific steps of step (2) are: If calculating the initial total daily dose TDD according to body weight, specifically: TDD = body weight (kg) × x% Among them, the basal total amount (initial basal amount = basal total amount / number of daily administrations) = TDD × y%, The total pre-meal bolus amount (initial dose per meal = total pre-meal dose / 3) = TDD - basal total amount x% is the percentage of the total dose, and y% is the percentage of the basal amount in the total dose.

5. The method for automatically recommending doses of the subcutaneous insulin injection system based on the big data AI algorithm according to claim 4, characterized in that, The specific calculation of the correction bolus factor CF is as follows: CF = 1700 / TDD.

6. The method for automatic dose recommendation of a subcutaneous insulin injection system based on a big data AI algorithm according to claim 4, wherein The specific steps of step (4) are: Compare the currently collected blood glucose with the target blood glucose. If the current blood glucose is within the target range, the adjustment factor AF is set to 1. If it is higher than the target range, the adjustment factor AF is set to 1.

2. If it is lower than the target range, the adjustment factor AF is set to 0.

8.

7. The method for automatic dose recommendation of a subcutaneous insulin injection system based on a big data AI algorithm according to claim 6, wherein, The specific steps of step (5) include the following steps: (5.1) If the current blood glucose is lower than the set hypoglycemia threshold, it is recommended to stop insulin injection, and it is recommended to re-check the patient's blood glucose after 15 minutes. At the same time, glucose injection is recommended to prevent hypoglycemia; (5.2) If it is time to inject the basal amount set by the system, it is recommended to directly give the basal injection amount; (5.3) If it is time for blood glucose monitoring, it is recommended to collect the patient's current blood glucose currentBG, with the blood glucose unit being mmol / l; (5.4) If the currentBG collected in step (5.3) is the pre-breakfast blood glucose, the currently recommended insulin dose = pre-meal bolus + correction bolus, where the correction bolus = (current blood glucose currentBG - median of the target blood glucose) × 18 / CF; (5.5) If the current blood glucose (currentBG) collected in step (5.3) is the pre-lunch blood glucose, then the currently recommended insulin dose = pre-meal bolus dose + correction bolus dose, where the correction bolus dose = (currentBG - median of the target blood glucose) × 18 / CF), and at the same time calculate the pre-breakfast bolus dose for the next day = the current day's pre-breakfast bolus dose × AF; (5.6) If the current blood glucose (currentBG) collected in step (5.3) is the pre-dinner blood glucose, then the currently recommended insulin dose = pre-meal bolus dose + correction bolus dose, where the correction bolus dose = (currentBG - median of the target blood glucose) × 18 / CF), and at the same time calculate the pre-lunch bolus dose for the next day = the current day's pre-lunch bolus dose × AF; (5.7) If the current blood glucose (currentBG) collected in step (5.3) is the bedtime blood glucose, then the currently recommended insulin dose = correction bolus dose, where the correction bolus dose = (currentBG - median of the target blood glucose) × 18 / CF), and at the same time calculate the pre-dinner bolus dose for the next day = the current day's pre-dinner bolus dose × AF; (5.8) If the current blood glucose (currentBG) collected in step (5.3) is the midnight blood glucose, then the currently recommended insulin dose = correction bolus dose, where the correction bolus dose = (currentBG - median of the target blood glucose) × 18 / CF); (5.9) Calculate the recommended basal insulin, where the basal insulin for the next day = AF × the amount of the previous basal insulin, and calculate the adjustment factor AF based on the lowest blood glucose value at breakfast or midnight; (5.10) Loop through the above steps until the dosage standard is met.

Images