JP2007525433A - Method and system for determining insulin administration schedule and carbohydrate-to-insulin ratio in diabetic patients - Google Patents

Abstract

  A method for digitally determining a daily insulin regimen for diabetics. The present invention divides a patient's day into adjustable time intervals that include a basal insulin dosage rate and a carbohydrate to insulin ratio (s) (to determine dietary insulin dosage). The present invention identifies adjusted insulin doses over a period of time as “error” of prescription insulin (basal insulin + meal insulin). The method first comprises estimating a change to one of these two components of prescription insulin, and then determining a change to the other by subtracting the one change from the prescription insulin error. Including. One method estimates the change in meal insulin allocated during the interval proportional to the old meal insulin. Another method combines postprandial insulin and meal insulin together. Another method divides the interval by the postprandial adjusted dose and determines the basal insulin from the time boundary adjusted dose. Data may be obtained from the previous day and a small fraction of the error can be added leading to an asymptotic reduction of the error. Data may also be obtained from recent history and a larger fraction of errors may be added by a physician or automated method.

Description

  The present invention relates generally to the field of digital supplements that aid in the treatment of diabetic patients using insulin pumps or multi-dose insulin dosing schedules, and provides a method for determining an insulin dosing schedule in diabetic patients.

  Diabetes has been treated with insulin injections for many years. However, three recent advances are changing the care of diabetes. An insulin pump, a new insulin prescription for multiple dose injection, and an inhaled insulin. These are discussed below.

  Insulin Pump: The insulin pump of the present invention has revolutionized diabetes care. It is a battery powered device approximately the size of a pager. Insulin cartridges are included and an “infusion set” (small plastic needle or “cannula” attached with an adhesive patch) pumps insulin through a flexible tube to a patient. The pump of the present invention makes it possible to adopt a typical insulin dosing regime as follows. That is, basal insulin is injected slowly and continuously at a rate that allows a program to change many times throughout the day (usually around 4 or 5 times per day). The basal insulin rate of infusion is constant between changes. The certain period is called “section”. Additionally, a bolus of insulin can be injected at the command of the patient. There are two main types of bolus injection:

  Instant meal administration is infused just before meals in an amount proportional to the blood glucose effect of the meal. This is generally proportional to the number of grams of carbohydrate in the meal. The proportionality constant is a personalized number called carbohydrate-to-insulin ratio, CIR. Use as follows:

Dietary insulin bolus = (grams of carbohydrate in the meal) / CIR (1)
This calculation is typically performed by the patient, but there are pump models that can store the patient's CIR in memory and require only the grams of carbohydrate in the diet as input.

  The corrected bolus is injected immediately after the blood glucose test is performed, and the amount of the corrected bolus is proportional to the error in blood glucose concentration from the patient's personalized target blood glucose. The proportionality constant is a personalized number called the correction factor, CF. Use as follows:

Adjusted insulin bolus dose = (blood glucose concentration-target) / CF (2)
There are two types of controlled bolus, each with different goals.

  Time boundary adjusted insulin bolus is administered on an empty stomach at the end of the time interval.

  Post-prandial bolus administration is most often administered 1-5 hours after a meal and is administered within a time interval.

  Recently, pump manufacturers are incorporating digital features into their pumps that make treatment easier. Some pumps can store CF and target values and require only blood glucose concentration (BG) as input. Among these new digital functions is the “insulin on board” function. This function mathematically models the amount of insulin still present in the body at a certain time after a bolus dose, and recommends a reduction relative to the bolus dose as appropriate. This feature makes postprandial bolus administration safer and more practical.

  Multiple dose injection (MDI): Progress is being made in the development of different types of insulin. Some work for very long periods and have no peaking. Insulin, which acts for a long time, can be injected at a low frequency of about once per day, with a regimen that closely resembles the basal insulin regimen of pump patients. Rapid action type insulin injections can be given as meals and corrected bolus doses. Both types act as a system. These insulins are available in portable “pens” (named as they are similar to writing instruments). The pen is combined with the BG meter in the form of a “kit”, these devices communicate with each other in the kit, and the combined memory is stored in one of the two devices in the “kit” .

  Inhaled insulin: Inhaled insulin delivery systems are being developed for short-acting insulin. The inhaler is expected to be combined in a “kit”, similar to the BG meter used for MDI, so the present invention could handle inhaled insulin as well. This development is expected in the future.

  The nature of diabetes care is very quantitative. Ironically, the increase in numbers makes the use of redundant algorithms in pocket calculators too time consuming and therefore expensive. Therefore, the majority of endocrinologists use subjective methods based on experience. Endocrinologists often use numbers simply to help identify trends and then treat those trends to give a greater subjective feel to the cases at hand. For example, they typically view blood glucose (BG) scatter charts and printouts, by subjective means such as the visual density of dots on the BG scatter chart and the relative location of the highest density areas. Often distinguishing trends. They translate these trends into changes in insulin dosage using their experience.

  Subjective methods based on experience are often not uniform among practitioners. There is a shortage of endocrinologists and other diabetes specialists. Therefore, the management of diabetes is performed in a chaotic manner by clinicians with vastly different levels of expertise. As a result, management of diabetes in most patients is satisfactory but not optimal. As a result of suboptimal BG management, the diabetic process may involve complications involving all body systems. These complications are associated with premature death and, in relation to costs, have reached 19% of health costs to care for 6% of the population.

The present invention addresses the above shortcomings of current care by providing a method for analyzing and prescribing changes to the daily insulin dosing schedule of diabetic patients using insulin pumps, multiple dose subcutaneous injections, or inhaled insulin. Overcoming. The method divides a patient's (one) day into selected time intervals, during which time the basal insulin dosage rate and carbohydrate to insulin ratio (s) (to determine dietary insulin dosage) ) Is provided with an adjustable schedule. Time boundaries and basal rate changes are generally set by the practitioner to match the patient's diet. Patients are generally advised to test their blood glucose (BG) at the time boundary just before they eat. The present invention uses the following two insulin nomenclature systems.
Traditional insulin nomenclature:
Total daily insulin = prescription insulin + adjusted insulin (3)
here,
Prescription Insulin = Basic Insulin + Dietary Insulin (4)
The present invention incorporates the idea of (1) using adjusted insulin over a selected time interval as the “error” of the patient's prescription insulin (basic insulin + meal insulin) for the entire day and each time interval. . A method is included for determining a change to one of the two components of the prescription insulin and then subtracting the one change from the prescription insulin error to determine a change to the other component of the prescription insulin. There are therefore two basic algorithmic forms, each called by the following "float" name.

MealIns float:
First, the change in basal insulin is estimated, and then the present invention calculates the change in dietary insulin to be the error in prescription insulin minus the change in basal insulin. Some of the ways to estimate changes in basal insulin are:

Borrow basal velocity from another interval; estimate basal velocity from another algorithm (eg, basal float); curve of time interval using time boundary adjusted insulin (at end of interval) as an indicator of error Estimate the change in basal insulin from the later part of the free; as an error, float in the first part of the interval using postprandial adjusted insulin. Or
Basic float:
First, the change in meal insulin is estimated, and then the present invention calculates the change in basal insulin to be the error in prescription insulin minus the change in meal insulin. Some of the ways to estimate dietary insulin changes are as follows.

  Borrowing Meal Insulin or CIR from another leg; change in Meal Insulin for one leg as share of total Meal Insulin Change for Total Day, equal to (Section Curve) / (Total Day Curve) The change in meal insulin for a section is estimated as the share of meal insulin change for the entire day at the same ratio as (diet meal insulin) / (total day meal insulin).

Enhanced insulin nomenclature:
In addition, in order to provide another method for describing postprandial insulin, the present invention identifies “enhancement variables” as follows as a system of insulin naming that combines postprandial insulin and dietary insulin together.

Total daily insulin = augmented prescription insulin + time boundary adjusted insulin (5)
here,
Enhanced prescription insulin = basal insulin + enhanced dietary insulin (6)
as well as,
Enhanced Meal Insulin = Meal Insulin + Postprandial Insulin (7)
The float is very similar:
MealIns float:
First, the change in basal insulin is estimated, and then the present invention calculates the change in enhanced dietary insulin to be the error in enhanced prescription insulin minus the change in basal insulin. Some of the ways to estimate changes in basal insulin are:

Borrow basic speed from another section;
Estimate basal velocity from another algorithm (eg, basal float).

Or
Basic float:
First, the change in enhanced dietary insulin is estimated, and then the present invention calculates the change in basal insulin to be the error in enhanced prescription insulin minus the change in enhanced dietary insulin. Some of the methods for estimating changes in augmented dietary insulin are as follows.

  Borrowing Enhanced Meal Insulin or CIR from another leg; Change in Augmented Meal Insulin in one leg as share of the change in Augmented Meal Insulin for the total day (Section Curve) / (Total Day Curve) The same ratio of (enhanced dietary insulin for the entire day) / (total enhanced dietary insulin) as the share of the total daily augmented dietary insulin change Estimated by

General:
It may not be desirable to apply all of the error terms to make the change. A method of applying a limited amount of error to prevent overshoot is included.

  The above selection of “float” algorithms was developed in two main versions:

  Daily update version: The above float algorithm is embodied in a program that calculates the current day using the previous day's data. In this version, the total amount of error terms is not applied, but instead a predetermined fraction of error terms is applied. If the error was valid, the reduced amount appears the next day. This reduced error for the next day is reduced by the same fraction, and so on until the error disappears asymptotically. This version is particularly suitable for installation in insulin pumps, inhalation insulin kits, MDI injection kits, PDAs and other portable devices.

  Multi-day data version: The above float algorithm is embodied in a program that uses a recent calendar period (eg, the last few weeks) for the data source. Therefore, the data is an average value. The total amount of error term is not necessarily applied, but instead the fraction or amount of error to be applied is entered by the practitioner. These versions are suitable for installation on a practitioner's computer. This version of the subversion automatically determines the fraction of errors. Suitable for patient computers, or websites, used by patients with insulin pumps or kits for multiple daily injections or inhaled insulin.

  Various aspects of the present disclosure can be more clearly understood and appreciated from a review of the following detailed description of the disclosed embodiments and by reference to the accompanying drawings.

1． Table of Contents The present invention is a set of algorithms used to determine various insulin dose rates and carbohydrate-to-insulin ratio (s), including a daily insulin dose schedule for a patient. The “float algorithm” applies to all preferred embodiments described herein. The description shown in the table of contents below is based on the “embodiment” (pump or MDI), then “version” (daily update or multi-day data), and finally “algorithm” (basic float 1, meal insulin) Float 1, etc.).
1. Table of Contents
2. Suffix and time index
3. Glossary
4. General guidance
4.1 Changes in carbohydrate to insulin ratio, CIR
4.2 Some important expressions
4.2.1 Traditional insulin nomenclature
4.2.2 Naming of enhanced insulin
4.2.3 Overview
4.3 Two-level basic system and Kf calculator
4.3.1 Statistical correlation
4.3.2 Average calculation
5. Type of insulin delivery system
6. Version, embodiment, description by algorithm
6. 1. Version with daily updates
6.1.1 Preliminary guidance
6.1.1.1 Time interval and adjusted insulin
6.1.1.2 KrxInsl: governs the magnitude of insulin changes
6.1.2 About the pump (type E)
6.1.2.1 Base-to-total ratio:
6.1.2.2 Basic float 1, (deMealIns is proportional to eMealIns or Carbs)
6.1.2.3 Base float 2, (deMealIns value from outside the interval)
6.1.2.4 Meal insulin float 1
6.1.2.5 Overview of 6.1.1.2 and 6.1.1.3
6.1.2.6 Meal float 2 (AMCorIns (i) is used as an error)
6.1.3 About multiple dose injection (MDI) and inhaled insulin
6.1.3.1 Base float 1, (not used for MDI or inhaled insulin)
6.1.3.2 Base float 2, (dMealIns values are from outside the interval)
6.1.3.3 Meal Insulin Float 1
6.1.3.4 Overview of 6.1.3
6.1.3.5 Meal insulin float 2, (using AMCorIns (i) as error term)
6.1.4 Omitted BG
6.1.5 Modified bulk instantaneous dose calculator
6.1.6 Changing patient schedule
6.2 Version with multi-day data
6.2.1 Multi-day data for medical practitioners (DIGITAL AD VISOR)
6.2.1.1 About the pump
6.2.1.1.1 Pump types D and E
6.2.1.1.1.1 Basic float 1, (deMealIns is proportional to eMealIns or Carbs)
6.2.1.1.1.2 Basic float 2, (not used for multi-day data)
6.2.1.1.1.3 Meal insulin float 1
6.2.1.1.1.4 6.2.1.1.1.1 and 6.2.1.1.1.3 Multi-day data, type D pump overview
6.2.1.1.1.5 Meal insulin float 2 (AMCorIns is used as an error)
6.2.1.1.2 Pump type C
6.2.1.1.3 Pump type B
6.2.1.1.4 Pump type A
6.2.1.2 Subcutaneous multiple doses and inhaled insulin
6.2.1.2.1 Basic float 1
6.2.1.2.2 Meal Insulin Float 1
6.2.1.2.3 Meal insulin float 2 (Use AMCorlhs (i) as error)
6.2.1.3 deRxInsl limited domain and multi-day data version safety net formula
6.2.1.3.1 deRxInsl restricted domain
6.2.1.3.2 Safety net formula
6.2.1.3.2.1 About Basic Float 1
6.2.1.3.2.2 About Meal Insulin Float 1
6.2.1.3.2.3 About Meal Insulin Float 2 (use AMCorIns (i) as an error)
6.2.2 Automatic multi-day data (DIGITAL ADVISOR)
6.2.2.1 deRxInsl automation
6.2.2.2 deMealIns automation
6.2.2.3 Automatic rounding of CIR for pump type A
6.3 Non-float algorithm
7. Time interval
7.1 Find the last entry for the day
2. Suffix and time index In the present invention, a 24-hour day can be divided into a plurality of time intervals. Time interval boundaries are numbered in the manner of time0, time1, time2, etc. These time boundaries are generally referred to as time (i), where “i” stands for “time index”.

The time interval, dt (i), is found by subtracting adjacent time boundaries. That is,
dt (i) = time (i + 1)-time (i) (8)
Each time interval is numbered the same as the upstream time boundary, as described above.

  It should be noted that the section surrounding midnight needs to be mentioned in particular.

dt0 = 24 + time1-Tmax (9)
Here, Tmax is the last time boundary in the day and is generally the bedtime. How to find Tmax is explained in a later section.

  Time boundaries are generally defined by changes in (pump) basal speed or by changes in average mealtime. Practitioners often schedule patients so that basal rate changes, meal times, BG testing, adjusted bolus doses of insulin, and bolus doses of meal insulin all occur at these time boundaries. This practice is not essential, but is preferred for use with the present invention. The base speed is numbered the same as the time interval, so BR1 belongs between time1 and time2.

  Other parameters (other than the basal rate) are identified at precise time boundaries, but in actual use, patients are not required to follow such a strict schedule. Therefore, another set of time periods (referred to as “bins” to distinguish them from intervals) is required, each of which envelops a time boundary. Parameters that are said to occur “at the time boundary” are actually given the same index number as the time boundary that occurs in the bin and is enveloped by the bin. Tests and bolus injections can be classified into these “time boundary bins”. For multi-day data versions, this is most often handled by downloading the blood glucose meter and insulin pump software, which is binned with the software. For the daily update version, these bin boundaries may be defined using the midpoint of the primary time interval.

  It is noted that time-aligned insulin measures the insulin error in the previous interval, and therefore when calculating the parameters for the i-th interval, the insulin error is “TBCorIns (d, i + 1)”. I want.

  Parameters that do not follow the time interval schedule do not have an integer time interval index. In the multi-day version, these parameters have no index at all.

  In the daily update version, all parameters are given a daily index such as “d”, “d-1”, etc. in the form of BF (d, i) or TDD (d).

  Figures 1 and 2 show page 2 of the preferred input form for version 6.2.1.1.1, a multi-day data version for Pump D. FIG. 1 shows an unindexed parameter with a single value in the dialog. FIG. 2 shows a time interval that includes an indexed parameter for the time interval.

The parameters related to the unique date and time are expressed in a function form, that is, “parameter (t)”. Here, some of these appear during guidance for explanation, but none is used in the present invention.
3. Glossary Average Glycemic Index (AGI): For average carbohydrates, the ratio of grams of glucose entering the blood within 2 hours divided by the total number of grams of sugar in the carbohydrate . For this number, studies have been published and range from 60% to 90%.

  AIM Formulas: “Accurate Insulin Management” acronyms (see Ref 1, which is incorporated by reference as if fully published here). A set of statistically derived formulas is used to estimate three parameters for the patient. There are three formulas. In each, a constant (starting with “K”) was determined statistically.

BasalAIM = Kb ＊ TDDavg (10)
CFaim = Kcf / TDDavg (11)
CIRaim = Kcir * BodyWeight (lbs) / TDDavg (12)
The values of these constants change from time to time as new research is conducted.

  The latest values (see Ref1) are Kb = 0.48, Kcf = 1700, Kcir = 2.8.

  After-Meal Corrective Insulin: Adjusted large-volume instant administration (updated daily) on the previous day, after meal (after meal) in the i-th segment.

  AMCorIns (d-1, i): “Postprandial adjusted insulin”: Adjusted massive bolus (daily updated version) given on the previous day, after meal (after meal) in the i-th section.

  AMCorIns (i): “Postprandial adjusted insulin in the i-th time interval”. Average of adjusted bolus doses taken after meals (after meals) in the i-th time interval (multiple past day version).

  Basal Rate: The rate at which basal insulin is administered (insulin units / hour).

  Basal Rate Float: CIR or dietary insulin is determined by the practitioner or estimation formula, and BR is the total required change, dRxInsl (enhanced or conventional) to dietary insulin (enhanced or conventional) Required by subtracting

Basal (d): “Basic insulin”. Basal insulin for the current day, calculated from the basal rate according to the present invention. (Daily update version)
Basal (d, i): Basal insulin for the current day and time interval (updated daily).

  Basal (i): Basis of time interval (multi-day version).

  Basal: “Basic insulin”. 1. General definition: Insulin administered continuously by a pump or manually injected insulin, which stays in the body for a long time due to its special chemical composition. 2. Computer variable: The current total of basal insulin administered during the day, calculated from the basal rate.

SUM (Basal (i)) is calculated by the present invention over a period of days. (Multi-day version)
BasalTot: “Total basic insulin”. The current basal insulin dose programmed into the pump at the time of dialogue with the practitioner. Calculated by the pump. Should be equal to the foundation. (Multi-day version)
BG (i): average of blood glucose tests over several days at approximately the same time boundary throughout the day, time (i), calculated by the glucose meter downloading software. (Multi-day version)
BGmean: The overall average of blood glucose tests since the last interaction or over the calendar period being analyzed, calculated by the BG meter downloading software. (Multi-day version)
BGpd: “Number of BGs per day”. Average number of BG tests per day over the calendar period being analyzed. (Multi-day version)
BGsd (i): “BG standard deviation of the i-th time interval”, which is obtained from the BG meter downloading software. (Multi-day version)
BGsd: “BG standard deviation” over the calendar period being studied. Obtained from BG meter downloading software. (Multi-day version)
BG, Blood Glucose Concentration: These studies are self-administered several times throughout the day, typically at time boundaries. In the current state of the art, they measure by placing a drop of blood on a test strip in a blood glucose meter. The meter stores the value and time in memory. At each practitioner / patient interaction, the value is typically downloaded by software from the blood glucose meter's memory, which graphs the data points and represents the overall average value represented by BGmean and BG (i) Calculate several values including the time boundary average. Several models of BG meters can be linked to an insulin pump to download combined and integrated data.

  Bin: A time period that envelops the time boundary of the primary time interval system. Any BG test or bolus given in the bottle is treated as if it were on a time boundary.

  Instant bolus (Bolus): (Source: Latin for ball). Insulin that is infused within a short period of time ordered by the patient, distinct from basal insulin. Basal insulin is infused slowly, continuously and automatically, by a pre-programmed pump or by slow-acting insulin injection.

  Bolus Time Period (i): Large instantaneous doses are identified by time boundaries, but their timing is variable. To resolve this, use a system with a large instantaneous dosing time period. This system has a boundary that alternates between the boundaries of the regular basal section in a rocking manner, so that any BG or bolus dose that falls within the massive bolus time period is precisely on the regular basal boundary. When done, it is automatically considered.

Boli (i): "Insulin with large instantaneous dose". This is called the mean sum of meal insulin + adjusted insulin over a time interval and is given as a bolus dose rather than continuously. (Multiple past days version)
BoT: “Basal over Total”, Basic / TDD.

  BoT (d): Daily “Basal over Total” basis / TDD.

  BoTFbk: “Basal over Total, Feedback”. Daily update version parameters. Multiplying this factor by the proposed insulin change for the day results in the amount of change to the basis.

  BoTTgt: “Basal over Total, Target”. Base / TDD required value (daily updated version).

  BoTTgtRec: “Basal over Total, Target Recommended”. Recommended BoTTgt value (updated daily).

BR (d, i): “Basic speed” in the i-th time interval. The current setting for the rate at which basal insulin is delivered. This speed is programmable by the pump so that it can be changed several times during the day. (Daily update version)
BR (i): “Basic speed” of the time interval. The current setting for the rate at which basal insulin is delivered. This speed is programmable by the pump so that it can be changed several times during the day. (Multi-day version)
BRavg: “Basal Rate Average”, Basics / 24
BRavgRec: “Basal Rate Average Recommended”
BRf: “Basal Rate, Fasting”. The patient's blood glucose is maintained in the target range if no meals are consumed within the previous 12 hours and measurements are not taken in the late sleep interval (to avoid complications of pre-dawn effects) Will BR.

  BRkey: “Key basic speed”. Other basal speeds are pegged to it.

  BRlateSlp: “Basal speed in late sleep interval”.

BRrec (i): “Recommended basic speed for the i-th time interval”. Calculated according to the present invention. (Multiple past days version)
BRreliable: A set of basic velocity data selected for the reliability and high frequency of related BG tests. Use with the Kf calculator.

BRrx (i): “Basic speed of the i-th time interval, prescription”. The value that the practitioner gives the patient after observing the case. (Multiple past days version)
BRsimilar: A set of basal velocity data chosen to be similar to or close to the fasting basal velocity. Use with the Kf calculator.

  Carbohydrate-to-Insulin Ratio (see CIR): Personalized conversion constant, current value.

  Carb-Counting: A method of determining dietary insulin by first counting the number of grams of carbohydrate in the diet that should be consumed and then dividing by the CIR.

  Carbm: “Corrected carbohydrate grams”, used in high volume instantaneous dose calculators that incorporate exercises.

Carbs: “grams of carbohydrates”
CarbSh (i): “Curve share”. For use with nearly Type A pumps, use grams of actual carbohydrates or 15 g of “extienzi”. When used with Type A pumps, without memory for MealIns (i) 5, this parameter is the relative size of carbohydrates consumed within each time interval, practitioner's or patient's Is an estimated value. For Type A pumps, any unit is acceptable because the parameters are only used as a percentage of the total, CarbShTot. Examples of units: grams of carbohydrate (preferred), percent of total, units of dietary insulin (if CIR is constant). (Multi-day version)
CarbShTot: Calculated as “Curve Share Total”, CarbShTot = sum (CarbSh (i)).

   CF: “Correction factor”. A personalized factor used to calculate a corrected bolus. See CorBol (t) and AIM.

  Changes to: Used as “changes to prescription insulin”, “changes to basal insulin”, “changes to enhanced dietary insulin”. This phrase represents the proposed change of the specified amount from the last measured value to the recommended value. For the daily update version, this is a change from one day to the next. For multi-day data versions, this is a change from the average to the recommended value over the past calendar period. The digital variable representing the change is preceded by a lowercase “d” as in calculus.

CIR (d, i): “Carbohydrate to insulin ratio”. CIR of time interval dt (i) for the current day. (Daily update version)
CIR (i): “Glucose to Insulin Ratio”. CIR of time interval dt (i) for the current day. (Multiple past days version)
CIR: “Glucose to Insulin Ratio”. Personalized conversion constant, current value.

CIR = (weight of carbohydrates consumed) / (insulin required to metabolize carbohydrates) (13)
See MealBol (t) and AIM. Calculation of this parameter is one of the important goals of PumpMaster. (Multiple past days version)
CIRrec (i): “Glucose to insulin ratio, recommended time interval”. (Multiple past days version)
CIRrec: “Sugar to insulin ratio, recommended”. (Multiple past days version)
CorBol (t): “Adjusted bolus at a unique date and time”. The patient regularly tests his / her blood glucose concentration. If it is too high, the patient calculates the bolus dose of insulin according to the following formula:

CorBol (t) = (BG (t)-target) / CF (14)
Here, the target is the required BG level, and CF is a personalized correction factor.

CorIns (i): “Adjusted insulin”. The sum of CorBol (t) at all times within the interval, including “after meal” and “time boundary” times. (Multiple past days version)
CorIns (d, i): “Regulated insulin”. Sum of CorBol (t) at all times in the interval, including “after meal” and “time boundary” times during the current day. (Multiple past days version)
CorTot (d): “Adjusted insulin total”. The sum of adjusted insulin at all times within the current day. (Daily update version)
CorTot: “Adjusted insulin total”. The average of the sum of adjusted insulin at all times during the day. (Multiple past days version)
dBasalAuto: An automated version of dBasal. (Automatic multi-day version)
dBaslToAIM: The basic change necessary to achieve a BasalAIM value.

deMealIns (d, i): “Change in Enhanced Meal Insulin” in the time interval of the current day. (Daily update version)
deMealIns (i): “Change in Enhanced Meal Insulin” in the time interval. (Multiple past days version)
deMealIns: "Total Enhanced Meal Insulin Change" throughout the day. (Multiple past days version)
deMealInsAuto: An automated version of dEmealIns. (Automatic multi-day version)
deRxInsAuto: "Enhanced prescription insulin change, automatic", automatically estimated. (Automatic multi-day version)
deRxInsl (d, i): “Change of augmented prescription insulin in the time interval of the current day” (updated daily)
deRxInsl (i): "Change of prescription insulin in time interval" (multiple past days version)
deRxInsl: “Change of total prescription insulin for total days” (multiple past days version)
deRxInslld (i): “Enhanced prescription insulin change, large domain in the interval”. "Safety-Net" formula. One of the two deRxInsl (i) formulas. (Multi-day version)
deRxInslsd (i): “Total augmented prescription insulin change, small domain”. One of two formulas of deRxInsl (i) for "Safety-Net" purposes. (Multi-day version)
dt (i): “Time difference”. The length of the “i” th time interval.

  Early-Sleep Time Interval: A time interval that begins at bedtime and ends at the Mid-Sleep time boundary. Most diabetics are encouraged not to take snacks at bedtime and are encouraged to have mid-sleep BG at approximately 3 am. If this advice is followed, the early sleep period provides a good indication of the true “inactive basal rate”, BRf, which is the basis for many other time periods.

eMealIns (d, i): “Enhanced Meal Insulin over time”. Sum of bolus bolus plus postprandial bolus during the current time interval dt (i). (Daily update version)
eMealIns (i): “Enhanced Meal Insulin over time”. Multi-day average of sum of postprandial insulin plus time boundary adjusted insulin over time interval. (Multiple past days version)
eMealInsRec (d, i): “Reinforced dietary insulin recommended for the current day and time interval”. (Daily update version)
eMealInsRec (i): “Enhanced dietary insulin for time intervals, recommended”. Calculated according to the present invention. (Multiple past days version)
eMealInsTot (d): “Total augmented meal insulin for the current day”. The sum of eMealIns (i) during the current day. (Daily update version)
eMealInsTot: “Enhanced Meal Insulin Total”. MealIns (i) sum plus postprandial adjusted insulin sum over the whole day averaged over several days. (Multi-day version)
Enhanced: A term used to describe a modified variable system, where postprandial insulin is incorporated as part of the meal insulin.

  eRxInsl: “Total Day Enhanced Prescription Insulin”. A subset of TDD is defined as follows: That is, the sum of dietary insulin + basal insulin during the day. It is generally a “prescription” by a practitioner. The present invention proposes a method for calculating it automatically.

  ExerCarbs: “Exercise curve”. Equivalent to exercises in grams of carbohydrates.

FInsAuto: A multiplier that limits dRxInsAuto. (Automatic multi-day version)
Fixed-Meal: Another simpler method for determining dietary insulin, other than curve counting. Includes having a fixed menu meal preceded by a fixed amount of insulin.

  Float: A change to one of the two main parameters, EmealIns or BR, is estimated or determined by the practitioner, and a change to the other parameter is estimated from the required change in total insulin Algorithm required by subtracting the insulin changes made. A “floating” parameter is that determined by subtraction.

  GlycemicIndex: Ratio of calories available from carbohydrates within 2 hours after consumption to total calories in carbohydrates. Different for different types of carbohydrates.

  Interaction: “Patient / Practitioner Dialogue”. Inclusive terms include person-to-person interactions and telecommunications-based interactions in which patient parameters are readjusted.

  Interval: The day is divided into time intervals so that the patient's insulin administration can be changed throughout the day and analyzed on a time-dependent basis. Sections are generally bounded by the base speed change time.

  Kb: A statistically derived constant used in the estimated basis formula. The latest publication of the “AIM” official lists the latest value of Kb = 0.48 (Ref 1). Used in the formula below.

Foundation = Kb ＊ TDDavg (15)
Kcf: A statistically derived constant used in the estimated CF formula. The latest publication of the “AIM” formula (Ref 1) lists the latest value of Kcir = 1700. Used in the formula below.

CF = Kcf / TDDavg (16)
Kcir: A statistically derived constant used in the estimated CIR formula. The latest publication of the “AIM” formula (Ref 1) lists the latest value of Kcir = 2.8. Used in the formula below.

CIR = Kcir ＊ BodyWeight / TDDavg (17)
KcirW: Constant used in the estimated CIR formula. The latest publication lists the latest value of Kcir = 500. Used in the formula below.

CIR = KcirW / TDDavg (18)
Kcycle: A constant representing the fraction with which you want to reduce CorTot over a specified number of days (Ncycle). (Daily update version)
Keyb: An identifier of a section that is considered to have data that can be relied upon, as in “Key Basic Speed”, and whose BR is also used for other sections.

  Keyc: An identifier of a section that is considered to have data that can be relied upon, as in “Key CIR”, and that CIR is also used for other sections.

  Kf = “K fasting”. Constant <1. Ratio of fasting basal speed to selected "reliable" basal speed.

Kf = BRf / BRreliable (19)
Kmauto: “K meal, automatic”. A positive constant equal to or less than 1 whose purpose is to determine deMealIns. (Automatic multi-day version)
Kfbk: “K Feedback”. A constant in the BoTFbk formula that adjusts the rate of convergence of the base-to-total ratio to the target value (in days). (Daily update version)
KrxInsl: A positive constant less than or equal to 1 whose purpose is to limit dRxInsl. (Daily update version)
KrxInslMax: Maximum allowable value of deRxInsl / TBCorTot. (Multi-day version)
Late-Sleep Time Interval: (same as the time interval before dawn). A time boundary that begins at the mid-sleep time boundary and ends at the breakfast time boundary. This time interval is characterized by a “pre-dawn effect”, which requires a somewhat greater amount of insulin than would normally be expected. This effect is thought to lead to the release of growth hormone. This time interval has the highest reliability of being curve-free among all time intervals. It may be the best candidate for “inactive fasting basal speed” even if it is not a candidate for the pre-dawn effect.

  Meal Insulin: Insulin taken with “instantaneous large doses” taken at the same time as or immediately before consumption of carbohydrates.

  Meal Insulin Float: BR is determined by the practitioner or by the formula of the estimate, and dietary insulin (enhanced or conventional) is based on the total change required, dRxInsl (enhanced or conventional) Sought by subtracting changes. The change in CIR is then sought from dietary insulin (enhanced or conventional). The use of this type of float makes it possible to keep the basal velocity more uniform. This is useful if the patient skips meals.

  MealBol (t): “Instant dietary meals at a unique date and time”. When a patient eats, he calculates the grams of carbohydrate consumed and calculates the high dose of insulin by the following formula:

MealBol (t) = (gm of carbohydrate) / CIR (20)
Where CIR is the personalized carbohydrate-to-insulin ratio.

MealIns (d, i) = Meal insulin during the time interval. (Daily update version)
MealIns (i) = Meal insulin during the time interval. (Updated version of multiple past days)
MealInsTot = total meal insulin for one day
NBGs (i): “Number of BGs in the time interval” over the calendar period being analyzed. (Multi-day update version)
NBGs: “Number of BGs”. The total number of BG meter BG tests since the last download or in the calendar period being analyzed. (Multi-day version)
Ncycle: fraction of insulin error Number of days before Kcycle is removed. (Daily update version)
NDbg: “Number of days of blood sugar”. The number of BG exam days in the calendar period being analyzed. (Multi-day version)
Peg: (verb). Multiply the parameter (CIR or BR) in one interval equal to a constant by the same parameter in another interval. The parameter is said to be “pegged” in the key interval. In general, the constant is equal to the ratio between parameters during the adjustment day by the practitioner, and therefore the same ratio is maintained thereafter.

  Pen: A pocket portable insulin delivery device, so named because it resembles a writing instrument.

PmPctBGsd: Database (population) mean value of quantity (BGsd / BGmean), used in deRxInsl automation. (Automatic multi-day version)
Practitioner: A doctor or nurse who analyzes the parameters of a diabetic patient and prescribes an insulin dosage regimen.

Pre-Dawn Time Interval: (Same as the time interval of late sleep)
PsdPctBGsd: Database (population) standard deviation of quantity (BGsd / BGmean), used in deRxInsl automation. (Automatic multi-day version)
Reliable data set: This term is specifically used to indicate intervals or data with good statistics for a parameter, ie, high BG test frequency and low standard deviation.

  RxInsl: “Total day prescription insulin”. A subset of TDD, defined as the sum of daily dietary insulin + basal insulin. Unlike regulated insulin, prescribed by a practitioner.

  Similar data set: This term is specifically used to indicate a data set that is likely to have parameters that are close to the values that are being statistically determined.

Swtch: “Switching parameters”. A binary parameter whose purpose is to evaluate deRxInsl against several domain limits and move between the two "domain" expressions of deRxInsl (i). (Multi-day version)
Target: A general term. A number used in the feedback algorithm that represents the desired result.

  TargetAM: “Target blood sugar after meals”. Adjusted insulin is calculated by how high the patient's BG is from TargetAM. See CorBol.

  TargetTB: “Target blood sugar before meals”. Adjusted insulin is calculated by how high the patient's BG is from TargetTB. See CorBol.

  Target Basal-to-Total Ratio: The required value divided by the daily basic insulin divided by the total daily insulin. Used in feedback algorithms.

TBCorIns (d-1, i + 1): “Time boundary adjustment insulin”. The patient is generally instructed to test his / her BG at a time boundary. This parameter is the sum of adjusted insulin taken at or near the time boundary, time (i + 1) the previous day. (Daily update version)
TBCorIns (i + 1): “Time boundary adjustment insulin in the (i + 1) th time interval”. The patient is generally instructed to test his / her BG at a time boundary. This parameter is an average over several days of the total amount of adjusted insulin taken at or near the time boundary, time (i + 1). (Multiple past days version)
TBCorTot (d): “Total time boundary adjusted insulin for the current day”. (Daily update version)
TBCorTot: “Time boundary adjusted insulin total”. Sum of TBCorIns (i) over the entire day. (Multiple past days version)
TDD (d-1): “Total daily dose” of insulin. Total amount of insulin the patient received during the previous day. (Daily update version)
TDDavg: “Total daily dose of insulin, average”. Average of TDD over several days. (Multiple past days version)
Time (i): “i” th time boundary.

  Time-Boundary Corrective Insulin: A large amount of adjusted insulin administered at or near the boundary of a time interval. In general, BG studies are performed at time boundaries (just before a meal if the next interval is a meal interval), and adjusted bolus doses are calculated from this BG study.

  TimeLabel (i): A short text phrase that labels each time boundary. PumpMaster provides several standard entries. In other words, mid-sleep (Pre-Breakfast), before lunch (Pre-Lunch), before dinner (Pre-Supper), bedtime, snacks, and basic speed changes ( Basal Rate Change).

Tmax: The patient's last time boundary. Generally bedtime.
4. General guidance
4.1 Changes in carbohydrate to insulin ratio, CIR
The definition of CIR can be reorganized in two different ways: in a way that suits the current situation and in a recommended way, assuming that the carbohydrates do not change.

Curve = CIR * Meal Insulin (21)
CIRrec = Curve / Recommended Meal Insulin (22)
If dietary insulin or curves are not known, differential methods can be used.

CIRrec = CIR + dCIR (23)
For dietary insulin to be used in the formula below, a CIR derivative is required (the derivative is in parentheses).

dCIR = (dCIR / d Meal Insulin) * d Meal Insulin (24)
as well as,
d Meal Insulin = dCIR / (dCIR / d Meal Insulin) (25)
The derivation below leads to a dCIR that results in a change in dietary insulin d (meal insulin), the formula for calculating the change in CIR. There are several ways to calculate or estimate the derivative. To get them, start by defining the CIR.

CIR = curve / (meal insulin) (26)
Differentiate using the calculus as follows.

(dCIR / d Meal Insulin) =-Curve / (Meal Insulin) ² (27)
Substituting equation (21) into equation (27), we obtain

(dCIR / d Meal Insulin) =-CIR / (Meal Insulin) (28)
This is a CIR dependent derivative estimate. By substituting correlations from statistical studies, the CIR can be eliminated from the equation. For example, using AIM correlation (Ref 1):

CIR = Kcir ＊ BodyWt / TDDavg (29)
Here, Kcir is a statistically determined constant.

  Substituting this into Eq. (28) for CIR, we get the following general formula:

(dCIR / d Meal Insulin) =-(Kcir * BodyWt / TDDavg) / (Meal Insulin) (30)
If dietary insulin is not available, it can be calculated as follows:

Meal Insulin = TDDavg-Basics-Adjusted Insulin (31)
Therefore,
(dCIR / d Meal Insulin) =-(Kcir * BodyWt / TDDavg) / (TDDavg-Basics-Adjusted Insulin) (32)
AIM's statistical studies show that the foundation is slightly smaller than 1/2 of TDDavg. Regulated insulin is generally small, so it can be said that dietary insulin is about 1/2 of TDDavg. Therefore, the following equation is derived from equation (30).

(dCIR / d Meal Insulin) =-(Kcir * BodyWt * 2 / TDDavg ² ) (33)
And finally,
dCIR = - (Kcir * BodyWt * 2 / TDDavg 2) * d meal insulin (34)
In another correlation (Ref 2), CIR = Kcirw / TDD is estimated, and the following equation is derived by the same procedure.

dCIR =-(Kcirw ＊ 2 / TDDavg ² ) ＊ d Meal Insulin (35)
Changes in meal insulin, dMealIns, can be obtained as follows.

dMealIns = MealInsNew-MealInsOld (36)
dMealIns = Curve * (1 / CIRnew-1 / CIRold) (37)
This may be entered by the practitioner.
4.2 Some important expressions
4.2.1 Traditional Insulin Naming The above section showing different types of insulin is rewritten as follows using variable names:

  Recall the following definition:

RxInsl: prescription insulin,
TBCorTot: Total day time boundary adjustment insulin,
AMCorTot: Total Day Postprandial Adjusted Insulin
CorTot: Total day adjustment insulin
TDD = RxInsl + CorTot (38)
Where RxInsl = Basic + MealIns (39)
If the change is small,
dRxInsl = dBasal + dMealIns (40)
Within the time interval, the same formula applies,
RxInsl (i) = Basal (i) + MealIns (i) (41)
dRxInsl (i) = dBasal (i) + dMealIns (i) (42)
The present invention incorporates the following ideas. That is, the total daily adjusted insulin, CorTot, represents the “error” in the patient's prescription insulin, and the goal is CorTot = 0. The practitioner (or automatic algorithm) determines the amount of CorTot that should be removed. This amount is called dRxInsl, mimicking the word “change in total prescription insulin”. By making this change, the program can reduce CorTot indirectly in advance in days. dRxInsl can be determined differently by version, as described below.

  Daily update version: KrxInsl is a predetermined constant equal to or less than 1 and is used to set dRxInsl in the following manner.

dMealIns + dBasal = dRxInsl = KrxInsl * CorTot (43)

Similarly, within a time interval,
dMealIns (i) + dBasal (i) = dRxInsl (i) = KrxInsl * CorIns (i + 1) (44)
For the daily update algorithm, currently KrxInsl = 0.16, but the value of KrxInsl should be adjusted for optimal safety and performance.

  Multi-day data version: In the manual subversion, dRxInsl is entered by the practitioner for each patient / practitioner interaction. The value is allowed to be any value up to the maximum value of KrxInslMax * CorTot. Here, KrxInslMax is a constant of fraction.

dMealIns + dBasal = dRxInsl <= KrxInslMax ＊ CorTot (45)
Each time interval obtains the share of “fix” in proportion to the total error share as follows:
dMealIns (i) + dBasal (i) = dRxInsl (i) = dRxInsl * (CorIns (i + 1) / CorTot) (46)
At present, KrxInslMax = 0.5, but the value of KrxInslMax is adjusted for optimal safety and performance.

  In the automatic subversion, dRxInsl is determined in an automatic manner (described in the section entitled “Automatic Multi-Day Data (Digital Advisor)”).

  The left hand side of equation (44) or (46) is performed on the interval in one of two ways: These are named “floats”.

MealIns float:
First, dBasal (i) is estimated. Then, according to the present invention, dMealIns (i) = dRxInsl (i) −dBasal (i) is calculated. One way to estimate dBasal (I) is to borrow a new base speed from another interval.

  Estimate basal velocity from another algorithm (eg, basal float).

Estimate dBasal (i) from the curve-free later part of the time interval, using TBCorIns (i) as an error indicator. The float is performed on the first part of the interval using AMCorIns (i) as the error. (47)
Or
Basic float:
First, dMealIns (i) is estimated. Then, according to the present invention, dBasal (i) = dRxInsl (i) −dMealIns (i) is calculated. One way to estimate dMealIns (i) is to borrow MealIns (i) or CIR (i) from another interval.

  Estimate dMealIns (i) as the share of total daily dMealIns at the same ratio as CarbSh (i) / CarbShTot.

Estimate dMealIns (i) as the share of total dMealIns for the day at the same ratio as MealIns (i) / MealInsTot. (48)
4.2.2 Enhanced Insulin Naming A type of insulin delivery system that incorporates insulin on-board calculations makes it safer for patients to use postprandial insulin administration (variable name: AMCorIns). The reason for calling “adjustment” is to use the adjustment formula, equation (14). However, because it is part of the insulin needed to deal with a particular meal curve, it is more convenient to bundle it with meal insulin by defining new variables.

Enhanced Meal Insulin: eMealIns = MealIns + AMCorIns
Enhanced prescription insulin: eRxInsl = Basic + eMealIns
This can be explained in more detail as follows.

TDD = eRxInsl + TBCorIns (49)
Where eRxInsl = Basic + eMealIns (50)
And eMealIns = MealIns + AMCorIns (51)
The two quantities in eMealIns are added at the start of these calculations and treated as a single variable.

If the change is small,
deRxInsl = dBasal + deMealIns (52)
Within the time interval, the following similar formula applies:

eRxInsl (i) = Basal (i) + eMealIns (i) (53)
deRxInsl (i) = dBasal (i) + deMealIns (i) (54)
The present invention incorporates the following ideas. That is, the total time boundary adjusted insulin, TBCorTot, represents the “error” in the patient's augmented prescription insulin, and the goal is TBCorTot = 0. The practitioner (or automatic algorithm) determines the amount of TBCorTot that should be removed. This amount is called deRxInsl, mimicking the word “change in total augmented prescription insulin”. By making this change, the program can indirectly reduce TBCorTot in days in advance. deRxInsl can be determined differently by version, as described below.

  Daily update version: KrxInsl is a predetermined constant equal to or less than 1 and is used to set deRxInsl in the following manner.

deMealIns + dBasal = deRxInsl = KrxInsl * TBCorTot (55)
Similarly, within a time interval,
deMealIns (i) + dBasal (i) = deRxInsl (i) = KrxInsl * TBCorIns (i + 1) (56)
For the daily update algorithm, currently KrxInsl = 0.16, but the value of KrxInsl should be adjusted for optimal safety and performance.

  Multi-day data version: In the manual subversion, deRxInsl is entered by the practitioner for each patient / practitioner interaction. The value is allowed to be any value up to the maximum value of KrxInslMax * TBCorTot. Here, KrxInslMax is a constant of a fraction equal to or less than 1.

deMealIns + dBasal = deRxInsl <= KrxInslMax ＊ TBCorTot (57)
Each time interval obtains the share of “fix” in proportion to the error as follows:

dMealIns (i) + dBasal (i) = deRxInsl (i) = deRxInsl *
(TBCorIns (i + 1) / TBCorTot) (58)
At present, KrxInslMax = 0.5, but the value of KrxInslMax is adjusted for optimal safety and performance. In the automatic subversion, deRxInsl is determined in an automatic way (denoted “Automatic Digital Assistant”).

  The left hand side of equation (56) or (58) is performed on the interval in one of two ways: These are named “floats”.

MealIns float:
First, dBasal (i) is estimated. Then, according to the present invention, deMealIns (i) = deRxInsl (i) −dBasal (i) is calculated. One way to estimate dBasal (i) is to borrow Basal (i) from another interval.

Estimate Basal (i) from another algorithm (eg, basic float). (59)
Or
Basic float:
First, deMealIns is estimated. Then, according to the present invention, dBasal (i) = deRxInsl (i) −deMealIns (i) is calculated. One way to estimate deMealIns is to borrow eMealIns (i) or CIR (i) from another interval.

deMealIns (i) is estimated as the share of total deMealIns for the day at the same ratio as CarbSh (i) / CarbShTot. (60)
deMealIns (i) is estimated as the share of total deMealIns for the day at the same ratio as eMealIns (i) / eMealInsTot.
4.2.3 Overview In certain versions, different time intervals may have different types of floats.

  The basic float has the advantage that it can be used for all time intervals in the day and can be easily modified to fit the patient's daily routine.

Meal insulin float can keep the basal rate schedule simple, and changes in dietary habits can only be handled by a CIR that tackles the change if a person consumes a curve and performs a massive bolus of meals, There is an advantage. However, since meal insulin float can only affect time intervals that have meals within the time interval, the basal float formula must still be used for non-meal intervals.
4.3 Two-level basal system and Kf calculator Depending on the practitioner, a “two-level basal” schedule, ie, a single basal rate, BRf (for BR fasting), except for the time period of late sleep immediately before waking Some people prefer to allocate during the last 24 hours. The late-sleep segment is different because of the “pre-dawn phenomenon” (which requires somewhat more insulin during the second half of the sleep period). Therefore, the basic speed is two levels BRf and BRlateSlp. Finding BRf is an important task.

  Non-meal intervals may be best for determining BRf. In general, the non-meal section is an early sleep period and a late sleep period. However, none of these are perfect candidates.

  The early sleep interval will give the best BRf value if it is truly mealless, but patients are often “fooled” by having a late night meal. Also, the patient must interrupt his or her sleep and perform the necessary late-sleep BG test, so data is often not available.

  The late sleep or pre-dawn section is most easily available and most often without meals, but this section is inappropriate due to the “dawn phenomenon”. None of these are perfect candidates for BRf.

The present invention incorporates two generalized methods designed to give several ways to calculate BRf. Here we have to explain some naming. BRf is intended to be used in most sections except for some sections called “excluded sections”. One interval or category of BR with good statistical reliability may be designated as a “reliable” (having basal speed BRreliable) interval. Another interval or data set may be designated as a “similar” interval because its base speed, BRsimilar, is similar to BRf. The present invention obtains BRf using two general methods.
4.3.1 Statistical correlation
BRf is estimated to fit a familiar least squares formula, ie the paired value of “reliable” and “similar” data over the recent calendar period to the formula BRsimilar = Kf * BRreliable. May be obtained by statistically relating to the correlation constant, Kf.

Kf = SUM [(BRsimilar) * (BRreliable)] / SUM [BRreliable ² ] (61)
Once calculated, the correlation constant, Kf, is used to convert a “reliable” basal rate value into a BRf value until the next practitioner interaction.

BRf = BRreliable ＊ Kf (62)
4.3.2 Average calculation
The BRf may be obtained by averaging several sets of BRsimilar data from the same or previous day (for daily update versions) or from the most recent calendar period (for multi-day versions). The smoothing effect of the averaging method helps to produce good values.

  In the following, some specific applications of these two methods will be described.

BR in early sleep at night when there was no midnight snack is used as BRsimilar. These are paired with the BR value of the late late sleep in the early morning of the next calendar day and are used as BRreliable. The formula used later is BRf = Kf * BRlateSlp. (63)
BR averages other than BRlateSlp may be paired with BRlateSlp as a reliable data set and named as a similar data set. The resulting formula looks the same as equation (63), but with different Kf.

  The average of all BRs except BRlateSlp may be used directly as BRf.

  The average of all BRs, ie (basic / 24), may be used directly as BRf.

  The average of all BRs, ie (basic / 24), may be paired with BRlateSlp as a reliable data set and named as a similar data set. Again, the resulting formula is the same as equation (63).

BR in early sleep at night when there was no midnight snack is used as BRsimilar. These are paired with the base / 24 “trusted” values. The formula used later is BRf = Kf * Basis / 24. (64)
BR averages other than BRlateSlp may be paired with all BRs as reliable data sets, ie, the base / 24 averages, and named as similar data sets. The resulting formula is the same as in equation (64), but Kf is different.

  The average of base float 1 results for all other intervals (other than late sleep intervals) may be paired with BRlateSlp as a reliable data set and named as a similar data set. Again, the resulting formula looks the same as equation (63).

  The average of the basic float 2 results for all other intervals (other than the late sleep interval) may be paired with BRlateSlp as a reliable data set and named as a similar data set. Again, the resulting formula looks the same as equation (63).

  The average of meal insulin float 2 results for all other intervals (other than late sleep intervals) may be paired with BRlateSlp as a reliable data set and named as a similar data set. Again, the resulting formula looks the same as equation (63).

  Additional uses of this type will be referred to throughout the text below.

General theory of Kf
The Kf calculator can be used outside of the insulin delivery device and is particularly suitable for multi-day data versions. Designed for use by practitioners during patient / practitioner interactions. This is because it depends on a sufficient number of nightly digital memories so that enough data points are obtained to calculate the exact value of Kf.
5. Types of Insulin Delivery Systems The following is a list of insulin delivery systems, showing the version of the invention applicable to each.

  Type A pump: TDD memory for several days. It has a basic program schedule, Basal (i) memory. There is no memory for CF, CIR, CIR (i), BG (i), MealBol (t), MealIns (i), TBCorIns, AMCorIns, CorBol (t), or CorIns (i). Therefore, daily average combined bolus should be obtained by subtracting the basis from TDDavg. Example: Medtronic MiniMed Paradigm 511.

  Version 6.2.1.1.4: This uses multi-day cumulative data. The program is external to the pump.

  Type B pumps: In addition to type A memory, these are BG (i), ie, the actual curve for CarbSh (i), CIR (i), CF, and combination bolus, for Boli (i) Have the memory. Example: Medtronic MiniMed Paradigm 5 12 and 712.

    Version 6.2.1.1.3: Use multi-day data. The program is external to the pump.

  Type C pumps: In addition to the other type B memories, these have memories for MealIns (i) and CorIns (i). Example: Deltec Cozmo.

    Version 6.2.1.1.2: Use multi-day data. The program is external to the pump.

  Type D pumps: In addition to Type C memories, these pumps have, as a minimum, memories for AMCorIns (i) and TBCorIns (i).

    Version 6.2.1.1.1: Use multi-day data. The program is external to the pump.

  Type E pump: As a minimum, it has all the parameters of a type D pump and has a program installed inside.

    Version 6.1.2: Use automatic daily updates. The program is inside the pump.

    Version 6.2.1.1.1: Use multi-day cumulative data. The program is external to the pump.

  Subcutaneous or inhaled insulin delivery where the insulin delivery device and the BG meter are either clipped together in the kit or otherwise communicated with each other and data is digitally recorded.

    Version 6.1.3: Use automatic daily updates. The program is inside the kit.

    Version 6.2.1.2: Use multi-day cumulative data. The program is outside the kit.

These have been reorganized in the table of contents to facilitate presentation. The difference between algorithm types is well understood.
6. Version, embodiment, description by algorithm
6. 1. Version with daily updates
6.1.1 Preliminary guidance
6.1.1.1 Time interval and adjusted insulin The index “d” represents the current day.

  The index “i” represents the “i” th time boundary or time interval (following the time boundary).

  The time-bound adjusted insulin, TBCorIns (d, i), is the sum of the adjusted bolus that occurs at the beginning of the i-th interval. Each bolus must be identified as a time-bound bolus and must be identified in interval i. This may be done by appropriately combining the following methods.

  If it falls between the midpoint of the previous section and the midpoint of the i section, it is identified by the i section.

  If it comes before a meal bolus having the same index number, it is identified as a time boundary adjusted bolus.

  One or both of these identifiers entered by the patient using the controller of the insulin delivery device during a bolus dose.

  Postprandial adjusted insulin AMCorIns (d, i) is the sum of adjusted bolus doses identified by an appropriate combination of the following methods.

  When entering the i-th section, it is identified by the i-th section.

  It is identified as a post-meal adjusted bolus by the fact that it comes after a bolus of meals marking the start of the interval.

One or both of these identifiers entered by the patient using the controller of the insulin delivery device during a bolus dose.
6.1.1.2 KrxInsl: governs the magnitude of insulin changes
If TBCorTot = TBCorTotStart, let us analyze the operation of the present invention, starting from any start. This represents an error of the present invention. The objective is to reduce the “error”, TBCorTot, for each day by a fractional amount, KrxInsl * TBCorTot (where KrxInsl <= 1). This means that starting with the “starting error”, TBCorTotStart, the error will decrease in the geometric sequence. Number of days, Ncycle, and after
Residual error, TBCorTot (Ncycle) = TBCorTotStart * [1 - KrxInsl] Ncycle (65)
This residual error approaches near zero as Ncycle increases. The phrase “near zero” can be defined by setting a reasonably small number for the percent residual error.

Percent residual error = [1-KrxInsl] ^Ncycle (66)
If the percent residual error and Ncycle are combined appropriately, the equation can be solved for KrxInsl. This is done according to the invention as follows.

KrxInsl = 1-(Percent residual error) ^{(1 / Ncycle)} (67)
The percent residual error is set by the inventor but is subject to change. (Typical value is 10%). The choice of Ncycle is left to the practitioner. (A typical value is 14 days). In this example, the preferred value KrxInsl = 0.16 is derived. For simplicity, selecting 0.16 for KrxInsl eliminates 90% of the error within two weeks. Used in the daily update algorithm as follows:

deRxInsl = KrxInsl * TBCorTot (68)
Or
dRxInsl = KrxInsl * TBCorTot (69)
Within the leg,
deRxInsl (i) = KrxInsl * TBCorIns (i + 1) (70)
Or
dRxInsl (i) = KrxInsl * TBCorIns (i + 1) (71)
6.1.2 About the pump (type E)
6.1.2.1 Base-to-total ratio:
The Basal (d) / TDD (d) ratio is called BoT (d) instead of “Basal over TDD”. This is calculated daily. Medical studies have determined that several Basal (d) / TDD (d) ratios lead to better management of diabetes. Basal (d) / TDD (d) ratio goals can be set in the program by the patient or practitioner. This is called BoTTgt instead of “Basal over TDD, Target”.

  The present invention includes a feedback coefficient to bring BoT (d) to BoTTgt. The feedback factor incorporates a constant Kfbk to adjust the speed of convergence.

BoTFbk (d) = BoT (d) + Kfbk * (BoTTgt-BoT (d)) * sign (deRxInsl) (72)
The resulting value is not allowed to be less than zero or greater than 1.

  Multiplying this factor by the proposed total insulin change for the day yields the amount of change to the basis. The “sign” function ensures that the magnitude of the change is correct for the “direction” of the change. The constant Kfbk is adjusted to optimize the speed of convergence to BoTTgt. The feedback factor is applied as follows.

dBasal = BoTFbk (d) * deRxInsl (73)
deMealIns = (1-BoTFbk (d)) * deRxInsl (74)
The present invention provides a recommended value for BoTTgt called BoTTgtRec, which is calculated as follows.

BoTTgtRec = 1-4 * CarbShTot * (mean glycemic index) / (statistical correlation to caloric intake as a function of height, weight, and other easily understood patient parameters)
This formula uses the conversion rate 4 to change the curve for calories, then multiply by the glycemic index to obtain calories immediately obtained from the curves, and then the patient's caloric intake estimated from body shape To break.

  This indicates the MealIns / TDD condition. The base to total ratio is 1 minus this amount.

Current versions of the present invention generally do not use this idea for both dBasal and deMealIns in the same time interval, but instead use one of these to state in the definition of “float”. Find the other parameters by subtracting from the sum, deRxInsl in the same way.
6.1.2.2 Basic float 1, (deMealIns is proportional to eMealIns or curve)
This algorithm estimates dietary insulin using the feedback factor BoTFbk (d). Use the enhanced insulin nomenclature system. The total change in augmented meal insulin is taken from equations (72) and (76).

deMealIns (d) = KrxInsl * (1-BoTFbk (d-1)) * TBCorTot (d-1) (75)
This portion of the amount (assigned to a section) is proportional to the share of that section of augmented meal insulin from the previous day. This idea is applied by the following formula.

deMealIns (d, i) = KrxInsl * (1-BoTFbk (d-1)) * TBCorTot (d-1) * eMealIns (d-1, i) / eMealInsTot (d-1) (76)
eMealIns (d, i) = eMealIns (d-1, i) + KrxInsl * (1-BoTFbk (d-1)) * TBCorTot (d-1) * eMealIns (d-1, i) / eMealInsTot (d-1 ) (77)
Using equation (22),
CIR (d, i) = CarbSh (d-1, i) / [eMealIns (d-1, i) + KrxInsl * (1-BoTFbk (d-1)) * TBCorTot (d-1) * eMealIns (d- 1, i) / eMealInsTot (d-1)] (78)
As in statement (60) in the definition of the basic float:
BR (d, i) = BR (d-1, i) + (deRxInsl (d, i)-deMealIns (d, i)) / dt (i) (79)
Using equation (70),
BR (d, i) = BR (d-1, i) + KrxInsl * [TBCorIns (d-1, i + 1)-(1-BoTFbk (d-1)) * TBCorTot (d-1) * eMealIns ( d-1, i) / eMealInsTot (d-1)] / dt (i) (80)
This algorithm can be used for all time intervals “across the board”. This algorithm has the effect of maintaining the first prescribed “shape” of the CIR or MealIns schedule when used in all time intervals. That is, each eMealIns (d, i) is multiplied by the same factor, so they increase or decrease consistently.
6.1.2.3 Base float 2, (deMealIns value from outside the interval)
This algorithm is used in sections where MealIns or CIR values are borrowed from outside the section (pegged to another section). This method is useful during non-meal time intervals. This is because the meal insulin data in these intervals is not reliable enough to be used for calculation. All CIRs increase or decrease in proportion to the “key CIR”. The CIR schedule maintains the same “shape” and each CIR maintains the same ratio to the same key CIR that it had at d = 1 for the first day. The key CIR may be selected from a number of sources, i.e., generally from a time interval that has a meal and uses a meal float algorithm. The number of the key CIR section is keyc. Use the enhanced insulin nomenclature system.

CIR (d, i) = [CIR (d-1, keyc) / CIR (1, keyc)] * CIR (1, i) (81)
Equation (37) is rewritten as shown below. Note that the augmented naming system is used, but the necessary changes are the same on both systems.

deMealIns (d, i) = dMealIns (d, i) = CarbSh (d-1, i) * [1 / CIR (d, i)-1 / CIR (d-1, i)] (82)
BR (d, i) = BR (d-1, i) + dBR (d, i) (83)
The “float” formula is obtained from statement (60) as follows:

BR (d, i) = BR (d-1, i) + (deRxInsl (d, i)-deMealIns (d, i)) / dt (i) (84)
Equation (56) is then applied as follows.

BR (d, i) = BR (d-1, i) + [KrxInsl * TBCorIns (d-1, i + l)-deMealIns (d, i)] / dt (i) (85)
BR (d, i) = BR (d-1, i) + [KrxInsl * TBCorIns (d-1, i + 1)-CarbSh (d-1, i) * [1 / CIR (d, i)-1 / CIR (d-1, i)]] / dt (i) (86)
6.1.2.4 Meal insulin float 1
This algorithm pegs BR to the “key” basal rate, ie, all basal rates increase or decrease in proportion to the key basal rate. The basal schedule maintains the same “shape” and each basal rate maintains the same ratio to the same key basal rate that d = 1 for the first day. The key basal rate may be selected from several sources, i.e., generally from a time interval that has a meal and uses a basal float algorithm. The key interval number is keyb. Use the enhanced insulin nomenclature system.

BR (d, i) = [BR (d-1, keyb) / BR (1, keyb)] * BR (1, i) (87)
The key basal velocity can also be obtained from the overall result of the basic float 1 algorithm, ie, calculation using Basal (d-1) plus correction. The formula uses d = 1 as in the above formula.

BR (d, i) = [Basal (d-1) + BoTFbk (d-1) * KrxInsl * TBCorTot (d-1)] / Basal (1) * BR (1, i) (88)
The key basic speed can also be obtained as follows.

From the sum of the basic speeds over all sections,
Basal (d) = SUM (BR (d, i) * dt (i)) (89)
BR (d, i) = [(Basal (d) / Basal (1)] * BR (1, i) (90)
Whatever the source of BR (d, i),
dBR (d, i) = BR (d, i)-BR (d-1, i) (91)
As in the statement (59) in the definition of dietary insulin float,
deMealIns (d, i) = KrxInsl * TBCorIns (d-1, i + 1)-dBR (d, i) * dt (i) (92)
deMealIns (d, i) = KrxInsl * TBCorIns (d-1, i + 1)-(BR (d, i)-BR (d-1, i)) * dt (i) (93)
eMealIns (d, i) = eMealIns (d-1, i) + KrxInsl * TBCorIns (d-1, i + 1)-(BR (d, i)-BR (d-1, i)) * dt (i ) (94)
To make equation (22) easier to understand,
CIR (d, i) = CarbSh (d-1, i) / eMealIns (d, i) (95)
CIR (d, i) = CarbSh (d-1, i) / (eMealIns (d-1, i) + KrxInsl * TBCorIns (d-1, i + 1)-(BR (d, i)-BR (d -1, i)) * dt (i)] (96)
6.1.2.5 Outline of 6.1.1.2 and 6.1.1.3 The basic float 2 algorithm is used in conjunction with a meal insulin 1 float as follows.

For the first day, for d = 1, enter the practitioner's latest prescription. This data is the starting point of the automatic daily update sequence. For non-meal intervals, the basic float 2 algorithm is given by pegging their CIR into the “Keyc” interval. One of the intervals including non-meal is designated as the “Keyb” basal rate (generally the interval of late sleep or average basal / 24). Basal velocities in the section containing meals are pegged to key basal rates to maintain their initial ratio to the key basal rate. A meal insulin float algorithm is given to a section including a meal.
6.1.2.6 Meal float 2 (AMCorIns (i) is used as an error)
This algorithm uses the postprandial BG test. These trials and related postprandial bolus doses divide the interval into two parts. The algorithm uses AMCorIns (d−1, i) as an indicator of “error” in the first part of the interval in which the float is calculated. The value of dBasal (d-1, i) is obtained from the curve-free second part of the interval using TBCorIns (d-1, i + 1) as the error term for dBasal (d-1, i). The term “enhanced” insulin is not used.

Define as follows. In other words, Timeb (d-1, i) = time for large-scale instantaneous postprandial administration after the i-th time boundary
dtb (d-1, i) = Time (i + 1)-Timeb (d-1, i) (97)
dta (d-1, i) = Timeb (d-1, i)-Time (i) (98)
In the second part of the section,
BR (d, i) = BR (d-1, i) + KrxInsl * TBCorIns (d-1, i + 1) / dtb (d-1, i) (99)
Once the basal speed is obtained from the second part of the leg (or another source), the following statement will be true for the first part of the leg.

dBR (d, i) = BR (d, i)-BR (d-1, i) (100)
MealIns (d, i) = MealIns (d-1, i) + KrxInsl * AMCorIns (d-1, i)-dBR (d, i) * dta (d-1, i) (101)
CIR (d, i) = CarbSh (d-1, i) / MealIns (d, i) (102)
6.1.3 About Multiple Dose Injection (MDI) and Inhaled Insulin Basal insulin is administered in the form of long-acting insulin as infrequently as once a day. Conditioned insulin and dietary insulin are administered as needed in the form of injection or inhaled short-acting insulin. This algorithm is suitable for a “kit” where the BG meter and insulin delivery device are clipped together or otherwise linked together to maintain a place where a combination of digital history and program resides. is there.
6.1.3.1 Base float 1, (not used for MDI or inhaled insulin)
6.1.3.2 Base float 2, (dMealIns values are from outside the interval)
This algorithm is used in sections where MealIns or CIR values are borrowed from outside the section (pegged to another section). This method is useful during non-meal time intervals. This is because in these intervals, meal insulin data is not reliable enough to be used for calculations. All CIRs increase or decrease in proportion to the “key CIR”. The CIR schedule maintains the same “shape”, and each CIR maintains the same ratio to the key CIR that it had at d = 1 during the day of the most recent interaction with the practitioner. The key CIR may be selected from a number of sources, i.e., generally from a time interval that has a meal and uses a meal float algorithm. The number of the key CIR section is keyc. Use the enhanced insulin nomenclature system.

CIR (d, i) = [CIR (d-1, keyc) / CIR (1, keyc)] * CIR (1, i) (103)
Equation (37) is rewritten as shown below. Note that the augmented naming system is used, but the necessary changes are the same on both systems.

deMealIns (d, i) = dMealIns (d, i) = CarbSh (d-1, i) * [1 / CIR (d, i)-1 / CIR (d-1, i)] (104)
BR (d, i) = Basal (d-1) / 24 + dBR (i) (105)
From equation (54), we can see that:

BR (d, i) = Basal (d-1) / 24 + (deRxInsl (d, i)-deMealIns (d, i)) / dt (i) (106)
BR (d, i) = Basal (d-1) / 24 + [KrxInsl * TBCorIns (d-1, i + 1)-deMealIns (d, i)] / dt (i) (107)
Non-meal intervals using this algorithm (or any other suitable source) are used as “reliable” data sets. The basal velocity resulting from equation (117) is used to determine the basis for the entire day.

Basal (d) = 24 * (some conversion rate) * BR (d-1, reliable) (108)
The recommended practice is to use the late sleep interval as the “reliable” interval. The following equations are shown without explanation because they are covered in a separate section.

Basal (d) = 24 * Kf * BR (d-1, late sleep) (109)
6.1.3.3 Meal Insulin Float 1
Use augmented insulin naming system. This algorithm is mainly used for intervals including meals. The overall daily Basal (d) can be derived from the key interval's basic float calculation, as in equation (119).
dBasal (d, i) = (Basal (d)-Basal (d-1)) * dt (i) / 24 (110)
Whatever the source of the underlying data, the meal insulin float continues as follows.

  The present invention calculates a new CIR (d, i) value schedule using the following equation similar to equation (103):

CIR (d, i) = CarbSh (d-1, i) / [eMealIns (d-1, i) + KrxInsl * TBCorIns (d-1, i + 1)-
(Basal (d)-Basal (d-1)) * dt (i) / 24] (111)
6.1.3.4 Overview of 6.1.3 The daily schedule is programmed as follows. That is, a certain section (generally a non-meal section) is designated as a “reliable” section, and a basic float 2 algorithm is given by pegging its CIR into a meal section. The BR from this “reliable” interval is used to determine the BR over 24 hours and the total daily basal dose (which is equal to a constant (eg, Kf or its hand constant) × BRreliable × 24) use. For the interval containing the meal, the meal insulin float algorithm is given and the same single value of the 24 hour BR is used. Non-meal intervals other than “reliable” intervals do not include calculations, their CIRs are pegged into the meal interval, and their BRs are the same as all other BRs.
6.1.3.5 Meal insulin float 2, (using AMCorIns (i) as error term)
This algorithm uses the postprandial BG test. These trials and related bolus doses divide the interval into two parts. The algorithm uses AMCorIns (d−1, i) as an indicator of “error” in the first part of the section where the float is performed.

Define as follows. In other words, Timeb (d-1, i) = time for large-scale instantaneous postprandial administration after the i-th time boundary
dtb (d-1, i) = Time (i + 1)-Timeb (d-1, i) (112)
dta (d-1, i) = Timeb (d-1, i)-Time (i) (113)
The base speed is the same in all sections, and only Basal (d) is required. This may be obtained from one of several sources.

TBCorIns (d-1, i + 1) is used as the error term of dBasal (d, i) for the second curve-free part of the section. The term “enhanced” insulin is not used. In the second part of the section,
BR (d, i) = BR (d-1, i) + KrxInsl * TBCorIns (d-1, i + 1) / dtb (d-1, i) (114)
This result, derived from a single “key” interval, results in Basal (d).

Basal (d) = [BR (d, key) / BR (1, key)] * Basal (1) (115)
The full schedule of results from the second part of the interval, equation (116), may be converted to a full day basis as follows:

SUM [BR (d, i) * dt (i)] over Basal (d) = i (117)
A basal float 1 algorithm, similar to version 6.1.2.1, within the “reliable” interval may be used to give a basal rate during 24 hours.

  The Basic Float 1 algorithm may be used in a comprehensive manner to obtain a daily total basis.

Basal (d) = [Basal (d-1) + BoTFbk (d-1) * KrxInsl * TBCorTot (d-1)] (118)
Whatever the source of Basal (d), for the first part of the interval, we can say

MealIns (d, i) = MealIns (d-1, i) + KrxInsl * AMCorIns (d-1, i)-(Basal (d)-Basal (d-1) * dta (i) / 24 (119)
If the source is the second part of the interval, this is

MealIns (d, i) = MealIns (d-1, i) + KrxInsl * AMCorIns (d-1, i)-BoTFbk (d-1) * KrxInsl * TBCorTot (dl) * dta (i) / 24 (120)
CIR (d, i) = CarbSh (d-1, i) / MealIns (d, i) (121)
6.1.4 Omitted BG
Version 1 and all sub-versions track a special “virtual interval” that consists of a combined interval of early sleep and late sleep. This is used if the patient skips the mid-sleep BG study. Over this combined interval, the basic float calculation is used to give basal velocity to both constituent sections, even if the currentity is maintained and the BG test for both constituent sections is omitted. This technique is used for frequently omitted BGs such as the mid-sleep BG described above. This technique can be used for any number of adjacent time intervals with BG omitted on the boundary. However, for BGs that are rarely omitted, it is more convenient to simply substitute BG = target and calculate a zero value for CorIns (i). This does not cause a change and the previous value can be used as is. Therefore, the BG (d, i) parameter will have the default value of TargetAM or TargetTB until it is entered.
6.1.5 Modified bolus calculator Many pump models have an internally programmed bolus calculator. This is calculated using equations (20) and (14). The present invention deals with exercises as a special application of a high dose instantaneous dose calculator. There is a practice input box in curve units that defines a new variable, ExerCarbs. The arithmetic difference between them is used as the variable Carbm shown below.

Carbm = Curve-ExerCarbs (122)
This can also be used only for exercises without curves. The modified bolus calculator can also calculate a corrected bolus. These must be designated as a “time boundary” or “after meal” by the patient flagging the memory record and selecting the correct one of the two goals, TargetAM and TargetTB. Corrected bolus doses (meal and correction) are summed. And if the result is positive, the pump infuses the calculated amount of insulin as a normal bolus. If the result is negative, the pump suspends basal pumping for the amount of time calculated as follows:

TimeOut =-NegativeBolus (t) / BR (d, i) (133)
Or reduce the basis for the next time.

TimeReduced =-NegativeBolus (t) / (BR (d, i)-BRreduced) (134)
6.1.6 Changing patient schedule In section 1.1 above, we described the usual 9-5 days. However, patients have many different schedules. To allow this, the type of interval is identified for each time interval by the parameter IntrvlType which can have the following values:

M: “Meal section”
Kb: “keyb interval”, generally the late sleep interval, or other non-meal interval where no food is consumed
Kc: "keyc interval", pegged CIR source
S: “Snack section”, food is occasionally consumed.

R: “Reliable”
The adjustment of this function is performed by the practitioner (the patient is not performed without the guidance of the practitioner). The practitioner will check whether there is always a value for IntrvlType for each time interval. He or she can change this if desired. There are no restrictions on other values. The present invention processes this data as follows.

  If IntrvlType = Kb, go to Algorithm 6.1.2.3, Basic Float 2.

  As a result, BR (d, keyb) is calculated.

  If IntrvlType = Kc, go to Algorithm 6.1.2.4, Meal Insulin Float 2.

  As a result, BR (d, keyb) is calculated.

  If IntrvlType = R, go to Algorithm 6.1.2.3, Basic Float 2.

  As a result, BRsimilar is calculated.

  If IntrvlType = S, go to Algorithm 6.1.2.3, Basic Float 2.

(THEN) if IntrvlType = M
Go to algorithm 6.1.4, without IF time boundary BG.

If there is IF postprandial adjusted dose, algorithm 6.1.2.6 Meal Insulin Float 2
go.

Go to ELSE Algorithm 6.1.2.4, Meal Insulin Float 1.
6.2 Versions with multi-day data These versions use average cumulative data over the calendar period prior to the patient / practitioner interaction, instead of re-adjusting patient parameters with the previous day's data. The data for each time interval during the day is averaged separately over all days. The daily index, “d”, is skipped and instead the new parameters are distinguished from the current parameters as follows: That is, the current parameter has no suffix. Parameters that are calculated and recommended have a suffix “rec”. Prescription parameters are suffixed with “rx”. The “rx” parameter is input after the practitioner considers the recommended value of the “rec” parameter. Parameters marked with “Parameter (i)” are parameters in the i-th section. Parameters with nothing in parentheses are unscheduled parameters that have only one value for each patient / practitioner interaction. Thus, “BRrec (i)” refers to the recommended basal rate for the i-th interval, eg, BRrecl, BRrec2..., And BRrx (i) refers to the prescribed basal rate.
6.2.1 Multi-day data for medical practitioners (DIGITAL ADVISOR)
KrxInsl is replaced by deRxInsl / TBCorTot and therefore (135)
deRxInsl (i) = deRxInsl / TBCorTot * TBCorIns (i + 1) (136)
Interactive input form:
A preferred form of the present invention is a two-table access database. There is a “one-to-many” relationship between the “patient table”, Tp, and the “dialogue table”, Ti. There is a digital interactive input form (discussed in more detail in Section 6.2.1.1.1).

The practitioner obtains some of the patient's data by examination or interview. Other data can be downloaded from the pump and BG meter, either directly into the present invention or using manufacturer software, to create a printed copy, which is then transferred to the interactive digital input form of the present invention. By manually transferring to A manufacturer's downloading software generally has a schedule of bins (a time period that envelops a primary time boundary) and includes several BG (i), AMCorIns (i), TBCorIns (i), and MealIns (i) The average of each parameter for each time boundary is calculated. (As an example, recall that MealIns (i) represents the average of the total meal bolus dose in bin i.)
6.2.1.1 About the pump
6.2.1.1.1 Pump types D and E
The formulas for pump types D and E are the simplest and are therefore discussed first.
Interactive input form:
The main input form is the outer panel of FIG. This is a collection of patient permanent demographic data for Table Tp. The subform (inner panel) collects data for the table Ti from the patient / practitioner interaction. The subform has two pages that can be reached with a scroll bar. The first page (FIG. 1) corresponds to an unindexed parameter that has a single value for each patient / practitioner interaction. The second page (FIG. 2) corresponds to a standard modal day schedule, which includes parameters having those similar to the time interval index referred to in the “parameter (i)” format.
6.2.1.1.1.1 Base float 1, (deMealIns is proportional to eMealIns or curve)
This algorithm is similar to the daily updated version of the basic float algorithm 6.1.2.2. Use the enhanced insulin nomenclature system.

  To show that dBasal is determined by the other two parameters, we transfer equation (52).

dBasal = deRxInsl-deMealIns (137)
dBasal (i) = deRxInsl (i)-deMealIns (i) (138)
deMealIns (i) is estimated by assuming that the distribution of deMealIns within the interval is proportional to eMealIns. That is,
deMealIns (i) = deMealIns * eMealIns (i) / eMealInsTot (139)
eMealInsRec (i) = eMealIns (i) + deMealIns * eMealIns (i) / eMealInsTot (140)
By equation (22)
CIRrec (i) = CarbSh (i) / eMealInsRec (i) (141)
CIRrec (i) = CarbSh (i) / [eMealIns (i) + deMealIns * eMealIns (i) / eMealInsTot] (142)
Note that both eMealInsRec (i) and CIRrec (i) are simply the initial value multiplied by the coefficient across the board. In the way of statement (60), the basis is “float” by subtracting deMealIns (i) from deRxInsl (i).

BRrec (i) = BR (i) + [deRxInsl (i)-deMealIns (i)] / dt (i) (143)
Substituting this into equations (133) and (136),
BRrec (i) = BR (i) + [deRxInsl / TBCorTot * TBCorIns (i + 1)-deMealIns * eMealIns (i) / eMealInsTot] / dt (i) (144)
The value of the following formula is also calculated:

BRaveRec = (Basal + deRxInsl-deMealIns) / 24 (145)
The practitioner uses this version of the invention as follows. That is, if all patient data is entered from a download or otherwise, the base float determines the main goals (schedule for recommended basal speed, BRrec (i) and recommended CIRrec (i)). Almost ready to calculate. The practitioner must enter at least two of the three quantities in equation (134). After that, the present invention performs. Therefore, the practitioner examines the data obtained so far, especially TBCorTot. Then he or she decides how much TBCorTot to add to the prescription insulin as a change. He or she then enters deRxInsl, which must be within the built-in input limits (see section Restricted Domains). Basal (i) goes “floating”, so the other amount needed from the practitioner is deMealIns. Before entering, he must ask himself, "How much of that amount do you want to allocate to enhanced dietary insulin when prescribing this change, deRxInsl?" He can make this decision by comparing the basis / TDDAvg (known as BoT in this embodiment) ratio to the optimal value from statistical studies (48% in the latest publication). . For example, if the basis is too high, he can use a value greater than half of deRxInsl as deMealIns. This increases deMealIns relative to deRxInsl, thus reducing the basis. There may be other medical types of considerations that can affect practitioner decisions. The selection of deMealIns may have input limits. The outputs are BRrec (i), BRavgRec, and CIRrec (i) calculated by the above equation. The practitioner takes these recommended values into account and enters an “rx” value based on his judgment.

6.2.1.1.1.2 Basic float 2, (not used for multi-day data)
6.2.1.1.1.3 Meal insulin float 1
This version is similar to version 6.1.2.4, Meal Insulin Float. For input, a prescribed basal rate, BRrx (i), entered by the practitioner is required. The practitioner may desire “advice” before entering BRrx (i). The values of BRrec (i) and BRaveRec from the results of version 6.2.1.1.1.1, basic float 1 above are good advice, so they are shown in the same computer screen input form. The BRf value from the Kf calculator is also good advice. Use the enhanced insulin nomenclature system.

  In order to show that deMealIns is determined by the other two parameters, equations (52) and (54) are transposed.

deMealIns = deRxInsl-dBasal (146)
deMealIns (i) = deRxInsl (i)-dBasal (i) (147)
BRrx (i) = practitioner input (148)

“Float” is very similar to equation (100) and is set up as follows:

  Using equation (132), modify equation (100) as follows:

deMealIns (i) = (deRxInsl / TBCorTot) * TBCorIns (i + 1)-(BRrx (i)-BR (i)) * dt (i) (149)
eMealInsRec (i) = eMealIns (i) + deRxInsl / TBCorTot * TBCorIns (i + 1)-(BRrx (i)-BR (i)) * dt (i) (150)
Using equation (22),
CIRrec (i) = CarbSh (i) / eMealInsRec (i) (151)
CIRrec (i) = CarbSh (i) / [eMealIns (i) + deRxInsl / TBCorTot * TBCorIns (i + 1)-
(BRrx (i)-BR (i)) * dt (i)] (152)
The practitioner considers this schedule for CIRrec (i) and then enters:

CIRrx (i) = practitioner input (153)
6.2.1.1.1.4 6.2.1.1.1.1 and 6.2.1.1.1.3 Overview of multi-day data, type D pump For basic float and meal insulin float, the recommended values are given on the same input form (see Figure 2). Therefore, the practitioner can consider the basic float recommendations, BRrec (i), BRaveRec, and BRf when writing to the prescribed BRrx (i) schedule. Meal insulin float uses these "BRrx (i)" values as input. CTRrec (i) is then calculated as the output for the interval containing the meal. The practitioner then fills in the blank for CIRrx (i). Looking back on this explanation, it can be said that the choice of BRrx (i) by a practitioner has a great influence. A description will be given below.

  The practitioner can use all of the BRrec (i) values as BRrx (i) if desired. This has the effect of changing the entire meal insulin (or CIR) schedule across the board by the same factor.

Or
As mentioned above, a simple basal schedule can be used to give patients greater flexibility in skipping meals. For example,
The practitioner can use BR for “as is” non-meal intervals. This will probably include the “early sleep” and “late sleep” intervals.

  The practitioner can support a meal interval with a single carefully determined fasting basal rate, BRf. This basic speed determination is one of the major tasks of a practitioner. The BRf calculator may be used at the practitioner's discretion.

6.2.1.1.1.5 Meal insulin float 2 (AMCorIns is used as an error)
This algorithm uses the postprandial BG test. These tests divide the interval into two parts. The algorithm uses AMCorIns (i) as an indicator of “error” in the first part of the section where the float is performed. The value of dBasal (i) is obtained from the curve-free second part of the interval using TBCorIns (i + 1) as the error term for dBasal (i). The term “enhanced” insulin is not used.

Define as follows. That is, Timeb (i) = time for large-scale instantaneous postprandial administration after the i-th time boundary
dta (i) = Timeb (i)-Time (i) (154)
dtb (i) = Time (i + 1)-Timeb (i) (155)
In the second part of the section,
BRrec (i) = BR (i) + KrxInsl * TBCorIns (i + 1) / dtb (i) (156)
dBR (I) = BRrec (i)-BR (i) (157)
Once the basal speed is obtained from the second part of the leg (or another source), the following statement will be true for the first part of the leg.

MealInsRec (i) = MealIns (i) + KrxInsl * AMCorIns (i)-dBR (i) * dta (i) (158)
If the second part of the interval is used, this is

MealInsRec (i) = MealIns (i) + KrxInsl * AMCorIns (i)-
dta (i) * KrxInsl * TBCorIns (i + 1) / dtb (i) (159)
CIRrec (i) = CarbSh (i) / MealInsRec (d, i) (160)
If desired, the coefficient KrxInsl may be obtained from dRxInsl input by the practitioner as described above.

KrxInsl = dRxInsl / (AMCorTot + TBCorTot) (161)
Alternatively, KrxInsl may be replaced with two “K coefficients”, that of each part of the time interval, KrxInsAM and KrxInsTB.

  First, the basic speed is determined from the second part of the section.

BRrec (i) = BR (i) + KrxInsTB * TBCorIns (i + 1) / dtb (i) (162)
The meal insulin is then determined from the first part of the interval.

MealInsRec (i) = MealIns (i) + KrxInsAM * AMCorIns (i)-dta (i) * KrxInsTB * TBCorIns (i + 1) / dtb (I) (163)
CIRrec (i) = CarbSh (i) / MealInsRec (d, i) (164)
This alternative may shift the basal / TDD ratio and may be useful if such results are intended.
6.2.1.1.2 Pump type C
These types of pumps are intermediate between types A and D. The algorithms in 6.2.1.1.1.1 and 6.2.1.1.1.3 and their overview apply to Type B pumps except as noted below.

  TBCorIns (i) is not available and must be calculated by formula.

AVG (165) over the calendar period of TBCorIns (i) = ((BG (t)-TargetTB) / CF)
AMCorIns (i) is not available, so meal float 2 cannot be used.
6.2.1.1.3 Pump type B
These types of pumps are intermediate between types A and C. The algorithms in 6.2.1.1.1.1 and 6.2.1.1.1.3 and their overview apply to type B pumps with the following exceptions in addition to the limitations of type C pumps.

  Since MealIns (i) is not available, dMealIns (i) is estimated by:

dMealIns (i) = dMealIns * CarbSh (i) / CarbShTot (166)
6.2.1.1.4 Pump type A
Type A pumps are the simplest but have complex formulas. This is because the value of the missing parameter must be calculated using the estimation formula. These estimation formulas are increasing in complexity. In addition to the type B pump limitations, type A pumps are limited as follows.

  Since there is no data, the only suitable algorithm for Type A pumps is “Basic Float 1” as in version 6.2.1.1.1.1.

  Since there is no memory for BG (i), the values of TBCorIns (i) and AMCorIns (i) must be calculated from the BG data downloaded from the BG meter. This data may contain abnormal or unused BG test results and is not as reliable as BG data from pump memory.

  Since there is no memory for CarbSh (i), the practitioner must make an estimate. The only use of the data is in the ratio CarbSh (i) / CarbShTot, so units are not an issue, and the practitioner will have any proportion proportional to grams, replacements, total percentage, insulin units, or dietary insulin Other units may be used.

  The lack of good data makes the CIR calculation difficult. The CIR used for 24 hours is only calculated. Use the differential formula (34).

dCIRcalcA = [-(Kcir ＊ Wt ＊ 2 / TDDavg ² ] ＊ deMealIns (167)
Somewhat complicated is that Type A pumps do not have memory for the CIR and therefore the patient must track it. To make this easier for the patient, only integers (or half-integer values with CIR <8) are prescribed. Therefore, two-stage manual input is used.

  According to the present invention, dCIRcalcA is calculated and appears on the screen. The practitioner rounds it up to a suitable rounding amount and returns it as the inverse of dCIR. The present invention then calculates as follows.

deMealInsCalcA =-dCIR / (Kcir * Wt * 2 / TDDavg ² ) (168)
The practitioner then enters this value in the deMealIns box. The resulting value of dCIRcalcA is an integer value, which is not surprising. The recommended value, CIRrecA, is automatically calculated as follows:

CIRrecA = dCIRcalcA + CIR (169)
6.2.1.2 Subcutaneous multiple doses and inhaled insulin Patients with manual injections generally use the following two types of insulin.

Long-acting insulin: For low-frequency “basal” injections, about once a day Short-acting insulin: For meals and corrected bolus.

  The insulin delivery device includes a “pen” and an inhaler.

  Currently, “kits” consisting of BG meters that clip or link to insulin delivery devices (insulin injection “pens” or insulin inhalers) are being developed in a way that data is shared. The BG test results are used to automatically calculate the adjusted insulin dose, and the insulin delivery device is automatically set up and used. The curve can also be entered manually so that a meal insulin bolus is calculated and preset as well. Digital memory is available in either the meter or the insulin delivery device so that a combination of BG and insulin history can be downloaded by the practitioner.

  There are several types of insulin available. The input form includes an input box for two prescription insulin brand names or generic names.

The formula deals with basal quantities within each time interval, not basal speed. Long-acting long-term response of insulin makes basal rate schedule adjustment impractical.
6.2.1.2.1 Basic float 1
For pump types D and E, a combination of basal float 1 and meal insulin float 1 may be used in a similar manner. Similarities and differences are as follows. That is, the practitioner enters deRxInsl and deMealIns as in the pump. The BR (i) input box is replaced with a single box for the base, and only one BR is calculated (1/24 of the base). BRrec (i) values are shown individually as in the pump. The value is calculated for BRaveRec using equation (141) as in the pump. All BRrx (i) are replaced with a single box for BRrx and Basalrx is calculated (24 × BRrx). The practitioner enters BRrx after reviewing the recommended values for basal and basal speed.
6.2.1.2.2 Meal Insulin Float 1
Meal insulin float 1 is similar to that used for Type D pumps. Use augmented insulin naming system.

  The value of BasalRx from the previous section (or some other source) is used as input by the Meal Insulin Float 1 algorithm.

dBasal (i) = (Basalrx-Basal) * dt (i) / 24 (170)
The present invention then uses the meal insulin float to calculate a schedule of recommended CIRrec (i) values. The formula is created from equation (132) by applying equation (103) as follows:

CIRrec (i) = CarbSh (i) / [eMealIns (i) + (deRxInsl * TBCorIns (i + 1) / TBCorTot-
(BaslRx-Basal) * dt (i) / 24] (171)
6.2.1.2.3 Meal insulin float 2 (Use AMCorlhs (i) as error)
This algorithm uses the postprandial BG test. These trials and the associated bolus dose divide the interval into two parts.

Define as follows. That is, Timeb (i) = time for large-scale instantaneous postprandial administration after the i-th time boundary
dtb (i) = Time (i + 1)-Timeb (i) (172)
dta (i) = Timeb (i)-Time (i) (173)
The algorithm uses AMCorIns (i) as an indicator of “error” in the first part of the interval in which the float is calculated. BasalRx number is required. The practitioner considers the calculated recommendation (one or more of the following) of the present invention and places it in the input box.

BasalRec may be obtained from a curve-free second part of the interval using TBCorIns (i + 1) as the error term for dBasal (i). The term “enhanced” insulin is not used. In the second part of the section,
BRrec (i) = BR (i) + KrxInsl * TBCorIns (i + 1) / dtb (i) (174)
Or
This result from the “trusted” interval may be converted to a full day basis as follows:

BasalRec = BRreliable * Kf * 24 (176)
Or
BasalRec may be obtained from the sum of the above results as follows:

SUM (176) over BasalRec = i (BRrec (i) * dt (i))
Or
BasalRec may be obtained from the calculation of basic float 1 in the “trusted” interval. This result is converted in the same manner as equation (171).

Or
Another source of BasalRec is a transformation of the total recommended basis from the Base Float 1 algorithm, similar to Equation (172).

  The practitioner enters BasalRx regardless of the source of BasalRec. The following statement can be said about the first part of the section.

MealInsRec (i) = MealIns (i) + KrxInsl * AMCorIns (i)-Basal * dta (i) / 24 (177)
CIRrec (i) = CarbSh (i) / MealInsRec (i) (178)
The coefficient KrxInsl may be obtained from dRxInsl entered as above by the practitioner.

KrxInsl = dRxInsl / (AMCorTot + TBCorTot) (179)
Alternatively, when using the method of Eq. (170), KrxInsl may be replaced with two “K coefficients”, that of each part of the time interval, KrxInsAM and KrxInsTB.

    First, the basic speed is determined from the second part of the section.

BRrec (i) = BR (i) + KrxInsTB * TBCorLis (i + 1) / dtb (i) (180)
The meal insulin is then determined from the first part of the interval.

MealInsRec (i) = MealIns (i) + KrxInsAM * AMCorIns (i)
-dta (i) * KrxInsTB * TBCorIns (i + 1) / dtb (i) (181)
CIRrec (i) = CarbSh (i) / MealInsRec (d, i) (182)
This alternative may shift the basis / TDD ratio and may be useful if such results are intended.
6.2.1.3 deRxInsl limited domain and multi-day data version safety net formula
6.2.1.3.1 deRxInsl limited domain Float formulas using multi-day data for the entire leg can only be used within certain limits in the same direction (same sign), with the required changes being smaller than TBCorTot.

  Both of these states are included in the following conditions.

TBCorTot / deRxInsl> 1 (183)
If the parameters exceed this limit, meaningless results will occur. To prevent this from happening and to provide a further margin of safety, an input limit is placed on the parameter to prevent the practitioner from entering a number outside the acceptable range. This input limit is provided in the form of the parameter KrxInslMax. That is, this limit is provided in the following form.

TBCorTot / deRxInsl> = 1 / KrxInslMax (184)
If KrxInslMax is equal to 1 or less than 1, meaningless results are avoided. Currently, a value of 0.5 is used.
6.2.1.3.2 Safety net formula
6.2.1.3.2.1 About Basic Float 1 As mentioned above, float formulas using multi-day data can only be used within certain limits. The first line of defense is the above input limit. However, there may be a way to get out of the input limits, intentionally or mistakenly. To protect against this possibility, there is a second line of defense, the safety net formula. Each float formula actually uses two formulas of deRxInsl (i) (each over a different domain of deRxInsl).

  Small Domain: This is an “acceptable” domain that uses the formulas discussed so far. If TBCorTot / deRxInsl is greater than 1, use the following formula: In the formula below, accurate results are produced over the limited domain where the required change is less than or equal to the total augmentation adjusted insulin and in the same direction. In the following formula, each time interval is assigned a small amount of deRxInsl that is proportional to the share of the TBCorTot time interval. Most patient / practitioner interactions are in a small domain. The “small domain” formula has already been discussed and is listed below again. The suffix “sd” is added to the name to indicate what the name is.

deRxInslsd (i) = deRxInsl * TBCorIns (i + 1) / TBCorTot (185)
Large Domain: The formula introduced below is a backup for the input limits set in deRxInsl. If the required change to deRxInsl is not in the small domain, ie if TBCorTot / deRxInsl is less than 1, use the following formula: In the formula below, if the required change is greater than TBCorTot or in the opposite direction of TBCorTot, the accuracy will be less accurate but a safe result will be produced. The right-hand side begins with a fully adjusted insulin for the time interval. The deRxInsl “overshoot” across the TBCorTot is then allocated during the time interval, but the allocation fraction includes the base as well as the augmented adjusted insulin, giving a less sensitive fraction, Accuracy is lower. Large domain formulas are designed so that backups against input limits are their role and are therefore never used. The safety net formula for type D pump, basic float 1 algorithm 6.2.1.1.1.1 is as follows.

deRxInslld (i) = TBCorIns (i + 1) + (deRxinsl-TBCorTot) * (TBCorIns (i + 1) +
Basal (i)) / (TBCorTot + Basal) (186)
The transition between these two formulas is made by a logical variable called Swtch, which has a value of zero (0) in the small domain and (1) in the large domain.

IF (TBCorTot / deRxInsl> = 1) THEN Swtch = 0 ELSE Swtch = 1 (187)
The present invention does this as follows.

Swtch = IF (TBCorTot / deRxinsl> = 1,0,1) (188)
Swtch helps to combine the two expressions into a single expression, switching from (236) to (238) as needed. This is shown below.
deRxInsl (i) = swtch * deRxInslld (i) + (1-swtch) * deRxInslsd (i) (189)
Substituting the large and small domain formulas into this reduces to the following equation, which requires less computer fields.

deRxInsl (i) = swtch * TBCorIns (i + 1) + (deRxInsl-Swtch * TBCorTot) *
(TBCorIns (i + 1) + Swtch ＊ Basal (i)) / (TBCorTot + Swtch ＊ Basal) (190)
This time, substituting this equation into one of the float equations, eg, equation (140), yields the following result and provides a schedule of recommended base speeds.

BRrec (i) = [Basal (i) + Swtch * TBCorIns (i + 1) + (deRxInsl-
Swtch * TBCorTot) * (TBCorIns (i + 1) + Swtch * Basal (i)) / (TBCorTot +
Swtch ＊ Basal)-dMealIns ＊ CarbSh (i) / CarbShTot] / dt (i) (191)
For pump type A, replace TBCorIns (i) with TBCorInsA (i) and TBCorTot with TBCorTotA in the above equation.
6.2.1.3.2.2 About Meal Insulin Float 1
The same definition of the two domains of deRxinsl is used in the Meal Insulin Float Formula, Algorithm 6.2.1.1.1.3 to derive the following equation:

eMealInsRec (i) = eMealIns (i) + Swtch * TBCorIns (i + 1) + (deRxInsl-
Swtch * TBCorTot) * [TBCorIns (i + 1) + Swtch * eMealIns (i)] /
[TBCorTot + Swtch * eMealInsTot]-(BRrx (i)-BR (i)) * dt (i) (192)
This is converted to CIRrec (i) by applying Eq. (22).

CIRrec (i) = CarbSh (i) / eMealInsRec (i) (193)
CIRrec (i) = CarbSh (i) / [eMealIns (i) + swtch ＊ TBCorTot + (deRxInsl-
swtch * TBCorTot) * (TBCorIns (i + 1) + swtch * eMealIns (i)) /
(TBCorTot + swtch * eMealInsTot)-dt (i) * (BRrx (i)-BR (i))] (194)
Where Swtch is the same “switching parameter” introduced previously by equation (183) to shift the formula between two different domains.
6.2.1.3.2.3 About Meal Insulin Float 2 (use AMCorIns (i) as error)
Small domain:
Over the whole section:
dRxInslsd (i) = [dRxInsl / (AMCorTot + TBCorTot)] * (AMCorIns (i) + TBCorIns (i)) (195)
This can be thought of as an expression in two parts.

For the last part of the interval,
dRxInslsdb (i) = [dRxlhsl / (AMCorTot + TBCorTot)] * TBCorIns (i) = dBasalb (i) (196)
Note that this is all basic insulin.

For the first part of the leg,
dRxInslsda (i) = [dRxInsl / (AMCorTot + TBCorTot)] * AMCorIns (i) (197)
Large domain:
For the last part of the interval,
dRxInslldb (i) = TBCorIns (i) + (dRxInsl-AMCorTot-TBCorTot) *
(TBCorIns (i) + BR (i) * dtb (i)) / (AMCorTot + TBCorTot + Basal) (198)
For the first part of the leg,
dRxInsllda (i) = AMCorIns (i) + (dRxInsl-AMCorTot-TB CorTot) *
(AMCorIns (i) + BR (i) * dta (i)) / (AMCorTot + TBCorTot + Basal) (199)
The sum of the two adjusted insulin sums must be used in the switch parameter.

Swtchm = IF ((AMCorTot + TBCorTot) / dRxInsl> = 1,0,1) (200)
6.2.2 Automatic multi-day data (DIGITAL ADVISOR)
The previous discussion in version 2 describes the “manual” version of the multiple past day algorithm. The automatic functions described below eliminate interruptions for human judgment and input.
6.2.2.1 Automating deRxInsl To automate the parameter deRxInsl, the present invention controls diabetes in patients who have a high percent standard deviation (BGsd / BGmean) compared to the norm of the database in the BG test. Take advantage of the fact that it is more difficult. The present invention does this using the simple ramp function shown below.

Some new parameters:
FInsAuto is a multiplier used as follows in the deRxInsAuto formula.
deRxInsAuto = FInsAuto * KrxInslMax * TBCorTot (201)
Where KrxInslMax is a positive constant introduced previously that is equal to or less than 1. If necessary, change by programmer. The current value is 0.5.

Two definitions:
PmPctBGsd: Average value of the database “population” of quantity (BGsd / BGmean).

    PsdPctBGsd: Standard deviation of database “population” of quantity (BGsd / BGmean), ie, database standard deviation of personal standard deviation.

  In the present invention, FinsAuto is a ramp function as follows.

IF BGsd / BGmean <PmPctBGsd + PsdPctBGsd then FInsAuto = 1
IF PmPctBGsd + PsdPctBGsd <= BGsd / Bgmean <= PmPctBGsd + 2
PsdPctBGsd,
THEN FInsAuto-(1-(BGsd / Bgmean-PmPctBGsd-PsdPctBGsd) /
PsdPctBGsd]
IF BGsd / BGmean> PmPctBGsd + 2 PsdPctBGsd then FInsAuto = 0 (202)
This function describes a flat area followed by a ramp to zero, as shown in FIG.
6.2.2.2 deMealIns Automation The present invention uses the parameter dBslToTgt, which is a general advisory parameter that indicates what fundamental changes are needed to achieve BoTTgt.

dBslToTgt = TDD ＊ BoTTgt-Basal (203)
The following two steps are used to automate dMealIns.

  The first equation below assigns an absolute value equal to the smallest absolute value of deRxInsl or dBaslToTgt to the change. The reason is to avoid overshooting either BoTTgt or deRxInsl. The second equation uses the result from the first equation together with equation (57) to calculate the maximum value of deMealInsAuto and then deMealInsAuto, if deMealInsAuto is positive (in the direction of adding insulin) Kmauto is to limit it to its maximum value. The parameter Kmauto is subject to adjustment by the programmer.

dBasalAuto = IF (ABS (dBaslToTgt / deRxInsl) <1, ABS (dBaslToTgt),
ABS (deRxInsl)] * sgn (dBaslToTgt) (204)
deMealInsAuto = IF [(deRxInsl> dBasalAuto, Kmauto * (deRxInsl-
dBasalAuto), (deRxInsl-dBasalAuto)] (205)
Here, Kmauto is a positive constant equal to or less than 1 and is currently set to 1.

BoTTgt values may be obtained from ATM study values for Kb, or from BoTTgtRec from Section 6.1.2.1 Base-to-Sum ratio.
6.2.2.3 Automatic rounding of CIR for pump type A In the manual mode of type A, a two-step process is required to obtain a half-integer value of CIR within a range of less than 8. This is accomplished by an automatic function by multiplying the number of floating points dCIRcalcA by 2, then rounding to an integer and then dividing by 2. The result is entered in the opposite formula and dMealIns is modified for change.
6.3 Non-float algorithm This algorithm uses the BoTFbk (d) factor to determine both the basic velocity and the CIR. This type of algorithm is shown in the daily updated version below.

eMealIns (d, i) = eMealIns (d-1, i) + KrxInsl * (l-BoTFbk (d-1)) *
TBCorIns (d-1, i + 1) (206)
CIR (d, i) = CTR (d-1, i)-KrxInsl * (1-BoTFbk (d-1)) * eCorIns (d-1, i + 1) *
CIR (d-1, i) / eMealIns (d-1, i) (207)
BR (d, i) = BR (d-1, i) + KrxInsl * BoTFbk (d-1) * TBCorIns (d-1, i + l) (208)
This method has the advantage of being very accurate, but has the disadvantage of assigning large changes in dietary insulin to time intervals that have little curve consumption. For this reason, this version is only suitable for sections that contain major meals.
7. Time intervals Theoretically, there should be no limit to the number of time intervals in a day. However, since the screen space is limited, the number of usable time intervals is limited. For individual patients, only the maximum number of screens can be used. The present invention requires finding the last one. The last time boundary of the day (midnight) is called Tmax. This is found by routines that pick up time boundaries with a clock time of up to 24 hours. This routine is incorporated into a single formula for nested IF statements: (Note: In MS Access, null and zero behave the same with the NZ function).
Tmax =
IIf (NZ ([Time1])> NZ ([Time2]), [Time1], IIf (NZ ([Time2])> NZ ([Time3]), [Time2], IIf (NZ ([Trnie3])> NZ ([Time4]), [Time3], IIf (NZ ([Time4])> NZ ([Time5]), [Time4], IIf (NZ ([Time5])> NZ ([Time6]), [Tmie5], IIf (NZ ([Time6])> NZ ([Time7]), [Time6], [Time7]))))))) (209)
The time interval, dt (i), is found by subtracting adjacent time boundaries.

dt (i) = time (i + 1)-copy of time (i) (4) Special mention is necessary for the interval surrounding midnight.

dt0 = 24 + time1-copy of Tmax (5)
7.1 Finding the last entry in a day People rarely go to bed exactly at midnight, so it is common for some parameters generally related to bedtime to affect early morning time intervals. For example, dietary insulin performed at bedtime affects adjusted insulin at the mid-sleep time boundary. The last entry of the day is found in the formula below.
[MealInsLast] = IIF ([Time2] = [Tmax],
[MealIns2], 0) + IIF ([Time3] = [Tmax], [MealIns3], 0)
+ IIF ([Time4] = [Tmax], [MealIns4], 0) + IIF ([Time5] = [Tmax], [MealIns5], 0) +
IIF ([Time6] = [Tmax], [MealIns6], 0) + IIF ([Time7] = [Tmax], [MealIns7], 0) (210)
The recommended value for the last time interval in the day must be relocated on the screen. The following formula is an example of the recommended basic speed, BR.

BRrec (i) = IIF ([Time (i)] = [Tmax], [BRrecLast], [Regular formula of BRrec (i)]) (211)
Other alternative embodiments will become apparent to those skilled in the art without departing from the scope and spirit of the preferred embodiment.
References
1. Paul C Davidson, Harry R Hebblewhite, Bruce W Bode, Pat L Richardson, R Denn is Steed, N Spencer Welch, and Joseph Johnson; "Statistical Estimates for CS II Parameters: Carbohydrate-to-Insulin Ratio (CIR, 2.8 Rule ); Correction Factor (CF, 1700 Rule); Basal Insulin "" CSII Parameters: Carbohydrate to Insulin Ratio (CI R, 2.8 Rule); Correction Factor (CF, 1700 Rule); and Statistical Estimation of Basal Insulin "; Diabetes Technology Meeting Poster, 10/31/2002. (Scheduled for Diabetes Technology and Theapeutics April 2003)
2. John Walsh, Ruth Roberts, Varma Chanrasekhar, and Timothy Bailey; USING INSU LIN, 2003
Appendix A
The following is a SQL list for multi-day data versions of the insulin pump (type C). This version of the invention is embodied in a Microsoft Access database for use by the practitioner when the practitioner interacts with the patient.
SELECT TI.Tp_PatientID, TP.SSN, TP.Name, TP.Endo, TP.DOB, TP.Sex,
TP.OnsetDMdate, TP.PumpStartDate, TP.PhoneNos, TP.Email,
TI.IntractionID, TI.IntrDate,
TI.Hgt, TI.Wt, TI.HbAlc, TI.Creatinine, TI.Microalb,
TI.BGmean, TI.BGsd, TI.BGpd, TI.NBGs,
TI.TDDa, TI.TDDb, TI.TDDc, TI.TDDd, TI.TDDe, TI.TDDf,
(NZ ([TDDa]) + NZ ([TDDb]) + NZ ([TDDc]) + NZ ([TDDd]) + NZ ([TDDe]) + NZ ([TDDf]))) / (-IsNumeric ([TDDa ])-IsNumeric ([TDDb])-IsNumeric ([TDDc])-IsNumeric ([TDDd])-IsNumeric ([TDDe])-IsNumeric ([TDDf])) AS TDDavg,
[Basal] / [TDDavg] AS BasalOvrTDD,
0.48 * [TDDavg] AS BasalAIM,
[BasalAIM]-[Basal] AS dBslToAIM,
TI.Basalrx,
TI.TargetBG, TI.TargetBGrx,
TI.CF, 1708 / [TDDavg] AS CFaim, TI.CFrx,
TI.CIR, 2.81 * [wt] / [TDDavg] AS CIRaim,
IIf ([sex] = "m", (([Hgt] -60) * 6 + 106) * 13/8, (([Hgt] -60) * 5 + 100) * 13/8) AS
[Carbspd (Hgt)],
1- [CarbShTot] * 0.8 / [Carbspd (Hgt)] AS BoTTgtRec,
[CarbShTot] / [eMealInsTot] AS [CIR (carb)], TI.MealInsTotARx,
[CarbShTot] / [MealInsTotARx] AS [CIR (MealInslRx)],
TI.TypeInsShort, TI.TypeInsLong,
TI.Diet, TI.PumpType, Tl.iComment,
TI.TimeLabl2, TI.TimeLabl3, TI.TimeLabl4, TI.TimeLabl5, TI.TimeLabl6,
TI.TimeLabl7, TI.TimeLabl8,
0 AS Time1, TI.Time2, TI.Time3, TI.Time4, TI.Time5, TI.Time6, TI.Time7,
TI.Time8,
IIF (NZ ([Time2])> NZ ([Time3]), [Time2], IIf (NZ ([Tinie3])> NZ ([Time4]), [Time3], IIf (NZ ([Time4])> NZ ([Time5]), [Time4], IIf (NZ ([Tirne5])> NZ ([Time6]), [Time5], IIf (NZ ([Time6])> NZ ([Time7]), [Time6], IIf (NZ ([Time7])> NZ ([Time8]), [Time7], [Time8]))))) AS Tmax,
TI.CarbSh2, TI.CarbSh3, TI.CarbSh.4, TI.CarbSh5, TI.CarbSh6,
TI.CarbSh7, TI.CarbSh8,
NZ ([CarbSh2]) + NZ ([CarbSh3]) + NZ ([CarbSh4]) + NZ ([CarbSh5]) + NZ ([CarbSh6]) + NZ ([CarbSh7]) + NZ ([CarbSh8]) AS CarbShTot ,
IIf ([Time2] = [Tmax], [CarbSh2], 0) + IIf ([Time3] = [Tmax], [CarbSh3], 0) + IIf ([Time4] = [Tmax], [Carbsh4], 0) + IIf ([Time5] = [Tmax], [CarbSh5], 0) + IIF ([Time6] = [Tmax], [CarbSh6], 0) + IIf ([Time7] = [Tmax], [CarbSh7], 0 ) + NZ ([CarbSh8]) AS
CarbShLast,
TI.MealIns2, TI.MealIns3, TI.MealIns4, TI.MealIns5, TI.MealIns6,
TI.MealIns7, TI.Meallns8,
TI.AMCorIns2, TI.AMCorIns3, TI.AMCorIns4, TI.AMCorIns5,
TI.AMCorIns6, TI.AMCorIns7, TI AMCorIns8,
TI.TBCorIns2, TI.TBCorIns3, TI.TBCorIns4, TI.TBCorIns5, TI.TBCorIns6,
TI.TBCorIns7, TI.TBCorIns8,
NZ ([TBCorIns2]) + NZ ([TBCorIns3]) + NZ ([TBCorIns4]) + NZ ([TBCorIns5]) + NZ ([TBCorIns6]) + NZ ([TBCorIns7]) + NZ ([TBCorIns8]) AS TBCorTot ,
[MealIns2] + [AMCorIns2] AS eMealIns2
[MealIns3] + [AMCorIns3] AS eMealIns3
[MealIns4] + [AMCorIns4] AS eMealIns4
[MealIns5] + [AMCorIns5] AS eMealIns5
[MealIns6] + [AMCorIns6] AS eMealIns6
[MealIns7] + [AMCorIns7] AS eMeallns7
[MealIns8] + [AMCorIns8] AS eMeallns8
NZ ([eMealIns2]) + NZ ([eMealIns3]) + NZ ([eMealIns4]) + NZ ([eMealIns5]) + NZ ([eMealIns6]) + NZ ([eMealIns7]) + NZ ([eMealIns8]) AS eMealInsTot ,
IIf ([Time2] = [Tmax], NZ ([eMealIns2]), 0) + IIf ([Time3] = [Tmax], NZ ([eMealIns3]), 0) + IIf ([Time4] = [Tmax], NZ ([eMealIns4]), 0) + IIf ([Time5] = [Tmax], NZ ([eMealIns5]), 0) + IIf ([Time6] = [Tmax], NZ ([eMealIns6]), 0) + IIf ([Time7] = [Tmax], NZ ([eMealIns7]), 0) + NZ ([eMealIns8]) AS eMeallnsLast,
[Time2] + 24- [tmax] AS dt0,
IIf (NZ ([Time2])> 0, [time2]-[time1], 0) AS dt1,
IIf (NZ ([Time2])> 0, [time3]-[time2], 0) AS dt2,
IIf (NZ ([Time4])> 0, [time4]-[time3], 0) AS dt3,
IIf (NZ ([Time5])> 0, [time5]-[time4], 0) AS dt4,
IIf (NZ ([Time6])> 0, [time6]-[time5], 0) AS dt5,
IIf (NZ ([Time7])> 0, [time7]-[time6], 0) AS dt6,
IIf (NZ ([Time8])> 0, [time8]-[time7], 0) AS dt7,
TI.BR1, TI.BR2, TI.BR3, TI.BR4, TI.BR5, TI.BR6, TI.BR7, TI.BR8,
[BR1] * [dt0] AS Basl0,
[BR2] * [dt2] AS Basl2,
[BR3] * [dt3] AS Basl3,
[BR4] * [dt4] AS Basl4,
[BR5] * [dt5] AS Basl5,
[BR6] * [dt6] AS Basl6,
[BR7] * [dt7] AS Basl7,
NZ ([Basl7]) + NZ ([Basl6]) + NZ ([Basl5]) + NZ ([Basl4]) + NZ ([Basl3]) + NZ ([Basl2]) + NZ ([Basl0]) AS Basal ,
TI.BG2, TI.BG3, TI.BG4, TI.BG5, TI.BG6, TI.BG7, TI.BG8,
TI.CIR1, TI.CIR2, TI.CIR3, TI.CIR4, TI.CIR5, TI.CIR6, TI.CIR7, TI.CIR.8,
IIf ([Time2] = [Tmax], [CIR2], 0) + IIf ([Time3] = [Tmax], [CIR3], 0) + IIf ([Time4] = [Tmax], [CIR4], 0) + IIf ([Time5] = [Tmax], [CIR5], 0) + IIf ([Time6] = [Tmax], [CIR6], 0) + IIf ([Time7] = [Tmax], [CIR7], 0 ) + NZ ([CIR8]) AS CIRLast,
TI.deRxInsl,
TI.deMealIns,
[deRxInsl]-[deMealIns] AS dBasal,
[CarbShTot] / ([eMealInsTot] + [deMealIns]) AS CIRrecC,
TI.CIRrx,
IIf (0 <= [deRxInsl] / [CorTot] <= 1,0,1) AS swtch,
([Basl0] + [swtch] * [TBCorIns2] + ([deRxInsl]-
[swtch] * [TBCorTot]) * ([TBCorIns2] + [swtch] * [Basl0])
/ ([TBCorTot] + [swtch] * [BaslTot])-[deMealIns] *
[eMealInsLast] / [eMealInsTot]) / [dt0]
AS BRrecA0,
[BRrecA0] AS BRrecA1,
([Basl2] + [swtch] * [TBCorIns3] + ([deRxInsl]-
[swtch] * [TBCorTot]) * ([TBCorIns3] + [swtch] * [Basl2])
/ ([TBCorTot] + [swtch] * [BaslTot])-
[deMealIns] * NZ ([eMealIns2]) / [eMealInsTot]) / [dt2] AS BRxecA2,
([Basl3] + [swtch] * [TBCorIns4] + ([deRxInsl]-[swtch] * [TBCorTot]) *
([TBCorIns4] + [swtch] * [Basl3])
/ ([TBCorTot] + [swtch] * [BaslTot])-
[deMealIns] * NZ ([eMealIns3]) / [eMealInsTot]) / [dt3] AS BRrecA3,
([Basl4] + [swtch] * [TBCorIns5] + ([deRxInsl]-
[swtch] * [TBCorTot]) * ([TBCorIns5] + [swtch] * [Basl4])
/ ([TBCorTot] + [swtch] * [BaslTot])-
[deMealIns] * NZ ([eMealIns4]) / [eMealInsTot]) / [dt4] AS BRrecA4,
([Basl5] + [swtch] * [TBCorIns6] + ([deRxInsl]-
[swtch] * [TBCorTot]) * ([TBCorIns6] + [swtch] * [Basl5])
/ ([TBCorTot] + [swtch] * [BaslTot])-
[deMealIns] * NZ ([eMealIns5]) / [eMealInsTot]) / [dt5] AS BRrecA5,
([Basl6] + [swtch] * [TBCorins7] + ([deRxInsl]-
[swtch] * [TBCorTot]) * ([TBCorIns7] + [swtch] * [Basl6])
/ ([TBCorTot] + [swtch] * [BaslTot])-
[deMealIns] * NZ ([eMealIns6]) / [eMealInsTot]) / [dt6] AS BRrecA6,
([Basl7] + [swtch] * [TBCorIns8] + ([deRxInsl]-
[swtch] * [TBCorTot]) * ([TBCorIns8] + [swtch] * [Basl7])
/ ([TBCorTot] + [swtch] * [BaslTot])-
[deMealIns] * NZ ([eMealIns7]) / [eMealInsTot]) / [dt7] AS BRrecA7,
TI.BRrx1, TI.BRrx2, TI.BRrx3, TI.BRrx4, TI.BRrx5, TI.BRrx6, TI.BRrx7,
TI.BRrx8,
0 AS TimeRxl,
TI.TimeRx2, TI.TimeRx3, TI.TimeRx4, TI.TimeRx5, TI.TiimeRx6,
TI.TimeRx7, TI.TimeRx8,
IIf (NZ ([TimeRx2])> NZ ([TimeRx3]), [TimeRx2], IIf (NZ ([TimeRx3])> NZ ([TimeRx4]), [TimeRx3], IIf (NZ ([TimeRx4])> NZ ([TimeRx5]), [TimeRx4], IIf (NZ ([TimeRx5])> NZ ([TimeRx6]), [TimeRx5], IIf (NZ ([TimeRx6])> NZ ([TimeRx7]), [TimeRx6], IIf (NZ ([TimeRx7])> NZ ([TimeRx8]), [TimeRx7], [TimeRx8]))))) AS
TRxmax,
IIf (NZ ([TimeRx2])> 0, [TimeRx2] + 24- [Trxmax], 0) AS dTrx0,
IIf (NZ ([TimeRx2])> 0, [TimeRx2]-[TimeRx l], 0) AS dTrx1,
IIf (NZ ([TimeRx3])> 0, [TimeRx3]-[TimeRx2], 0) AS dTrx2,
IIf (NZ ([TimeRx4])> 0, [TimeRx4]-[TimeRx3], 0) AS dTrx3,
IIf (NZ ([TimeRx5])> 0, [TimeRx5]-[TimeRx4], 0) AS dTrx4,
IIf (NZ ([TimeRx6])> 0, [TimeRx6]-[TimeRx5], 0) AS dTrx5,
IIf (NZ ([TimeRx7])> 0, [TimeRx7]-[TimeRx6], 0) AS dTrx6,
IIf (NZ ([TimeRx8])> 0, [TimeRx8]-[TimeRx7], 0) AS dTrx7,
[CarbShLast] / [eMealInsLast] + [deRxInsl] * [TBCorIns2] / [TBCorTot]-([BRrxl]-
[BR1]) * [dt0]) AS CIRrec0,
[CarbSh2] / ([eMealIns2] + [deRxInsl] * [TBCorIns3] / [TBCorTot]-([BRrx2]-
[BR2]) * [dt2]) AS CIRrec2,
[CarbSh3] / ([eMealIns3] + [deRxInsl] * [TBCorIns4] / [TBCorTot]-([BRra3]-
[BR3]) * [dt3]) AS CIRrec3,
[CarbSh4] / ([eMealIns4] + [deRxInsl] * [TBCorIns5] / [TBCorTot]-([BRrx4]-
[BR4]) * [dt4]) AS CIRrec4,
[CarbSh5] / ([eMealIns5] + [deRxInsl] * [TBCorIns6] / [TBCorTot]-([BRrx5]-
[BR5]) * [dt5]) AS CIRrec5,
[CarbSh6] / ([eMealIns6] + [deRxInsl] * [TBCorIns7] / [TBCorTot]-([BRrx6]-
[BR6]) * [dt6]) AS CIRrec6,
[CarbSh7] / ([eMealIns7] + [deRxInsl] * [TBCorIns8] / [TBCorTot]-([BRrx7]-
[BR7]) * [dt7]) AS CIRrec7,
TI.eMealInsRx1, TI.eMealInsRx2, TI.eMealInsRx3, TI.eMealInsRx4,
TI.eMealInsRx5, TI.eMealInsRx6, TI.eMealInsRx7,
TI.CIRrx1, TI.CIRrx2, TI.CIRrx3, TI.ClRrx4, TI.CIRrx5, TI.CIRrx6,
TI.CIRxx7, TI.CIRrx8, TI.InptFormType
FROM TP INNER JOIN TI ON TP.PatientID = TI.Tp_PatientID
ORDER BY TI.IntrDate DESC;

Fig. 4 illustrates a preferred embodiment of a multi-day data version input form for an insulin pump (type D). The outer panel is demographic data from Table Tp. The inner panel is SubForm, for patient-practitioner interaction on a date. There are 2 pages (2 screens) and can be reached by scrolling. The first page of these is shown. Contains univalued data (unscheduled data). In other words, this is the first screen of version 6.2.1.1.3, Digital Advisor for Practitioners using multi-day data from pump type D. Fig. 4 shows a preferred embodiment of a multi-day data version input form for an insulin pump (type D). The inner panel is a SubForm. There are 2 pages and can be reached by scrolling. The second page (second screen) is shown. Includes daily schedule data. This is the second screen of version 6.2.1.1.3, the Digital Advisor for Practitioners, using multi-day data from pump type D. A preferred embodiment of the source of the multiplier FinsAuto, which is an automated multi-day data version (Automatic Digital Advisor) by comparing the standard deviation of the patient's BG to the average standard deviation from the database. Automates dRxInsAuto in If the patient number is high, reduce the multiplier (always <1). Automated change of augmented prescription insulin is another factor that multiplies FinsAuto time boundary adjustment insulin. That is, the multiplier FInsAuto automates deRxInsAuto in an automated multi-day data version by comparing the standard deviation of the patient's BG to the average standard deviation from the database. The result is used as follows. DeRxInsAuto = FInsAuto ^* KrxInslMax ^* TBCorTot

Claims (129)

  A dose of basal insulin administered over a period of time (basic insulin) and an additional dose of insulin administered in connection with a meal taken during the period of time (meal insulin). In a method of adjusting an insulin dosing schedule that includes a combined combination, estimating the required change in dietary insulin, and then changing the total insulin prescribed for the time interval (prescription insulin); Determining a change in basal insulin in relation to a dose of insulin given during the same time interval of the previous day from a difference from an estimated change in meal insulin. Method.   Divide the determined change in basal insulin by the elapsed time of the certain time interval to obtain a change in the rate of basal insulin administration (basal rate), and then change the basal rate during the same time interval of the previous day The method according to claim 1, characterized in that, in addition to the basal rate, a new dose of basal insulin to be administered over the period of time is obtained.   The estimated required change in dietary insulin is the cumulative total of dietary insulin for a given day over several time intervals, at the same ratio as the ratio of dietary insulin to the sum of all daily dietary insulin doses. The method of claim 1, wherein the method is determined as a share of change.   The estimated necessary change in dietary insulin is the same ratio as the ratio of grams of carbohydrates taken orally during the time interval to the total grams of carbohydrates taken orally over the day The method of claim 1, wherein the method is determined as a share of a cumulative change in total of meal insulin for a given day. The estimated necessary change in dietary insulin is the following formula:
(Grams of carbohydrate taken orally during the current time interval) / (ratio of carbohydrate to insulin during the current time interval) minus (grams of carbohydrate taken orally during the current time interval) / ( Carbohydrate to insulin ratio in the same time interval of the previous day)
The method of claim 1, wherein the method is determined from:   6. The method of claim 5, wherein the estimated necessary change in dietary insulin is determined from a ratio of insulin doses administered during different time intervals in grams of carbohydrate.   7. The method of claim 6, wherein the number of grams of carbohydrates orally ingested during the current time interval is determined as the meal insulin for the time interval multiplied by the carbohydrate to insulin ratio.   A dose of basal insulin administered over a period of time (basic insulin) and an additional dose of insulin administered in connection with a meal taken during the period of time (meal insulin). In a method of adjusting an insulin dosing schedule comprising a combined combination, estimating a necessary change in basal insulin, and then changing the prescribed total insulin (prescription insulin) for the fixed time interval; Determining a difference between the estimated required change in basal insulin for the fixed time interval.   9. The method of claim 8, wherein the estimated necessary change in basal insulin is determined from basal insulin in the same time interval on different days.   9. The method of claim 8, wherein the estimated necessary change in basal insulin is determined from an average of basal insulin administered during the same time interval over a plurality of different days.   A conversion rate is added to the basal insulin taken during the same time interval in one or more different days, and the conversion rate is calculated between the basal insulin of the current time interval and the time interval of the previous day or previous day group. 11. Method according to claim 9 and 10, characterized in that it is a statistically determined correlation between them.   12. The method of claim 11, wherein the statistical correlation is taken from the patient's own past data.   The method of claim 11, wherein the statistical correlation is taken from a plurality of patient samplings.   The change to the basal speed of the time interval is determined by the method of claim 16 and multiplied by the elapsed time over the time interval to obtain a change in basal insulin. The method described.   9. The method of claim 8, wherein the required change in basal insulin is determined by using the method of claim 1.   A dose of basal insulin administered over a period of time (basic insulin) and an additional dose of insulin administered in connection with a meal taken during the period of time (meal insulin). In a method of adjusting an insulin dosing schedule comprising a combined combination, a bolus of postprandial insulin is administered between a time interval beginning near the end of a meal and between the beginning and end of the constant time interval. And a change to the basal insulin of the time interval is determined based on a later portion of the time interval, the change of the basal insulin being used to determine a change of the basal insulin in an earlier portion of the time interval. Used, subtracting the change from the required change in prescribed insulin dose, Method characterized by including the time interval to obtain a change of the phosphorous.   17. The method of claim 16, wherein the change to basal insulin in the late part of the time interval is a predetermined fraction of an adjusted insulin dose at the end of the time interval.   18. The method of claim 17, comprising the step of dividing a change to basal insulin in a later portion of the time interval by the elapsed time over the time interval to obtain a basal velocity for the entire time interval. .   17. The change in prescribed insulin dose in an earlier portion of the time interval is a predetermined fraction of an intermediate adjusted insulin dose in the time interval after a meal. the method of.   20. A method according to claim 19, wherein the predetermined fraction is the same as that of claim 17.   21. The method of claim 20, wherein the predetermined fraction is equal to a total day's prescribed insulin change divided by a total day's adjusted insulin.   Further comprising the step of converting the meal insulin change to a new meal insulin dose by adding the meal insulin change to the meal insulin dose previously determined for the certain time interval of the previous day. 17. A method according to claim 1, 8, and 16.   2. The method of claim 1 wherein the new carbohydrate to insulin ratio is determined by dividing the grams of carbohydrate taken orally during the time interval by the new dietary insulin for the time interval. .   24. The method of claim 23, wherein the gram of carbohydrates orally ingested is determined by multiplying the carbohydrate to insulin ratio by the meal insulin dose for the time interval of the previous day.   The method of claim 23, wherein the fixed time interval comprises an entire day.   24. The method of claim 23, wherein the fixed time interval constitutes a time interval that does not span an entire day.   Taking a calculus derivative of the carbohydrate-to-insulin ratio for meal insulin over a period of time to change the insulin dose (meal insulin) administered in relation to a meal taken during a period of time To convert to a change in carbohydrate to insulin ratio.   28. The method of claim 27, comprising calculating the derivative as a negative number obtained by dividing the grams of carbohydrate taken orally during a period of time by the square of dietary insulin. .   29. The method of claim 28, further comprising: calculating the number of grams of carbohydrates taken orally during a given time interval as a meal insulin for said given time interval multiplied by the carbohydrate to insulin ratio. Method.   31. The method of claim 30, comprising estimating a carbohydrate to insulin ratio based on a statistical correlation between the carbohydrate to insulin ratio and the total daily dose of body weight or insulin. .   31. The method of claim 30, comprising estimating meal insulin as a preset fraction of the total daily dose of insulin.   32. The method of claim 31, wherein the preset fraction is 1/2.   2. The method of claim 1, wherein the sum of the meal insulin and the adjusted post-meal insulin dose is used in place of the meal insulin.   The method of claim 1, wherein a feedback mechanism is used to adjust the relative ratio of the daily sum of basal insulin and dietary insulin as part of prescription insulin.   35. The method of claim 34, wherein the feedback mechanism is a coefficient that represents a change in prescription insulin days divided by a change in basic insulin days. The feedback factor is as follows:
BoTFbk (d) = BoT (d) + Kfbk * (BoTTgt-BoT (d)) * sign (deRxInsl)
36. The method of claim 35, wherein the method is calculated as:   17. A method according to claim 1, 8, and 16, wherein the data for determining the value is data taken from the corresponding time interval of the previous day.   17. The method according to claim 1, 8, and 16, wherein the data for determining the values are data taken from corresponding time intervals of a plurality of days and averaged.   Calculate the adjusted insulin dose for the time interval based on the difference in averaged blood glucose measurements taken at or near a specified time over multiple days and a correction factor that divides the target blood glucose level The method according to claim 1, comprising the steps of:   40. The method of claim 39, wherein the change to prescription insulin is determined as a number minus the adjusted insulin dose for the time interval.   41. The method of claim 40, wherein the change to prescription insulin is a fraction of adjusted insulin for the constant time interval equal to adjusted insulin for the time interval multiplied by a multiplier.   42. The method of claim 41, wherein the multiplier is a number that provides a response time to reduce the amount of correction over a selected time period for correction.   43. The multiplier of claim 42, wherein the multiplier is determined as 1 minus (the remaining percentage of the original error allowed at the end of response time) to the power of (divide by 10 days of response time). Method.   41. The method of claim 40, wherein the total daily prescription insulin change value is divided by the total daily adjusted insulin to obtain the multiplication factor.   (Delete)   45. The method of claim 44, wherein the change to the applied prescription insulin is limited to a maximum change value based on a set fraction of the total adjusted insulin sum for the day.   41. The method of claim 40, wherein changes to prescription insulin are limited to a maximum change value based on a set fraction of adjusted insulin determined at the end of a certain time interval.   Changes to prescription insulin are automatically adjusted based on a comparison of the patient's percent blood glucose standard deviation to the population sample's average percent standard deviation over the recent calendar period, and the patient's standard deviation is 45. The method of claim 44, wherein the prescription insulin change is made less than the determined change if greater than the average standard deviation of the sample.   49. The method of claim 48, wherein the change in prescription insulin is equal to the determined maximum change multiplied by a fractional multiplier.   The multiplier is determined as follows, i.e., if the percent standard deviation of the patient's blood glucose measurement is less than the population mean of the population percent standard deviation plus one standard deviation of the population percent standard deviation. The percent standard deviation of the patient's blood glucose measurement is 1 to 2 percent standard deviation of the population standard deviation, greater than the mean of the population percent standard deviation, the multiplier is Set to slope down linearly until it reaches zero at the upper boundary of this interval; and if the percent standard deviation of the patient's blood glucose measurement is greater than this, the multiplier is set to zero 50. The method of claim 49, determined in such a way that no change in prescription insulin is allowed.   Change in total daily basal insulin dose based on the absolute value of the minimum of the absolute changes in prescribed basal insulin dose for the total day, and the current basis for a time interval shorter than the total day A method of automatically estimating a target basal insulin dose that is less than an insulin dose.   The target ratio of total insulin to basal insulin is the average glycemic index divided by the amount of grams of total carbohydrates taken orally over 1 minus day (patient body measurements or The method according to claim 36, characterized in that it is determined as a result of a formula based on any known statistics about the patient's daily energy condition (if other parameters are given).   37. The method of claim 36, wherein the AIM correlation may be used as a target ratio of basal insulin to basal to total insulin.   52. The estimated value of a change in basal insulin is converted to an automatic value for a change in dietary insulin by subtracting the change in basal insulin from a change to a prescription insulin for a total day. the method of.   55. The method of claim 54, wherein if the change is in a positive direction, the estimated value of meal insulin change is multiplied by a fractional reduction factor.   Omitted adjusted insulin bolus at a time boundary determined based on the sum of insulin bolus doses from the time interval on either side of the omitted blood glucose measurement corresponding to the omitted adjusted bolus insulin dose 2. The method of claim 1, further comprising administration.   17. The method of claim 1, 8, and 16, further comprising the determination of an insulin bolus based on the amount of exercise during the fixed time interval.   Further comprising the determination of an instantaneous insulin dose based on the amount of exercise over the period of time, wherein a negative insulin dose results in the basal rate infusion being interrupted for an appropriate amount of time. 17. A method according to claim 1, 8, and 16.   39. The method of claim 38, wherein if the uncorrected change to prescription insulin divided by the total adjusted insulin is less than 1, the prescription insulin day change is distributed among various time intervals using a large domain formula. The method described.   The large domain formula gives a change in prescription insulin within an interval or sub-interval as follows: at each end of the interval, a total amount of adjusted insulin calculated based on blood glucose measurements is given to each interval, The difference between the required change in daily total prescription insulin and the daily total adjusted insulin is allocated proportionally to the share of each segment of the amount of old basic insulin plus adjusted insulin during the interval. 60. The method of claim 59, wherein:   Within an interval or sub-interval, the large domain formula that determines the change in prescription insulin is the amount of adjusted insulin plus amount (change in total prescription insulin minus total adjusted insulin) to be administered at the end of the time interval multiplied by the amount 61. Method according to claim 60, characterized in that it is equal to the basal insulin of the time interval plus the adjusted insulin to be administered at the end of the time interval) divided by the total adjusted insulin plus the total daily basal insulin.   The formula depends on the use of the switching parameter as follows: a change in prescription insulin for the certain time interval is multiplied by the switching parameter multiplied by the large domain formula plus an amount (1 minus the switching parameter). 62. The method of claim 61, wherein the method is performed to be equal to.   The switching parameter value is 1 if the daily change in prescription insulin divided by the total adjusted insulin divided by 1 to be administered at the end of the time interval of the day is 1; 64. The method of claim 62, comprising a value.   60. The method of claim 59, wherein two large domain formulas are used, one for the first portion of the time interval and one for the late portion of the interval. .   The large domain formula for the first part of the time interval is as follows: change in prescription insulin in the first part of the time interval is the postprandial adjusted insulin plus amount of the interval (prescription insulin day Change minus day total postprandial adjusted insulin minus total day of adjusted insulin to be administered at the end of the time interval for the day) Multiply (postprandial adjusted insulin for the interval plus basal insulin in the interval) Divide amount (for the day 66. A total postprandial adjusted insulin plus day of adjusted insulin to be administered at the end of the time interval plus a total sum of days of baseline insulin). the method of.   The large domain formula for the last part of the interval is as follows: a change in prescription insulin for the last part of the time interval is the adjusted insulin plus amount (prescription) to be administered at the end of the time interval. Insulin day change minus total post-prandial adjusted insulin minus day total adjusted insulin to be administered at the end of the time interval for the day) multiplied by (adjusted insulin plus basal insulin in the interval) divided by (total for the day 65. The method of claim 64, calculated to be equal to post-prandial adjusted insulin plus adjusted insulin total days plus total daily basal insulin).   Switching parameters are used for both parts of the time interval, and the value of the switching parameter is divided by the ratio of (daily total postprandial adjusted insulin plus adjusted insulin to be administered at the end of the time interval) ( 65. The method of claim 64, wherein the prescription insulin day change) is 1 if greater than 1, otherwise it has a value of zero.   The method of claim 37, wherein the program is installed in an insulin pump.   38. The method of claim 37, wherein the program is installed in a kit including a blood glucose measurement device, an insulin delivery device, and a microprocessor, one of which can store data.   70. The method of claim 69, wherein the microprocessor is in the blood glucose measurement device.   70. The method of claim 69, wherein the microprocessor is in the insulin delivery device.   70. The method of claim 69, wherein the microprocessor is housed separately.   40. The method of claim 38, wherein the program is used for a pump patient.   40. The method of claim 38, wherein the program is used for multiple daily injections or inhaled insulin.   A method, wherein one of several algorithms described herein is applied to a time interval according to the characteristics of the time interval.   76. The method of claim 75, wherein each time interval is digitally flagged as a cue with a parameter of "interval type" and a certain dosing algorithm is applied.   76. The method of claim 75, wherein a time interval that includes a meal is identified.   76. The method of claim 75, wherein time intervals that include occasional small snacks are identified.   76. The method of claim 75, wherein a time interval is identified as a source of basal rate for use in determining insulin administration during another time interval.   76. The method of claim 75, wherein a time interval is identified as a source of a carbohydrate to insulin ratio for use in determining insulin administration during another time interval.   76. The method of claim 75, wherein the time interval comprising postprandial adjusted insulin is identified by the presence of postprandial adjusted insulin.   76. The method of claim 75, wherein an interval for which there is no adjusted insulin value to be administered at the end of the time interval is identified by a lack of adjusted insulin.   The version that takes data from the corresponding time interval of the previous day uses automatic daily updates of the pump and selects the appropriate algorithm for each time interval, and each time interval is only one from the previous day. 76. The method of claim 75, representing either one time interval or a corresponding time interval taken and averaged from multiple days.   The method according to claims 83 and 5, characterized in that the algorithm according to claim 4 is applied to a time interval containing small snacks.   The method according to claims 83 and 8, wherein the algorithm according to claim 8 is applied to a time interval including a meal.   The algorithm of claim 4 is applied to a time interval designated as a source of basal rate for use in determining insulin administration during another time interval. The method described in 1.   The algorithm of claim 8 is applied to a time interval designated as a source of a carbohydrate to insulin ratio for use in determining insulin administration during another time interval. 83. The method according to 83 and 8.   The method according to claim 83 and 16, wherein the algorithm according to claim 16 is applied to a time interval comprising postprandial adjusted insulin.   The version taking data from the corresponding time interval of the previous day uses multiple daily injections and automatic daily updates of inhaled insulin, and applies an appropriate algorithm to each interval. the method of.   The method according to claims 89 and 5, wherein the method according to claim 5 is applied to a section containing a small snack.   The method according to claims 89 and 8, wherein the method according to claim 8 is applied to a section including a meal.   The method of claim 5 is applied to a time interval designated as a source of basal rate for use in determining insulin administration during another time interval. The method described in 1.   9. The method of claim 8, wherein the method is applied to a time interval designated as a source of a carbohydrate to insulin ratio for use in determining insulin administration during another time interval. The method according to 89 and 8.   The method of claim 75, wherein the version whose input is in the form of an average value over a calendar period selects an appropriate algorithm for each time interval.   The method according to claims 94 and 3, characterized in that the method according to claim 3 is applied to a time interval containing small snacks.   The method according to claims 94 and 8, wherein the method according to claim 8 is applied to a time interval including a meal.   The method of claim 3 is applied to a time interval designated as a source of basal rate for use in determining insulin administration during another time interval. The method described in 1.   9. The method of claim 8, wherein the method is applied to a time interval designated as a source of a carbohydrate to insulin ratio for use in determining insulin administration during another time interval. 94. The method according to 94 and 8.   The method according to claims 94 and 16, wherein the method according to claim 16 is applied to a time interval comprising postprandial adjusted insulin.   76. The method of claim 75, wherein the version of the present invention, whose input is in the form of an average over a calendar period of multiple daily injections, selects an appropriate algorithm for each time interval.   6. The method according to claim 100 and 5, wherein the method according to claim 5 is applied to a time interval including a small snack.   9. The method according to claim 100 and 8, wherein the method according to claim 8 is applied to a time interval including a meal.   The method of claim 5 is applied to a time interval designated as a source of basal rate for use in determining insulin administration during another time interval. The method described in 1.   9. The method of claim 8, wherein the method is applied to a time interval designated as a source of a carbohydrate to insulin ratio for use in determining insulin administration during another time interval. 100. The method according to 100 and 8.   4. The method according to claim 3, wherein the multi-day data version of the present invention for a pump applies the method according to claim 3 to all time intervals.   Basal insulin within a certain time interval is calculated as the adjusted insulin to be administered at the end of the certain time interval and the meal insulin for that time interval is multiplied by the amount (1 minus the feedback factor) A method characterized in that it is calculated as a section adjusted insulin.   The method of claim 36, wherein the constant is selected for the purpose of optimal convergence rate to a target basal insulin to total ratio.   The method further comprises the step of determining a new carbohydrate to insulin ratio by dividing the grams of carbohydrate taken orally during the time interval by the new meal insulin for the time interval. 9. The method according to 8.   109. The method of claim 108, wherein the gram of sugar taken orally is determined by multiplying a carbohydrate to insulin ratio by a meal insulin dose for the time interval of the previous day. .   110. The method of claim 109, wherein the fixed time interval comprises an entire day.   111. The method of claim 110, wherein the constant time interval constitutes a time interval that does not span an entire day.   9. The method of claim 8, wherein the sum of the meal insulin and the adjusted post-meal insulin dose is used in place of the meal insulin.   113. The method of claim 112, wherein a feedback mechanism is used to adjust the relative ratio of the daily sum of basal insulin and dietary insulin as part of prescription insulin.   114. The method of claim 113, wherein the feedback mechanism is a coefficient that represents a change in prescription insulin days divided by a change in basic insulin days. The feedback factor is as follows:
BoTFbk (d) = BoT (d) + Kfbk * (BoTTgt-BoT (d)) * sign (deRxInsl)
115. The method of claim 114, calculated as:   Calculate the adjusted insulin dose for the time interval based on the difference in averaged blood glucose measurements taken at or near a specified time over multiple days and a correction factor that divides the target blood glucose level 9. The method of claim 8, comprising the steps of:   117. The method of claim 116, wherein the change to prescription insulin is determined as a number minus the adjusted insulin dose for the time interval.   118. The method of claim 117, wherein the change to prescription insulin is a fraction of adjusted insulin for the constant time interval equal to adjusted insulin for the time interval multiplied by a multiplier.   119. The method of claim 118, wherein the multiplier is a number that provides a response time to reduce the amount of correction over a selected time period for correction.   120. The multiplier of claim 119, wherein the multiplier is determined as 1 minus (the remaining percentage of the original error allowed at the end of response time) to the power of (divide by 10 days of response time). Method.   118. The method of claim 117, wherein the total day prescription insulin change value is divided by the total day adjusted insulin to obtain the multiplication factor.   (Delete)   122. The method of claim 121, wherein changes to the added prescription insulin are limited to a maximum change value based on a set fraction of the total adjusted insulin sum for the day.   118. The method of claim 117, wherein changes to prescription insulin are limited to a maximum change value based on a set fraction of adjusted insulin determined at the end of a certain time interval.   Changes to prescription insulin are automatically adjusted based on a comparison of the patient's percent blood glucose standard deviation to the population sample's average percent standard deviation over the recent calendar period, and the patient's standard deviation is 122. The method of claim 121, wherein the prescription insulin change is made less than the determined change if greater than the average standard deviation of the sample.   126. The method of claim 125, wherein the change in prescription insulin is equal to the determined maximum change multiplied by a fractional multiplier.   The multiplier is as follows: if the patient's blood glucose measurement percent standard deviation is less than the population average of the population percent standard deviation plus one standard deviation of the population percent standard deviation, the multiplier The percent standard deviation of the patient's blood glucose measurement is 1 to 2 percent standard deviation of the population standard deviation, greater than the mean of the population percent standard deviation, the multiplier is Set to slope down linearly until it reaches zero at the upper boundary of this interval; and if the percent standard deviation of the patient's blood glucose measurement is greater than this, the multiplier is set to zero 127. The method of claim 126, determined in such a way that no change in prescription insulin is allowed.   Calculate the adjusted insulin dose for the time interval based on the difference in averaged blood glucose measurements taken at or near a specified time over multiple days and a correction factor that divides the target blood glucose level The method according to claim 16, further comprising a step. The total daily basis changes are as follows: (the total daily prescription insulin change) and (the target basic insulin minus the current basic insulin) absolute value of the minimum value 9. The method of claim 8, wherein the method automatically estimates and multiplies the resulting minimum by the latter amount of sign.