A system and method for dynamic titration of an antidiabetic

Abstract

A method and system for establishing and executing a dynamic titration protocol for administration of an antidiabetic is disclosed. The method may include establishing an initial titration protocol and then applying an aggressiveness level to the initial titration protocol to obtain an initial use titration protocol. The initial use titration protocol may then be executed with a patient and glucose monitoring data may be received from the patient. The monitoring data may be used to perform a failsafe monitoring procedure and a revised aggressiveness level and titration protocol may be determined based on the results of the failsafe monitoring.

Description

Attorney Docket Nos.: 9134-0785 P37880-WO-1 A SYSTEM AND METHOD FOR DYNAMIC TITRATION OF AN ANTIDIABETIC RELATED APPLICATIONS

[0001] This application claims priority to PCT / US2024 / 19107, filed March 8, 2024, and isrelated to co-pending application nos. PCT / US2023 / 70698, filed July 21, 2023, PCT / US2024 / 038824, filed July 19, 2024, the entire disclosures of which are hereby incorporated herein by reference. TECHNICAL FIELD

[0002] The teachings of this disclosure generally relate to a system and a method fordynamic titration of an antidiabetic. In particular, this disclosure relates to a system and a method for titrating an antidiabetic. BACKGROUND

[0003] In general, titration is the process of adjusting the dose of a medicine to achieveoptimal therapeutic benefit with minimal adverse effects. The objective of a titration protocol is to determine the dosing strategy that produces the desired therapeutic effect for the particular patient as quickly as possible without causing adverse events. Titration is particularly important for medications having a narrow therapeutic index because the difference between a therapeutic dose and a dose that may cause significant side effects is comparatively small. Antidiabetics, including insulin and biosimilar insulins, are among the medications that commonly require titration to achieve proper glycemic control without causing hypoglycemic events.

[0004] Existing antidiabetic titration systems have largely pre-set titration schedules that maybe based on drug manufacturer guidelines and / or basic characteristics of therapy (i.e., varying levels of settings corresponding to titration aggressiveness), but these systems do not take into account the characteristics of the patient. Alternatively, titration systems can be customized manually by the healthcare provider before the protocol is initiated, such that the starting dose, total dose, dose step adjustments, time to adjustments, and target ranges can be customized before titration begins. Whatever the selected titration protocol, if it fails, the healthcare provider adjusts the parameter settings and starts over with another titration protocol.

[0005] A typical procedure for titrating an antidiabetic (10) is shown in FIG. 1. The stepsinclude: (a) Diagnosing a need for antidiabetic administration (12).Attorney Docket Nos.: 9134-0785 P37880-WO-1 (b) Prescribing a particular antidiabetic (14).(c) Obtaining generic titration instructions for the prescribed antidiabetic (16).(d) Establishing an administration regimen based on the generic titration instructions (18).(e) Teaching the patient the administration regimen (20).(f) Following the administration regimen (22).(g) Evaluating patient results (24).(h) If the titration is successful, the administration regimen is continued (28).(i) If the titration is unsuccessful, the healthcare provider manually adjusts theadministration regimen (30) and begins a new titration protocol.

[0006] In practice, such generic titration procedures are suboptimal. This is because patientsstarting on any particular antidiabetic, such as insulin, may respond to the treatment differently based on a number of patient-specific factors, which the generic titration procedures fail to take into account. The generic titration protocol is medication-specific, not patient-specific. The healthcare provider rarely has the tools to determine a patient’s likely response and to adjust either the selected antidiabetic or the generic titration parameters to something more appropriate for the particular patient. While the healthcare provider may be able to adjust certain parameters once the titration protocol has been in place for some time and failed to achieve the desired glycemic control, this process is slow and adds a cognitive burden for the provider.

[0007] One key issue that providers face is balancing the time required to titrate a medicationagainst the inherent risk of creating an adverse event by increasing the dose too much. In the case of an antidiabetic, the challenge is to increase the dose of the antidiabetic at periodic intervals until the blood glucose level comes down to the target without inducing hypoglycemia by, for example, increasing the dose too quickly or too high. The tendency for most titration protocols is to err on the side of caution, which can result in an undesirably long amount of time required to titrate.

[0008] One problem with existing antidiabetic titration systems is that they have statictitration schedules based on drug manufacturer guidelines and / or basic characteristics of therapy (i.e. varying levels of settings corresponding to titration aggressiveness). These static titration schedules do not allow for adjustment of the titration protocol while titration is in progress. Consequently, titration protocols tend to err on the side of caution, setting the titration protocol parameters in a way that will most likely avoid any adverse events. For example, in the case ofAttorney Docket Nos.: 9134-0785 P37880-WO-1 titration protocols for antidiabetics, the adverse event to be avoided is hypoglycemia. To avoid causing a hypoglycemic event, the dose increases should be smaller and the adjustments should be well spaced apart. But this creates unnecessarily long delays in the patient reaching the target blood glucose level.

[0009] Further, if the titration protocol fails, the healthcare provider must adjust theparameter settings and begin a new titration protocol. Since, in many cases, the state of the art solutions do not produce a protocol tailored to the patient and likely to succeed on the first try, the process become iterative, reducing the likelihood of successful and timely health outcomes. This delays needed therapy optimization for the patient and increases the risk of chronic comorbid disease progression. Thus, there is a need for a titration system that increases the probability of a successful and timely titration. SUMMARY

[0010] This disclosure teaches a dynamic titration system that increases the probability of asuccessful and timely titration by allowing for adjustments to the titration protocol parameters while the titration is in progress.

[0011] A system and method for dynamically adjusting the parameters of an antidiabetictitration protocol are described. Initialization of the protocol may begin with default titration protocol parameters (i.e., antidiabetic type, initial dose, step adjustment, and cycle time). Alternatively, diagnoses and related learnings from previous protocol failures and / or a digital twin / similarity analysis may be performed for the patient to select starting titration protocol parameters. Furthermore, a level of aggressiveness may be applied to titration protocol parameters as a secondary adjustment to increase the chance of successfully and safely reaching the target glucose threshold within an acceptable time frame. Once the titration protocol begins and after at least one titration cycle has occurred, a time to target glucose prediction and insulin sensitivity may be calculated to determine if the aggressiveness level remains appropriate. At this point in the protocol, aggressiveness may be increased or decreased. For each subsequent titration cycle, the time to target glucose prediction, insulin sensitivity calculation and concomitant adjustment to the aggressiveness of the titration protocol may be repeated with continuous failsafe monitoring until the titration protocol ends due to the target glucose thresholdAttorney Docket Nos.: 9134-0785 P37880-WO-1 being reached, a failsafe being triggered, dangerously low insulin sensitivity being detected, or low adherence to the titration protocol by the patient being detected.

[0012] As used in the following, the terms “have,” “comprise” or “include” or any arbitrarygrammatical variations thereof are generally open-ended terms. Thus, these terms may both refer to a situation in which, besides the feature introduced by these terms, no further features happen to be present in the entity described in this context and to a situation in which one or more further features are present. As an example, the expressions “A has B,” “A comprises B” and “A includes B” may both refer to a situation in which, besides B, no other element is present in A (i.e., a situation in which A solely and exclusively consists of B) and to a situation in which, besides B, one or more further elements are present in entity A, such as element C, elements C and D or even further elements.

[0013] Further, it shall be noted that the terms “at least one”, “one or more” or similarexpressions indicating that a feature or element may be present once or more than once typically will be used only once, if at all, when introducing the respective feature or element. In the following, in most cases, when referring to the respective feature or element, the expressions “at least one” or “one or more” will not be repeated, notwithstanding the fact that the respective feature or element may be present once or more than once. It shall also be understood for purposes of this disclosure and appended claims that, regardless of whether the phrases “one or more” or “at least one” precede an element or feature appearing in this disclosure or claims, such element or feature shall not receive a singular interpretation unless it is made explicit herein. By way of non-limiting example, the terms “antidiabetic”, “personal parameter”, and “titration protocol parameter”, to name just a few, should be interpreted wherever they appear in this disclosure and claims to mean “at least one” or “one or more” regardless of whether they are introduced with the expressions “at least one” or “one or more.” All other terms used herein should be similarly interpreted unless it is made explicit that a singular interpretation is intended.

[0014] Further, as used in the following, the terms “preferably,” “more preferably,”“particularly,” “more particularly,” “specifically,” “more specifically” or similar terms are used in conjunction with optional features, without restricting alternative possibilities. Thus, features introduced by these terms are optional features and are not intended to restrict the scope of the claims in any way. The invention may, as the skilled person will recognize, be performed by using alternative features. Similarly, features introduced by “in an embodiment of the invention”Attorney Docket Nos.: 9134-0785 P37880-WO-1 or similar expressions are intended to be optional features, without any restriction regarding alternative embodiments of the invention, without any restrictions regarding the scope of the invention and without any restriction regarding the possibility of combining the features introduced in such way with other optional or non-optional features of the invention.

[0015] The terms “patient” and “subject” may be used interchangeably herein. Both refer toa person with diabetes or a person with pre-diabetes.

[0016] TERMS

[0017] The term “titration” as used herein is a broad term and is to be given its ordinary andcustomary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “titration” specifically may refer, without limitation, to a procedure, system or method used to adjust the dose and / or timing of a particular medication to achieve therapeutic effect in a patient while minimizing the adverse effects of the medication on the patient.

[0018] The term “antidiabetic” as used herein is a broad term and is to be given its ordinaryand customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “antidiabetic” specifically may refer, without limitation, to insulin as defined below, as well as, for example, amylinomimetic injectables, alpha-glucosidase inhibitors, biguanides, dopamine-2 agonists, dipeptidyl peptidase-4 (DPP-4) inhibitors, glucagon-like peptide-1 receptor agonists (GLP-1 receptor agonists), meglitinides, sodium- glucose transporter (SGLT) 2 inhibitors, sulfonylureas, thiazolidinediones, and other medications with similar therapeutic effects.

[0019] The term “insulin” as used herein is a broad term and is to be given its ordinary andcustomary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “insulin” specifically may refer, without limitation, to naturally occurring human or animal insulin, partially or wholly biosynthetic insulins such as biosimilar insulins, , fast-acting insulin, or the like. Insulin may be delivered orally, by inhalation or by injection.

[0020] The term “personal parameter” as used herein is a broad term and is to be given itsordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “personal parameter” specifically may refer, withoutAttorney Docket Nos.: 9134-0785 P37880-WO-1 limitation, to parameters that describe the health, demographic or other attributes of a particular person, subject or patient.

[0021] The term “health parameter” as used herein is a broad term and is to be given itsordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “health parameter” specifically may refer, without limitation, to HbA1c, comorbid conditions, age, height, weight, body mass index, other medications, hypoglycemia risk level, blood glucose values, vital signs, etc.

[0022] The term “demographic parameter” as used herein is a broad term and is to be givenits ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “demographic parameter” specifically may refer, without limitation, to ethnicity, age, gender, socioeconomic status, preferred modes of communication, etc.

[0023] The term “digital twin” as used herein is a broad term and is to be given its ordinaryand customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “digital twin” specifically may refer, without limitation, to a digital representation of a patient that serves as a digital counterpart to the patient for purposes of statistical analysis or the like. A digital twin need not be precisely identical for the purposes of this disclosure. A digital twin may be individually created using continuous glucose monitor data, activity data from, e.g., smart devices, electronic medical records or the like. Alternatively, the twin may be selected from an existing set of common profiles. The common profile that most closely parallels the patient’s data can be selected and used as the digital twin.

[0024] The term “cohort” as used herein is a broad term and is to be given its ordinary andcustomary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “cohort” specifically may refer, without limitation, to a group of people who share one or more common characteristics of interest (also referred to herein as “commonalities” or “similarities”). More specifically, “cohort” may refer to a group of patients sharing one or more personal parameters, such as health and / or demographic parameters.

[0025] The term “patient similar cohort” as used herein is a broad term and is to be given itsordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “patient similar cohort” specifically may refer,Attorney Docket Nos.: 9134-0785 P37880-WO-1 without limitation, to the cohort sharing common characteristics of interest with the patient for whom the custom titration protocol is being developed.

[0026] The term “titration protocol” as used herein is a broad term and is to be given itsordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “titration protocol” specifically may refer, without limitation, to the procedure used to titrate a medication, including the various input parameters such as selection of an antidiabetic, starting dose, total dose, dose step adjustments, time to adjustments, target ranges, etc.

[0027] The term “titration protocol parameter” as used herein is a broad term and is to begiven its ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “titration protocol parameter” specifically may refer, without limitation, to various input parameters such as selection of the antidiabetic, starting dose, total dose, dose step adjustments, time to adjustments, target ranges, etc.

[0028] The term “titration output parameter” as used herein is a broad term and is to be givenits ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “titration output parameter” specifically may refer, without limitation, to parameters resulting from performing the titration protocol such as, e.g., time to achieve glycemic control, number of hypoglycemic events, number of hyperglycemic events, HbA1c, fasting blood glucose level, the percentage of time within a target blood glucose range during the titration period, etc.

[0029] The term “successful titration” as used herein is a broad term and is to be given itsordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “successful titration” specifically may refer, without limitation, to a titration meeting the titration output parameter goals or threshold levels.

[0030] The term “unsuccessful titration” as used herein is a broad term and is to be given itsordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “unsuccessful titration” specifically may refer, without limitation, to a titration that does not meet the titration output parameter goals or threshold levels.

[0031] The term “information delivery medium” as used herein is a broad term and is to begiven its ordinary and customary meaning to a person of ordinary skill in the art and is not to beAttorney Docket Nos.: 9134-0785 P37880-WO-1 limited to a special or customized meaning. The term “information delivery medium” specifically may refer, without limitation, to various devices or modes that may be used alone or in combination to convey custom titration protocol instructions to the patient. For example, “information delivery medium” may refer to SMS, lightweight titration service app, integration in other app (e.g., mySugr), voice-skill Alexa, Siri, Cortana, Google Home (smartphone vs. home device), bot or assistant calling the patient, personal computers, enterprise computers, dumb terminals, television screens, bot assistants, network communication devices, tablets, smart phones, smart watches and the like.

[0032] The term “aggressiveness scheme” as used herein is a broad term and is to be givenits ordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “aggressiveness scheme” specifically may refer, without limitation, to a set of aggressiveness levels that correspond to different adjustments that may be made to a dynamic titration protocol while it is in progress. An aggressiveness scheme may include levels that correspond to a single adjustment to the dynamic titration protocol and the same aggressiveness scheme may also include levels that correspond to adjustments in multiple titration parameters. By way of a non-limiting example, an “aggressiveness scheme” may have three levels and each level may correspond to either a dose adjustment or a titration cycle time adjustment or a frequency of administration adjustment or some combination thereof.

[0033] The term “aggressiveness level”, as used herein is a broad term and is to be given itsordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “aggressiveness level” specifically may refer, without limitation, to one of several levels within an aggressiveness scheme that correspond to a set of dynamic adjustments that may be made to a titration protocol while it is in progress.

[0034] The term “use titration protocol” as used herein is a broad term and is to be given itsordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “use titration protocol” specifically may refer, without limitation, to a “first use titration protocol” or a “revised use titration protocol”. A “use titration protocol” may refer to a titration protocol that has been modified using the aggressiveness scheme. It may also refer to a titration protocol that remains the same when applying the aggressiveness scheme to a titration protocol does not result in changes to the dose, dose step adjustment, or frequency of adjustments.Attorney Docket Nos.: 9134-0785 P37880-WO-1

[0035] The term “target glucose level” as used herein is a broad term and is to be given itsordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “target glucose level” specifically may refer, without limitation, to a specific numerical target for a user’s blood glucose level, a targeted range of blood glucose levels for the user, a targeted percentage of time within a particular target range of blood glucose levels, or the like.

[0036] The term “insulin sensitivity” as used herein is a broad term and is to be given itsordinary and customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “insulin sensitivity” specifically may refer, without limitation, to the body’s ability to remove glucose from the blood in response to insulin.

[0037] The term “risk scheme” as used herein is a broad term and is to be given its ordinaryand customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “risk scheme” specifically may refer, without limitation, to a set of risk levels that correspond to different health parameters. A risk scheme may include levels that correspond to a one or more health parameter values such as, e.g., insulin sensitivity, blood glucose level, stability of insulin sensitivity, stability of blood glucose level, rate of change of one or more health parameters, and the like. By way of a non-limiting example, an “risk scheme” may have three levels and each level may correspond to a level of risk for adverse events occurring in connection with the dynamic titration protocol.

[0038] The term “risk level”, as used herein is a broad term and is to be given its ordinaryand customary meaning to a person of ordinary skill in the art and is not to be limited to a special or customized meaning. The term “risk level” specifically may refer, without limitation, to one of several levels within a risk scheme that correspond to a set of health parameter values.

[0039] METHOD

[0040] The method steps disclosed herein can be carried out in the illustrated sequence.However, alternative sequences are also possible. Further, individual or multiple method steps can be carried out in parallel, simultaneously, or repeatedly, either on their own or in groups. For example, the steps for determining titration protocol parameters need not be carried out in the precise order described below. Furthermore, the method can comprise additional method steps that are not illustrated. Independently of the fact that the term method step is used, the termAttorney Docket Nos.: 9134-0785 P37880-WO-1 “step” says nothing about the duration of the method steps. Thus, the specified method steps can, individually or in groups, be carried out briefly, but can also be carried out over a longer time period, for example, over time intervals of a number of minutes, hours, days, weeks or even months, for example, continuously or repeatedly.

[0041] The present disclosure relates to a method of establishing and executing a dynamictitration protocol for administration of an antidiabetic including the steps of: (a) establishing an initial titration protocol including, but not limited to selecting anantidiabetic, an initial dose, an initial titration cycle time and initial titration step adjustments (b) applying an aggressiveness level to the titration protocol to obtain an initial usetitration protocol (c) executing the initial use titration protocol with a patient(d) performing a failsafe monitoring procedure, which may comprise:^ predicting the probability of a hypoglycemic event^ calculating insulin sensitivity^ predicting the length of time for the patient to reach the target glucose level basedon patient glucose data (e) revising the aggressiveness level based on the results of steps (d)(f) applying the revised aggressiveness level to the initial use titration protocol to obtaina revised use titration protocol (g) executing the revised use titration protocol with a patient and(h) Repeating steps (d)-(h) until the target glucose level is reached

[0042] Establishing an initial titration protocol:

[0043] The initial titration protocol may be established in a variety of ways. For example,since the disclosed method will dynamically adjust the titration protocol while in use, the initial titration protocol may simply be the default protocol according to the antidiabetic manufacturer or industry guidelines for certain demographics.

[0044] Alternatively, if a previous titration protocol failed, it may be advantageous todiagnose the reason for failure and incorporate it into the initial titration protocol. Failure analysis results may be used if, for example, the previous titration timed out or a maximum basalAttorney Docket Nos.: 9134-0785 P37880-WO-1 dose was reached, but the titration protocol was not stopped due to hypoglycemia. One non- limiting example of incorporating failure analysis results into the initial titration protocol may be using an ending titration dose from a previous titration protocol failure as the starting dose for a new initial titration protocol. This may reduce the time required to reach the target glucose level. Further, if the previous titration failed due to small dosage increments causing it to take too long to reach the target glucose level, the initial titration protocol may include higher dosage increments.

[0045] Another alternative for establishing the initial titration protocol is to perform digitaltwin or similarity analysis to determine the personal parameters impacting the time to reach the target glucose level (e.g., age, BMI, baseline HbA1c, etc.). Such an analysis may begin by identifying a cohort of previously managed subjects who share certain characteristics with the current patient and then determining initial titration protocol parameters that are most likely to result in reaching the target glucose level within a target length of time.

[0046] The titration system may include anonymized personal parameters, such as, forexample, health and demographic parameters, for previously managed subjects. The previously managed subjects may be grouped into cohorts based on a similarity analysis, which may be performed using any one or a combination of techniques to determine statistical or learned similarity of data sets. Examples of techniques suitable for use in the disclosed method include, but are not limited to:

[0047] Cosine Similarity: Similarity between a patient and a database of previously managedsubjects can be determined by modeling previously managed subjects as vectors of defined health and demographic parametric data points and comparing vectors using similarity measures. Vectors are feature embeddings composed of binary or numeric features representing health and demographic parametric data points such as but not limited to existing conditions, admitted medications, vital signs, lab observations and temporal relationships of those data points and clinical events.

[0048] Knowledge Graph Databases / Algorithms: Similarity between a patient and a databaseof previously managed subjects can be determined using knowledge graph databases. Health and demographic parameters can be vertices in the knowledge graph. For example, age, HbA1c, comorbidities, medications and body mass index may form vertices in a knowledge graph. Edges can be defined accordingly, e.g., a patient having diabetes and taking Metformin wouldAttorney Docket Nos.: 9134-0785 P37880-WO-1 have edges to the diabetes and Metformin vertex. Similarity between patients may be determined using, for example, metric similarity or vertex similarity. Metric similarity may be determined based on a normalized distance metric such as Euclidean distance, Manhattan distance, Levenshtein distance, Mahalanobis distance, Minkowski distance, Hamming distance, etc. Vertex similarity may be determined using, for example, neighborhood count, neighborhood selectivity, neighborhood rarity, SimRank, etc.

[0049] Artificial Intelligence / Machine Learning Modeling: Alternatively, similarity betweenone or more patients and a database of previously managed subjects can be determined using a machine-learning based AI model. Non-limiting examples of such learning algorithms include: K-nearest neighbor, support vector machines, naive bayes, decision trees such as random forest, logistic regression such as multinominal logistic regression, neuronal network, decision trees and Bayes network. Exemplary methods are described, below (Sammut et al., Encyclopedia of Machine Learning, 1st. Springer Publishing Company, Incorporated, 2011). Table 1 summarizes advantages and disadvantages of the methods.

[0050] Clustering: Clustering uses one or more analytical techniques for grouping a set ofobjects in such a way that objects in the same group are more similar to each other than to those in other groups.

[0051] K-nearest Neighbor: This is one example of a clustering method. Other deepclustering methods may also be used with this disclosure. The goal of this method is to place an object into a class with similar objects. The class for a particular object is determined based on which class appears most frequently for objects with similar parametric values. In order to determine the proximity of the objects, a similarity measure, such as, for example, the Euclidian distance is used. This method is very well suited for significantly larger data quantities.

[0052] Support Vector Machines: In this method, a hyper plane is calculated, whichclassifies objects into classes. For calculating the hyper plane, the distance around the class boundaries is to be maximized, which is why the Support Vector Machine is one of the ‘Large Margin Classifiers’. An important assumption of this method is the linear separability of the data, which, however, can be expanded to higher dimensional vector spaces by means of the Kernel trick. Large data quantities are required for a classification with less overfitting.

[0053] Naive Bayes: The naive assumption is that the present variables are statisticallyindependent from one another. This assumption is not true for most cases. In many cases,Attorney Docket Nos.: 9134-0785 P37880-WO-1 Naive Bayes nonetheless reaches a high rate of correct classification even if the attributes correlate slightly. Naive Bayes analysis is relatively simple to perform.

[0054] Regression: In a regression analysis, the relationships between a dependent variableand one or more independent variables are determined using a statistical process.

[0055] Logistic Regression: This is one example of a regression method. Other regressionmethods may also be used with this disclosure. In a logistic regression, the likelihood that values of a dependent variable can be attributed to values of independent variables is calculated.

[0056] Deep Learning: Deep learning is part of the broader family of machine learningmethods, which is based on artificial neural networks with representation learning. The term “deep” refers to the use of multiple layers in the network. Methods of deep learning can be either supervised, semi-supervised or unsupervised. Deep learning architectures may include, but are not limited to deep neural networks, deep belief networks, deep reinforcement learning, recurrent neural networks, convolutional neural networks and transformers, etc.

[0057] Neuronal Networks: This is one example of a deep learning method. Other deeplearning methods may also be used with this disclosure. Artificial neuronal networks are based on the biological structure of neurons in the brain. A simple neuronal network consists of neurons arranged in three layers. These layers are the input layer, the hidden layer and the output layer. Between the layers, all neurons are connected to one another via weights, which are optimized during a training phase.

[0058] Decision Trees: Decision trees are sorted, layered trees, which are characterized bytheir simple and easily comprehensible appearance. Nodes which are located close to the root are more significant for the classification of an object than nodes located close to the leaf. Decision trees often experience problems caused by overfitting. Consequently, the random forest methodology can be useful. A random forest consists of a plurality of decision trees, whereby each tree represents a subset of variables.

[0059] Bayes Networks: A Bayes network is a directed graph, which illustrates multi-variable likelihood distributions. The nodes of the network correspond to random variables and the edges show the relationships between them. For developing a Bayes network, it is helpful to describe the dependencies between the variables in as much detail as possible.Attorney Docket Nos.: 9134-0785 P37880-WO-1

[0060] Table 1: Summary of advantages and disadvantages of AI / machine learning methodsMethod Advantage Disadvantage ow s ts, orAttorney Docket Nos.: 9134-0785 P37880-WO-1 Method Advantage Disadvantage[0] e persona parame ers, e.g., ea an emograp c parame ers or e pa en maybe received into the dynamic titration system. For example, the parameters may be obtained by the dynamic titration system from an electronic medical records system or a healthcare provider may enter the patient’s personal parameters using a healthcare provider interface. Non-limiting examples of healthcare provider interfaces include personal computers, enterprise computers, dumb terminals, network communication devices, tablets, smart phones, smart watches and the like. Non-limiting examples of health parameters include HbA1c, comorbid conditions, age, height, weight, body mass index, other medications, etc. Non-limiting examples of demographic parameters include ethnicity, age, gender, socioeconomic status, preferred modes of communication, etc.

[0062] The patient’s personal parameters may be used to create a digital twin for the patient.The patient may then be placed into a cohort of previously managed subjects by analyzing the digital twin and the previously managed subject cohorts using similarity analysis techniques such as those discussed above. Alternatively, the patient may be placed into a cohort by using similarity analysis techniques to identify a digital twin from a set of pre-existing digital profiles. The patient may then be assigned to a cohort according to the selected digital twin.

[0063] Data from previous patients with titrations meeting the threshold time to reach thetarget glucose level can be isolated and correlative titration protocol parameters may be determined. At the same time, data from previous patients with titrations that did not meet theAttorney Docket Nos.: 9134-0785 P37880-WO-1 threshold time to reach the target glucose level can also be isolated and the correlative protocol parameters leading to failure may also be determined.

[0064] For example, each titration protocol parameter may be assessed in terms of itsstatistical correlation to the time required to reach the target glucose level. This analysis may be done for the entire patient population pool, the similar patient cohort pool, or some combination thereof. Within acceptable confidence and significance levels, titration parameters at or above a minimum statistical correlation threshold (e.g., 0.7, 0.8, 0.9 or 0.95) could be considered for the initial titration protocol.

[0065] Alternatively, a regression and clustering analysis to compare the parameters of asuccessful titration could be used to identify which characteristics were deterministic of success. At the same time, a regression and clustering analysis to compare the parameters of an unsuccessful titration could be used to identify which characteristics were deterministic of failure. A logistic regression approach may include, for example: 1. Collecting patient personal parameters, including the time to reach the target glucose level, which may be reduced to, e.g., a binary result based on whether the time required meets or does not meet a target threshold. 2. Using the time to reach the target glucose level as the response / target variable. 3. Initially assuming all titration protocol parameters are predictors for the regression. 4. Identifying titration protocol parameters with a significant p-value, meaning that such parameters have a significant effect on the outcome and are to some extent deterministic of success. 5. Carrying titration protocol parameters identified in step (4) as meeting the significance criteria to the next step for analysis with the patient similar cohort.

[0066] To determine which parameters to adjust in the initial titration protocol, thesignificant / correlative parameters can be assessed against their respective protocol outcomes within the patient cohort. This may include: 1. Identifying a patient cohort (discussed above). 2. Identifying correlative / significant titration protocol parameters. 3. Interpolating correlative / significant titration protocol parameters to determine which have the highest number of the patients successfully / unsuccessfully completing the titration protocol within the threshold time.Attorney Docket Nos.: 9134-0785 P37880-WO-1 4. Adjusting the initial titration protocol based on parameter values with the greatest success rates and / or fewest failures.

[0067] It is also conceivable that the initial titration protocol may be a combination, ifavailable and applicable, of the above failure analysis and similarity analysis approaches. For example, a combination may be used to select initial titration parameters if there is a low confidence level from either method or the combination may be used to reconcile incompatible or contraindicated parameters based on, e.g., the patient’s personal parameters such as comorbid conditions.

[0068] Applying an aggressiveness level to the titration protocol to obtain an initial usetitration protocol:

[0069] Once initial titration parameters from the previous step are selected, anaggressiveness level may be determined to further optimize the titration speed according to an aggressiveness scheme.

[0070] In some non-limiting embodiments of this disclosure, the aggressiveness scheme mayinclude some number of levels, e.g., 2 levels, 3, levels, 4 levels, or more. Each level of the aggressiveness scheme may correspond to a set of modifications that may be made to the dynamic titration protocol for patient’s falling within that aggressiveness level. For example, an aggressiveness scheme may include three levels. In such a case, the aggressiveness parameter would be the level selected, e.g., level 1, level 2 or level 3. Aggressiveness level 1 may correspond to higher dose adjustments, level 2 may correspond to shorter titration cycle time and level 3 may correspond to both dose and titration cycle time adjustments. One of skill in the art would recognize that the number of levels in such an embodiment may vary and the specific corresponding modifications may also vary. Moreover, the levels need not be represented by numerals. For example, an aggressiveness scheme may have its levels defined as “low, medium, high” or “standard, accelerated”.

[0071] In an alternative embodiment of this disclosure, the aggressiveness scheme may be amatrix and the aggressiveness level may be further refined as the intersection of the relevant parameters on the matrix. For example, one dimension of the matrix may relate to a set of personal parameters, such as, for example, health parameters. Another dimension of the matrix may relate to a different set of personal parameters, such as, for example, demographic parameters. Another dimension may relate to external factors, such as, for example, time ofAttorney Docket Nos.: 9134-0785 P37880-WO-1 year. One of skill in the art will appreciate that any number of dimensions may be used in such a matrix.

[0072] In another non-limiting embodiment, the initial aggressiveness level for a patient maybe determined using analytical techniques such as digital twin and similarity analysis described in connection with determining the initial titration protocol parameters above. Alternatively, the aggressiveness level may be derived from the values of one or more health parameters of the patient. Alternatively, the initial aggressiveness level may be a default level or a level defined for different demographic parameters. Alternatively, the initial aggressiveness level may be set by the healthcare provider.

[0073] Circumstances that may affect the selection of an aggressiveness level include certaincalendar events, such as holidays, suggesting behaviors that would result in above average glucose values, women’s health considerations, such as menstrual cycles, menopause, etc., or illnesses, such as viral infections, that may affect the desired aggressiveness for a titration protocol.

[0074] Once the aggressiveness level is determined, the aggressiveness scheme may beconsulted for the corresponding modifications to the dynamic titration protocol. Modifications may be made to any parameter of the dynamic titration protocol including, but not limited to, the dose, the dose step adjustment, the frequency of administration, the titration cycle time, etc.

[0075] In one embodiment, the dose step adjustment may be modified. As a non-limitingexample of dose step adjustments, the default dose step adjustment profile for hyper and extreme hyper values may be +1 units, +2 units, respectively. Using an exemplary three-level aggressiveness scheme, aggressiveness level 1 may call for +2 units for hyperglycemic values and +3 units for extreme hyperglycemic values. Aggressiveness level 2 may call for +3 units for hyperglycemic values and +4 units for extreme hyperglycemic values. Aggressiveness level 3 may call for +4 units for hyperglycemic values and +5 units for extreme hyperglycemic values.

[0076] Further, the titration cycle time may be modified. Ordinarily, dosage increases mayoccur after, e.g., days 2, 4, 6, 8 and 10. Again referring to a three-level aggressiveness scheme as an example, levels 2 and 3 may increase the dosage after 2, 3, 4, 5, 7 and 10 days. One of skill in the art will understand that the precise number of days to adjust the timing of dosage increases may be varied.Attorney Docket Nos.: 9134-0785 P37880-WO-1

[0077] Further, the frequency of administration may be modified. In some aggressivenessschemes, for example, a less aggressive level might suggest splitting the dose into two per day instead of one larger dose at the end of the day.

[0078] Executing the initial use titration protocol with a patient:

[0079] The initial use titration protocol may then be communicated to the patient. In onenon-limiting embodiment, the titration system may determine a preferred information delivery medium for communicating the titration protocol to the patient. A similarity analysis may be conducted for demographic parameters within the similar patient cohort to determine titration success rates for the different types of information delivery medium. The information delivery medium with the greatest success rate for the patient’s cohort may be recommended. Optimizing the titration protocol information delivery medium increases the likelihood of patient compliance with the titration protocol, which increases the probability of a successful titration outcome.

[0080] The possible information delivery media for the titration protocol, monitoring anddata entry based on patient cohort may include: SMS, lightweight titration service app, integration in other app (e.g., mySugr), voice-skill Alexa, Siri, Cortana, Google Home (smartphone vs. home device), bot or assistant calling the patient.

[0081] Performing a failsafe monitoring procedure:

[0082] One of skill in the art will recognize that a failsafe monitoring procedure may includevarious tests and analysis to ensure that the dynamic titration protocol is not placing the user at increased risk for an adverse event. Traditionally, such monitoring has been performed either ad hoc by the patient in consultation with a healthcare provider or in a hospital when more aggressive titration is required. In dynamic titration protocols according to the instant disclosure, the failsafe monitoring procedures may be built into the system to allow for a more aggressive titration protocol without requiring inpatient services.

[0083] In one non-limiting embodiment, the failsafe monitoring procedure may include, forexample, evaluating the risk of hypoglycemia, evaluating insulin sensitivity and evaluating the time required to reach the target blood glucose level.

[0084] Evaluating the risk of hypoglycemia may begin with a prediction of the probabilitythat a hypoglycemic event will occur. It may also include the probability of such an event occurring at different times of the day, such as when the patient is normally sleeping. TheAttorney Docket Nos.: 9134-0785 P37880-WO-1 probability of a hypoglycemic event may be determined using various statistical methods including, but not limited to, correlative analysis, pattern recognition, machine learning, etc.

[0085] Once the probability is determined, the dynamic titration system may determine if theprobability of a hypoglycemic event is within an acceptable level. For example, the system may have a predetermined threshold level of acceptable risk set to a range selected from the group consisting of between 0% and about 60%, between about 0% and 50%, between about 0% and 33%, between about 0% and 25% and between about 0% and 10%. The range of the predetermined threshold level of acceptable risk may be set based on such things as the patient’s ability to detect hypoglycemia, comorbid conditions that may cause more serious consequences in the event of hypoglycemia, the patient’s medical history, similar factors learned from the patient similar cohort, etc.

[0086] If the risk of hypoglycemia is above the acceptable level, then the aggressivenesslevel may be reduced if possible. The system may then re-evaluate the risk of hypoglycemia using a new titration protocol according to the reduced aggressiveness level.

[0087] If, on the other hand, the risk of hypoglycemia is above the acceptable level and theaggressiveness level is already at the lowest possible level, the dynamic titration protocol may be ended.

[0088] Finally, if the risk of a hypoglycemic event is within acceptable levels, then thedynamic titration protocol may be continued.

[0089] Insulin sensitivity may be evaluated as part of the failsafe monitoring procedure.After the first dose of antidiabetic and the second blood glucose measurement, it is possible to calculate insulin-sensitivity. Insulin sensitivity may have a large margin of statistical uncertainty at the beginning of the titration protocol and the statistical uncertainty may decrease by the end of the titration protocol. Any remaining uncertainty of the insulin sensitivity at the end of titration may be due to the patient’s health, age, BMI, comorbidities, maladies, etc.

[0090] Insulin sensitivity may be calculated using the following formula:IS = 1500 / total daily dose where IS = Insulin Sensitivity in U / mg / dLAttorney Docket Nos.: 9134-0785 P37880-WO-1

[0091] As the patient progresses through the titration protocol the insulin sensitivity can beupdated on a daily basis to provide a trajectory of insulin sensitivity that changes daily until a steady state is reached. The following formula may be used: ISt = ISstart + (ISsteadystate - ISstart) * log(k * t) where t = current day ISStart = Starting Basal Insulin Dose ISSteadyState= Expected steady state insulin dose ISt= Basal Insulin Dose on day t k = aggressiveness of insulin titration, also a surrogate for risk of hypoglycemia 1 ≤ k*t ≤ 10

[0092] Insulin sensitivity may be used to determine a risk level. Similar to theaggressiveness level, described above, a risk level represents a particular level of risk within an overall risk scheme. A risk scheme may be constructed in a variety of ways. It may include any number of risk levels (e.g., 2 levels, 3 levels, 4 levels, … n levels). Risk levels within a risk scheme may be defined according to various criteria including, but not limited to, the absolute value of insulin sensitivity, the rate of change of insulin sensitivity, the stability of insulin sensitivity, the absolute value of the fasting blood glucose level, the rate of change of the fasting blood glucose level, the average blood glucose level, the variability of the blood glucose level, the frequency of hypoglycemic events, the probability of a night time hypoglycemic event, etc.

[0093] In one embodiment, the risk scheme may have three levels determined by insulinsensitivity. A three dimensional table incorporating starting basal insulin value x continuous insulin sensitivity level for various risk levels may be used to guide the determination of a user’s risk level. Table 1 below presents 3 risk levels based on the relationship between starting bolus dose, time to target and different insulin sensitivities for a given risk level. Table 1: Risk Level 1 Risk Level 2 Risk Level 3 3 3Attorney Docket Nos.: 9134-0785 P37880-WO-1

[0094] Once the risk level is determined, it may be used to adjust the aggressiveness level.For example, users with a risk level 3 may be assigned a higher aggressiveness level as compared to users with a risk level 1. In such a case, the risk level may be used to determine the best aggressiveness level going forward in the dynamic titration protocol. In this example, the risk level is determined based on the value of the insulin-sensitivity with the restriction that even with the worst statistical uncertainty the expected blood glucose level is not lower than the upper limit of the target-range, the middle of the target range and the lower limit of the target range, which corresponds to risk level 1, risk level 2 and risk level 3.

[0095] Using the risk level to guide adjustment of the dynamic titration protocol may resultin the adjustments to the titration protocol that are smaller for users assigned to the lowest risk level, but even in the worst case the target range is just reached. For higher risk level settings, the target range might be reached faster but there is a risk that the low limit of the target range or even lower (hypoglycemic) values are reached.

[0096] This is one example using continuously calculated insulin sensitivity as a failsafemonitor in a dynamic titration protocol. One of skill in the art will appreciate that there are other ways to incorporate insulin sensitivity into the aggressiveness level determination so as to ensure the target blood glucose level is reached as quickly as possible without creating undue risk of adverse health events. For example, if the calculated insulin sensitivity is uncertain or trending higher, then a planned upward adjustment of aggressiveness may be delayed.

[0097] Also as part of the failsafe monitoring procedure, the time remaining to reach thetarget glucose level may be evaluated. Once the titration protocol has begun, a time series or trend analysis of measured blood glucose values may be used to predict the time remaining to reach the target blood glucose level at the end of each titration cycle. One of skill in the art will recognize that there are several ways to predict the time remaining to reach the target blood glucose level.

[0098] In one example, the initial prediction of the time remaining to reach the target bloodglucose level may be made, even without any blood glucose values, based on the digital twin or similarity analysis discussed above. The prediction may be made, for example, according to significant patient parameters (i.e., age, BMI, etc.). This predictive method may also be combined with additional methods so that, with every additional blood glucose value, the time remaining prediction may be recalculated to improve accuracy.Attorney Docket Nos.: 9134-0785 P37880-WO-1

[0099] In an alternative embodiment, the time remaining to reach the target blood glucose levelmay be determined using linear regression techniques, described elsewhere herein.

[0100] In an alternative embodiment, the time remaining to reach the target blood glucoselevel may be determined using multiple linear regression techniques, described elsewhere herein. A multiple regression model may be made more robust by adding patient parameters, including both health and demographic parameters. A regression model may be used to make predictions for multiple days / weeks in the future. The accuracy increases with the number of days a user has been following the dynamic titration protocol.

[0101] In an alternative embodiment, the time remaining to reach the target blood glucoselevel may be determined using stochastic modelling techniques.

[0102] In an alternative embodiment, the time remaining to reach the target blood glucoselevel may be determined by transforming blood glucose values into a reduced rank space and comparing them to Eigen vectors derived from blood glucose values of previous subjects.

[0103] In an alternative embodiment, the time remaining to reach the target blood glucoselevel may be determined using machine learning techniques, described elsewhere herein. For example, decision tree based models or neural networks may be used to predict the time to reach the target blood glucose.

[0104] Additional failsafe monitoring procedures are also contemplated for use with theinstant disclosure to ensure parameter adjustments do not present an unacceptable risk to the patient. Examples include, but are not limited to:

[0105] Evening blood glucose: If the measured blood glucose value in the evening before theinjection of insulin differs significantly from the other days, then dynamic titration adjustments based on the aggressiveness level may be paused to avoid hypoglycemia.

[0106] Health Care Provider Monitoring: exemplary methods for allowing the health careprovider to influence or even over-ride the dynamic titration system include: ^Providing the health care provider with a summary of all patients terminated lastday / week with option to connect and continue protocol ^Providing the health care provider with an option to override dynamic titration modeadjustments by modulating user interface sliders for each protocol parameter to see what the hypo risk and time to target would be with different titration parametersAttorney Docket Nos.: 9134-0785 P37880-WO-1 ^The health care provider may set a governing range of ‘safe’ parameter levels for thedynamic titration protocol system

[0107] Revising the initial aggressiveness factor based on the results of the failsafemonitoring procedure to obtain a revised aggressiveness factor

[0108] The aggressiveness level may be revisited and revised based on the results of thefailsafe monitoring, e.g., the probability of a hypoglycemic event, the predicted time remaining to reach the target blood glucose level and the uncertainty of the insulin sensitivity. The uncertainty in the insulin sensitivity calculation may provide an estimation of how the titration is going to evolve.

[0109] In one non-limiting embodiment, a recommended time to complete titration for thepatient may be defined. The recommended time may be based on a statistical analysis of similar patients according to the above discussed techniques. Alternatively, the recommended time may be based on arbitrary standards set by the antidiabetic manufacturer or health care provider, etc. Alternatively, the recommended time may be based on prior personal titration history.

[0110] Once the recommended time is established, the system may evaluate the 95%confidence interval for the predicted time remaining to reach the target blood glucose level. The dynamic titration system may then evaluate if the predicted time remaining is acceptable in light of the recommended time to complete titration and adjust accordingly.

[0111] For example, if the patient is expected to reach the target blood glucose level withinthe recommended time, then the aggressiveness level need not be adjusted. However, if predicted time to reach the target blood glucose level exceeds the recommended time, then the aggressiveness level may be increased. Alternatively, if the predicted time is significantly longer than the recommended time, a change in antidiabetic may be indicated.

[0112] Applying the revised aggressiveness level to the initial use titration protocol to obtaina revised use titration protocol:

[0113] In one alternative embodiment for dynamic titration with failsafe monitoring, thetitration parameters may be adjusted using, e.g., the progress of fasting blood glucose towards the target, as follows: ^Fasting blood glucose at start = 250 mg / dL, goal: <100 mg / dL^ phases 1-2: titration starts with standard titration stepsAttorney Docket Nos.: 9134-0785 P37880-WO-1 ^phases 2-4: if the blood glucose decrease per time is below a predefined threshold (e.g., -40 mg / dL per week), the dynamic titration system may automatically increase the aggressiveness level, leading to a higher blood glucose decrease / week. One of skill in the art will recognize that, in the alternative, if the blood glucose decrease is still below a certain threshold, this may indicate high insulin resistance and the titration might be stopped. ^phases 4-5: the dynamic titration protocol reduce the aggressiveness to slow bloodglucose reductions as fasting blood glucose approaches the target range of 100 mg / dL. ^phases 5-6: the dynamic titration protocol may further decrease titration steps to reducethe probability of hypoglycemic events. ^phases 6-7: blood glucose target may be reached and titration successfully ended^ phases 7-8: post-titration, fixed dose

[0114] In another example, if the failsafe monitoring produces positive and stable results, thedynamic titration protocol may adjust the dosage steps at a higher adjustment interval to approach the target blood glucose level faster. For example, instead of titrating with a steady dose step adjustment along the lines of: 2-2-2-2-2-2-2-2-2-2-2-2-2-2-2, the dynamic titration system may use dose step adjustments along the lines of: 2-2-3-4-5-5-3-3-2-1. In this example, the patient would reach the target blood glucose level 5 titration cycles earlier, reducing overall titration protocol length.

[0115] TITRATION SYSTEM

[0116] A dynamic titration system includes a database having anonymized data for aplurality of previously managed subjects. The anonymized data may include personal parameters, e.g., demographic parameters and health parameters for each previously managed patient, the titration protocol used for each previously managed patient, and the titration output parameters for each previously managed patient.

[0117] The titration system may further include a healthcare provider interface configured toreceive personal parameters for the patient. The healthcare provider interface may be used to initiate the dynamic titration protocol. The healthcare provider interface may further be configured to receive, e.g., a target blood glucose level and a target time to reach the target blood glucose level. Non-limiting examples of healthcare provider interfaces include personalAttorney Docket Nos.: 9134-0785 P37880-WO-1 computers, enterprise computers, dumb terminals, network communication devices, tablets, smart phones, smart watches and the like.

[0118] The methods of the disclosed dynamic titration protocol may be stored and executedon or by the healthcare provider interface. Additional aspects of the disclosed methods including the time to target predictions, insulin sensitivity calculations, aggressiveness adjustments and fail safe monitoring may be stored and executed by the healthcare provider interface. Alternatively, the methods of the disclosed dynamic titration protocol and additional aspects may be stored and executed on or by a device remote from the healthcare provider interface, such as, for example, an internal server, a cloud-based server, or a similar remote device.

[0119] In some embodiments, the titration system may include a data processing deviceconfigured to perform some or all of the disclosed dynamic titration protocol and additional aspects of the disclosed method such as, but not limited to, predicting the time to reach the target blood glucose level, insulin sensitivity calculations, aggressiveness adjustments and fail safe monitoring.

[0120] Finally, a messaging server may correspond with the patient. For example, themessaging server may facilitate exchanging blood glucose measurements, measurement and dosing reminders, administered antidiabetic doses, and titrated dose recommendations between the patient and the healthcare provider.

[0121] The titration system may also include a patient user interface configured to receiveadministration instructions from the messaging server. Devices that may be used as a patient user interface are non-exclusive. For example, suitable patient user interfaces could include personal computers, enterprise computers, dumb terminals, television screens, bot assistants, network communication devices, tablets, smart phones, smart watches and the like.

[0122] Advantageously, dynamic titration systems and methods according to this disclosureincrease the speed of titration while minimizing the risk of hypoglycemic events, which increases the likelihood of achieving a successful titration outcome (e.g., timely titration, minimal adverse side effects and sustained normoglycemic levels).

[0123] Embodiment 1: A method of establishing and executing a dynamic titration protocolfor administration of an antidiabetic, the method comprising: (a) establishing an initial titration protocol;Attorney Docket Nos.: 9134-0785 P37880-WO-1 (b) applying an aggressiveness level to the initial titration protocol to obtain an initial use titration protocol; (c) executing the initial use titration protocol with a patient; (d) performing a failsafe monitoring procedure; (e) revising the aggressiveness level based on the results of steps (d); (f) applying the revised aggressiveness level to the initial use titration protocol to obtain a revised use titration protocol; (g) executing the revised use titration protocol with a patient; and (h) repeating steps (d)-(g) until the target glucose level is reached.

[0124] Embodiment 2: The method of embodiment 1, wherein step (a) comprises selectingan antidiabetic, an initial dose, an initial titration cycle time and initial titration step adjustments.

[0125] Embodiment 3: The method of embodiment 2, wherein the initial titration protocol isestablished by identifying a patient similar cohort and selecting the antidiabetic, initial dose, initial titration cycle time and initial titration protocol step adjustments with the highest number of successful titrations.

[0126] Embodiment 4: The method of embodiment 3, wherein the patient similar cohort isidentified using one or more of cosine similarity, knowledge graph, artificial intelligence, machine learning, clustering, k-nearest neighbor, support vector machine, naïve Bayes, Bayes network, regression, logistic regression, deep learning, neuronal network, decision tree, random forest.

[0127] Embodiment 5: The method of embodiment 3, wherein a digital twin of the patient isused to identify the patient similar cohort.

[0128] Embodiment 6: The method of embodiment 1, wherein the aggressiveness level isestablished by identifying a patient similar cohort and selecting the aggressiveness level with the highest number of successful titrations.

[0129] Embodiment 7: The method of embodiment 6, wherein the patient similar cohort isidentified using one or more of cosine similarity, knowledge graph, artificial intelligence, machine learning, clustering, k-nearest neighbor, support vector machine, naïve Bayes, Bayes network, regression, logistic regression, deep learning, neuronal network, decision tree, random forest.Attorney Docket Nos.: 9134-0785 P37880-WO-1

[0130] Embodiment 8: The method of embodiment 6, wherein a digital twin of the patient isused to identify the patient similar cohort.

[0131] Embodiment 9: The method of embodiment 1, wherein step (c) further comprises:i) determining an information delivery medium most likely to result in a successful titration for the patient based on the patient cohort; and ii) delivering the titration protocol instructions to the patient using the information delivery medium determined in step (i).

[0132] Embodiment 10: The method of embodiment 9, wherein the information deliverymedium is one or more of SMS, an application, voice-skill Alexa, Siri, Cortana, Google Home, smartphone, email, home device, bot or assistant calling the patient.

[0133] Embodiment 11: The method of embodiment 1, wherein the failsafe monitoringprocedure comprises at least one of (i) predicting the probability of a hypoglycemic event, (ii) calculating insulin sensitivity and (iii) predicting the length of time for the patient to reach the target glucose level based on patient glucose data.

[0134] Embodiment 12: The method of embodiment 11, wherein the failsafe monitoringcomprises establishing a risk level for the patient.

[0135] Embodiment 13: The method of embodiment 12, wherein the aggressiveness level isrevised based on the risk level.

[0136] Embodiment 14: A system for dynamic titration of an antidiabetic for a patient,comprising a processor configured to: (i) access a database having anonymized personal parameters for a plurality of previously managed subjects; (ii) receive data from a healthcare provider interface configured to receive patient- specific personal parameters; (iii) generate patient cohorts from the plurality of previously managed subjects based on commonalities in the anonymized personal parameters, (iv) identify from the generated cohorts a patient similar cohort corresponding to the patient specific personal parameters, (v) based on the patient similar cohort, establish an initial titration protocol and aggressiveness level for the patient, (vi) communicate the dynamic titration protocol to the patient,Attorney Docket Nos.: 9134-0785 P37880-WO-1 (vii) receive a blood glucose level from the patient, (viii) perform a failsafe monitoring procedure, (ix) revise the aggressiveness level based on the results of the failsafe monitoring procedure, and (x) revise the dynamic titration protocol based on the revised aggressiveness level.

[0137] Embodiment 15: The system of embodiment 14, wherein the personal parametercomprises at least one of HbA1c, comorbid conditions, age, height, weight, body mass index, other medications, hypoglycemia risk level, blood glucose values, vital signs.

[0138] Embodiment 16: The system of embodiment 14, wherein the personal parametercomprises at least one of ethnicity, age, gender, socioeconomic status, preferred mode of communication.

[0139] Embodiment 17: The system of embodiment 14, wherein the processor is configuredto identify the patient similar cohort using one or more of cosine similarity, knowledge graph, artificial intelligence, machine learning, clustering, k-nearest neighbor, support vector machine, naïve Bayes, Bayes network, regression, logistic regression, deep learning, neuronal network, decision tree, random forest.

[0140] Embodiment 18: The system of embodiment 14, wherein the processor is configuredto identify the patient similar cohort using a digital twin of the patient.

[0141] Embodiment 19: The system of embodiment 14, wherein the processor is configuredto establish the initial titration protocol and the aggressiveness level based on a correlation between a success rate and the initial titration protocol parameters and the aggressiveness levels used in the patient similar cohort.

[0142] Embodiment 20: The system of embodiment 19, wherein the processor is configuredto establish the initial titration protocol and the aggressiveness level based on a regression and clustering analysis.

[0143] Embodiment 21: The system of embodiment 14, further comprising a patient userinterface configured to receive the initial titration protocol and / or the revised titration protocol from the data processing device.

[0144] Embodiment 22: The system of embodiment 14, wherein the processor is furtherconfigured to determine an information delivery medium most likely to result in a successful titration for the patient based on the patient similar cohort.Attorney Docket Nos.: 9134-0785 P37880-WO-1

[0145] Embodiment 23: The system of embodiment 22, wherein the information deliverymedium is one or more of SMS, an application, voice-skill Alexa, Siri, Cortana, Google Home, smartphone, email, home device, television, bot or assistant calling the patient.

[0146] Embodiment 24: A system for reducing the time to titrate an antidiabetic, comprisinga processor configured to: (a) communicate a titration protocol to a patient; (b) receive blood glucose monitoring data of the patient; (c) evaluate a risk level of an adverse health event if the titration protocol is continued; (d) predict the length of time for the patient to reach a target glucose level if the titration protocol is continued; (e) adjust the parameters of the titration protocol to increase the dose of the antidiabetic at a faster rate or at a slower rate depending on the outcome of steps (c) and (d); (f) repeat steps (a)-(e) until the blood glucose level received at step (b) meets a target level.

[0147] Embodiment 25: The system of embodiment 24, wherein the processor is furtherconfigured to access a database having anonymized personal parameters for a plurality of previously managed subjects.

[0148] Embodiment 26: The system of embodiment 25, wherein the personal parametercomprises at least one of HbA1c, comorbid conditions, age, height, weight, body mass index, other medications, hypoglycemia risk level, blood glucose values, vital signs.

[0149] Embodiment 27: The system of embodiment 25, wherein the personal parametercomprises at least one of ethnicity, age, gender, socioeconomic status, preferred mode of communication.

[0150] Embodiment 28: The system of any one of embodiments 24-27, wherein theprocessor is further configured to establish an initial titration protocol by identifying a patient similar cohort and selecting the antidiabetic, initial dose, initial titration cycle time and initial titration protocol step adjustments with the highest number of successful titrations.

[0151] Embodiment 29: The system of embodiment 28, wherein the processor is configuredto identify the patient similar cohort using one or more of cosine similarity, knowledge graph, artificial intelligence, machine learning, clustering, k-nearest neighbor, support vector machine,Attorney Docket Nos.: 9134-0785 P37880-WO-1 naïve Bayes, Bayes network, regression, logistic regression, deep learning, neuronal network, decision tree, random forest.

[0152] Embodiment 30: The system of embodiment 28, wherein the processor is configuredto identify the patient similar cohort using a digital twin of the patient.

[0153] Embodiment 31: The system of any of embodiments 28-30, wherein the processor isconfigured to establish the initial titration protocol and the aggressiveness level based on a correlation between a success rate and the initial titration protocol parameters and the aggressiveness levels used in the patient similar cohort.

[0154] Embodiment 32: The system of any of embodiments 28-30, wherein the processor isconfigured to establish the initial titration protocol and the aggressiveness level based on a regression and clustering analysis.

[0155] Embodiment 33: The system of any one of embodiments 24-32, wherein theprocessor is further configured to receive data from a healthcare provider interface comprising patient-specific personal parameters.

[0156] Embodiment 34: The system of any one of embodiments 24-33, wherein theprocessor is further configured to generate patient cohorts from the plurality of previously managed subjects based on commonalities in the anonymized personal parameters.

[0157] Embodiment 35: The system of any one of embodiments 24-34, wherein the titrationstep adjustment is adjusted in step (e).

[0158] Embodiment 36: The system of any one of embodiments 24-35, wherein the titrationcycle time is adjusted in step (e).

[0159] Embodiment 37: The system of any one of embodiments 24-36, further comprising apatient user interface configured to receive the initial titration protocol and / or the revised titration protocol from the data processing device.

[0160] Embodiment 38: The system of any one of embodiments 24-37, wherein theprocessor is further configured to determine an information delivery medium most likely to result in a successful titration for the patient based on the patient similar cohort.

[0161] Embodiment 39: The system of embodiment 38, wherein the information deliverymedium is one or more of SMS, an application, voice-skill Alexa, Siri, Cortana, Google Home, smartphone, email, home device, television, bot or assistant calling the patient.Attorney Docket Nos.: 9134-0785 P37880-WO-1

[0162] Embodiment 40: A system for dynamic titration of an antidiabetic for a patient,comprising a processor configured to: (i) establish an initial titration protocol by identifying a patient similar cohort and selecting the antidiabetic, initial dose, initial titration cycle time and initial titration protocol step adjustments with the highest number of successful titrations; communicate a dynamic titration protocol to the patient, (ii) receive a blood glucose level from the patient, (iii) perform a failsafe monitoring procedure, (iv) revise the aggressiveness level based on the results of the failsafe monitoring procedure, and (v) revise the dynamic titration protocol based on the revised aggressiveness level.

[0163] Embodiment 41: The system of embodiment 40, wherein the processor is furtherconfigured to access a database having anonymized personal parameters for a plurality of previously managed subjects.

[0164] Embodiment 42: The system of embodiment 41, wherein the personal parametercomprises at least one of HbA1c, comorbid conditions, age, height, weight, body mass index, other medications, hypoglycemia risk level, blood glucose values, vital signs.

[0165] Embodiment 43: The system of embodiment 41, wherein the personal parametercomprises at least one of ethnicity, age, gender, socioeconomic status, preferred mode of communication.

[0166] Embodiment 44: The system of any one of embodiments 40-43, wherein theprocessor is further configured to establish an initial titration protocol by identifying a patient similar cohort and selecting the antidiabetic, initial dose, initial titration cycle time and initial titration protocol step adjustments with the highest number of successful titrations.

[0167] Embodiment 45: The system of embodiment 44, wherein the processor is configuredto identify the patient similar cohort using one or more of cosine similarity, knowledge graph, artificial intelligence, machine learning, clustering, k-nearest neighbor, support vector machine, naïve Bayes, Bayes network, regression, logistic regression, deep learning, neuronal network, decision tree, random forest.

[0168] Embodiment 46: The system of embodiment 44, wherein the processor is configuredto identify the patient similar cohort using a digital twin of the patient.Attorney Docket Nos.: 9134-0785 P37880-WO-1

[0169] Embodiment 47: The system of any of embodiments 44-46, wherein the processor isconfigured to establish the initial titration protocol and the aggressiveness level based on a correlation between a success rate and the initial titration protocol parameters and the aggressiveness levels used in the patient similar cohort.

[0170] Embodiment 48: The system of any of embodiments 44-46, wherein the processor isconfigured to establish the initial titration protocol and the aggressiveness level based on a regression and clustering analysis.

[0171] Embodiment 49: The system of any one of embodiments 40-48, wherein theprocessor is further configured to receive data from a healthcare provider interface comprising patient-specific personal parameters.

[0172] Embodiment 50: The system of any one of embodiments 40-49, wherein theprocessor is further configured to generate patient cohorts from the plurality of previously managed subjects based on commonalities in the anonymized personal parameters.

[0173] Embodiment 51: The system of any one of embodiments 40-50, further comprising apatient user interface configured to receive the initial titration protocol and / or the revised titration protocol from the data processing device.

[0174] Embodiment 52: The system of any one of embodiments 40-51, wherein theprocessor is further configured to determine an information delivery medium most likely to result in a successful titration for the patient based on the patient similar cohort.

[0175] Embodiment 53: The system of embodiment 52, wherein the information deliverymedium is one or more of SMS, an application, voice-skill Alexa, Siri, Cortana, Google Home, smartphone, email, home device, television, bot or assistant calling the patient.

[0176] Embodiment 54: The method according to any of embodiments 1-13, wherein theantidiabetic is selected from the group consisting of insulin, amylinomimetic injectables, alpha- glucosidase inhibitors, biguanides, dopamine-2 agonists, dipeptidyl peptidase-4 (“DPP-4”) inhibitors, glucagon-like peptide-1 (“GLP-1”) receptor agonists, meglitinides, sodium-glucose transporter 2 (“SGLT 2”) inhibitors, sulfonylureas and thiazolidinediones.

[0177] Embodiment 55: The system according to any of embodiments 14-53, wherein theantidiabetic is selected from the group consisting of insulin, amylinomimetic injectables, alpha- glucosidase inhibitors, biguanides, dopamine-2 agonists, dipeptidyl peptidase-4 (“DPP-4”)Attorney Docket Nos.: 9134-0785 P37880-WO-1 inhibitors, glucagon-like peptide-1 (“GLP-1”) receptor agonists, meglitinides, sodium-glucose transporter 2 (“SGLT 2”) inhibitors, sulfonylureas and thiazolidinediones. BRIEF DESCRIPTION OF THE DRAWINGS

[0178] The above-mentioned aspects of exemplary embodiments will become more apparentand will be better understood by reference to the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0179] FIG. 1 is a schematic representation of a prior art titration process;

[0180] FIG. 2 shows a schematic diagram of a dynamic titration system;

[0181] FIG. 3 shows a schematic diagram of a method of dynamic titration;

[0182] FIG. 4 shows a schematic diagram of a method of creating a first use titrationprotocol according to the embodiment of FIG.3;

[0183] FIG. 5 shows a schematic diagram of a process for communicating the dynamictitration protocol to a patient according to the embodiment of FIG.3;

[0184] FIG. 6 shows a schematic diagram of a failsafe monitoring procedure according to theembodiment of FIG.3;

[0185] FIG. 7 shows a schematic diagram of a method of dynamic titration;

[0186] FIG. 8 is a chart showing the time to reach the target blood glucose level usingdifferent aggressiveness levels;

[0187] FIG. 9 is a chart showing a 95% confidence interval about the time to targetprediction; and

[0188] FIG. 10 is a schematic representation of blood glucose levels in different phases oftitration.Attorney Docket Nos.: 9134-0785 P37880-WO-1 DESCRIPTION

[0189] The embodiments described below are not intended to be exhaustive or to limit theinvention to the precise forms disclosed in the following detailed description. Rather, the embodiments are chosen and described so that others skilled in the art may appreciate and understand the principles and practices of this disclosure.

[0190] Fig. 2 is a schematic representation of a dynamic titration system 100 for creating andmanaging a dynamic titration protocol for a patient. In the embodiment shown, dynamic titration system 100 includes a healthcare provider interface 104 that allows healthcare provider 102 to interact with the dynamic titration system 100. The healthcare provider interface 104 may include both an input and an output device such as, for example, a keyboard, a mouse, a touchscreen, a display, etc. Non-limiting examples of a healthcare provider interfaces include personal computers, enterprise computers, dumb terminals, network communication devices, tablets, smart phones, smart watches and the like.

[0191] The healthcare provider interface 104 may include a processor and memory. Theprocessor and memory may be configured to execute the method for creating an initial titration protocol 106. (Also see FIG.4, dashed lines.) Alternatively, a separate device (not shown) having a processor and memory may be configured to execute the method for creating an initial titration protocol 106. In embodiments having such a separate device, the separate device may be configured to communicate with the healthcare provider interface and perform some or all of the functions of the healthcare provider interface described herein.

[0192] A population titration data pool 110 may be housed in an electronic medical recordssystem 108. The healthcare provider interface 104 may be connected to the electronic medical records system 108 using wired or wireless communication technology.

[0193] The healthcare provider interface 104 may also be connected to a messaging server112. The messaging server 112 may be configured to communicate instructions for the titration protocol to the patient 116 using one or more information delivery media 114. For example, instructions may be sent to patient 116 using SMS 114a, a mobile app 114b, a smart home assistant 114c, a bot assistant 114d, or any other suitable information delivery medium or combinations thereof.Attorney Docket Nos.: 9134-0785 P37880-WO-1

[0194] FIG. 3 shows a block diagram of a non-limiting embodiment of the method for usinga dynamic titration protocol to treat a patient 120. The method shown in FIG.3 may be implemented using the dynamic titration system 100 shown in FIG.2.

[0195] The method steps disclosed herein can be carried out in the illustrated sequence.However, alternative sequences are also possible. Further, individual or multiple method steps can be carried out in parallel, simultaneously, or repeatedly, either on their own or in groups. For example, the steps for determining the first use titration protocol parameters need not be carried out in the precise order described below. Furthermore, the method can comprise additional method steps that are not illustrated. Independently of the fact that the term method step is used, the term “step” says nothing about the duration of the method steps. Thus, the specified method steps can, individually or in groups, be carried out briefly, but can also be carried out over a longer time period, for example, over time intervals of a number of minutes, hours, days, weeks or even months, for example, continuously or repeatedly.

[0196] In step 122, an initial titration protocol is established. One of skill in the art willrecognize that this can be accomplished in a variety of ways. For example, since the disclosed method will dynamically adjust the titration protocol while it is in use, the initial titration protocol may simply be the default protocol according to the antidiabetic manufacturer or industry guidelines for certain demographics.

[0197] Alternatively, if a previous titration protocol failed, it may be advantageous todiagnose the reason for failure and incorporate it into the initial titration protocol. Failure analysis results may be used if, for example, the previous titration timed out or a maximum basal dose was reached, but the titration protocol was not stopped due to hypoglycemia. One non- limiting example of incorporating failure analysis results into the initial titration protocol may be using an ending titration dose from a previous titration protocol failure as the starting dose for a new initial titration protocol. This may reduce the time required to reach the target glucose level. Further, if the previous titration failed due to small dosage increments causing it to take too long to reach the target glucose level, the initial titration protocol may include higher dosage increments.

[0198] Another alternative for establishing the initial titration protocol is to perform digitaltwin or similarity analysis to determine the personal parameters impacting the time to reach theAttorney Docket Nos.: 9134-0785 P37880-WO-1 target glucose level (e.g. age, BMI, baseline HbA1c, etc.). Techniques for performing digital twin and similarity analysis are discussed above.

[0199] It is also conceivable that the initial titration protocol may be a combination, ifavailable and applicable, of the above failure analysis and similarity analysis approaches. For example, a combination may be used to select initial titration parameters if there is a low confidence level from either method or the combination may be used to reconcile incompatible or contraindicated parameters based on, e.g., the patient’s personal parameters such as comorbid conditions.

[0200] At step 124, the dynamic titration system applies an initial aggressiveness parameterto create a first use titration protocol. In one embodiment, the aggressiveness parameter may simply be an aggressiveness level selected from several possible levels in an aggressiveness scheme. For example, an aggressiveness scheme may include three levels. In such a case, the aggressiveness parameter would be the level selected, e.g., level 1, level 2 or level 3. Alternatively, the aggressiveness scheme may include 4 levels and the aggressiveness parameter would be, e.g., level 1, level 2, level 3 or level 4. Alternatively, the aggressiveness scheme may be binary, in which the aggressiveness parameter may be set to either “standard” or “accelerated”. Indeed, the aggressiveness scheme may have any number of levels and the aggressiveness parameter will correspond to the level selected from the aggressiveness scheme.

[0201] Circumstances that may affect the determination of an aggressiveness level includecertain calendar events, such as holidays, suggesting behaviors that would result in above average glucose values, women’s health considerations, such as menstrual cycles, menopause, etc., or illnesses, such as viral infections, that may affect the desired aggressiveness for a titration protocol.

[0202] The aggressiveness parameter may be determined, at least initially, by performing adigital twin and / or other form of similarity analysis to identify the aggressiveness parameter most correlated to success in patients with similar personal parameters. Alternatively, the aggressiveness parameter may be selected based on patient-specific considerations such as history of hypoglycemic events, past failed titration attempts, etc.

[0203] Once the aggressiveness parameter is determined, it may be used to adjust parametersof the initial titration protocol to create a first use titration protocol. For example, the aggressiveness parameter may form the basis for increases in antidiabetic dose adjustments forAttorney Docket Nos.: 9134-0785 P37880-WO-1 each titration cycle to varying degrees, or the aggressiveness parameter may be the basis to shorten the length of the titration cycle, or both. For example, in the case of a three-level aggressiveness scheme, selecting aggressiveness level 1 may result in dose adjustments, level 2 may result in shorter titration cycle time and level 3 may result in both dose and titration cycle time adjustments. One of skill in the art would recognize various other schemes that would be possible.

[0204] Further, the timing of the dose administration may be varied in accordance with theaggressiveness parameter. In some aggressiveness schemes, for example, a less aggressive level might suggest splitting the dose into two smaller doses per day instead of one larger dose at the end of the day.

[0205] Turning briefly to FIG. 4, FIG. 4 illustrates one embodiment of a method of creatingthe initial use titration protocol, also shown as box 106 in FIG.3. In the non-limiting embodiment according to FIG.4, a digital twin is created for the patient at step 402. The dynamic titration system then identifies a patient similar cohort for the digital twin at step 404. In step 406, the initial titration protocol and the aggressiveness parameter may be derived based on the patient similar cohort results. Finally, at step 408, the aggressiveness parameter may be used to modify the initial protocol and create the first use titration protocol. An exemplary plot is shown in FIG.8 for the length of time required to titrate an antidiabetic at different aggressiveness levels. Line 808 indicates the least aggressive, most conservative aggressiveness level. As can be seen, it takes substantially longer to reach the target blood glucose level using the most conservative aggressiveness level. Line 806 shows an expected trend line for a normal or moderate aggressiveness level. Line 804 shows an exemplary titration trajectory for the highest aggressiveness level. The X axis in FIG.8 represents the number of days of titration and the Y axis represents the percentage of the target dose of antidiabetic.

[0206] Returning now to FIG. 3, step 126, the first use titration protocol may now beexecuted with a patient. FIG.5 shows an exemplary process for communication with the patient during execution of the dynamic titration protocol (step 126) in more detail. At step 502, the dynamic titration system may determine a preferred information delivery medium for communicating the titration protocol to the patient. In one embodiment, a similarity analysis may be conducted for demographic parameters within the similar patient cohort to determine titration success rates for the different types of information delivery media. The informationAttorney Docket Nos.: 9134-0785 P37880-WO-1 delivery medium with the greatest success rate for the patient’s cohort may be recommended via the healthcare provider interface 104. Optimizing the titration protocol information delivery medium increases the likelihood of patient compliance during the protocol, which increases the probability of a successful titration outcome.

[0207] The possible information delivery media for the titration protocol, monitoring anddata entry based on patient cohort may include: SMS, lightweight titration service app, integration in other app (e.g. mySugr), voice-skill Alexa, Siri, Cortana, Google Home (smartphone vs. home device), bot or assistant calling the patient. Since there is always the possibility that a particular patient does not have access to the preferred information delivery medium for his or her cohort, the healthcare provider and / or the patient may optionally confirm the selection of the information delivery medium at step 504.

[0208] At step 506, the first use titration protocol is delivered to the patient via themessaging server 112 so that the patient may follow the administration regimen. Results, such as fasting blood glucose, hyperglycemic and hypoglycemic events, time in range, etc. can then be returned to the healthcare provider interface via the messaging server 112 at step 508.

[0209] Returning to FIG. 3, while the patient is using the dynamic titration protocol, it isnecessary to perform failsafe monitoring 134, shown generally as steps 128, 130, 132 and 136 in FIG.3. For example, step 128 is calculating insulin sensitivity. The calculation is made according to the procedure described above. One inquiry at the failsafe monitoring stage 134 is whether the patient’s insulin sensitivity is stable. Step 130 is predicting the time for the patient to reach a target blood glucose level. The target blood glucose level may be, for example, below about 140 mg / dL; below about 130 mg / dL; below about 120 mg / dL or a patient-specific target determined by the healthcare provider. In step 130, the method may include further assessing whether the time to reach the target blood glucose level is acceptable. In light of the length of time to reach the target blood glucose level, the method may also assess the aggressiveness level. Step 132, determining the revised aggressiveness factor, may involve, for example, lowering the aggressiveness level if there is an unacceptable risk of hypoglycemia or increasing the aggressiveness level if the time to reach the target blood glucose level is too long.

[0210] Turning now to FIG. 6, a more detailed non-limiting embodiment of the failsafemonitoring procedure 134 (steps 128-136 of FIG.3) is shown, beginning with step 602.Attorney Docket Nos.: 9134-0785 P37880-WO-1

[0211] At step 604, the probability of a hypoglycemic event may be predicted using variousstatistical methods including, but not limited to, correlative analysis, pattern recognition, machine learning, etc. At step 606, the system determines if the probability of a hypoglycemic event is within an acceptable level. For example, the system may have a predetermined threshold level of acceptable risk set to a range selected from the group consisting of between 0% and about 60%, between about 0% and 50%, between about 0% and 33%, between about 0% and 25% and between about 0% and 10%. The range of the predetermined threshold level of acceptable risk may be set based on such things as the patient’s ability to detect hypoglycemia, comorbid conditions that may cause more serious consequences in the event of hypoglycemia, the patient’s medical history, similar factors learned from the patient similar cohort, etc.

[0212] If the risk of hypoglycemia is above the acceptable level, then it must be determinedwhether the aggressiveness level can be reduced at step 608. If the aggressiveness level can be reduced, then it should be reduced at step 132a and the first use titration protocol may be revised with the new aggressiveness factor at step 136. The system may then start the failsafe monitoring procedure again at step 604 by predicting the probability of a hypoglycemic event based on the revised use titration protocol with the lower aggressiveness level.

[0213] If the aggressiveness factor cannot be reduced because it is determined at step 608that it is already at the lowest level, then the titration protocol should be ended at step 142, show also in FIG.3.

[0214] If the risk of a hypoglycemic event is within acceptable levels at step 606, then thesystem may proceed to step 128, where the insulin sensitivity may be calculated. The insulin sensitivity calculation is discussed elsewhere herein. If the insulin sensitivity is determined to be unstable at step 610, then the dynamic titration protocol should be ended at step 142.

[0215] If the insulin sensitivity is determined to be stable at step 610, then the system mayproceed to step 130. The time remaining for the patient to reach the target blood glucose level may be determined at step 130 using the methods described elsewhere herein or other methods known in the art. FIG.9 shows an exemplary plot 902 of the 95% confidence interval 904, 908 surrounding the predicted time remaining to reach the target blood glucose level 906. As can be seen in FIG.9, the time remaining should decrease over the time that the dynamic titration protocol is in use. If it is determined at step 612 that the predicted time to reach the target blood glucose level is acceptable, then the failsafe monitoring procedure is complete and the systemAttorney Docket Nos.: 9134-0785 P37880-WO-1 may proceed to step 614, where it will return to the dynamic titration protocol shown in FIG.3 at step 138.

[0216] Alternatively, if it is determined at step 612 that the time to reach the target bloodglucose level exceeds the acceptable limit, then the system may proceed to step 614 to determine whether the aggressiveness factor is set to the highest level. If the aggressiveness factor is not set to the highest level, then the aggressiveness level may be increased at step 132b and the system and the first use titration protocol may be revised with the new aggressiveness factor at step 136. The system may then return to step 130 to repeat the prediction of time to reach the target blood glucose level based on the revised use titration protocol. If, on the other hand, the aggressiveness factor is already set to the highest level, the dynamic titration protocol should be stopped at step 142.

[0217] Returning to FIG. 3, at step 138 the revised use titration protocol should be executed.This can be accomplished using the same tools and procedures discussed with reference to step 126. One of skill in the art will recognize that if no revision is made to the aggressiveness factor at step 132, then the revised use titration protocol will be the same as the first use titration protocol.

[0218] At step 140, the dynamic titration system checks whether the target blood glucoselevel has been reached. If it has, the dynamic titration protocol ends at step 142. If the target blood glucose has not been reached at step 140, then the system returns to the failsafe monitoring procedure 134 at, e.g., step 128.

[0219] FIG. 7 shows an alternative embodiment of the disclosed method. The method shownin FIG.7 may also be implemented using the dynamic titration system 100 shown in FIG.2.

[0220] The dynamic titration protocol begins with the initial setup 702. Default insulin type,initial dose, step and cycle time values are selected at step 704.

[0221] The system then checks for previous failure modes at step 706. If there are previousfailure modes available, the system diagnoses the cause of the previous titration failures and applies the learned parameters to adjust the initial titration protocol at step 708.

[0222] If there are no previous failure modes available at step 706, they system checks fordigital twin / similarity analysis availability at step 710. If the digital twin / similarity analysis is available, the system applies adjusted parameters to the titration protocol based on the digital twin / similarity analysis at step 712.Attorney Docket Nos.: 9134-0785 P37880-WO-1

[0223] If there is no digital twin / similarity analysis available at step 710, then the systemselects at titration aggressiveness level at step 714 and checks for a history of nighttime hypoglycemic events, high coefficient of variability, etc. at step 716. If the risk factors are present in step 716, then the dynamic titration system defaults back to a standard titration protocol at step 718.

[0224] If the risk factors are not present at step 716, then the dynamic titration systemidentifies the appropriate risk level for the patient in steps 720, 722 and 724. The titration parameters can then be adjusted according to the aggressiveness level corresponding to the risk level. The titration protocol can be initiated at step 726.

[0225] While the dynamic titration protocol is in use, the system performs the continuousfailsafe monitoring procedure 728. In particular, the continuous failsafe monitoring protocol checks the evening blood glucose measurements and predictions for hypoglycemia at step 736. At step 730, the time to reach the target blood glucose and insulin sensitivity are checked. If the results at step 730 are unfavorable, the aggressiveness level may be adjusted at step 732 and the titration protocol may be revised. The system then returns to the failsafe monitoring until the target blood glucose level is reached at step 734 and the titration protocol ends at step 740.

[0226] On the other hand, if the aggressiveness level cannot be adjusted either because thehighest or lowest level has been reached and the insulin sensitivity threshold is reached, then the dynamic titration protocol ends with step 740.

[0227] FIG. 10 shows an exemplary graph of a predicted fasting blood glucose by titrationphase 1002. As can be seen, the dynamic titration protocol should result in faster reductions in blood glucose 1004 during the initial phases of the titration and a slower rate of reduction as the fasting blood glucose level 1004 approaches the target blood glucose level 1006.

[0228] While exemplary embodiments have been disclosed hereinabove, the presentinvention is not limited to the disclosed embodiments. Instead, this application is intended to cover any variations, uses, or adaptations of this disclosure using its general principles. Further, this application is intended to cover such departures from the present disclosure as come within known or customary practice in the art to which this invention pertains and which fall within the limits of the appended claims.

Claims

Attorney Docket Nos.: 9134-0785 P37880-WO-1 WHAT IS CLAIMED IS:

1. A method of establishing and executing a dynamic titration protocol for administration ofan antidiabetic, the method comprising: (a) establishing an initial titration protocol; (b) applying an aggressiveness level to the initial titration protocol to obtain an initial use titration protocol; (c) executing the initial use titration protocol with a patient; (d) performing a failsafe monitoring procedure; (e) revising the aggressiveness level based on the results of steps (d); (f) applying the revised aggressiveness level to the initial use titration protocol to obtain a revised use titration protocol; (g) executing the revised use titration protocol with a patient; and (h) repeating steps (d)-(g) until the target glucose level is reached.

2. The method of claim 1, wherein step (a) comprises selecting an antidiabetic, an initial dose, an initial titration cycle time and initial titration step adjustments.

3. The method of claim 2, wherein the initial titration protocol is established by identifying a patient similar cohort and selecting the antidiabetic, initial dose, initial titration cycle time and initial titration protocol step adjustments with the highest number of successful titrations.

4. The method of claim 3, wherein the patient similar cohort is identified using one or more of cosine similarity, knowledge graph, artificial intelligence, machine learning, clustering, k- nearest neighbor, support vector machine, naïve Bayes, Bayes network, regression, logistic regression, deep learning, neuronal network, decision tree, random forest.

5. The method of claim 3, wherein a digital twin of the patient is used to identify the patient similar cohort.Attorney Docket Nos.: 9134-0785 P37880-WO-1 6. The method of claim 1, wherein the aggressiveness level is established by identifying a patient similar cohort and selecting the aggressiveness level with the highest number of successful titrations.

7. The method of claim 6, wherein the patient similar cohort is identified using one or more of cosine similarity, knowledge graph, artificial intelligence, machine learning, clustering, k- nearest neighbor, support vector machine, naïve Bayes, Bayes network, regression, logistic regression, deep learning, neuronal network, decision tree, random forest.

8. The method of claim 6, wherein a digital twin of the patient is used to identify the patient similar cohort.

9. The method of claim 1, wherein step (c) further comprises: i) determining an information delivery medium most likely to result in a successful titration for the patient based on the patient cohort; and ii) delivering the titration protocol instructions to the patient using the information delivery medium determined in step (i).

10. The method of claim 9, wherein the information delivery medium is one or more of SMS, an application, voice-skill Alexa, Siri, Cortana, Google Home, smartphone, email, home device, bot or assistant calling the patient.

11. The method of claim 1, wherein the failsafe monitoring procedure comprises at least one of (i) predicting the probability of a hypoglycemic event, (ii) calculating insulin sensitivity and (iii) predicting the length of time for the patient to reach the target glucose level based on patient glucose data.

12. The method of claim 11, wherein the failsafe monitoring comprises establishing a risk level for the patient.Attorney Docket Nos.: 9134-0785 P37880-WO-1 13. The method of claim 12, wherein the aggressiveness level is revised based on the risk level.

14. A system for dynamic titration of an antidiabetic for a patient, comprising a processor configured to: (i) access a database having anonymized personal parameters for a plurality of previously managed subjects; (ii) receive data from a healthcare provider interface configured to receive patient- specific personal parameters; (iii) generate patient cohorts from the plurality of previously managed subjects based on commonalities in the anonymized personal parameters, (iv) identify from the generated cohorts a patient similar cohort corresponding to the patient specific personal parameters, (v) based on the patient similar cohort, establish an initial titration protocol and aggressiveness level for the patient, (vi) communicate the dynamic titration protocol to the patient, (vii) receive a blood glucose level from the patient, (viii) perform a failsafe monitoring procedure, (ix) revise the aggressiveness level based on the results of the failsafe monitoring procedure, and (x) revise the dynamic titration protocol based on the revised aggressiveness level.

15. The system of claim 14, wherein the personal parameter comprises at least one of HbA1c, comorbid conditions, age, height, weight, body mass index, other medications, hypoglycemia risk level, blood glucose values, vital signs.

16. The system of claim 14, wherein the personal parameter comprises at least one of ethnicity, age, gender, socioeconomic status, preferred mode of communication.

17. The system of claim 14, wherein the processor is configured to identify the patient similar cohort using one or more of cosine similarity, knowledge graph, artificial intelligence,Attorney Docket Nos.: 9134-0785 P37880-WO-1 machine learning, clustering, k-nearest neighbor, support vector machine, naïve Bayes, Bayes network, regression, logistic regression, deep learning, neuronal network, decision tree, random forest.

18. The system of claim 14, wherein the processor is configured to identify the patient similar cohort using a digital twin of the patient.

19. The system of claim 14, wherein the processor is configured to establish the initial titration protocol and the aggressiveness level based on a correlation between a success rate and the initial titration protocol parameters and the aggressiveness levels used in the patient similar cohort.

20. The system of claim 19, wherein the processor is configured to establish the initial titration protocol and the aggressiveness level based on a regression and clustering analysis.

21. The system of claim 14, further comprising a patient user interface configured to receive the initial titration protocol and / or the revised titration protocol from the data processing device.

22. The system of claim 14, wherein the processor is further configured to determine an information delivery medium most likely to result in a successful titration for the patient based on the patient similar cohort.

23. The system of claim 22, wherein the information delivery medium is one or more of SMS, an application, voice-skill Alexa, Siri, Cortana, Google Home, smartphone, email, home device, television, bot or assistant calling the patient.

24. A system for reducing the time to titrate an antidiabetic, comprising a processor configured to: (a) communicate a titration protocol to a patient; (b) receive blood glucose monitoring data of the patient;Attorney Docket Nos.: 9134-0785 P37880-WO-1 (c) evaluate a risk level of an adverse health event if the titration protocol is continued; (d) predict the length of time for the patient to reach a target glucose level if the titration protocol is continued; (e) adjust the parameters of the titration protocol to increase the dose of the antidiabetic at a faster rate or at a slower rate depending on the outcome of steps (c) and (d); (f) repeat steps (a)-(e) until the blood glucose level received at step (b) meets a target level.

25. The system of claim 24, wherein the processor is further configured to access a database having anonymized personal parameters for a plurality of previously managed subjects.

26. The system of claim 25, wherein the personal parameter comprises at least one of HbA1c, comorbid conditions, age, height, weight, body mass index, other medications, hypoglycemia risk level, blood glucose values, vital signs.

27. The system of claim 25, wherein the personal parameter comprises at least one of ethnicity, age, gender, socioeconomic status, preferred mode of communication.

28. The system of any one of claims 24-27, wherein the processor is further configured to establish an initial titration protocol by identifying a patient similar cohort and selecting the antidiabetic, initial dose, initial titration cycle time and initial titration protocol step adjustments with the highest number of successful titrations.

29. The system of claim 28, wherein the processor is configured to identify the patient similar cohort using one or more of cosine similarity, knowledge graph, artificial intelligence, machine learning, clustering, k-nearest neighbor, support vector machine, naïve Bayes, Bayes network, regression, logistic regression, deep learning, neuronal network, decision tree, random forest.Attorney Docket Nos.: 9134-0785 P37880-WO-1 30. The system of claim 28, wherein the processor is configured to identify the patient similar cohort using a digital twin of the patient.

31. The system of any of claims 28-30, wherein the processor is configured to establish the initial titration protocol and the aggressiveness level based on a correlation between a success rate and the initial titration protocol parameters and the aggressiveness levels used in the patient similar cohort.

32. The system of any of claims 28-30, wherein the processor is configured to establish the initial titration protocol and the aggressiveness level based on a regression and clustering analysis.

33. The system of any one of claims 24-32, wherein the processor is further configured to receive data from a healthcare provider interface comprising patient-specific personal parameters.

34. The system of any one of claims 24-33, wherein the processor is further configured to generate patient cohorts from the plurality of previously managed subjects based on commonalities in the anonymized personal parameters.

35. The system of any one of claims 24-34, wherein the titration step adjustment is adjusted in step (e).

36. The system of any one of claims 24-35, wherein the titration cycle time is adjusted in step (e).

37. The system of any one of claims 24-36, further comprising a patient user interface configured to receive the initial titration protocol and / or the revised titration protocol from the data processing device.Attorney Docket Nos.: 9134-0785 P37880-WO-1 38. The system of any one of claims 24-37, wherein the processor is further configured to determine an information delivery medium most likely to result in a successful titration for the patient based on the patient similar cohort.

39. The system of claim 38, wherein the information delivery medium is one or more of SMS, an application, voice-skill Alexa, Siri, Cortana, Google Home, smartphone, email, home device, television, bot or assistant calling the patient.

40. A system for dynamic titration of an antidiabetic for a patient, comprising a processor configured to: (i) establish an initial titration protocol by identifying a patient similar cohort and selecting the antidiabetic, initial dose, initial titration cycle time and initial titration protocol step adjustments with the highest number of successful titrations; communicate a dynamic titration protocol to the patient, (ii) receive a blood glucose level from the patient, (iii) perform a failsafe monitoring procedure, (iv) revise the aggressiveness level based on the results of the failsafe monitoring procedure, and (v) revise the dynamic titration protocol based on the revised aggressiveness level.

41. The system of claim 40, wherein the processor is further configured to access a database having anonymized personal parameters for a plurality of previously managed subjects.

42. The system of claim 41, wherein the personal parameter comprises at least one of HbA1c, comorbid conditions, age, height, weight, body mass index, other medications, hypoglycemia risk level, blood glucose values, vital signs.

43. The system of claim 41, wherein the personal parameter comprises at least one of ethnicity, age, gender, socioeconomic status, preferred mode of communication.Attorney Docket Nos.: 9134-0785 P37880-WO-1 44. The system of any one of claims 40-43, wherein the processor is further configured to establish an initial titration protocol by identifying a patient similar cohort and selecting the antidiabetic, initial dose, initial titration cycle time and initial titration protocol step adjustments with the highest number of successful titrations.

45. The system of claim 44, wherein the processor is configured to identify the patient similar cohort using one or more of cosine similarity, knowledge graph, artificial intelligence, machine learning, clustering, k-nearest neighbor, support vector machine, naïve Bayes, Bayes network, regression, logistic regression, deep learning, neuronal network, decision tree, random forest.

46. The system of claim 44, wherein the processor is configured to identify the patient similar cohort using a digital twin of the patient.

47. The system of any of claims 44-46, wherein the processor is configured to establish the initial titration protocol and the aggressiveness level based on a correlation between a success rate and the initial titration protocol parameters and the aggressiveness levels used in the patient similar cohort.

48. The system of any of claims 44-46, wherein the processor is configured to establish the initial titration protocol and the aggressiveness level based on a regression and clustering analysis.

49. The system of any one of claims 40-48, wherein the processor is further configured to receive data from a healthcare provider interface comprising patient-specific personal parameters.

50. The system of any one of claims 40-49, wherein the processor is further configured to generate patient cohorts from the plurality of previously managed subjects based on commonalities in the anonymized personal parameters.Attorney Docket Nos.: 9134-0785 P37880-WO-1 51. The system of any one of claims 40-50, further comprising a patient user interface configured to receive the initial titration protocol and / or the revised titration protocol from the data processing device.

52. The system of any one of claims 40-51, wherein the processor is further configured to determine an information delivery medium most likely to result in a successful titration for the patient based on the patient similar cohort.

53. The system of claim 52, wherein the information delivery medium is one or more of SMS, an application, voice-skill Alexa, Siri, Cortana, Google Home, smartphone, email, home device, television, bot or assistant calling the patient.

54. The method according to any of claims 1-13, wherein the antidiabetic is selected from the group consisting of insulin, amylinomimetic injectables, alpha-glucosidase inhibitors, biguanides, dopamine-2 agonists, dipeptidyl peptidase-4 (“DPP-4”) inhibitors, glucagon-like peptide-1 (“GLP-1”) receptor agonists, meglitinides, sodium-glucose transporter 2 (“SGLT 2”) inhibitors, sulfonylureas and thiazolidinediones.

55. The system according to any of claims 14-53, wherein the antidiabetic is selected from the group consisting of insulin, amylinomimetic injectables, alpha-glucosidase inhibitors, biguanides, dopamine-2 agonists, dipeptidyl peptidase-4 (“DPP-4”) inhibitors, glucagon-like peptide-1 (“GLP-1”) receptor agonists, meglitinides, sodium-glucose transporter 2 (“SGLT 2”) inhibitors, sulfonylureas and thiazolidinediones.