This application generally relates to medical devices (e.g., analyte sensors), and more specifically to systems, devices, and methods for data collection and development as well as providing user interaction policies.
Diabetes is a metabolic condition relating to the production or use of insulin by the body. Insulin is a hormone that allows the body to use glucose for energy, or store glucose as fat.
When a person eats a meal that contains carbohydrates, the food is processed by the digestive system, which produces glucose in the person's blood. Blood glucose can be used for energy or stored as fat. The body normally maintains blood glucose levels in a range that provides sufficient energy to support bodily functions and avoids problems that can arise when glucose levels are too high, or too low. Regulation of blood glucose levels depends on the production and use of insulin, which regulates the movement of blood glucose into cells.
When the body does not produce enough insulin, or when the body is unable to effectively use insulin that is present, blood sugar levels can elevate beyond normal ranges. The state of having a higher than normal blood sugar level is called “hyperglycemia.” Chronic hyperglycemia can lead to a number of health problems, such as cardiovascular disease, cataract and other eye problems, nerve damage (neuropathy), and kidney damage. Hyperglycemia can also lead to acute problems, such as diabetic ketoacidosis—a state in which the body becomes excessively acidic due to the presence of blood glucose and ketones, which are produced when the body cannot use glucose. The state of having lower than normal blood glucose levels is called “hypoglycemia.” Severe hypoglycemia can lead to acute crises that can result in seizures or death.
A diabetes patient can receive insulin to manage blood glucose levels. Insulin can be received, for example, through a manual injection with a needle. Wearable insulin pumps may also be utilized to receive insulin. Diet and exercise also affect blood glucose levels.
Diabetes conditions may be referred to as “Type 1” and “Type 2.” A Type 1 diabetes patient is typically able to use insulin when it is present, but the body is unable to produce sufficient amounts of insulin, because of a problem with the insulin-producing beta cells of the pancreas. A Type 2 diabetes patient may produce some insulin, but the patient has become “insulin resistant” due to a reduced sensitivity to insulin. The result is that even though insulin is present in the body, the insulin is not sufficiently used by the patient's body to effectively regulate blood sugar levels
For diabetes patients, monitoring blood glucose levels and regulating those levels to be within an acceptable range is important not only to mitigate long-term issues such as heart disease and vision loss, but also to avoid the effects of hyperglycemia and hypoglycemia. Maintaining blood glucose levels within an acceptable range can be challenging, as this level is almost constantly changing over time and in response to everyday events, such as eating or exercising. Advances in medical technologies have enabled development of various systems for monitoring blood glucose, including continuous glucose monitoring (CGM) systems, which measure and record glucose concentrations in substantially real-time. CGM systems are important tools for users of these systems to ensure that measured glucose values are within the acceptable range.
This background is provided to introduce a brief context for the summary and detailed description that follow. This background is not intended to be an aid in determining the scope of the claimed subject matter nor be viewed as limiting the claimed subject matter to implementations that solve any or all of the disadvantages or problems presented above.
The various embodiments of the present systems, devices, and methods comprise several features, no single one of which is solely responsible for their desirable attributes. Without limiting the scope of the present embodiments, their more prominent features will now be discussed below. After considering this discussion, and particularly after reading the section entitled “Detailed Description,” one will understand how the features of the present embodiments provide the advantages described here.
In a first aspect, a method comprises: dividing a plurality of users into an exploration subset of users and an exploitation subset of users; randomly assigning at least one user interaction policy to each of the exploration subset of users; and determining at least one user interaction policy for each of the exploitation subset of users using one or more contextual models trained using contextual data corresponding to the exploitation subset of users, wherein the contextual data corresponding to the exploitation subset of users comprises at least some of a first set of contextual profiles and a second set of contextual profiles.
In a second aspect, a method comprises collecting contextual data for a first subset of a plurality of users; generating a first set of contextual profiles for the first subset of the plurality of users based on the collected contextual data; determining that contextual data for a second subset of the plurality of users is incomplete or not available; training one or more imputation models based on the contextual data for the first subset of the plurality of users to develop the contextual data for the second subset of the plurality of users; generating the contextual data for the second subset of the plurality of users using the one or more imputation models; and generating the second set of contextual profiles for the second subset of the plurality of users based on the generated contextual data for the second subset of the plurality of users.
Also described herein are embodiments of a non-transitory computer readable medium comprising instructions to be executed in a computer system, wherein the instructions when executed in the computer system perform the methods described above.
Also described herein are embodiments of a computer system, wherein software for the computer system is programmed to execute the methods described above.
Also described herein are embodiments of a computer system comprising means for executing the methods described above.
Portable and/or wearable health monitoring devices (also referred to herein as “health monitoring devices”) and mobile health applications (also referred to herein as “applications”), have rapidly become renowned for their capabilities to support user-centered care. For example, management of diabetes can present complex challenges for patients, clinicians, and caregivers, as a confluence of many factors can impact a patient's glucose level and glucose trends. To assist patients with better managing this condition, health monitoring devices (e.g., sensors and other types of monitoring and diagnostic devices) as well as a variety of mobile health applications such as, but not limited to, health and fitness monitoring applications, which can assist with diabetes management for type 1 and type 2 diabetes patients have been developed. The wide dissemination of health monitoring devices and the increase in the development and distribution of mobile health applications has improved health management, and more specifically chronic disease management, in the healthcare domain. In particular, the use of mobile health applications in conjunction with these health monitoring devices, represents a more scalable and potentially more cost-effective alternative to traditional interventions, offering a means of improving health and chronic disease management by expanding the reach of healthcare services and improving users' access to health-related information and interventions.
Mobile health applications enable users to be much more involved in the users' own medical care by granting them access to and control over their health information. In particular, mobile health applications enable users to access, monitor, record, and update their health information regardless of physical constraints, such as time and location. In particular, a variety of intervention applications have been developed to deliver guidance that may assist patients, caregivers, healthcare providers, or other users in improving lifestyle or clinical/patient outcomes by meeting a variety of challenges, such as analyte control, exercise, and/or other health factors. For example, diabetes intervention applications may assist patients, caregivers, healthcare providers, or other users in overnight glucose control (e.g., reduce incidence of hypoglycemic events or hyperglycemic excursions), glucose control during and after meals (e.g., use historical information and trends to increase glycemic control), hyperglycemia corrections (e.g., increase time in target zone while avoiding hypoglycemic events from over-correction), and/or hypoglycemia treatments (e.g., address hypoglycemia while avoiding “rebound” hyperglycemia), to name a few.
Unfortunately, many mobile health applications designed to support the management of chronic diseases or health conditions have been plagued with low user engagement and high user attrition rates. A reason for low user engagement and/or high user attrition rates may include failure of mobile health applications to provide individualized or personalized user interaction policies (as defined further below) to the users. When a mobile health application fails to provide individualized and/or personalized user interaction policies, users of the application may find the interaction to be ineffective in enabling them to take a holistic approach to managing their health (e.g., diseases, conditions, fitness, etc.). Further, user interaction policies that are not tailored to an individual (i.e., not “individualized”) may result in sub-optimal health outcomes. In addition, user interaction policies that are not tailored to a cohort that the individual is a member (i.e., not “personalized”) may also result in sub-optimal health outcomes. Furthermore, user interaction polices that are not individualized and/or personalized, may result in user interaction policies that are not relevant to the user. Thus, user engagement associated with such mobile health applications may decrease and, thereby, user attrition rates may increase.
Accordingly, certain embodiments described herein are directed to a health monitoring platform that uses contextual multi-armed bandit (MAB) models to provide individualized or personalized user interaction policies to users, for example, based on their corresponding contextual data. User interaction policies refer to the various ways or means through which the software application interacts with users. For example, user interaction policies may include recommendations (e.g., increase your steps, get more sleep, you might be interested in intermittent fasting, etc.), insights, content, information, etc., provided to a user. User interaction policies may also include delivery methods (e.g., video, audio, text, phone call, emails, push notification, in app messages, etc.) for delivering such recommendations, insights, content, information, etc. to the user. User interaction policies may further include UI configurations associated with, for example, different layouts of the software application used by the user. An objective of the software application is to individualize or personalize the user interaction policies for each user by selecting and providing, to each user, user interaction policies that maximize user engagement by that specific user and/or the likelihood that the specific user achieves their goals.
In order to achieve this level of individualization and/or personalization of the user interaction policies, the health monitoring platform may utilize contextual MAB models that take, as input, contextual data corresponding to a user, and output individualized and/or personalized user interaction policies to each user. A user's contextual data may include the user's demographic information (e.g., age, gender, ethnicity, etc.), physiological information (e.g., analyte information generated by one or more analyte sensors), non-analyte health information (e.g., heart rate, temperature, or other data generated by non-analyte sensors), disease information, and any other health related information), and any other relevant user-specific information.
A user's contextual data also includes psychographic data. As used herein, psychographic data may include various data related to goals, interests, values, attitudes, personality traits, behavior related data, etc. For example, user goals may include sleep goals, exercise goals, eating goals, glucose management goals, etc.). User interests may include interested in certain types of exercises, certain types of activities, etc. Having access to a user's psychographic data plays an important role in providing individualized or personalized user interaction policies to users that maximize user engagement. If a user's goal is unknown, user interaction policies (e.g., recommendations) provided to the user may not be beneficial in helping the user achieve the goal. For example, if a diabetic user's goal of getting uninterrupted sleep is unknown by the health monitoring platform, providing recommendations that help the user achieve that goal may be unlikely. Similarly, if a user's interest in running (as opposed to other types of exercise) is unknown by the health monitoring platform, recommending that the user engage in weightlifting to achieve a weight loss goal may be unhelpful to the user. Thus, obtaining user's psychographic data and utilizing it as part of the contextual data that is used as input into contextual MAB models for providing user interaction policies is advantageous.
However, there is a technical challenge associated with obtaining psychographic data, as it may not always be fully available for each user. For example, unlike demographic and/or physiological information which may be available for a new user who just started using the health monitoring platform, psychographic data may not be available. Further, even when users are asked to provide psychographic data, they may not necessarily be responsive to such requests. At the same time, overburdening users to provide inputs by continuously, or frequently, seeking such information may not even be desirable, as it results in a negative experience and thus higher user attrition rates. Further, continuously asking a user to provide input about their psychographic data and updating such information for processing may be technically disadvantageous, as such a process would use the limited processing, memory, battery, and network resources that are available to the computing device used by the user.
As such, certain embodiments herein are directed to performing a psychographic data collection and development phase that collects psychographic data for a plurality of users as well as trains imputations models for inferring user psychographic data that is not available for other users, as further described below. Obtaining psychographic data about users allows for creating a more complete training dataset that can then be used in training the contextual MAB models, discussed above, for determining user interaction policies during a later exploration-exploitation phase.
By utilizing imputation models, the psychographics data collection and development phase is able to generate psychographic data that is not available and provide a more useful and complete training data for use by the contextual MAB models. In particular, the imputation models may be used to infer psychographic data for new users and/or for users where information is not available or limited. For example, for a new user, while demographic information may be available, information about the user's goals and/or interests may not be available. For each of a plurality of psychographic features (e.g., goals, behaviors (e.g., diabetic distress, fear of hyperglycemia, psychological well-being), interests, concerns, psychological well-being, etc.) that may be used by the exploration-exploitation to provide user interaction policies, there may be a different imputation model for inferring the corresponding value for the psychographic feature, as further discussed below.
The systems, devices, and methods of the embodiments described herein can be used in conjunction with any type of analyte sensor for any measurable analyte. Further, the system, devices, and methods of the embodiment described herein may be used in conjunction with any health-related application that is provided to the user to improve the user's health. For example, a health-related application may help the user with treating a certain disease or just help with improving the overall health of a user who is not necessarily diagnosed with a disease.
Alternately or additionally, one or more of the analyte monitoring device 104, the medicament delivery system 106, and the computing device 108 may be communicatively coupled in other ways, such as using one or more short range communication protocols or techniques. For example, the analyte monitoring device 104, the medicament delivery system 106, and the computing device 108 may communicate with one another using one or more of Bluetooth, near-field communication (NFC), 5G, and so forth. The analyte monitoring device 104, the medicament delivery system 106, and the computing device 108 may leverage these types of communication to form a closed-loop system between one another. In this way, the medicament delivery system 106 may deliver a medicament (e.g., insulin) based on predictions of an analyte level (e.g., glucose level) computed in real-time (e.g., by the computing device 108) as analyte measurements are obtained by the analyte monitoring device 104.
In one or more implementations, the analyte monitoring device 104 is a continuous glucose monitoring (“CGM”) device. As used herein, the term “continuous” when used in connection with analyte monitoring may refer to an ability of a device to produce measurements substantially continuously, such that the device may be configured to produce the analyte measurements at regular or irregular intervals of time (e.g., approximately every hour, approximately every 30 minutes, approximately every 5 minutes, and so forth), responsive to establishing a communicative coupling with a different device (e.g., when the computing device 108 establishes a wireless connection with the analyte monitoring device 104 to retrieve one or more of the measurements), and so forth. This functionality along with further aspects of the configuration of the analyte monitoring device 104 are discussed in more detail in relation to
In one or more implementations, the analyte monitoring device 104 transmits the analyte measurements 118 to one or more of the computing device 108, the medicament delivery system 106, or some other device, such as via Bluetooth. The analyte monitoring device 104 may communicate these measurements in real-time, e.g., as they are produced using an analyte sensor. Alternately or in addition, the analyte monitoring device 104 may communicate the analyte measurements 118 at set time intervals, e.g., every 30 seconds, every minute, every 5 minutes, every hour, every 6 hours, every day, and so forth. Further still, the analyte monitoring device 104 may communicate these measurements responsive to a request for the measurements, e.g., communicated to the analyte monitoring device 104 when the computing device 108 causes display of a user interface having information about the person 102's glucose level, updates such a display, predicts the person 102's upcoming glucose level for the purpose of delivering insulin, and so forth. Accordingly, the computing device 108 may maintain the analyte measurements 118 of the person 102 at least temporarily, e.g., in computer readable storage media of the computing device 108.
Note that, while in certain examples the analyte monitoring device 104 is assumed to be a glucose monitoring device, analyte monitoring device 104 may operate to monitor one or more additional or alternative analytes. The term “analyte” as used herein is a broad term, and is to be given 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), and refers without limitation to a substance or chemical constituent in the body or a biological sample (e.g., bodily fluids, including, blood, serum, plasma, interstitial fluid, cerebral spinal fluid, lymph fluid, ocular fluid, saliva, oral fluid, urine, excretions, or exudates). Analytes can include naturally occurring substances, artificial substances, metabolites, and/or reaction products. In some embodiments, the analyte for measurement by the sensing regions, devices, and methods is albumin, alkaline phosphatase, alanine transaminase, aspartate aminotransferase, bilirubin, blood urea nitrogen, calcium, CO2, chloride, creatinine, glucose, gamma-glutamyl transpeptidase, hematocrit, lactate, lactate dehydrogenase, magnesium, oxygen, pH, phosphorus, potassium, sodium, total protein, uric acid, metabolic markers, and drugs.
Other analytes are contemplated as well, including but not limited to acetaminophen, dopamine, ephedrine, terbutaline, ascorbate, uric acid, oxygen, d-amino acid oxidase, plasma amine oxidase, xanthine oxidase, NADPH oxidase, alcohol oxidase, alcohol dehydrogenase, pyruvate dehydrogenase, diols, Ros, NO, bilirubin, cholesterol, triglycerides, gentisic acid, ibuprophen, L-Dopa, methyl dopa, salicylates, tetracycline, tolazamide, tolbutamide, acarboxyprothrombin; acylcarnitine; adenine phosphoribosyl transferase; adenosine deaminase; albumin; alpha-fetoprotein; amino acid profiles (arginine (Krebs cycle), histidine/urocanic acid, homocysteine, phenylalanine/tyrosine, tryptophan); andrenostenedione; antipyrine; arabinitol enantiomers; arginase; benzoylecgonine (cocaine); biotinidase; biopterin; c-reactive protein; carnitine; carnosinase; CD4; ceruloplasmin; chenodeoxycholic acid; chloroquine; cholesterol; cholinesterase; conjugated 1-β hydroxy-cholic acid; cortisol; creatine kinase; creatine kinase MM isoenzyme; cyclosporin A; d-penicillamine; de-ethylchloroquine; dehydroepiandrosterone sulfate; DNA (acetylator polymorphism, alcohol dehydrogenase, alpha 1-antitrypsin, cystic fibrosis, Duchenne/Becker muscular dystrophy, glucose-6-phosphate dehydrogenase, hemoglobin A, hemoglobin S, hemoglobin C, hemoglobin D, hemoglobin E, hemoglobin F, D-Punjab, beta-thalassemia, hepatitis B virus, HCMV, HIV-1, HTLV-1, Leber hereditary optic neuropathy, MCAD, RNA, PKU, Plasmodium vivax, sexual differentiation, 21-deoxycortisol); desbutylhalofantrine; dihydropteridine reductase; diptheria/tetanus antitoxin; erythrocyte arginase; erythrocyte protoporphyrin; esterase D; fatty acids/acylglycines; free (3-human chorionic gonadotropin; free erythrocyte porphyrin; free thyroxine (FT4); free tri-iodothyronine (FT3); fumarylacetocetase; galactose/gal-1-phosphate; galactose-1-phosphate uridyltransferase; gentamicin; glucose-6-phosphate dehydrogenase; glutathione; glutathione perioxidase; glycocholic acid; glycosylated hemoglobin; halofantrine; hemoglobin variants; hexosaminidase A; human erythrocyte carbonic anhydrase I; 17-alpha-hydroxyprogesterone; hypoxanthine phosphoribosyl transferase; immunoreactive trypsin; lactate; lead; lipoproteins ((a), B/A-1, β); lysozyme; mefloquine; netilmicin; phenobarbitone; phenyloin; phytanic/pristanic acid; progesterone; prolactin; prolidase; purine nucleoside phosphorylase; quinine; reverse tri-iodothyronine (rT3); selenium; serum pancreatic lipase; sissomicin; somatomedin C; specific antibodies (adenovirus, anti-nuclear antibody, anti-zeta antibody, arbovirus, Aujeszky's disease virus, dengue virus, Dracunculus medinensis, Echinococcus granulosus, Entamoeba histolytica, enterovirus, Giardia duodenalisa, Helicobacter pylori, hepatitis B virus, herpes virus, HIV-1, IgE (atopic disease), influenza virus, Leishmania donovani, leptospira, measles/mumps/rubella, Mycobacterium leprae, Mycoplasma pneumoniae, Myoglobin, Onchocerca volvulus, parainfluenza virus, Plasmodium falciparum, poliovirus, Pseudomonas aeruginosa, respiratory syncytial virus, rickettsia (scrub typhus), Schistosoma mansoni, Toxoplasma gondii, Trepenoma pallidium, Trypanosoma cruzi/rangeli, vesicular stomatis virus, Wuchereria bancrofti, yellow fever virus); specific antigens (hepatitis B virus, HIV-1); succinylacetone; sulfadoxine; theophylline; thyrotropin (TSH); thyroxine (T4); thyroxine-binding globulin; trace elements; transferrin; UDP-galactose-4-epimerase; urea; uroporphyrinogen I synthase; vitamin A; white blood cells; and zinc protoporphyrin. Salts, sugar, protein, fat, vitamins, and hormones naturally occurring in blood or interstitial fluids can also constitute analytes in certain embodiments.
The analyte can be naturally present in the biological fluid, for example, a metabolic product, a hormone, an antigen, an antibody, and the like. Alternatively, the analyte can be introduced into the body, for example, a contrast agent for imaging, a radioisotope, a chemical agent, a fluorocarbon-based synthetic blood, or a drug or pharmaceutical composition, including but not limited to insulin; ethanol; cannabis (marijuana, tetrahydrocannabinol, hashish); inhalants (nitrous oxide, amyl nitrite, butyl nitrite, chlorohydrocarbons, hydrocarbons); cocaine (crack cocaine); stimulants (amphetamines, methamphetamines, Ritalin, Cylert, Preludin, Didrex, PreState, Voranil, Sandrex, Plegine); depressants (barbituates, methaqualone, tranquilizers such as Valium, Librium, Miltown, Serax, Equanil, Tranxene); hallucinogens (phencyclidine, lysergic acid, mescaline, peyote, psilocybin); narcotics (heroin, codeine, morphine, opium, meperidine, Percocet, Percodan, Tussionex, Fentanyl, Darvon, Talwin, Lomotil); designer drugs (analogs of fentanyl, meperidine, amphetamines, methamphetamines, and phencyclidine, for example, Ecstasy); anabolic steroids; and nicotine. The metabolic products of drugs and pharmaceutical compositions are also contemplated analytes. Analytes such as neurochemicals and other chemicals generated within the body can also be analyzed, such as, for example, ascorbic acid, uric acid, dopamine, noradrenaline, 3-methoxytyramine (3MT), 3,4-dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA), 5-hydroxytryptamine (5HT), histamine, Advanced Glycation End Products (AGEs) and 5-hydroxyindoleacetic acid (FHIAA).
Although illustrated as a wearable device (e.g., a smart watch), the computing device 108 may be configured in a variety of ways without departing from the spirit or scope of the described techniques. By way of example and not limitation, the computing device 108 may be configured as a different type of mobile device (e.g., a mobile phone, tablet device, smart ring, or other wearable device, the computing device 108 may be configured as a dedicated device associated with the health monitoring platform 112, e.g., with functionality to obtain the analyte measurements 118 from the analyte monitoring device 104, perform various computations in relation to the analyte measurements 118, display information related to the analyte measurements 118 and the health monitoring platform 112, communicate the analyte measurements 118 to the health monitoring platform 112, and so forth. In contrast to implementations where the computing device 108 is configured as a mobile phone, however, the computing device 108 may not include some functionality available with mobile phone or wearable configurations when configured as a dedicated analyte monitoring device, such as the ability to make phone calls, camera functionality, the ability to utilize social networking applications, and so on.
Additionally, the computing device 108 may be representative of more than one device in accordance with the described techniques. In one or more scenarios, for instance, the computing device 108 may correspond to both a wearable device (e.g., a smart watch) and a mobile phone. In such scenarios, both of these devices may be capable of performing at least some of the same operations, such as to receive the analyte measurements 118 from the analyte monitoring device 104, communicate them via the network 116 to the health monitoring platform 112, display information related to the analyte measurements 118, and so forth.
Alternately or in addition, different devices may have different capabilities that other devices do not have or that are limited through computing instructions to specified devices. In the scenario where the computing device 108 corresponds to a separate smart watch and a mobile phone, for instance, the smart watch may be configured with various sensors and functionality to measure a variety of physiological markers (e.g., heartrate, breathing, rate of blood flow, and so on) and activities (e.g., steps, movement, etc.) of the person 102. In this scenario, the mobile phone may not be configured with these sensors and functionality or may include a limited amount of that functionality—although in other scenarios a mobile phone may be able to provide the same functionality. Continuing with this particular scenario, the mobile phone may have capabilities that the smart watch does not have, such as an amount of computing resources (e.g., battery and processing speed) that enables the mobile phone to more efficiently carry out computations in relation to the analyte measurements 118. Even in scenarios where a smart watch is capable of carrying out such computations, computing instructions may limit performance of those computations to the mobile phone so as not to burden both devices and to utilize available resources efficiently. To this extent, the computing device 108 may be configured in different way and represent different numbers of devices than discussed herein without departing from the spirit and scope of the described techniques.
As mentioned above, the computing device 108 communicates the analyte measurements 118 to the health monitoring platform 112. In the illustrated environment 100, the analyte measurements 118 are shown stored in storage device 122 of the health monitoring platform 112 as part of analyte data 120. The storage device 122 may represent one or more databases and also other type of storage capable of storing the analyte data 120. The analyte data 120 also includes user profile 124. In accordance with the described techniques, the person 102 corresponds to a user of at least the health monitoring platform 112 and may also be a user of one or more other, third-party service providers. To this end, the person 102 may be associated with a username and be required, at some time, to provide authentication information (e.g., password, biometric data, and so forth) to access the health monitoring platform 112 using the username. This information may be captured in the user profile 124. The user profile 124 may also include a variety of other information about the user, such as demographic information describing the person 102, information about a health care provider, payment information, prescription information, determined health indicators, user preferences, account information for other service provider systems (e.g., a service provider associated with a wearable, social networking systems, and so on), and so forth. The user profile 124 may include different information about a user within the spirit and scope of the described techniques.
Further, the analyte data 120 not only represents data of a user that corresponds to the person 102, but also represents data of the other users in the user population 110. Given this, the analyte measurements 118 in the storage device 122 include the analyte measurements from an analyte sensor of the analyte monitoring device 104 worn by the person 102 and also include analyte measurements from analyte sensors of analyte systems worn by persons corresponding to the other users in the user population 110. It follows also that the analyte measurements 118 of these other users are communicated by their respective devices via the network 116 to the health monitoring platform 112 and that these other users have respective user profiles 124 with the health monitoring platform 112.
The data analytics platform 126 represents functionality to process the analyte data 120 to generate a variety of predictions, such as by using various machine learning models. Based on these predictions, the health monitoring platform 112 may provide a recommended action. For instance, the health monitoring platform 112 may provide the decision support output directly to the user, to a medical professional associated with the user, and so forth. Although depicted as separate from the computing device 108, portions or an entirety of the data analytics platform 126 may alternately or additionally be implemented at the computing device 108. The data analytics platform 126 is also configured to generate these predictions using data in addition to the analyte measurements 118, such as additional data obtained via the IoT 114.
It is to be appreciated that the IoT 114 represents various sources capable of providing data that describes the person 102 and the person 102's activity as a user of one or more service providers and activity with the real world. By way of example, the IoT 114 may include various devices of the user, e.g., cameras, mobile phones, laptops, and so forth. To this end, the IoT 114 may provide information about interaction of the user with various devices, e.g., interaction with web-based applications, photos taken, communications with other users, and so forth. The IoT 114 may also include various real-world articles (e.g., shoes, clothing, sporting equipment, appliances, automobiles, etc.) configured with sensors to provide information describing behavior, such as steps taken, force of a foot striking the ground, length of stride, temperature of a user (and other physiological measurements), temperature of a user's environment, types of food stored in a refrigerator, types of food removed from a refrigerator, driving habits, and so forth.
The IoT 114 may also include third parties to the health monitoring platform 112, such as medical providers (e.g., a medical provider of the person 102) and manufacturers (e.g., a manufacturer of the analyte monitoring device 104, the medicament delivery system 106, or the computing device 108) capable of providing medical and manufacturing data, respectively, that can be leveraged by the data analytics platform 126. Certainly, the IoT 114 may include devices and sensors capable of providing a wealth of data in connection with recommendations based on analyte monitoring (e.g., continuous glucose monitoring) without departing from the spirit or scope of the described techniques. In the context of measuring an analyte, e.g., continuously, and obtaining data describing such measurements, consider the following discussion of
In this example 200, the analyte monitoring device 104 is illustrated to include an analyte sensor 202 (e.g., a glucose sensor) and a sensor module 204. Here, the analyte sensor 202 is depicted in the side view having been inserted subcutaneously into skin 206, e.g., of the person 102. The sensor module 204 is approximated in the top view as a dashed rectangle. The analyte monitoring device 104 also includes a transmitter 208 in the illustrated example 200. Use of the dashed rectangle for the sensor module 204 indicates that it may be housed or otherwise implemented within a housing of the transmitter 208. Antennae and/or other hardware used to enable the transmitter 208 to produce signals for communicating data, e.g., over a wireless connection to the computing device 108, may also be housed or otherwise implemented within the housing of the transmitter 208. In this example 200, the analyte monitoring device 104 further includes adhesive pad 210.
In operation, the analyte sensor 202 and the adhesive pad 210 may be assembled to form an application assembly, where the application assembly is configured to be applied to the skin 206 so that the analyte sensor 202 is subcutaneously inserted as depicted. In such scenarios, the transmitter 208 may be attached to the assembly after application to the skin 206 via an attachment mechanism (not shown). Alternatively, the transmitter 208 may be incorporated as part of the application assembly, such that the analyte sensor 202, the adhesive pad 210, and the transmitter 208 (with the sensor module 204) can all be applied substantially at once to the skin 206. In one or more implementations, this application assembly is applied to the skin 206 using a separate sensor applicator (not shown). Unlike the finger sticks required by conventional blood glucose meters, user-initiated application of the analyte monitoring device 104 with a sensor applicator is nearly painless and does not require the withdrawal of blood. Moreover, the automatic sensor applicator generally enables the person 102 to embed the analyte sensor 202 subcutaneously into the skin 206 without the assistance of a clinician or healthcare provider.
The analyte monitoring device 104 may also be removed by peeling the adhesive pad 210 from the skin 206. It is to be appreciated that the analyte monitoring device 104 and its various components as illustrated are simply one example form factor, and the analyte monitoring device 104 and its components may have different form factors without departing from the spirit or scope of the described techniques.
In operation, the analyte sensor 202 is communicably coupled to the sensor module 204 via at least one communication channel which can be a wireless connection or a wired connection. Communications from the analyte sensor 202 to the sensor module 204 or from the sensor module 204 to the analyte sensor 202 can be implemented actively or passively and these communications can be continuous (e.g., analog) or discrete (e.g., digital).
The analyte sensor 202 may be a device, a molecule, and/or a chemical which changes or causes a change in response to an event which is at least partially independent of the analyte sensor 202. The sensor module 204 is implemented to receive indications of changes to the analyte sensor 202 or caused by the analyte sensor 202. For example, the analyte sensor 202 can include glucose oxidase which reacts with glucose and oxygen to form hydrogen peroxide that is electrochemically detectable by the sensor module 204 which may include an electrode. In this example, the analyte sensor 202 may be configured as or include a glucose sensor configured to detect analytes in blood or interstitial fluid that are indicative of glucose level using one or more measurement techniques. In one or more implementations, the analyte sensor 202 may also be configured to detect analytes in the blood or the interstitial fluid that are indicative of other markers, such as lactate levels, ketones, or ionic potassium, which may improve accuracy in identifying or predicting glucose-based events. Additionally or alternatively, the analyte monitoring device 104 may include additional sensors and/or architectures to the analyte sensor 202 to detect those analytes indicative of the other markers.
In another example, the analyte sensor 202 (or an additional sensor of the analyte monitoring device 104 — not shown) can include a first and second electrical conductor and the sensor module 204 can electrically detect changes in electric potential across the first and second electrical conductor of the analyte sensor 202. In this example, the sensor module 204 and the analyte sensor 202 are configured as a thermocouple such that the changes in electric potential correspond to temperature changes. In some examples, the sensor module 204 and the analyte sensor 202 are configured to detect a single analyte, e.g., glucose. In other examples, the sensor module 204 and the analyte sensor 202 are configured to use diverse sensing modes to detect multiple analytes, e.g., ionic sodium, ionic potassium, carbon dioxide, and glucose. Alternatively or additionally, the analyte monitoring device 104 includes multiple sensors to detect not only one or more analytes (e.g., ionic sodium, ionic potassium, carbon dioxide, glucose, and insulin) but also one or more environmental conditions (e.g., temperature). Thus, the sensor module 204 and the analyte sensor 202 (as well as any additional sensors) may detect the presence of one or more analytes, the absence of one or more analytes, and/or changes in one or more environmental conditions. As noted above, the analyte monitoring device 104 may be configured to produce data describing a single analyte (e.g., glucose) or multiple analytes. Further, a combination of the analytes for which analyte monitoring devices are configured may vary across different lots of the monitoring devices manufactured (e.g., by the health monitoring platform 112), such that analyte monitoring devices having different architectures may be configured for use by different patient populations and/or for different health conditions.
In one or more implementations, the sensor module 204 may include a processor and memory (not shown). The sensor module 204, by leveraging the processor, may generate the analyte measurements 118 based on the communications with the analyte sensor 202 that are indicative of the above-discussed changes. Based on the above-noted communications from the analyte sensor 202, the sensor module 204 is further configured to generate communicable packages of data that include at least one analyte measurement 118. In this example 200, the analyte data 116 represents these packages of data. Additionally or alternatively, the sensor module 204 may configure the analyte data 116 to include additional data, including, by way of example, supplemental sensor information 212. The supplemental sensor information 212 may include a sensor identifier, a sensor status, temperatures that correspond to the analyte measurements 118, measurements of other analytes that correspond to the analyte measurements 118, and so forth. It is to be appreciated that supplemental sensor information 212 may include a variety of data that supplements at least one analyte measurement 118 without departing from the spirit or scope of the described techniques.
In implementations where the analyte monitoring device 104 is configured for wireless transmission, the transmitter 208 may transmit the analyte data 116 as a stream of data to a computing device. Alternatively or additionally, the sensor module 204 may buffer the analyte measurements 118 and/or the supplemental sensor information 212 (e.g., in memory of the sensor module 204 and/or other physical computer-readable storage media of the analyte monitoring device 104) and cause the transmitter 208 to transmit the buffered analyte data 120 later at various regular or irregular intervals, e.g., time intervals (approximately every second, approximately every thirty seconds, approximately every minute, approximately every five minutes, approximately every hour, and so on), storage intervals (when the buffered analyte measurements 118 and/or supplemental sensor information 212 reach a threshold amount of data or a number of measurements), when requested by another device, and so forth.
Having considered an example environment and example analyte monitoring system, consider now a discussion of some example details of the techniques in accordance with one or more implementations.
At block 302, in some embodiments, the process 300 includes performing a psychographic data collection phase. The data collection phase includes, a health monitoring platform (e.g., health monitoring platform 112) collecting psychographic data from users such as by providing users with a digital survey or quiz asking them to identify their goals, interests, and other types of psychographic data. In such an example, users may be provided with various user interfaces (“UIs”) on their mobile health applications that ask for psychographic data. For example, a UI may be presented to the user asking what the user's goal is and listing various goals (e.g., lower A1C, spend more time in range, feel better, lower stress, better sleep, more energy) that the user may select from. Other types of data (e.g., non-psychographic data) about the user may be collected as part of block 302 or separately. For example, the health monitoring platform may separately collect various other information about the users such as, but not limited to, device information, account information, demographics information, food consumption information, activity information (e.g., sleep and exercise), patient status information, medication (e.g., insulin) information, information from connected sensors such as blood glucose.
In some embodiments, the data collection phase (e.g., including psychographic data collection or collection of other types of data) may be continuous and ongoing or based on triggering events. For example, the psychographic data collection phase may be a progressive reveal where the user's mobile health application requests information from the user as the user uses the application. Other types of data, such as sensor data, may continuously be collected. In another example, the data collection phase may be triggered when the user's mobile health application determines that a confidence score for the data, e.g., psychographic data, that is available or inferred, is low. In such examples, the mobile health application may explicitly request user input.
As described above, not all users may respond to requests for psychographic data. For example, surveys, quizzes, or other types of request for psychographic data may be sent to 100,000 users. However, only 20,000 users may respond with responses to all psychographic related questions. As such, certain users may have incomplete psychographic profiles that are only partially complete or not complete at all. Therefore, to infer or predict psychographic data for the remaining 80,000 users, a plurality of imputation models may be trained, each configured to infer or predict a different psychographic feature, such as a goal, interest, behavior, concern, etc. Note that a psychographic profile is a type of contextual profile (e.g., a user profile comprising contextual information about the user, which may also include psychographic data).
At block 302, the process 300 includes training imputation models using a training dataset that includes the psychographic data, and/or other types of contextual data, collected during block 301. For example, a training dataset may be prepared using the data collected from the 20,000 users, the training dataset include all sorts of demographic, physiological, and other types of relevant data, as well as psychographic data. Using that training dataset, a plurality of imputation models may be trained, each configured to infer or predict a value for a different psychographic feature, such as a goal, interest, behavior, concern, etc. Depending on what an imputation model is configured to predict, the training dataset may be labeled differently. For example, if an imputation model is to be trained to predict a diabetic distress score, the training dataset is labeled with diabetic distress scores. The imputation models may be trained using one of various artificial intelligence/machine learning algorithms, such as supervised learning algorithms. In certain embodiments, the imputation models may be trained as classification, regression, or other types of models.
At block 303, the process 300 includes developing psychographic data, and/or other types of contextual data, for one or more users using the trained imputation models. Continuing with the example above, the trained imputation models may be used to develop psychographic profiles for the 80,000 remaining users whose psychographics profiles were incomplete. For example, an imputation model that is trained to predict a diabetic distress score for users is used to output predicted diabetic distress scores for the 80,000 remaining users. In another example, an imputation model that is trained to predict a hyperglycemia fear score for users is used to output predicted hyperglycemia fear scores for the 80,000 remaining users. In yet another example, an imputation model that is trained to predict the type of exercise users are interested in is used to output the type of exercise each of the 80,000 remaining users is interested in. Similarly, an imputation model that is trained to predict the type of glucose control goal users are interested in is used to output a predicted glucose control goal for each of the 80,000 remaining users.
The psychographic data that is imputed for the remaining 80,000 users is then used to build a training dataset that can be used to train the contextual MAB models utilized in block 306 for determining user interaction policies.
At block 306, the process 300 may include performing an exploration-exploitation phase for determining optimal user interaction policies. The exploration-exploitation phase may implemented using contextual MAB models and involve continuous exploration on some fraction of the user population to refine the contextual MAB models. The exploration-exploitation phase includes individualized or personalized contextual models to determine user interaction policies. In certain embodiments, the exploration-exploitation phase also includes utilizing observed outcomes to train and/or update the contextual MAB models and/or the imputation models discussed above. For example, the health monitoring platform may use feedback telemetry associated with the user interaction action policies provided to users to maximize user engagement and/or optimize health outcomes resulting from the user interaction policies.
As described below in relation to
At block 308, the process 300 may also include the health monitoring platform retraining the contextual and/or imputation models. In some embodiments, the health monitoring platform may determine to continuously train the models. In some embodiments, the health monitoring platform may determine to retrain models in response to a trigger. For example, the health monitoring platform may determine to retrain periodically (e.g., every month, regardless of performance). In some embodiments, the process 300 may determine to retrain models every time the sample size of the exploration set of users' user interaction policies and observed outcomes hits a threshold number (e.g. 10,000). In certain embodiments, the health monitoring platform retrains the imputation and/or the contextual MAB models using user feedback telemetry and/or measured outcomes/rewards associated with the user interaction policies.
Below, blocks 304 and 306 are described in more detail and by reference to subsequent
As described above in reference to
At block 406, psychographic data is collected for one or more users. For example, psychographic data may be collected using a quiz provided to each user, where the questions may be provided to a user via various UIs, as further described below. In some embodiments, each user's engagement with a UI may also provide psychographic data. For example, if the user engages with content (e.g., video, audio, text) that includes a particular subject matter, that subject matter may be used to collect psychographic data of the user, which can be indicative of the user's interest in such content. In certain embodiments, any metadata (e.g., user's search terms, time spent on a particular UI element, etc.) may be used to collect psychographic data of the user. As described above, psychographic data may be collected for a subset of the one or more users while remaining users may be unresponsive or provide incomplete information.
Once psychographic data is collected for the subset of users, at block 408, the psychographic data may be provided for constructing psychographic profiles (i.e., psychographic segmentations). For example, if a user has provided answers to a quiz, then the answers may be used for generating a psychographic profile including separating the user's psychographic data into one or more data entries that influence the user's behaviors. For example, a survey instrument (e.g. quiz) may include a series of questions that users can answer to determine a user's diabetes distress score. In another example, a survey instrument may be provided to a user regarding their fear of hyperglycemia or psychological well-being, etc. By issuing one or more surveys, if the data is available, then the answers for a user may be used to generate the training data set that then becomes input to subsequent contextual MAB models, as further described below.
For the remaining users for whom psychographic data is not available or is incomplete, the psychographic data collected for the subset of responsive users may be used in a training dataset to build imputation models. Thus, for users without psychographic data, the health monitoring platform may predict their psychographic data. For example, at block 410, the health monitoring platform may execute the one or more imputation models to infer the psychographic data for the remaining users. There may be an imputation model corresponding to each particular feature (e.g., diabetes distress, fear of hyperglycemia, psychological well-being, etc.) in the psychographic profile of the users. In other words, for each of these features (i.e., diabetes distress, fear of hyperglycemia, psychological well-being), there may be an imputation model for predicting a corresponding feature value. Each imputation model may be trained to take, as input, a user's demographic, physiological information, and other types of relevant information to predict a feature value of a certain psychographic feature in the user's psychographic profile.
In certain embodiments, the imputation models may infer psychographic data (and/or other types of contextual data) based on the best data available. If the inferred psychographic data is low confidence, or if there's an opportune time, the health monitoring platform may ask the user for that information to validate or retrain the imputation models on a regular basis. For example, for users that have indicated type 2 diabetes, the imputation models may infer that their goal is better sleep. The confidence level for the predicted goal of better sleep for this user, however, may be low. Therefore, the health monitoring platform may ask the user to validate whether sleeping better is a goal for the user. In some embodiments, even if the confidence level for the predicted goal of better sleep is high, the health monitoring platform may still confirm it with the user.
The imputation models may also be trained and retrained using user feedback telemetry from the exploration-exploitation phase 306, as shown in block 412. For example, after providing a user interaction policy (e.g., insight, recommendation, etc.) to a user, the user may be provided with a feedback element (e.g., a thumbs up, thumbs down, one through five star, etc.), and the user's feedback telemetry may be used to refine the imputation models. As an example, an imputation model may predict an interest for a user, such as running, and the exercise recommendations to the user may constantly indicate that the user should engage in running. However, the user may be interested in weightlifting instead. As such, user feedback telemetry may indicate that the user is not interested in running and, therefore used for retraining the imputation model.
As described above in reference to
At block 422, the exploration-exploitation phase 306 may include dividing a set of users into an exploration subset and an exploitation subset. As illustrated in block 424, the set of users are divided into an exploration subset and an exploitation subset using a sampling criteria. For example, the set of users may be divided where a ε fraction of the users are randomly selected and used to identify the exploration subset, and the remaining (1−ε) fraction of the users are used to develop an exploitation subset. In some embodiments, the magnitude of the exploration ratio ε decreases after each iteration of the exploration-exploitation phase. In some embodiments, the magnitude of the exploration ratio ε is set to zero after a predefined number of iterations of the exploration-exploitation phase. In certain embodiments, the magnitude of the decrease in exploration ratio ε may be a function of the user interaction policy accuracy (which may improve over time). In addition, the exploration ratio c may also have a lower bound >0, to ensure some degree of exploration is always performed since what works best for users may change over time.
For the exploration subset of users, at block 426, the health monitoring platform may randomly assign a user interaction policy (e.g., “random” action). At block 428, the health monitoring platform may measure outcomes that may be used to train/retrain the contextual MAB model at block 430. In some embodiments, user feedback, at block 434, may also be received and used to train/retrain the contextual MAB and/or imputation models at block 410.
For the exploitation subset of users, at block 430, the exploration-exploitation phase may include executing contextual models using contextual data (e.g., including psychographic data) for each user or user cohort to determine user interaction policies. For example, one of the many contextual models may be a decision support model that is a user-facing algorithm that determines user interaction policy, such as a decision support recommendation, for a user based on the user's contextual data.
At block 432, user interaction policies may be provided for each user of the exploitation subset of users. At block 434, the health monitoring platform may receive user feedback telemetry indicate of the effectiveness of the corresponding user interaction policy, for example, on user engagement, health, etc. For example, when the user interaction policy recommends a video, the user's engagement with the video may indicate positive user feedback telemetry. However, if the user does not engage with the video, then the user feedback telemetry may be determined to be negative. In another example, the recommend video may also include rating element (e.g., a thumbs up or thumbs down element) allowing the user to provide direct indication of positive or negative user feedback telemetry. In a further example, the user interaction policy may recommend the user to perform some action (e.g., walk 5,000 steps today). The user may perform the action thereby indicating a positive user feedback telemetry or the user may decline to perform the action thereby indicating a negative user feedback telemetry.
In some embodiments, after implementing a user interaction policy (e.g., watching the recommended video or performing a recommended action), outcomes (e.g., user metrics, such as physiological metrics (e.g., glucose trends)) may be observed at block 428 to measure if the user's metrics have improved or deteriorated. For example, if the user has been provided a recommendation, but the user's metrics have deteriorated, then the measured outcomes may indicate a low reward. In another example, if the user has been provided a recommendation, but the user's metrics have not improved or have not improved enough, then the measured outcomes may indicate a low reward. As described above, user feedback telemetry
Implementations and examples of CMAB models are further defined in U.S. application Ser. No. 17/931,531, which is incorporated herein by reference. In addition, implementations and examples of providing decision support recommendation using recommender models based on users' goals and interests are further defined in U.S. application Ser. No. 17/241919, which is incorporated herein by reference.
Some other UIs may be used to collect data. As described further above, the present embodiments may include a data collection phase (and other data collection steps) to collect psychographic data and/or other types of data from users. Such data may be (1) used to train imputation models for inferring psychographic data (or other types of data) for users for whom psychographic data is not available (e.g., new users or users who were not responsive to data collection efforts) and (2) used as contextual data to provide optimal user interaction policies to users. Therefore, for example, a user may provide psychographic data directly via one or more UIs, as further described below.
The exemplary UIs illustrated in
In implementations, such as those depicted in
In implementations, such as those depicted in
Additionally, the implementations depicted in
In implementations, such as those depicted in
In implementations, such as those depicted in
In implementations, such as those depicted in
In one or more implementations, the mobile health application enables a user to send other users, e.g., family and friends, praise based on their monitored analyte data. The mobile health application enables supporting users to follow a particular user's analyte data and to see daily highlights with explanations of what went well that day for the particular user. Supporting users can also receive real-time notifications a user the supporting user follows is experiencing a good glucose moment. The mobile health application allows supporting users to quickly send a message congratulating a followed user to show your support the followed user in the right moments.
In implementations, such as those depicted in
In implementations, such as those depicted in
After selection, the user may be provided with personalization UI interface 2510 that displays decision support outputs (e.g., tips and resources) curated for the user based on the user's selected goal as well other data associated with the user's goal. For example, personalization UI 2510 includes a first decision support output 25122 (i.e., “Low-carb recipes”) and a second decision support output 2514 (i.e., “Hear from others who share the same goal”). In certain embodiments, the tips and resources may be curated based on other user data. In one or more implementations, the mobile health application acknowledges when the user achieves a goal, e.g., by presenting a congratulatory message via the user interface.
Personalization Uls may enable a user to update and/or provide additional data related to their goals. For example, personalization UI 2520 displays elements that enable a user to take a subsequent quiz at a subsequent time, where the subsequent quiz enables the user selection one or more health condition related goals. In one or more implementations, the application prompts the user to take the subsequent quiz. The personalization UI 2520 includes a quiz element 2522 that enables the user to take a quiz to update one or more health condition (e.g., diabetes) related goals. In the depicted example, the user use a selection element 2524 to select “Better sleep” as an updated goal.
After selection, the user may be provided with personalization UI interface 2530 that displays decision support outputs (e.g., tips and resources) curated for the user based on the user's updated goal, data associated with the user's goal, or other user data. For example, personalization UI 2530 includes a first decision support output 2534 (i.e., “Preparing for sleep”) and a second decision support output 2536 (i.e., “Hear from others who share the same goal”).
In implementations, such as those depicted in
In implementations, such as those depicted in
In one or more implementations, the mobile health application outputs notifications when it is determined that at least one item of the digital content might provide helpful insight to the user, e.g., based on the analyte data of the user over a time period and/or based on metrics (e.g., trends) derived from the analyte data over time.
In implementations, such as those depicted in
Having described example procedures in accordance with one or more implementations, consider now an example system and device that can be utilized to implement the various techniques described herein.
The example computing device 3902 as illustrated includes a processing system 3904, one or more computer-readable media 3906, and one or more I/O interfaces 3908 that are communicatively coupled, one to another. Although not shown, the computing device 3902 may further include a system bus or other data and command transfer system that couples the various components, one to another. A system bus can include any one or combination of different bus structures, such as a memory bus or memory controller, a peripheral bus, a universal serial bus, and/or a processor or local bus that utilizes any of a variety of bus architectures. A variety of other examples are also contemplated, such as control and data lines.
The processing system 3904 is representative of functionality to perform one or more operations using hardware. Accordingly, the processing system 3904 is illustrated as including hardware elements 3910 that may be configured as processors, functional blocks, and so forth. This may include implementation in hardware as an application specific integrated circuit or other logic device formed using one or more semiconductors. The hardware elements 3910 are not limited by the materials from which they are formed or the processing mechanisms employed therein. For example, processors may be comprised of semiconductor(s) and/or transistors (e.g., electronic integrated circuits (ICs)). In such a context, processor-executable instructions may be electronically-executable instructions.
The computer-readable media 3906 is illustrated as including memory/storage 3912. The memory/storage 3912 represents memory/storage capacity associated with one or more computer-readable media. The memory/storage component 3912 may include volatile media (such as random access memory (RAM)) and/or nonvolatile media (such as read only memory (ROM), Flash memory, optical disks, magnetic disks, and so forth). The memory/storage component 3912 may include fixed media (e.g., RAM, ROM, a fixed hard drive, and so on) as well as removable media (e.g., Flash memory, a removable hard drive, an optical disc, and so forth). The computer-readable media 3906 may be configured in a variety of other ways as further described below.
Input/output interface(s) 3908 are representative of functionality to allow a user to enter commands and information to computing device 3902, and also allow information to be presented to the user and/or other components or devices using various input/output devices. Examples of input devices include a keyboard, a cursor control device (e.g., a mouse), a microphone, a scanner, touch functionality (e.g., capacitive or other sensors that are configured to detect physical touch), a camera (e.g., which may employ visible or non-visible wavelengths such as infrared frequencies to recognize movement as gestures that do not involve touch), and so forth. Examples of output devices include a display device (e.g., a monitor or projector), speakers, a printer, a network card, tactile-response device, and so forth. Thus, the computing device 3902 may be configured in a variety of ways as further described below to support user interaction.
Various techniques may be described herein in the general context of software, hardware elements, or program modules. Generally, such modules include routines, programs, objects, elements, components, data structures, and so forth that perform particular tasks or implement particular abstract data types. The terms “module,” “functionality,” and “component” as used herein generally represent software, firmware, hardware, or a combination thereof. The features of the techniques described herein are platform-independent, meaning that the techniques may be implemented on a variety of commercial computing platforms having a variety of processors.
An implementation of the described modules and techniques may be stored on or transmitted across some form of computer-readable media. The computer-readable media may include a variety of media that may be accessed by the computing device 3902. By way of example, and not limitation, computer-readable media may include “computer-readable storage media” and “computer-readable signal media.”
“Computer-readable storage media” may refer to media and/or devices that enable persistent and/or non-transitory storage of information in contrast to mere signal transmission, carrier waves, or signals per se. Thus, computer-readable storage media refers to non-signal bearing media. The computer-readable storage media includes hardware such as volatile and non-volatile, removable and non-removable media and/or storage devices implemented in a method or technology suitable for storage of information such as computer readable instructions, data structures, program modules, logic elements/circuits, or other data. Examples of computer-readable storage media may include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical storage, hard disks, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or other storage device, tangible media, or article of manufacture suitable to store the desired information and which may be accessed by a computer.
“Computer-readable signal media” may refer to a signal-bearing medium that is configured to transmit instructions to the hardware of the computing device 3902, such as via a network. Signal media typically may embody computer readable instructions, data structures, program modules, or other data in a modulated data signal, such as carrier waves, data signals, or other transport mechanism. Signal media also include any information delivery media. The term “modulated data signal” means a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal. By way of example, and not limitation, communication media include wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, RF, infrared, and other wireless media.
As previously described, hardware elements 3910 and computer-readable media 3906 are representative of modules, programmable device logic and/or fixed device logic implemented in a hardware form that may be employed in some embodiments to implement at least some aspects of the techniques described herein, such as to perform one or more instructions. Hardware may include components of an integrated circuit or on-chip system, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a complex programmable logic device (CPLD), and other implementations in silicon or other hardware. In this context, hardware may operate as a processing device that performs program tasks defined by instructions and/or logic embodied by the hardware as well as a hardware utilized to store instructions for execution, e.g., the computer-readable storage media described previously.
Combinations of the foregoing may also be employed to implement various techniques described herein. Accordingly, software, hardware, or executable modules may be implemented as one or more instructions and/or logic embodied on some form of computer-readable storage media and/or by one or more hardware elements 3910. The computing device 3902 may be configured to implement particular instructions and/or functions corresponding to the software and/or hardware modules. Accordingly, implementation of a module that is executable by the computing device 3902 as software may be achieved at least partially in hardware, e.g., through use of computer-readable storage media and/or hardware elements 3910 of the processing system 3904. The instructions and/or functions may be executable/operable by one or more articles of manufacture (for example, one or more computing devices 3902 and/or processing systems 3904) to implement techniques, modules, and examples described herein.
The techniques described herein may be supported by various configurations of the computing device 3902 and are not limited to the specific examples of the techniques described herein. This functionality may also be implemented all or in part through use of a distributed system, such as over a “cloud” 3914 via a platform 3916 as described below.
The cloud 3914 includes and/or is representative of a platform 3916 for resources 3918. The platform 3916 abstracts underlying functionality of hardware (e.g., servers) and software resources of the cloud 3914. The resources 3918 may include applications and/or data that can be utilized while computer processing is executed on servers that are remote from the computing device 3902. Resources 3918 can also include services provided over the Internet and/or through a subscriber network, such as a cellular or Wi-Fi network.
The platform 3916 may abstract resources and functions to connect the computing device 3902 with other computing devices. The platform 3916 may also serve to abstract scaling of resources to provide a corresponding level of scale to encountered demand for the resources 3918 that are implemented via the platform 3916. Accordingly, in an interconnected device embodiment, implementation of functionality described herein may be distributed throughout the system 3900. For example, the functionality may be implemented in part on the computing device 3902 as well as via the platform 3916 that abstracts the functionality of the cloud 3914.
Embodiment 1: A non-transitory computer readable medium comprising instructions that, when executed by at least one processor, cause the at least one processor to perform a method including: dividing a plurality of users into an exploration subset of users and an exploitation subset of users; randomly assigning at least one user interaction policy to each of the exploration subset of users; and determining at least one user interaction policy for each of the exploitation subset of users using one or more contextual models trained using contextual data corresponding to the exploitation subset of users, wherein the contextual data corresponding to the exploitation subset of users comprises at least some of a first set of contextual profiles and a second set of contextual profiles.
Embodiment 2: the non-transitory computer readable medium of Embodiment 1, wherein the exploration-exploitation phase is further performed by: receiving user feedback telemetry from the exploitation subset of users, wherein the feedback telemetry provides information regarding effectiveness of the at least one user interaction policy assigned to each user of the exploitation subset of users.
Embodiment 3: The non-transitory computer readable medium of Embodiment 2, wherein at least one of the one or more imputation models or at least one of the contextual models is retrained using the user feedback telemetry.
Embodiment 4: The non-transitory computer readable medium of Embodiment 1, wherein the exploration-exploitation phase is further performed by: measuring outcomes associated with the exploitation subset of users, wherein the measured outcomes provide information regarding effectiveness of the at least one user interaction policy assigned to each user of the exploitation subset of users.
Embodiment 5: The non-transitory computer readable medium of Embodiment 4, wherein at least one of the contextual models is retrained using the measured outcomes.
Embodiment 6: The non-transitory computer readable medium of Embodiment 1, wherein the method further comprises: collecting contextual data for a first subset of the plurality of users; generating the first set of contextual profiles for the first subset of the plurality of users based on the collected contextual data; determining that contextual data for a second subset of the plurality of users is incomplete or not available; training one or more imputation models based on the contextual data for the first subset of the plurality of users to develop the contextual data for the second subset of the plurality of users; generating the contextual data for the second subset of the plurality of users using the one or more imputation models; and generating the second set of contextual profiles for the second subset of the plurality of users based on the generated contextual data for the second subset of the plurality of users.
Embodiment 7: The non-transitory computer readable medium of Embodiment 6, wherein: the contextual data for the first subset of the plurality of users corresponds to psychographic data for the first subset of the plurality of users; the first set of contextual profiles for the first subset of the plurality of users corresponds to a first set of psychographic profiles for the first subset of the plurality of users; the contextual data for the second subset of the plurality of users corresponds to psychographic data for the second subset of the plurality of users; and the second set of contextual profiles for the second subset of the plurality of users corresponds to a second set of psychographic profiles for the second subset of the plurality of users.
Embodiment 8: The non-transitory computer readable medium of Embodiment 6, wherein determining that contextual data for the second subset of the plurality of users is incomplete or not available comprises: requesting the contextual data from the second subset of the plurality of users; and determining that at least one of (1) some of the second subset of the plurality of users provided incomplete data in response to the requesting or (2) some of the second subset of the plurality of users provided no data in response to the requesting.
Embodiment 9: A non-transitory computer readable medium comprising instructions that, when executed by at least one processor, cause the at least one processor to perform a method including: collecting contextual data for a first subset of a plurality of users; generating a first set of contextual profiles for the first subset of the plurality of users based on the collected contextual data; determining that contextual data for a second subset of the plurality of users is incomplete or not available; training one or more imputation models based on the contextual data for the first subset of the plurality of users to develop the contextual data for the second subset of the plurality of users; generating the contextual data for the second subset of the plurality of users using the one or more imputation models; and generating the second set of contextual profiles for the second subset of the plurality of users based on the generated contextual data for the second subset of the plurality of users.
Embodiment 10: The non-transitory computer readable medium of Embodiment 9, wherein: the contextual data for the first subset of the plurality of users corresponds to psychographic data for the first subset of the plurality of users; the first set of contextual profiles for the first subset of the plurality of users corresponds to a first set of psychographic profiles for the first subset of the plurality of users; the contextual data for the second subset of the plurality of users corresponds to psychographic data for the second subset of the plurality of users; and the second set of contextual profiles for the second subset of the plurality of users corresponds to a second set of psychographic profiles for the second subset of the plurality of users.
Embodiment 11: The non-transitory computer readable medium of Embodiment 9, wherein determining that contextual data for the second subset of the plurality of users is incomplete or not available comprises: requesting the contextual data from the second subset of the plurality of users; and determining that at least one of (1) some of the second subset of the plurality of users provided incomplete data in response to the requesting or (2) some of the second subset of the plurality of users provided no data in response to the requesting.
Embodiment 12: The non-transitory computer readable medium of Embodiment 9, wherein the method further comprises: dividing the plurality of users into an exploration subset of users and an exploitation subset of users; randomly assigning at least one user interaction policy to each of the exploration subset of users; and determining at least one user interaction policy for each of the exploitation subset of users using one or more contextual models trained using contextual data corresponding to the exploitation subset of users, wherein the contextual data corresponding to the exploitation subset of users comprises at least some of the first set of contextual profiles and the second set of contextual profiles.
Embodiment 13: the non-transitory computer readable medium of Embodiment 12, wherein the exploration-exploitation phase is further performed by: receiving user feedback telemetry from the exploitation subset of users, wherein the feedback telemetry provides information regarding effectiveness of the at least one user interaction policy assigned to each user of the exploitation subset of users.
Embodiment 14: The non-transitory computer readable medium of Embodiment 13, wherein at least one of the one or more imputation models or at least one of the contextual models is retrained using the user feedback telemetry.
Embodiment 15: The non-transitory computer readable medium of Embodiment 12, wherein the exploration-exploitation phase is further performed by: measuring outcomes associated with the exploitation subset of users, wherein the measured outcomes provide information regarding effectiveness of the at least one user interaction policy assigned to each user of the exploitation subset of users.
Embodiment 16: The non-transitory computer readable medium of Embodiment 15, wherein at least one of the contextual models is retrained using the measured outcomes.
This application claims priority to U.S. Provisional Patent Application No. 63/289,376, filed on Dec. 14, 2021, the disclosure of which is incorporated herein by reference.
Number | Date | Country | |
---|---|---|---|
63289376 | Dec 2021 | US |