Patent Application
20230044000
This invention relates to an automated system, bot, and method to promote medication adherence and perform remote patient monitoring (RPM).
With advancing technology and increasing use of that technology, such as smart phones, an increasing number of users now access information and services via websites or downloaded client applications provided by respective service providers. Such remote communication provides numerous benefits to both the service providers and the end users as compared to in person or over the phone communications, including the ability to offer information and services to end users at any time of day and without the cost associated with providing a human representative. Increasingly, service providers employ a “virtual assistant” to act as an interface between end users and the information on the service provider site. The virtual assistant can be in the form of a “bot,” a software application that is programmed to do certain tasks without specific instructions from humans. Some virtual assistants or bots can embody a human representative of the service provider displayed on a website, client application, or the like, and can also include an interface (e.g., a text box) that allows users to input queries, and the service provider or a third party can identify the contents of the user's query and provide a response. These virtual assistants or bots act as an effective interface that allows users to seek information and services of interest while still allowing service providers to realize cost savings associated with providing information online rather than via a human representative.
Now, consider the healthcare industry. Patients afflicted with acute or chronic illnesses may not find a cure for the condition despite ongoing treatment. Some patients may receive medical treatment including medication prescribed by a physician to relieve symptoms or prevent the illness from worsening. Some patients may need acute or limited time treatment care after release from hospital inpatient treatment, while the patient is still recovering. Treatment of some illnesses may require medications that require critical adherence to time of intake because of a narrow therapeutic window to improve efficacy or avoid toxicity. Some patients may experience loss of short term memory (for example, forgetfulness in early dementia, in an elderly or Alzheimer patient), or physically debilitating conditions, and hence, may be dependent on homecare without the benefit of expensive daily licensed nursing services.
Some medications used to treat chronic or acute illness may be expensive. The therapeutic effect of some medicine prescribed to treat a patient's chronic or acute illness may be limited by the patient's adherence to the dosing protocol prescribed by the patient's doctor. Elderly patients simply taking multiple drugs or patients with debilitating mental or physical illness (for example, diseases like Alzheimer's, Parkinson's, dementia, or multiple sclerosis) may not remember or be physically able to take drugs on time from multiple bottles traditionally dispensed by pharmacies.
In various examples, a patient's adherence to the dosing protocol prescribed by the patient's doctor may be a crucial component of caring for the patient's illness. Lack of adequate medication adherence may result in preventable disease progression and unnecessary expense.
Furthermore, during a dosing protocol, certain side effects may arise they may be harmful to the patient or adversely affect the treatment. Also, the dosing protocol may not be effective and may need to be adjusted or titrated or ceased altogether. This may entail switching medications or other therapy, but may be difficult to analyze if the patient is remote.
In order to support targeted individuals on their healthcare journeys, focusing first on medication adherence and remote patient monitoring, an adherence bot is provided for use as a system with external devices, such as a smart cap, a hub, a blood pressure reader, or a scale, and can use Internet of Things (IoT) connectivity. A bot is a software application that is programmed to do certain tasks without specific instructions from humans. The adherence bot of the present invention is preferably automated and runs according to programmed instructions without a human user. The adherence bot will identify through analysis of aggregated data (from general population data and a targeted individual's own remote patient monitoring (RPM) devices and individual interactions) to assist the individual in maximizing his or her medication adherence. The bot will aid in identifying high risk individuals who would most likely need more coaching/assistance/interactions/touch points and tailor the interactions with the individual to help in improving and maintaining high levels of adherence/compliance. Additionally, the system can direct or focus caregivers (physicians, prescribers, pharmacists, healthcare entities) to provide more impactful attention to those individuals that are at risk of low adherence/compliance. Furthermore, the bot can determine whether the dosing protocol is effective and whether it may need to be adjusted or ceased altogether. This may entail titrating the medication, switching medications, or suggesting other therapy. After such adjustment, the adherence bot can continue performing monitoring and provide further recommendations for adjustment as necessary.
In one implementation, the adherence bot employs a conversation user interface to convey a representation of a conversation between the virtual healthcare assistant and the target individual (patient, user). The conversation UI presents a series of dialog representations, such as dialog bubbles, which include user-originated dialog representations associated with input from a user (verbal, textual, or otherwise) and device-originated dialog representations associated with response from the bot.
According to one aspect of the present invention, a system for remotely managing a medication regimen, includes a communication device usable by a target individual, the communication device communicating over a network; at least one peripheral device that obtains biometric information regarding the target individual, the at least one peripheral device communicating over the network to transmit the biometric information; a medication dispenser usable by the target individual, the medication dispenser communicating over the network to transmit information regarding the medication in the medication dispenser; and at least one processor communicating with the communication device, the at least one peripheral device, and the medication dispenser over the network. The at least one processor can execute computer-executable instructions to cause the communication device to enable a conversation interface associated with a virtual assistant, the virtual assistant providing at least one of information, queries, and directions to the target individual through the conversation interface; receive input from the target individual through the conversation interface; receive the information regarding the medication from the medication dispenser; receive the biometric information regarding the target individual from the at least one peripheral device; determine a status of management of the medication regimen based on at least one of the input from the target individual, the information regarding the medication from the medication dispenser, and the biometric information regarding the target individual from the at least one peripheral device; determine a response based at least in part on the determined status of management medication regimen; and transmit the response.
According to another aspect of the present invention, a method of remotely managing a medication regimen in a system including a communication device usable by a target individual, the communication device communicating over a network, at least one peripheral device that obtains biometric information regarding the target individual, the at least one peripheral device communicating over the network to transmit the biometric information, and a medication dispenser usable by the target individual, the medication dispenser communicating over the network to transmit information regarding the medication in the medication dispenser, includes causing the communication device to enable a conversation interface associated with a virtual assistant, the virtual assistant providing at least one of information, queries, and directions to the target individual through the conversation interface; receiving input from the target individual through the conversation interface; receiving the information regarding the medication from the medication dispenser; receiving the biometric information regarding the target individual from the at least one peripheral device; determining a status of management of the medication regimen based on at least one of the input from the target individual, the information regarding the medication from the medication dispenser, and the biometric information regarding the target individual from the at least one peripheral device; determining a response based at least in part on the determined status of management medication regimen; and transmitting the response.
According to yet another aspect of the present invention, a non-transitory, computer-readable medium stores computer-executable instructions that, when executed on one or more processors, cause the one or more processors to perform a method to remotely manage a medication regimen in a system including a communication device usable by a target individual, the communication device communicating over a network, at least one peripheral device that obtains biometric information regarding the target individual, the at least one peripheral device communicating over the network to transmit the biometric information, and a medication dispenser usable by the target individual, the medication dispenser communicating over the network to transmit information regarding the medication in the medication dispenser, the method including causing the communication device to enable a conversation interface associated with a virtual assistant, the virtual assistant providing at least one of information, queries, and directions to the target individual through the conversation interface; receiving input from the target individual through the conversation interface; receiving the information regarding the medication from the medication dispenser; receiving the biometric information regarding the target individual from the at least one peripheral device; determining a status of management of the medication regimen based on at least one of the input from the target individual, the information regarding the medication from the medication dispenser, and the biometric information regarding the target individual from the at least one peripheral device; determining a response based at least in part on the determined status of management medication regimen; and transmitting the response.
These and other aspects, objects, features, and advantages of the invention will become apparent from the following detailed description of illustrative embodiments thereof, which is to be read in connection with the accompanying drawings.
This disclosure describes a system, method, and bot for assisting targeted individuals (patients, users) with their healthcare. The techniques described herein provide for a personal virtual healthcare assistant or adherence bot that engages in dialogs with the targeted individual to help with medication adherence and other aspects of healthcare. To facilitate the exchanges between the targeted individual and adherence bot, a conversation user interface (UI) is provided to enable the targeted individual to intuitively understand his or her interactions with the adherence bot.
With the present invention, the adherence bot is enabled to:
While
The electronic device 104 renders a conversation user interface (UI) 110 that displays conversation with an adherence bot (virtual-assistant) service 116. The conversation UI 110 may be served from servers of the healthcare provider 106 or servers of the adherence bot service 116.
The conversation UI 110 engages the target individual 102 in a conversation that emulates human conversation. In some cases, the conversation UI 110 may include a virtual assistant that has a human-like personality and persona. The virtual assistant may include an avatar and the conversation UI 110 conveys a visual representation of a conversation between the target individual 102 and the avatar or adherence bot service 116. The conversation UI 110 presents a series of dialog representations 112, 114, such as graphical content bubbles, which are designated as representing dialog from either the target individual 102 or the adherence bot. In this illustration, the target individual-originated dialog representations 114 contain input from the target individual 102 (via text or otherwise) and the device- or avatar-originated dialog representations 112 contain responses from the device or adherence bot. The representations 112, 114 may be visually distinguished in the conversation UI 110 in any known manner to visually convey which entity is associated with the content. The conversation UI 110 may also include an interface area 118 that captures input from the target individual 102, including via typed input, audio, or speech input, as well as touch input and gesture input. Gesture or emotive input may be captured if the electronic device 104 is equipped with a camera or other sensor.
The target individual 102 may enter a query into the interface area 118 of the conversation UI 110. The electronic device 104 transmits this query over the network 108 to the adherence bot service 116. In response, the adherence bot service 116 may identify a response to provide to the target individual 102. The response may be added to a dialog representation of the conversation UI 110.
The adherence bot service 116 may comprise one or more computing devices (e.g., one or more servers) that include or otherwise have access to one or more processors 130, one or more network interfaces 132, and memory 134. The healthcare provider 106 may comprise one or more computing devices (e.g., one or more servers) that include or otherwise have access to one or more processors 136, one or more network interfaces 138, and memory 140, which stores or has access to medical records 142 of the target individual 102, medical research 144, nutrition information 146, insurance information 148, and/or general information 150.
The electronic device 104 of the target individual 102 may include or otherwise have access to one or more processors, one or more network interfaces, and memory, which stores a conversation application for rendering the UI 110 and a healthcare application for providing information from the healthcare provider 106 to the target individual 102. The client application may comprise a browser for rendering a site, a downloaded application provided by the healthcare provider 106, or any other client application configured to output content from the healthcare provider 106. While
The various memories 132, 140, and 156 store modules and data, and may include volatile and/or nonvolatile memory, removable and/or non-removable media, and the like, which may be implemented in any method or technology for storage of information, such as computer-readable instructions, data structures, program modules, or other data. Such memory includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage, or other magnetic storage devices, RAID storage systems, or any other tangible medium which can be used to store the desired information and which can be accessed by a computing device.
While
The conversation UI 110 emulates human-to-human interaction between the target individual 102 and the healthcare entity 106. The conversation UI 110 includes one or more adherence bot-originated dialog representations associated with the healthcare entity. The adherence bot image may be associated with the healthcare entity or as a personal digital assistant personalized for the target individual 102.
Each query sent to the healthcare entity 106 and/or the adherence bot service 116 may comprise the words and phrases within the string of text entered by the target individual 102, from which concepts may be derived. In some implementations, the concepts may be derived at least partly by the electronic device 104 through some natural language pre-preprocessing. In other implementations, the concepts may be derived as part of the adherence bot service 116 or a combination of the device and service.
A query sent to the healthcare entity 106 and/or the adherence bot service 116 may further comprise one or more pieces of context. The context may be based on any additional factors associated with the target individual 102, the electronic device 104, the peripheral device(s) 109, the medication dispenser 111, or the like. The context may include whether or not the target individual 102 is signed in with the healthcare entity 106, a health status of the target individual 102, an age of the target individual 102, a type of the peripheral device(s) 109 or medication dispenser 111 used by the target individual 102, or the like.
The memory 134 may store or otherwise have access to the conversation UI 110 and a response module 126. The response module 126 may include an expert system module 402, a device module 404, a knowledge base module 406, an algorithmic module 408, a behavior modeling module 410, a predictive analytics module 412, a user engagement module 414, and a feedback module 416. The expert system module 402 employs some combination of machine learning (context aware supervised/semi-supervised or unsupervised) and natural language processing. Suitable applications are available from Rasa and IBM, for example. The device module 404 can interface with devices to gather health/medication related activities. Suitable applications are available from RxCap and Omron. The knowledge base module 406 can interface with medication knowledge databases such as those provided by First DataBank and Epic. The algorithmic module 408 can function as a natural language generation platform, such as Wordsmith from Automated Insights. The behavioral modeling module 410 can tailor messages for specific individuals or groups. This module can combine a machine learning (context aware supervised/semi-supervised or unsupervised) agent with a human.
The predictive analytics module 412 can aid in determining preemptive coaching strategies to prevent reduction of compliance/adherence and may be a combination of any of the above modules. The predictive analytics module 412 can observe target individual activity and attempt to learn characteristics about the target individual that can be used as input to the response module 126. The predictive analytics module 412 may initially access a user profile database to find any preferences that the target individual may have provided. Then, over time, the predictive analytics module 412 may learn any number of characteristics about the target individual, such as health status (generally healthy or very sick), treatment regimens of the target individual (e.g., dialysis, chemotherapy, etc.), a current location of the target individual, insurance eligibility for particular procedures, and the like. In particular, predictive analytics module 412 can learn reasons for the target individual's lack of medication adherence.
The predictive analytics module 412 may also track patterns (e.g., the target individual has been losing weight over a certain time period, the target individual's blood pressure is higher when at work, etc.). Patterns may be stored and each of these observed behaviors, patterns, and navigation history may be useful to the response module 126 by providing additional context to the input of the target individual through the adherence bot 116. Such analysis can be performed through another module.
The user engagement module 414 can utilize other known third-party services to help the user have a more fulfilling experience. The feedback module 416 allows caregivers (and other healthcare entities) to provide feedback to the individual with connection to patient health records to provide a full 360-degree view to the primary care entity. For example, during a dosing protocol, certain side effects may arise they may be harmful to the patient or adversely affect the treatment. Also, the adherence bot may determine that the dosing protocol is not effective and may need to be adjusted or titrated or ceased altogether. This may entail switching medications or other therapy. This can be relayed to the healthcare entity 106, such as a physician, who can then decide whether to provide further instructions to the target individual 102, either in person, over another communication channel (e.g., in a telephone call), or t.
While
At 202, the healthcare entities 106 and/or the adherence bot service 116 causes display of the conversation UI on the electronic device 104. The conversation UI may be the sole graphics on a screen, or it may on or adjacent to other content.
At 204, and in response, the electronic device 104 renders the conversation UI 110. At 206, the electronic device 104 receives input from the target individual interacting with the conversation UI. The input may comprise a string of text, verbal input, or some other input (e.g., gesture, video images, etc.). At 208, the electronic device 104 provides the input to the healthcare entities 106 and/or the adherence bot service 116, which receives the input at 210.
At 212, the peripheral device(s) 109 provide acquired biometric information to the healthcare entities 106 and/or the adherence bot service 116, which receives the input at 214. At 216, the medication dispenser 111 provides dispenser data to the healthcare entities 106 and/or the adherence bot service 116, which receives the input at 218. At 220, the healthcare entities 106 and/or the adherence bot service 116 analyze the received information. That is, the healthcare entities 106 and/or the adherence bot service 116 may use language processing and machine learning techniques to identify queries, patterns, behaviors, anomalies, and other information from the received data. In some examples, the concept(s) of a query are determined at least partly with reference to one or more keywords expressed within the input. For instance, the concepts may be determined using relatively basic keyword matching in some instances. This matching can be improved with the adherence bot modules, so that specific words or phrases can be mapped to a given concept based on learned specific user behavior.
At 222, the healthcare entities 106 and/or the adherence bot service 116 may also determine a level of medication adherence, the existence of any side effects, and any adverse drug events, based on the data analyzed at 220. Depending on the adherence level, the existence of any side effects, and any adverse drug events, at 224, the healthcare entities 106 and/or the adherence bot service 116 provides feedback regarding adherence, the existence of any side effects, and any adverse drug events to the electronic device 104 at 226 and/or the medication dispenser 111 at 228. This feedback can also be provided to a third party, such as healthcare entity 106, for example, the target individual's primary care physician. The feedback can include a summary of the analysis, such as the level of medication adherence, the existence of any side effects, and any adverse drug events, as well as recommendations for further treatment including switching, titrating, or ceasing the medication or therapy. This can include recommendations regarding other medications that may be having interactions with the target medication.
Using a combination of the modules, the adherence bot can ensure that specific questions/responses will be used to identify non-adherent users and address the problems. Based on the received and processed data, the adherence bot can:
The timing of questions is also important. The adherence bot will ask “health related questions” at the same time the target individual is interacting with the medication. Similarly, the peripheral devices 109 can sample the biometric data at this timing to receive relevant data. Because the medication dispenser will have the medication information the user is taking, the system can help monitor for side effects using biometric data (from the hub, other peripheral devices, and/or self-reporting). As medication adherence increases or decreases, the system can update both the doctor and patient on the impact of the adherence levels. For example, a medication might cause a side effect of causing trouble sleeping. The system can monitor the target individual's medication intake as well as the sleep duration automatically and report the findings/correlations back to the user or doctor to help them make a health valuation on the impact. Another example is with regard to mental health surveys, such as PHQ-9, which includes a set of several patient questions. These questions can be sent through the adherence bot at the right time (such as when a smart cap having antidepressant medication is opened) and biometric data (lack of sleep, etc.) can be analyzed. If the analysis results in a high alert, for example, the adherence bot can escalate to alert a provider/family member/suicide hotline, etc.
The adherence bot can initiate several processes in its dialog with the target individual 102. Some of these processes are shown in
Glow: A user can text “Glow” or similar command into the electronic device to prompt the medication dispensers, such as a smart caps, to light up the caps corresponding to the medications that are due to be taken. The smart caps can have memory that stores the scheduling information.
Location: Shows the user via the screen of the electronic device Where the medication dispenser is located.
Enhanced Security: Allows a user to receive a text message each time the cap is opened.
Double Dose Alert: Alerts the user if the user opens the cap after the initial dose was due, i.e., alerts the user if the medication dispenser was accessed already.
Pharmacist—Links a user with a pharmacist or live nurse/caregiver.
Caregiver/Group Monitoring—Allows a user to add an individual to the SMS chat to provide access to see when the user has taken his or her medication (e.g., friends/family).
Change Schedule: Allows a user to adjust the reminders and medication schedule.
Change Medication: Allows a user to change the medication name.
Missed: Allows a user to see what medications were missed that day.
Help—Sends a list of commands and how to use the adherence bot.
Battery level—Sends battery reading of the medical dispenser.
Did I take my meds/Last take—Sends information regarding when the user last took the medication.
Medication info—Sends medication information from third parties.
Side effects—Sends side effect information regarding the medication the user is taking.
Drug to drug interactions—User can ask if a medication interacts with the current medication being taken.
Education—Sends relevant drug information to the user.
Chime—Causes the medication dispenser to make an audible sound
The adherence bot can also set up the medication schedule with the target individual 102 as shown in
Although this invention has been described in certain specific exemplary embodiments, many additional modifications and variations will be apparent to those skilled in the art in light of this disclosure. It is, therefore, to be understood that this invention may be practiced otherwise than as specifically described. Thus, the exemplary embodiments of the invention should be considered in all respects to be illustrative and not restrictive, and the scope of the invention to be determined by any claims supportable by this application and the equivalents thereof, rather than by the foregoing description.
| Number | Date | Country | |
|---|---|---|---|
| 63230963 | Aug 2021 | US |
