Droplet delivery device implementing ai

Abstract

A droplet delivery device implementing artificial intelligence (AI) to provide drug reduction therapy, such as nicotine, specifically adapted to a user.

Description

DROPLET DELIVERY DEVICE IMPLEMENTING AlCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of priority of U.S. Provisional Patent Application No. 63 / 658,765, filed June 11, 2024, and U.S. Provisional Patent Application No. 63 / 523.194, filed June 26, 2023, which are incorporated herein by reference in their entirety.FIELD OF THE INVENTION

[0002] This disclosure relates to droplet delivery7devices and more specifically to droplet delivery devices for the delivery of fluids that are inhaled into mouth, throat, nose, and / or lungs. While the present invention particularly pertains to the delivery of therapeutic compositions, in some embodiments non-therapeutic compositions that are capable of being aerosolized can also be used.BACKGROUND

[0003] The present application incorporates herein by reference in their entirety the contents of WO 2020 / 264501 (describing “Ring Mode” ejection), PCT / US2022 / 034552 (describing “Push Mode” ejection), U.S. Pat. No. 10,449,314 (describing dose verification) and US Patent Application Pub. No. 20190134330 (describing user feedback and instructions).

[0004] There is a need for droplet delivery devices, such as those devices described in Applicant’s referenced patent publications, to implement artificial intelligence (Al) for improved dosing and personalization of aerosol treatments. Specifically, the present invention can determine through Al based on numerous parameters associated with the user, time of day, dosing habits, inhalation data, and the like, how effective droplet delivery7treatments are for a user, optimal dosing plans for the user, adjustments to dosing plans and medications, and how aggregated data associated with use and results of droplet delivery' device treatments can be applied to others, such as those having similar lifestyles, physical characteristics and / or demographics.

[0005] In various embodiments, Al is implemented to provide nicotine reduction therapies with droplet delivery devices. In particular, “Push Mode” ejection devices may be particularly well-suited for implementing Al as compared to prior ejection technologies that do not use “push mode” technologies. Further, Al in combination with “Push Mode” ejection devices appear particularly advantageous for nicotine reduction therapy.SUMMARY OF THE INVENTION

[0006] BRIEF DESCRIPTION OF THE DRAWINGS

[0007] FIGS 1A and IB show block diagrams of a network-based system that encourages cessation of a drug delivered to the lungs via a droplet delivery device in an embodiment of the invention.

[0008] FIG. 2 shows a display screen of an electronic device including exemplary responses the user receives when communicating with a Generative Pre-trained Transformer (GPT) chat-bot in an embodiment of the invention.

[0009] FIG. 3 shows a display screen of an electronic device including an exemplary motivational story provided by a GPT chat-bot to a user in an embodiment of the invention.

[0010] FIG. 4 shows a display screen of an electronic device including a prompt and response for a fact from a scientific article provided to a user in an embodiment of the invention.

[0011] FIG. 5. shows a flow chart for a process to create a plan to gradually reduce a user's consumption of a drug by implementing artificial intelligence (Al) with an electronoic portable drug droplet delivery device in an embodiment of the invention.

[0012] FIG. 6 shows a flow chart for a process of generating a drug cessation or reduction plan implementing Al with an electronic portable drug droplet delivery device in an embodiment of the invention.DETAILED DESCRIPTION

[0013] An embodiment of the invention features Al that is used in conjunction with a mobile app that is paired to a droplet delivery device to support treatment protocols, treatment personalization, and connectivity.

[0014] In embodiments where Al is used in combination with droplet delivery devices for nicotine reduction therapy, Al can consider a large number of factors that affect the amount of nicotine required to satisfy a user. These factors can include, but are not limited to, attributes such as body height, body weight, and gender. This data can be input into a paired mobile app by the user. Al and the app can then use the data to calculate the amount of nicotine required to satisfy cravings based on predicted blood concentration for a given dose. Via Al and the app, this can aid in developing a personalized nicotine consumption control or reduction plan for the user. In some embodiments this can be used to develop a smoking cessation plan to aid in getting the user off cigarettes.

[0015] In other embodiments, the mobile app can prompt cigarette smokers to enter their current cigarette brand and smoking usage, alone or in addition to exemplary datadescribed in this disclosure. The embedded Al in the app can search through databases to find the nicotine content of the user’s entered cigarette brand. This data can be used to provide a recommended dose of nicotine that should satisfy the user’s cravings. In some cases, this dose can be the starting dose from which a nicotine reduction plan is developed through the app. In other cases, this can be a controlled dose that is used in smoking cessation to get the user off cigarettes.

[0016] In embodiments where a nicotine amount is recommended through Al and the app, the app can prompt users to provide feedback on their level of craving satisfaction. Based on this feedback, the Al will learn and improve its calculations and recommendations over time. In other embodiments where Al is used in combination with droplet delivery devices for nicotine reduction therapy, Al can consider an additional number of factors that affect the amount of nicotine required to satisfy a user. These factors can include time of day, heart rate, steps taken in a day, length and frequency of workouts, and sleep quality. For example, most users may require less nicotine when they wake up in the morning and are already in nicotine withdrawal. After eating or working out, the amount of nicotine required may vary’. Over time, the Al will learn the user’s habits and needs and tailor the amount of nicotine delivered. The user can utilize this functionality to decrease their nicotine intake over time.

[0017] In another embodiment, a droplet delivery system is used in conjunction with a mobile application to generate data for supporting a model to predict whether the user will succeed or fail a plan to quit or reduce their nicotine consumption. Data collected on the user’s health data (weight, height, BMI, gender, etc.), their nicotine consumption data (nicotine intake, time of day of intake, location of usage, inspiratory flow during intake, etc.), and Pulmonary function data (forced expiratory volume in 1 second, forced vital capacity7, peak expiratory flow, FEVi / FVC, etc.) and their effects on the user’s nicotine reduction and cessation rates can be used in a supervised learning model. The supervised learning model can then be used to predict the likelihood of a user quitting or reducing their nicotine consumption.

[0018] In order to leam the user’s habits, the device administers test doses based on a number of factors (i.e., time of day, time after / before eating, cardiovascular workouts, etc.). The inhalation strength, topography, and frequency of the user during the test doses is recorded. The information recorded from the test doses is used to determine the minimum amount of nicotine required to satisfy the user. This information is used to reduce the amount of nicotine used over time and individualize the device to specific user’s needs.

[0019] One embodiment includes the implementation of Al with a droplet delivery device to include connection to a user’s phone, smartwatch and / or health trackers, and healthtracking apps. These connections provide further data for the Al to learn a user’s activity and other factors in connection with the user’s nicotine use requirements. Additionally, when the droplet delivery device is paired with its own mobile app, the app can provide alerts or prompts through the smartwatch, phone, or health tracking apps on usage (i.e., higher or lower than average usage) and treatment progress. As an example, with nicotine reduction therapy, the paired app with Al will track a user’s nicotine consumption through the device. If the app / Al registers unusually high consumption or consumption that is above what is recommended in the user’s established reduction plan, the app can send an alert to the user’s phone or smartwatch / health tracker. The app can then prompt the user to provide insight on whether something in particular is causing the increase in consumption (e.g., stress, feeling low, social event).

[0020] In the above embodiments where a droplet delivery device is connected to a mobile app with embedded Al, the app communicates with the device and can be used to set the amount of nicotine delivered through the device.

[0021] In some embodiments, Al is used in conjunction with a mobile app connected to a droplet delivery device to function as a virtual counselor throughout treatment programs. The app can act as a counselor, prompting the user with regular check-ins and providing personalized support, such as cognitive behavioral therapy. The app / AI can monitor the progress of a user’s treatment; if a treatment plan is not going well for a user, the app / AI can provide suggestions for modifying the treatment plan. For example, in the case of nicotine reduction, the app / AI can suggest a slower taper off nicotine. As another example, in the case of smoking cessation, the app / AI can suggest a higher dosage of nicotine from the paired device to help the user get off cigarettes.

[0022] In some embodiments, Al is used to personalize treatment through the way a mobile app or the device communicates with a user. Over time, Al will learn the way a user speaks and incorporate their speech mannerisms into voice prompts and communications. This may allow the user to feel more comfortable and connected in their treatment plan.

[0023] In other embodiments, Al can be used to help monitor smoking cessation plans. Keeping track of external data of when a user is likely to smoke to help keep the user informed. This could be a specific location, a time of day, or when the user’s heart rate reaches a certain level. This can be used as an early interv ention to ensure the user is aware of what is happening. This can be in combination with the Al therapy.

[0024] In some embodiments, when Al is used in a mobile app paired with a droplet delivery device, Al can generate educational or motivational content for a user. This contentcan be personalized based on historic data from the user or can be based on specific prompts input by a user. Prompts can be given through a chat-style section of the app. In some embodiments, ChatGPT is utilized for this chat section.Al can learn from databases of stories or educational material to generate appropriate output for the prompt. As an example, during a smoking cessation program, a user can prompt the app to tell a story about quitting cigarettes to provide them with motivation. The app can then generate a story about someone quitting cigarettes and the positive impact it had on their life. Continually reinforcing the positive benefits of living a nicotine-free and addiction -free life can affect the user’s willingness to quit.In addition to motivational stories, Al can supply the user with educational materials and facts. As an example, Al can be trained on information from official sites, such as the US CDC and use this information to provide the user with education on the short term and long term impacts of stopping smoking, such as the impact on increased life expectancy and improved lung functioning.

[0025] Key Al parameters

[0026] While Al based on machine learning with app notifications is generally known, the present invention provides new ways to implement these Al techniques for nicotine reduction therapy.

[0027] Machine learning algorithms:

[0028] Acquire databases of smoker's behavior, smoking habits and known smoking cessation models.

[0029] The Al model can be trained on the databases to improve the accuracy.

[0030] The algorithms can be continuously updated with new data to improve their performance and ensure they remain relevant.

[0031] Personalization:

[0032] The Al model can collect user's smoking habits to provide personalized recommendations.

[0033] Providing customized forms to present personal data.

[0034] Infographics for the user to visualize their smoking and their quitting.

[0035] Measuring the user’s exhale, similar to spirometry, to collect more data on the user as well as to give the user feedback on .

[0036]

[0037] This personalization can increase the effectiveness of the device and make it more attractive to users.

[0038] Gaming:

[0039] People can be particularly receptive to gaming, including where gaming can assist with lifestyle changes and modifying one’s activities toward healthier choices.

[0040] A droplet device with Al can incorporate gamification elements, such as rewards and challenges to incentivize users to quit smoking and increase their motivation.

[0041] The games can be self-challenged, interpersonal, intergroup, or intercommunity. A multiple gaming system can enrich society, and encourage sociologists, psychologists, and doctors to create new gaming systems or doing experiments. “A game for making games.”

[0042] Real-time feedback can be provided.

[0043] Real-time statistic data - personal and macro data.

[0044] Real-time behavior interference.

[0045] Integration with other health devices:

[0046] The device can be integrated with other health devices, such as fitness trackers, to provide a more holistic approach to quitting smoking and improve overall health outcomes.

[0047] Using data from the integration to help users know when to take nicotine and when they have enough nicotine. This can also adjust the amount of nicotine to deliver. This can also change the smoking cessation plan for the user.

[0048] Using data from the integration to help us know when a user will quit and if a user will quit.

[0049] Cross-platform app.

[0050] CO monitoring.

[0051] Predictive modeling:

[0052] Collecting existing predictive model, including heredity or non-heredity model.

[0053] The machine learning algorithms can be used to develop predictive models that can identify users who are at high risk of relapse and provide targeted support to prevent it.

[0054] The algorithm may also predict next smoking related action and provide recommendation in advance. It might be possible that the advance notification may reduce the overall consumption of nicotine.

[0055] Data can be collected to predict if the user’s lungs are getting better or if there is a problem with their lungs.

[0056] Referring to FIGS. 1A and IB, the diagrams show a system 100 designed to encourage cessation of a drug delivered to the lungs via a droplet delivery device 105. The system comprises several processing devices connected via a network 120 to displayinformation to the user about their drug usage and health data. The analyzation of the data includes using an Al engine 230 with a machine learning model 232 trained to support the cessation of the delivered drug. The system utilizes a generative pre-trained transformer (GPT) chat-bot platform 215 to interact with the system 100 within the user interface 210.

[0057] The network 100 receives and transmits data to and from all processing devices via a wired or wireless connection (e.g. Bluetooth, Zigbee, Wi-fi. Cellular data). The network 100 is enabled by the data communication systems 150 on the processing devices.

[0058] Drug usage data 107 collected from the portable electronic droplet delivery device 105 contains information on the time of day of the drug dosages from time keeping system 132 that is operably coupled to microcontroller 130. This information can be used to approximate the plasma concentration of the drug in the user’s body in real time. The data 107 is collected and stored locally in the memory 1 11 on the portable electronic droplet delivery device until it is connected to a network 100 to transmit the data to the cloud for permanent storage and analyzation. Instructions 113 for operation and control of the droplet delivery device 105, including the drug delivery system 134, are also included in memory 113.

[0059] A portable electronic diagnostic device 110 can be used to collect disease diagnostic data 114 on the health of the user via a disease diagnostic system 117 operably coupled to a microcontroller 118 that is operably coupled to a time keeping system 119. This data 114 can be used in conjunction with the drug usage data 107 collected from the portable drug delivery’ device 150 to encourage cessation of a drug. The data is collected and stored locally in memory 112 on the portable electronic diagnostic device 110 until it is connected to a network to transmit the data to the cloud for permanent storage and analyzation. Instructions 116 for operation and control of the diagnostic device 1110 are also included in memory 112.

[0060] The server 220. connected to network 120, includes a computer processor 270 and an Al engine 230 supporting machine learning model(s) 232 which is trained to determine user’s likelihood of cessation and generate new treatment plans to be displayed to the user and loaded onto the droplet delivery’ device 105. This engine 230 uses drug usage data 107 collected from the portable electronic droplet delivery device 105 and disease diagnostic data 114 from the portable electronic diagnostic device 110, such data stored in memory 260. to assist the user in the cessation of the drug. Server memory 260 also includes user data 108 from the user interface 210. Memory 260 also includes instructions 280 for processing the diagnostic data 114, usage data 107 and user 108, in coordination with the other devices connected to network 100. The training engine 240 is used to further improve the machine learning models on theserv er 220, producing more accurate predictions and generating more appropriate treatment plans.

[0061] User interface 210, such as included on a mobile application of a user’s mobile computing device (e.g. mobile phones, tablets and the like) or as can be integrated as a computing and display device together with droplet delivery7device 105, is connected via network 100 to server 220. A device with an application providing user interface 210, includes a computer processor 203 operably coupled to GPT chat-bot 215 for receiving input and providing outputs back to the user from the Al Engine 230 on server 220. The device including user interface 210 also includes memory 270 with instructions 275 for storing and retrieving drug usage data 114, disease diagnostic data 107 and user data 108 in the memory 270 associated with user interface 210.

[0062] The administration interface 205 is used to manage users and deliver insights on the performance of the system 100. The administration 205 interface helps improve the system to increase the chances of user cessation.

[0063] FIG. 2 illustrates an example of the type of responses the user receives when communicating with the GPT chat-bot 215 in user interface 210.

[0064] FIG. 3 illustrates an example of a motivational story provided by the GPT chatbot 215 in user interface 210.

[0065] FIG. 4 illustrates an example of prompt and response for a fact from a scientific article provided by the GPT chat-bot 215 in user interface 210.

[0066] FIG. 5 illustrates a flow chart describing a process to create a plan to gradually reduce the user’s consumption of the drug with Al on server 220. At step 500, server processor 270 initiates the process to create the plan. At step 510, data from the drug delivery system 134 and diagnostic system 117 are processed by processor 270 to create an initial dataset. At step 515, the artificial intelligence engine 230 generates an initial plan for reducing drug intake gradually over time based on drug usage data 114 and diagnostic data 107. At step 525, a message is transmitted to the user interface 210 displaying the plan. At step 530, data is further received over the course of the drug consumption reduction that includes user compliance, drug concentration levels, time of day drug dosages, and like pertinent data from drug delivery system 134 and diagnostic system 1 17 to determining adjustments to the plan. At step 535, a modified plan is generated based on the collected data to increase user compliance and the modified plan is transmitted to the user interface by returning to step 525. When the user has reached full compliance the plan(s) will be completed.

[0067] FIG. 6 illustrates a flow chart describing a process of how a plan generated through Al on server 220 can control the portable electronic droplet delivery device 105. At portable drug delivery device 105 the process begins at step 600. At step 610, the delivery device 105 receives a gradual drug reduction plan from user interface 210, including receiving such plan from a mobile device that is running an application providing the user interface 210. At step 620, drug delivery device tracks the total drug consumption on time-determined basis, such daily and hourly, throughout the course of the plan. At step 630, when the user reaches a time-based limit, the drug delivery device 105 will not dispense more drug until is reset be an elapsed time.

[0068] Various embodiments of the invention have been described. It will, however, be evident that various modifications and changes may be made thereto, and additional embodiments may be implemented, without departing from the broader scope of the invention as set forth by the disclosure. This specification is to be regarded in an illustrative rather than a restrictive sense.

Claims

What is Claimed:

1. A droplet delivery device control system comprising an artificial intelligence engine that is remotely and operably coupled via a computer network to implement a drug consumption reduction plan on a droplet delivery7device dispensing the drug based on one or more of the user’s habits, user's physical characteristics and user’s test doses.

2. The droplet delivery device control system of claim 1, wherein the drug is nicotine.

3. The droplet delivery7device control system of claim 2, wherein the droplet delivery device includes a push mode ejector.

4. The droplet delivery device control system of claim 1, wherein the droplet delivery device includes a push mode ejector.

5. The droplet delivery7device control system of clam 4, further comprising a portable electronic diagnostic device operably coupled to the computer network and providing disease diagnostic data to the artificial intelligence engine.

6. The droplet delivery device control system of clam 3, further comprising a portable electronic diagnostic device operably coupled to the computer network and providing disease diagnostic data to the artificial intelligence engine.

7. The droplet delivery device control system of clam 2, further comprising a portable electronic diagnostic device operably coupled to the computer network and providing disease diagnostic data to the artificial intelligence engine.

8. The droplet delivery7device control system of clam 1, further comprising a portable electronic diagnostic device operably coupled to the computer network and providing disease diagnostic data to the artificial intelligence engine.

9. The droplet delivery device control system of claim 9, further comprising a mobile computing device configured to display a user interface via a mobile application connected via the network to the artificial intelligence engine, wherein the user interface presents a treatment protocol based on the drug consumption reduction plan.

10. The droplet delivery device control system of claim 8, further comprising a mobile computing device configured to display a user interface via a mobile application connected via the network to the artificial intelligence engine, whereinthe user interface presents a treatment protocol based on the drug consumption reduction plan.

11. The droplet delivery device control system of claim 7, further comprising a mobile computing device configured to display a user interface via a mobile application connected via the network to the artificial intelligence engine, wherein the user interface presents a treatment protocol based on the drug consumption reduction plan.

12. The droplet delivery device control system of claim 6, further comprising a mobile computing device configured to display a user interface via a mobile application connected via the network to the artificial intelligence engine, wherein the user interface presents a treatment protocol based on the drug consumption reduction plan.

13. The droplet delivery device control system of claim 5, further comprising a mobile computing device configured to display a user interface via a mobile application connected via the network to the artificial intelligence engine, wherein the user interface presents a treatment protocol based on the drug consumption reduction plan.

14. The droplet delivery device control system of claim 4, further comprising a mobile computing device configured to display a user interface via a mobile application connected via the network to the artificial intelligence engine, wherein the user interface presents a treatment protocol based on the drug consumption reduction plan.

15. The droplet delivery' device control system of claim 3, further comprising a mobile computing device configured to display a user interface via a mobile application connected via the network to the artificial intelligence engine, wherein the user interface presents a treatment protocol based on the drug consumption reduction plan.

16. The droplet delivery device control system of claim 2, further comprising a mobile computing device configured to display a user interface via a mobile application connected via the network to the artificial intelligence engine, wherein the user interface presents a treatment protocol based on the drug consumption reduction plan.

17. The droplet delivery device control system of claim 1. further comprising a mobile computing device configured to display a user interface via a mobileapplication connected via the network to the artificial intelligence engine, wherein the user interface presents a treatment protocol based on the drug consumption reduction plan.

18. A mobile computing device configured to display a user interface via a mobile application connected via a computer network to a remote server with an artificial intelligence engine and to a drug droplet delivery device, wherein the user interface presents information for a drug consumption reduction plan determined by the artificial intelligence engine.

19. The mobile computing device of claim 18, wherein drug consumption reduction plan is to reduce use of nicotine.

20. The mobile computing device of claim 18. further operably coupled to a portable electronic diagnostic device via the network.

Images