Method for managing drug prescription and dosage based on digital biomarker and apparatus for performing same method

Abstract

The present invention relates to a digital biomarker-based prescription management method and an apparatus for performing the method. The digital biomarker-based prescription management method may comprise the steps of: receiving, by a prescription management device, user data; and generating, by the prescription management device, prescription data on the basis of the user data. The present invention is a technology developed through the Seoul Business Agency's 2021 Bio-Medical Technology Commercialization Support Project (BT210029) "Technology Development of Precision Digital Therapeutic Agent for Alcohol Addiction."

Description

Digital biomarker-based drug prescription and dosage management method and device for performing such method

[0001] The present invention relates to a digital biomarker-based drug prescription and dosage management method and a device for performing such a method. More specifically, the present invention relates to a digital biomarker-based prescription and dosage management method for more adaptively prescribing to a user based on user data including user biomarkers, and a device for performing such a method.

[0002] The development of various smart technologies has enabled the recording of data on individuals' daily activities, enabling them to manage their lives more efficiently. Among these, health-related data logging is gaining attention due to the growing interest in health. Many users are already generating and utilizing various health-related data, such as exercise, diet, and sleep, through their smartphones and wearable devices. This shift is a shift from the past, when health-related data was solely generated and managed by medical institutions, to users themselves, through their own devices like smartphones and wearables.

[0003] Health-related data logging is often accomplished through wearable devices. Wearable devices are devices worn or attached to the user's body. Thanks to advancements in technologies like the Internet of Things, wearable devices are increasingly being used to collect health-related data. Wearable devices can collect information about the user's physical changes and the environment surrounding them, and based on this data, they can provide necessary health advice.

[0004] A user's health-related data may include user biomarkers, and research is underway on methods to provide more adaptive medical prescriptions to users based on their health-related data.

[0005] This invention is a technology developed through the Seoul Business Agency's 2021 Bio-Medical Technology Commercialization Support Project (BT210029) "Development of Precision Digital Treatment Technology for Alcohol Addiction."

[0006] Related technology includes Korean registered patent No. 10-2425479.

[0007] The purpose of the present invention is to solve all of the above-described problems.

[0008] In addition, the present invention aims to predict prescription effects and side effects based on artificial intelligence and generate prescription data that is adaptively tailored to the user.

[0009] In addition, the present invention aims to generate prescription data that is adaptively suited to the user by predicting the prescription effect and side effects of the prescription considering the user data through an artificial intelligence engine tuning optimized for the user.

[0010] A representative configuration of the present invention to achieve the above purpose is as follows.

[0011] According to one embodiment of the present invention, a digital biomarker-based prescription management method may include a step in which a prescription management device receives user data and a step in which the prescription management device generates prescription data based on the user data.

[0012] Meanwhile, the prescription management device generates the prescription data based on an artificial intelligence engine, and the artificial intelligence engine determines the prescription effect and the prescription side effects based on the user data, and generates the prescription data based on the prescription effect and the prescription side effects.

[0013] In addition, the artificial intelligence engine includes a default artificial intelligence engine and a user-optimized artificial intelligence engine, and the user-optimized artificial intelligence engine can be tuned using the user data based on the default artificial intelligence engine.

[0014] According to another embodiment of the present invention, a prescription management device that performs digital biomarker-based prescription management can be implemented to receive user data and generate prescription data based on the user data.

[0015] Meanwhile, the prescription management device generates the prescription data based on an artificial intelligence engine, and the artificial intelligence engine determines the prescription effect and the prescription side effects based on the user data, and generates the prescription data based on the prescription effect and the prescription side effects.

[0016] In addition, the artificial intelligence engine includes a default artificial intelligence engine and a user-optimized artificial intelligence engine, and the user-optimized artificial intelligence engine can be tuned using the user data based on the default artificial intelligence engine.

[0017] According to the present invention, prescription data adaptively tailored to the user can be generated by predicting prescription effects and side effects based on artificial intelligence.

[0018] In addition, according to the present invention, prescription data adaptively suited to the user can be generated by predicting the prescription effect and side effects of the prescription by considering the user data through artificial intelligence engine tuning optimized for the user.

[0019] Figure 1 is a conceptual diagram illustrating a digital biomarker-based prescription management system according to an embodiment of the present invention.

[0020] Figure 2 is a conceptual diagram illustrating the operation of a prescription management device according to an embodiment of the present invention.

[0021] Figures 3 and 4 are conceptual diagrams showing the differentiation of an artificial intelligence engine according to an embodiment of the present invention.

[0022] Figure 5 is a conceptual diagram illustrating an artificial intelligence engine for generating prescription data for a user according to an embodiment of the present invention.

[0023] Figure 6 is a conceptual diagram illustrating a method for creating a user-optimized artificial intelligence engine according to an embodiment of the present invention.

[0024] Figure 7 is a conceptual diagram illustrating a method for creating a user-optimized artificial intelligence engine according to an embodiment of the present invention.

[0025] The following detailed description of the present invention refers to the accompanying drawings, which illustrate specific embodiments in which the present invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the present invention. It should be understood that the various embodiments of the present invention, while different from each other, are not necessarily mutually exclusive. For example, specific shapes, structures, and characteristics described herein may be modified and implemented from one embodiment to another without departing from the spirit and scope of the present invention. Furthermore, it should be understood that the positions or arrangements of individual components within each embodiment may also be modified without departing from the spirit and scope of the present invention. Accordingly, the following detailed description is not to be taken in a limiting sense, and the scope of the present invention is to be construed to encompass the scope of the claims and all equivalents thereof. Like reference numerals in the drawings represent the same or similar elements throughout the several aspects.

[0026] Hereinafter, various preferred embodiments of the present invention will be described in detail with reference to the attached drawings so that a person having ordinary skill in the art to which the present invention pertains can easily practice the present invention.

[0027]

[0028] Figure 1 is a conceptual diagram illustrating a digital biomarker-based prescription management system according to an embodiment of the present invention.

[0029] In FIG. 1, a method for performing adaptive prescriptions for each user based on user data is disclosed.

[0030] Referring to FIG. 1, a digital biomarker-based prescription management system may include a user device (100), a prescription management device (120), and an information security device (140).

[0031] The user device (100) can be implemented to generate user data for a prescription for the user, transmit the generated data to the prescription management device (120), and receive prescription data adaptively determined according to the user.

[0032] User data may include various data used for prescribing to a user. For example, user data may include biomarker data acquired through a user device (100) (e.g., a smartphone, a wearable device), input data separately entered by the user, and data regarding the user's personal information (gender, age, etc.).

[0033] Prescription data may include data prescribed to a user by a prescription management device (120). The prescription management device (120) may generate prescription data based on user data and transmit the prescription data to a user device.

[0034] Prescription data can include information about medications taken, dosages, and timing of administration. For common medications, information about the user's dosage and timing of administration can be included in the prescription data. Prescription data can also include prescription algorithms based on digital therapeutics. For example, if a digital therapeutic exists for treating insomnia or eating disorders, prescriptions based on the digital therapeutic for the user can also be included in the user's prescription data.

[0035] The prescription management device (120) may be implemented to determine prescription data for a user based on user data. For example, the prescription management device (120) may be implemented as an artificial intelligence engine to generate optimal prescription data for a user based on user data. The artificial intelligence engine may learn changes in the user according to the prescription based on the user data according to the prescription data, and perform adaptive prescriptions for each user. The prescription management device (120) may include multiple artificial intelligence engines, and each of the multiple artificial intelligence engines may generate different data for the user's prescription (e.g., predicted prescription effect values, predicted prescription side effects, medications, dosages, timing of medication administration, prescription algorithms, prescription data, etc.), and prescription data may be generated based on this. In addition, the prescription management device (120) may also generate prescription data based on a prescription algorithm other than artificial intelligence.

[0036] According to an embodiment of the present invention, at least one of the artificial intelligence engines included in the prescription management device (120) may be divided into a default artificial intelligence engine before being differentiated for the user and a user-optimized artificial intelligence engine differentiated for the user.

[0037] The default AI engine is an AI engine that has been optimized for the user based on user data before being tuned, and can generate default prescription data. The user-optimized AI engine is an AI engine that has been optimized for the user based on user data after being tuned, and can generate optimized prescription data.

[0038] Additional learning is performed based on the default AI engine through user data to create a user-optimized AI engine, and the user-optimized AI engine can create optimized prescription data.

[0039] Additionally, the prescription management device (120) can generate prescription data based on drug regulation data. The drug regulation data can be learned on a separate artificial intelligence engine or implemented using a separate algorithm to ensure that the prescription data does not violate drug regulations.

[0040] An information security device (140) can be implemented to secure information generated from a user device (100) and a prescription management device (120).

[0041]

[0042] Figure 2 is a conceptual diagram illustrating the operation of a prescription management device according to an embodiment of the present invention.

[0043] In Fig. 2, a method for generating prescription data based on an artificial intelligence engine in a prescription management device is disclosed.

[0044] Referring to FIG. 2, the prescription management device may include multiple AI engines, each of which may generate multiple sub-prescription data sets. Prescription data may be generated based on the multiple sub-prescription data sets.

[0045] User data may be received from various devices in addition to the user device, but for convenience of explanation, it is assumed that user data generated through the user device is transmitted to the prescription management device.

[0046] User data can be preprocessed for transmission to each of multiple AI engines. For example, multiple user data in various formats can be grouped into multiple user data groups, and each of the multiple user groups can be transmitted to each of the multiple AI engines. Each of the multiple AI engines can generate multiple sub-prescription data sets.

[0047] Prescription data to be delivered to the user can be generated based on multiple sub-prescription data.

[0048] For example, artificial intelligence engine 1 (210) can be implemented to predict prescription effects.

[0049] AI Engine 2 (220) can be implemented to predict prescription side effects. AI Engine 3 (230) can be implemented to determine prescription data (type of medication, dosage, timing of administration, and prescription data based on a digital therapeutics prescription algorithm).

[0050] User data group 1 to user data group 3 corresponding to each of artificial intelligence engine 1 (210) to artificial intelligence engine 3 (230) can be defined, and user data group 1 to user data group 3 generated based on user data can be transmitted to each of artificial intelligence engine 1 to artificial intelligence engine 3.

[0051] Additionally, according to an embodiment of the present invention, a specific AI engine may receive output values ​​generated by other AI engines as input values. For example, prescription data generated by AI engine 3 (230) may be generated by receiving not only a user data group, but also prescription effect prediction data and prescription side effect prediction data generated by other AI engines as input values ​​for AI engine 3 (230).

[0052]

[0053] Figures 3 and 4 are conceptual diagrams showing the differentiation of an artificial intelligence engine according to an embodiment of the present invention.

[0054] Figures 3 and 4 are conceptual diagrams showing the process by which an artificial intelligence engine is differentiated based on user data.

[0055] Referring to FIG. 3, the default artificial intelligence engine (prescription effect) (300) that predicts the prescription effect can be tuned to a user-optimized artificial intelligence engine (prescription effect) (320) based on user data.

[0056] The default AI engine (prescription effect) (300) may be an AI engine that predicts the effectiveness of prescribing medication to a user or the effectiveness of prescribing a digital treatment algorithm to a user. The default AI engine (prescription effect) (300) may be an AI engine prior to being optimized for the user based on user data. The user-optimized AI engine (prescription effect) (320) is an AI engine optimized for the user based on user data.

[0057] For example, the default artificial intelligence engine (prescription effect) (300) can be tuned based on data set A to generate a user-optimized artificial intelligence engine (prescription effect) (320). Data set A can include data for optimizing prescriptions for users by considering user characteristics, such as user prescription data, user biomarker data, and user prescription effect data. Based on the user's data set A, the default artificial intelligence engine (prescription effect) (300) can be differentiated to generate a user-optimized artificial intelligence engine (prescription effect) (320).

[0058] The default AI engine (prescription effect) (300) may be further divided into a user-optimized AI engine (prescription effect) (320) for each user group determined by grouping multiple users with different prescription effects according to their prescription effects. Alternatively, the default AI engine (prescription effect) (300) may be further divided into a user-optimized AI engine (prescription effect) (320) in which separate parameter adjustments are performed solely for the user.

[0059] The user-optimized AI engine (prescription effect) (320) may also generate hourly drug effect prediction values ​​for the user. Alternatively, the default AI engine (prescription effect) (300) may be separately segmented by time to generate hourly prescription effects defined by the user-optimized AI engine (prescription effect) (320).

[0060] The default AI engine (prescription effect) (300) and the user-optimized AI engine (prescription effect) (320) can generate predicted prescription effects for multiple prescriptions. For example, if Prescription A, Prescription B, and Prescription C are administered simultaneously, predicted prescription effects can be generated and provided based on the timing of Prescription A, Prescription B, and Prescription C.

[0061] Referring to FIG. 4, the default artificial intelligence engine (prescription side effects) (400) can be differentiated based on user data and created into a user-optimized artificial intelligence engine (prescription side effects) (420).

[0062] The default AI engine (prescription side effects) (400) can be tuned based on data set B to generate a user-optimized AI engine (prescription side effects) (420). Data set B can include user prescription data, user biomarker data, user prescription side effects data, etc. Based on the user's data set B, the default AI engine (prescription side effects) (400) can be differentiated to generate a user-optimized AI engine (prescription side effects) (420).

[0063] The user-optimized artificial intelligence engine (prescription side effects) (420) may be generated for each user or user group, or may be generated in a time-separated manner.

[0064] The default AI engine (prescription side effects) (400) and the user-optimized AI engine (prescription side effects) (420) can generate predicted prescription side effects for multiple prescriptions. For example, if Prescription A, Prescription B, and Prescription C are administered simultaneously, predicted prescription side effects can be generated and provided based on the timing of Prescription A, Prescription B, and Prescription C.

[0065] A default AI engine can be created based on user data for all users, and a user-optimized AI engine can be created by differentiating the default AI engine and optimizing it based on user data for a specific user / group of users.

[0066] For the convenience of explanation, the artificial intelligence engine for prescription effects and side effects is assumed and explained, but other artificial intelligence engines can also operate as a default artificial intelligence engine and a user-optimized artificial intelligence engine differentiated from the default artificial intelligence engine.

[0067]

[0068] Figure 5 is a conceptual diagram illustrating an artificial intelligence engine for generating prescription data for a user according to an embodiment of the present invention.

[0069] In Fig. 5, a method for generating an artificial intelligence engine (prescription data) for generating prescription data for a user is disclosed.

[0070] Referring to Figure 5, the artificial intelligence engine (prescription data) can be learned and generated based on data based on other artificial intelligence engines, such as the artificial intelligence engine (prescription effect) and the artificial intelligence engine (prescription side effects).

[0071] The default artificial intelligence engine (prescription data) (530) can generate prescription data for a user based on the predicted prescription effect data generated by the default artificial intelligence engine (prescription effect) (510) and the predicted prescription side effect data generated by the artificial intelligence engine (prescription side effect) (520). The prescription data may include the medication taken, the dosage of the medication taken, the timing of taking, a digital therapeutic prescription algorithm, digital therapeutic prescription data, etc.

[0072] That is, the default artificial intelligence engine (prescription data) (530) is linked with the default artificial intelligence engine (prescription effect) (510) and the artificial intelligence engine (prescription side effect) (520) to generate prescription data based on the output values ​​output by the default artificial intelligence engine (prescription effect) (510) and the artificial intelligence engine (prescription side effect) (520).

[0073] Additionally, the default AI engine (prescription data) (530) can output prescription data that satisfies the drug regulation data through learning about the drug regulation data. For example, learning about the drug regulation data can be performed through feedback on the output prescription data, and the default AI engine (prescription data) (530) can output prescription data that satisfies the drug regulation data.

[0074] The default artificial intelligence engine (prescription data) (530) can be linked with the default artificial intelligence engine (prescription effect) (510) and the default artificial intelligence engine (prescription side effects) (520) to generate prescription data by considering the effects of multiple prescriptions and the side effects of multiple prescriptions.

[0075] When a user-optimized AI engine (prescription effect) (550) and / or a user-optimized AI engine (prescription side effect) (560) are determined according to differentiation of the default AI engine (prescription effect) (510) and / or the default AI engine (prescription side effect) (520), the default AI engine (prescription data) (530) connected to the user-optimized AI engine (prescription effect) (510) and / or the user-optimized AI engine (prescription side effect) (520) may be changed to the user-optimized AI engine (prescription data) (570).

[0076]

[0077] Figure 6 is a conceptual diagram illustrating a method for creating a user-optimized artificial intelligence engine according to an embodiment of the present invention.

[0078] In Fig. 6, a method for generating prescription data by generating a user-optimized artificial intelligence engine for each drug type is disclosed.

[0079] Referring to Figure 6, a user-optimized AI engine can be created for each drug type. For example, prescription efficacy and side effects can be determined by drug type, defined by drug properties rather than individual drugs, and prescription data can be generated for each drug type.

[0080] To determine the drug type, a primary classification can be performed based on the drug components that make up the drug. Among the multiple drug components that make up the drug, the drug components associated with the drug's effects (effects) can be extracted. Furthermore, among the multiple drug components that make up the drug, the drug components associated with side effects (side effects) can be extracted.

[0081] Drug types (effects) can be classified based on their effects, and drug types (side effects) can be classified based on their side effects, based on their composition (side effects). Drug types can be determined based on both drug types (effects) and drug types (side effects). Each drug type (effect) and drug type (side effects) can be embedded in a two-dimensional space.

[0082] For example, a drug type (effect) can be embedded based on a first dimension value set according to the effect on the first dimension. The drug type (effect) can be embedded so that the similar effects have closer first dimension values. The drug type (side effect) can be embedded based on a second dimension value set according to the side effect on the second dimension. The drug type (side effect) can be embedded so that the similar side effects have closer second dimension values. More specifically, on the first dimension, effect A can be located on x1~x2 depending on the degree of effect. On the first dimension, effect B can be located on x3~x4 depending on the degree of effect. As another example, on the second dimension, side effect C can be located on y1~y2 depending on the degree of side effect. On the second dimension, side effect D can be located on y3~y4 depending on the degree of side effect.

[0083] A single drug may have multiple different drug effects and side effects, and a single drug may have multiple individual embedding values. The drug type may be determined based on multiple individual embedding values ​​set for each drug. The drug effect / side effect similarity (600) may be determined based on the similarity of the multiple individual embedding values ​​(e.g., the distance between the individual embedding values). Based on the drug effect / side effect similarity (600), drugs with similar drug effects and side effects may be classified into the same drug type. In this case, the drug type may be determined by excluding some of the multiple individual embedding values ​​based on the degree of effect and side effect.

[0084] Furthermore, considering drug effects and side effects, weights can be adjusted across each dimension so that relatively strong drug effects and relatively strong side effects are classified as the same drug type. Furthermore, relatively important drug effects and relatively fatal side effects can be assigned higher weights, ensuring that drugs with similar important drug effects and similar potential for fatal side effects are classified as the same drug type.

[0085] In the present invention, a default artificial intelligence engine can be created by considering the drug type determined based on individual embedding values ​​as described above.

[0086] The default AI engine is differentiated based on user data, but can be differentiated into a user-optimized AI engine by forming user groups based on user data.

[0087] User groups can be established by drug type, taking into account the relationships between user data, including prescription efficacy and side effects for actual users. Based on user data, if prescription efficacy differs beyond a threshold value for specific user data values, or prescription side effects differ beyond a threshold value, users with those user data values ​​can be assigned to a user group, allowing for differentiation through a user-optimized AI engine.

[0088] Through this method, predictions of prescription efficacy and prescription adverse effects can be made for each user based on individual embedding values ​​of the drug composition (effect) and drug composition (side effects) of the new drug.

[0089] In the present invention, for convenience of explanation, only two dimensions (drug effects, drug side effects) are disclosed, but the drug type may be determined based on multiple dimensions beyond two dimensions or one dimension, and prediction of prescription effects and prescription side effects based on the drug type may also be performed.

[0090]

[0091] Figure 7 is a conceptual diagram illustrating a method for creating a user-optimized artificial intelligence engine according to an embodiment of the present invention.

[0092] In FIG. 7, a method for generating prescription data by considering drug combinations is disclosed.

[0093] Referring to FIG. 7, a method for generating prescription data by predicting drug effects and drug side effects for a combination of multiple drugs is disclosed.

[0094] In order to consider drug effects and drug side effects for a combination of multiple drugs, a combination embedding value (700) for the drug combination can be determined based on the individual embedding value (710) of each of the multiple drugs.

[0095] According to an embodiment of the present invention, a combined embedding value (700) is generated based on a combination of individual embedding values, and may be determined by considering the individual embedding value (710), the individual embedding value density (720), and the relationship between individual embedding values ​​(730).

[0096] Step 1: To determine a combined embedding value (700), individual embedding values ​​(710) of each of multiple drugs may be positioned on the embedding plane. At this time, if a specific individual embedding value (710) has a characteristic of having an effect below a threshold effect or a side effect not being strong enough to be below a threshold side effect, the specific individual embedding value may be excluded.

[0097] Step 2: By considering the relationship (730) between individual embedding values, the individual embedding values ​​(710) of each of the multiple drugs can be adjusted by considering whether the combination of drugs causes a change in effect (e.g., a decrease in effect, an increase in effect) or a change in side effects (e.g., a decrease in side effects, an increase in side effects). For example, in the case of an increase in effect, in the case of a decrease in side effects, etc., the individual embedding values ​​(730) representing the degree of effect and side effects can be adjusted to adjust the position on the embedding plane.

[0098] Step 3: The combined embedding value (700) can form clusters that reflect effects and side effects by clustering individual embedding values ​​(710). Dense clusters with a density greater than a threshold can be determined based on the density of individual embedding values ​​(710) within the cluster. Clustering can be performed by effect or side effect. That is, the clustering scope can be set to be performed between identical / similar effects and identical / similar side effects, thereby determining dense clusters by effect and side effect.

[0099] Based on the density of individual embedding values ​​within a dense cluster, a final combined embedding value is determined, and predictions can be made regarding the increase or decrease in specific side effects or effects when taking a combination of multiple drugs. For example, if the density of a dense cluster corresponding to a specific effect exceeds a critical density, an increase in effect can be predicted. Similarly, if the density of a dense cluster corresponding to a specific side effect exceeds a critical density, an increase in side effects can be predicted.

[0100] According to embodiments of the present invention, predictions regarding the increase or decrease of specific side effects or effects when taking a combination of multiple drugs can be made, taking into account changes in efficacy and side effects over time. For example, individual embedding values ​​can be adjusted based on a decrease in efficacy over time, and individual embedding values ​​can be adjusted based on a decrease in the likelihood of side effects over time. Based on these changes in individual embedding values ​​over time, a user optimization AI engine can determine and provide a user with a dosage routine that relatively increases efficacy and relatively reduces side effects when taking a combination of multiple drugs.

[0101]

[0102] The embodiments of the present invention described above may be implemented in the form of program commands that can be executed through various computer components and recorded on a computer-readable recording medium. The computer-readable recording medium may include program commands, data files, data structures, etc., either singly or in combination. The program commands recorded on the computer-readable recording medium may be specially designed and configured for the present invention or may be known and available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical recording media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and hardware devices specifically configured to store and execute program commands, such as ROMs, RAMs, and flash memories. Examples of program commands include not only machine language codes generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter, etc. Hardware devices may be changed into one or more software modules to perform processing according to the present invention, and vice versa.

[0103] Although the present invention has been described above with specific details such as specific components and limited examples and drawings, these are provided only to help a more general understanding of the present invention, and the present invention is not limited to the above examples, and those with ordinary knowledge in the technical field to which the present invention pertains can make various modifications and changes based on this description.

[0104] Therefore, the idea of ​​the present invention should not be limited to the embodiments described above, and not only the scope of the patent claims described below but also all scopes equivalent to or equivalently modified from the scope of the patent claims are considered to fall within the scope of the idea of ​​the present invention.

Claims

1. Digital biomarker-based prescription management method, Step of the prescription management device receiving user data; and A method characterized in that the above prescription management device comprises a step of generating prescription data based on the above user data.

2. In paragraph 1, The above prescription management device generates the above prescription data based on an artificial intelligence engine, A method characterized in that the artificial intelligence engine determines the prescription effect and the prescription side effects based on the user data, and generates the prescription data based on the prescription effect and the prescription side effects.

3. In paragraph 2, The above artificial intelligence engine includes a default artificial intelligence engine and a user-optimized artificial intelligence engine. A method characterized in that the user-optimized artificial intelligence engine is tuned using the user data based on the default artificial intelligence engine.

4. A prescription management device that performs digital biomarker-based prescription management, Receive user data, A prescription management device characterized in that it is implemented to generate prescription data based on the above user data.

5. In paragraph 4, The above prescription management device generates the above prescription data based on an artificial intelligence engine, A prescription management device characterized in that the artificial intelligence engine determines the prescription effect and the prescription side effects based on the user data, and generates the prescription data based on the prescription effect and the prescription side effects.

6. In paragraph 5, The above artificial intelligence engine includes a default artificial intelligence engine and a user-optimized artificial intelligence engine. A prescription management device characterized in that the user-optimized artificial intelligence engine is tuned using the user data based on the default artificial intelligence engine.

Images