KR20240026603A - System and method for providing personalized drug interaction analysis and health functional food recommendation - Google Patents

Abstract

This disclosure provides a system and method for personalized drug interaction analysis and nutraceutical recommendation. The method includes constructing user data including the user's health checkup information, survey information, information on medications taken, and information on health functional foods taken; Obtaining target health functional food information using a user terminal; Analyzing drug interactions between the target health functional food and the taking medicine or the taking health functional food and predicting the possibility of side effects; determining a health state type of the user corresponding to the health state of the user obtained from the health checkup information and the survey information and determining candidate health functional foods for nutrients required for the health state type; And it may include a step of determining recommended health functional foods that reflect preferences in the health status type among the candidate health functional foods.
According to the present disclosure, there is an effect of providing an application to which algorithmic techniques are applied that can provide personalized health functional food drug interaction information and health functional food recommendation information.

Description

System and method for personalized drug interaction analysis and health functional food recommendation {SYSTEM AND METHOD FOR PROVIDING PERSONALIZED DRUG INTERACTION ANALYSIS AND HEALTH FUNCTIONAL FOOD RECOMMENDATION}

This disclosure relates to a health functional food recommendation application, and more specifically, to a system and method for personalized drug interaction analysis and health functional food recommendation.

As social interest in well-being increases, consumer demand for health functional foods is increasing. Health functional foods refer to foods manufactured and processed using raw materials or ingredients with functional properties useful to the human body. Health functional foods do not directly treat or prevent diseases like pharmaceuticals, but rather aim to maintain and improve health through maintaining the normal functions of the human body or activating physiological functions. For example, health functional foods may include various vitamin supplements, foods containing nutrients such as magnesium, zinc, and manganese, and foods containing extracts such as green tea, ginkgo biloba, and aloe.

However, health functional foods may cause abnormal symptoms due to misuse or abuse if indiscriminately consumed depending on the user's health condition, constitution, etc., and may contain ingredients that may cause health problems when consumed together with medicines for disease treatment. There is. Since it is realistically difficult for general users who lack medical knowledge to identify these side effects, we have developed an application that can provide personalized target health functional food drug interaction information and suitable/unsuitable health functional food guide information in a user-friendly manner. There is a need to develop.

Patent Document 1: Public Patent Publication No. 10-2020-0117218 (2020.10.14)

The purpose of this disclosure to solve these problems is to provide a system and method for personalized drug interaction analysis and health functional food recommendation.

According to one embodiment of the present disclosure, a personalized health functional food recommendation system is provided. The system includes a server configured to provide drug interaction analysis information and recommended health functional food information customized to the user to the user terminal based on user data; And it may include a database configured to store and manage the user data, drug information, and health functional food information. The server includes a user data collection unit configured to construct the user data including the user's health checkup information, survey information, information on medications taken, and information on health functional foods taken; a drug interaction analysis unit configured to analyze drug interactions between the target health functional food, information about which is obtained using the user terminal, and the taking medicine or the taking health functional food and predict the possibility of side effects; and determining a health state type of the user corresponding to the health state of the user obtained from the health checkup information and the survey information, and determining candidate health functional foods for nutrients required for the health state type, It may include a health functional food recommendation unit configured to determine recommended health functional foods that reflect preferences in the health status type among candidate health functional foods.

In addition, the drug interaction analysis unit uses each of the target health functional food, the medicine to be taken, and the health food to be taken as a node, the interaction between the two nodes as an edge, and interaction on each edge. It may be configured to analyze drug interactions for the user by constructing an undirected weighted graph in which the risk is set, and determine that there is a possibility of side effects between two nodes connected by an edge where the interaction risk is more than a preset threshold.

In addition, the health functional food recommendation unit defines health state types that match nutrient information in which normalized values are assigned to each nutrient according to the degree of need for each type of health state, and K-Nearest Neighbor (KNN) It may be configured to determine the user's health state type corresponding to the user's health state through an algorithm and determine the necessary nutrients corresponding to the determined health state type.

In addition, the health functional food recommendation unit sorts the candidate health functional foods in the order of highest preference predicted in the user group belonging to the user's health condition type according to a user-based collaborative filtering (UCF) algorithm. It may be configured to determine the recommended health functional foods.

In addition, the health functional food recommendation unit excludes, from among the determined recommended health functional foods, health functional foods determined by the drug interaction analysis unit to have the possibility of side effects with the drug or the health functional food to be taken, or the possibility of the side effects. It may be configured to provide information on recommended health functional foods.

Additionally, the server may provide an application for providing the drug interaction analysis information and the recommended health functional food information to the user terminal, and the application may include a health functional food scan function. The user data collection unit may be configured to acquire health functional food information in the database that matches information optically recognized through the camera of the user terminal by the health functional food scan function as target health food information.

In addition, the system is configured to provide dosage information for each nutrient compared to the recommended intake amount of the nutrients contained in the health functional foods being taken in the health functional food information being taken, and to indicate whether each health functional food being taken is necessary. It may further include a nutritional information provision section.

According to one embodiment of the present disclosure, a personalized health functional food recommendation method that can be performed by a health functional food recommendation system is provided. The method includes constructing user data including the user's health checkup information, survey information, information on medications taken, and information on health functional foods taken; Obtaining target health functional food information using a user terminal; Analyzing drug interactions between the target health functional food and the taking medicine or the taking health functional food and predicting the possibility of side effects; determining a health state type of the user corresponding to the health state of the user obtained from the health checkup information and the survey information and determining candidate health functional foods for nutrients required for the health state type; And it may include a step of determining recommended health functional foods that reflect preferences in the health status type among the candidate health functional foods.

In addition, the step of analyzing the drug interaction and predicting the possibility of side effects includes the target health functional food, the taking medicine, and the taking health functional food each as nodes, the interaction between the two nodes as an edge, and each edge Constructing an undirected weight graph in which interaction risk is set; And it may include determining that there is a side effect between two nodes connected by an edge in the undirected weighted graph where the interaction risk is greater than or equal to a preset threshold.

In addition, the step of determining the candidate health functional foods includes defining health state types that match nutrient information in which normalized values are assigned to each nutrient according to the degree of need for each health state type; and determining the user's health state type corresponding to the user's health state through a K-least nearest (KNN) algorithm and determining the necessary nutrients corresponding to the determined health state type.

In addition, the step of determining the recommended health functional foods includes recommending the candidate health functional foods in the order of highest preference predicted in the user group belonging to the user's health condition type according to a user-based collaborative filtering (UCF) algorithm. It may include the step of determining health functional foods.

In addition, the step of determining the recommended health functional foods includes excluding, from among the determined recommended health functional foods, health functional foods determined to have a possibility of side effects with the drug or the health functional food to be taken, or recommending the health functional foods together with the possibility of side effects. A step of providing health functional food information may be further included.

Additionally, the health functional food recommendation system may provide a health functional food recommendation application to the user terminal, and the application may include a health functional food scan function. The step of acquiring the target health functional food information includes converting health functional food information in the database of the system, which is matched with information optically recognized through the camera of the user terminal by the health functional food scan function, into target health food information. It may include an acquisition step.

In addition, the method includes providing dosage information for each nutrient compared to the recommended intake amount of the nutrients contained in the health functional foods being taken in the health functional food information being taken, and indicating whether each health functional food being taken is necessary. It may further include.

According to one embodiment of the present disclosure, a computer program stored on a computer-readable medium is provided, including computer-executable instructions for executing the personalized health food recommendation method.

According to the present disclosure, there is an effect of providing an application to which algorithmic techniques are applied that can provide personalized health functional food drug interaction information and health functional food recommendation information.

In addition, according to the present disclosure, there is an effect that the user can conveniently check not only information on the health functional food of interest but also the possibility of side effects with the medicine or health functional food being taken by just scanning the health functional food of interest.

In addition, according to the present disclosure, the level of intake of nutrients contained in the health functional foods the user is taking compared to the recommended intake is clearly provided in a graph for each nutrient, as well as a guide as to whether each health functional food being taken is necessary. It has the effect of providing.

1 is a schematic diagram illustrating an example system for personalized drug interaction analysis and nutraceutical recommendation according to an embodiment of the present disclosure.
2 is a schematic diagram illustrating example components of a server according to one embodiment of the present disclosure.
Figure 3 is an exemplary flowchart showing a personalized health functional food recommendation method according to an embodiment of the present disclosure.
4 is an exemplary diagram illustrating an undirected weight graph constructed for drug interaction analysis according to an embodiment of the present disclosure.
Figure 5 shows an example service screen of a health functional food recommendation application to which a user is logged in according to an embodiment of the present disclosure.
6 illustrates an example service screen providing a user nutritional graph according to an embodiment of the present disclosure.
Figure 7 is an exemplary diagram illustrating scanning a target health functional food through the scan function of an application according to an embodiment of the present disclosure.
FIG. 8 shows an exemplary service screen for obtaining and providing information on health functional foods scanned in FIG. 7 according to an embodiment of the present disclosure.
FIG. 9 illustrates an example service screen that provides information on the user's potential side effects for the health functional food scanned in FIG. 7 according to an embodiment of the present disclosure.
Figure 10 shows an example service screen providing information on nutrients and recommended health functional foods necessary for the user's health condition type according to an embodiment of the present disclosure.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings. First, when adding reference numerals to components in each drawing, it should be noted that identical components are given the same reference numerals as much as possible even if they are shown in different drawings. Additionally, in describing the present invention, if it is determined that a detailed description of a related known configuration or function may obscure the gist of the present invention, the detailed description will be omitted.

Various aspects of the invention are described below. It is to be understood that the inventions presented herein may be implemented in a wide variety of forms and that any particular structure, function, or both, presented herein is illustrative only. Based on the inventions presented herein, those skilled in the art will understand that one aspect presented herein can be implemented independently of any other aspects and that two or more of these aspects can be implemented in various ways. You will understand that it can be combined with . For example, an apparatus may be implemented or a method may be practiced using any number of the aspects set forth herein. Additionally, such an apparatus may be implemented or such a method may be practiced using other structure, functionality, or structure and functionality in addition to or other than one or more aspects described herein.

1 is a schematic diagram illustrating an example system for personalized drug interaction analysis and nutraceutical recommendation according to an embodiment of the present disclosure.

As shown in FIG. 1 , an exemplary system 100 according to the present disclosure may be configured to include a communication network 300, a service system 400, and a user terminal 200.

The communication network 300 may provide a communication path through which the system 400 and the user terminal 200 can transmit and receive signals and data to each other, and may include a wired or wireless communication network. Additionally, the communication network 300 may also provide a communication path for the system 400 to communicate with an external communication entity, such as a health checkup information provider 600. The communication network 300 is not limited to a specific communication method, and an appropriate communication method may be used depending on the implementation. For example, when configured as an Internet Protocol (IP)-based system, the communication network 300 may be implemented as an Internet network, and when the user terminal 200 is implemented as a mobile communication terminal, the communication network 300 may be implemented as a cellular network or It can be implemented as a wireless network such as a WLAN (Wireless LAN) network.

The user terminal 200 can access the service system 400 and use the personalized drug interaction analysis and health functional food recommendation service according to the present disclosure. Depending on the implementation, the user terminal 200 may be a fixed computer such as a desktop computer, and may access the system 400 through an Internet browser through a wired/wireless communication network 300 or a dedicated computer provided by the system 400. You can access the system 400 by installing a program or application (App). Additionally, the user terminal 200 may be a portable mobile communication terminal such as a cellular phone, smartphone, laptop computer, tablet PC, etc. For example, when implemented as a smartphone, the user terminal 200 accesses the system 400 through a mobile Internet browser or uses a dedicated program or application (App) provided by the system 400 from the system 400. You can download and install it directly or through a mobile application platform such as Google Play or App Store.

The service system 400 can provide personalized drug interaction analysis and health functional food recommendation services according to the present disclosure to the user terminal 200, and for this purpose, the server 500 and database (DB) 550 are used. It can be included.

The server 500 may be configured to provide drug interaction analysis information and recommended health functional food information customized to the user to the user terminal 200 based on user data. In one implementation, the server 500 may be an application server for providing application services to the user terminal 200, and as described above, an application for providing personalized drug interaction analysis and health functional food recommendation services. It can be provided to the user terminal 200. The database 550 may be configured to store and manage data including user data, drug information, health functional food information, etc. In one implementation, the database 550 may be a relational database in which stored data can be multiplexed for each field to form data.

The health checkup information provider 600 may be configured to provide the requested user's health checkup information to the server 500 in connection with the server 500. For example, if the health checkup information is national health checkup information, the provider 600 may be a national health insurance company, and if the health checkup information is private health checkup information, the provider 600 may provide the information. It may be a private health checkup center, hospital, etc.

2 is a schematic diagram illustrating example components of a server according to one embodiment of the present disclosure.

The server 500 according to an embodiment of the present disclosure may include a user data collection unit 510, a drug interaction analysis unit 520, a health functional food recommendation unit 530, and a nutritional information provision unit 540. You can.

The user data collection unit 510 may be configured to construct user data including the user's health checkup information, survey information, information on medications taken, and information on health functional foods taken. For example, the user data collection unit 510 can collect user information, survey information, information on medications taken, information on health functional foods, etc. from the user terminal 200 by providing a corresponding information input menu, and health checkup information. In connection with the provider 600, the user's health checkup information can be collected from the health checkup information provider 600 through a predetermined user authentication procedure (for example, public authentication, mobile authentication, etc.). In one implementation, the user data collection unit 510 may provide a medicine/health functional food search menu, and register the medicine/health functional food searched and selected by the user terminal 200 as a medicine/health functional food to be taken. You can do it.

The application provided by the server 500 to the user terminal 200 may provide a health functional food scanning function as illustrated in FIG. 7 . In one implementation, this scan function can be implemented as a function that can recognize a specific product using information recognized as characters in a scanned image, such as Google Lens API. As an example, the user terminal 200 can project the target health functional food 210 for which product information is desired through a camera through the scan function of the application, and the user data collection unit 510 can detect the user by using the scan function. It may be configured to acquire health functional food information in the database 550 that matches optically recognized information through the camera of the terminal 200 as target health functional food information. In addition, if there is no health functional food information matching the optically recognized information in the database 550, the user data collection unit 510 determines similar health functional foods based on nutrient information, health functional food classification information, etc. in the optically recognized information. It may be configured to provide information to the user terminal 200. Additionally, in one embodiment, the user terminal 200 may use the above-described scan function to register the medicine/health function food taken, and the user data collection unit 510 may use the medicine/health function obtained through the scan function. It may be configured to register food information as a user's medicine/health functional food.

In one implementation, to build user data, the database 550 may include a user table, a health record table, a survey table, a medicine table, a dietary supplement table, a nutrient recommended intake table, and a user-dose table; This information can be linked to each other as a relational database.

As illustrated in Table 1, the user table can store user information such as ID, name, email, and password.

user table number One User ID (ID) abcde name abc email abcde@*****.com password ******

As shown in Table 2, the health record table is linked to the user ID in Table 1 and contains essential health checkup information (e.g., age, gender, height, weight, vision, hearing, blood pressure, cholesterol, bone density, etc.) for the user. ) as well as additional health examination information (e.g., endoscopy, ultrasound examination, etc.).

health record table number One User ID (ID) abcde age ** gender other key *** weight ** Blood pressure *** .....

The user data collection unit 510 may provide survey items to the user terminal 200 to obtain survey information such as the user's exercise frequency, smoking status, eating habits, family diseases, health goals, etc. By way of example, the survey results for each survey item may be labeled with a predetermined classification and recorded in a survey table. For example, the number of workouts may be labeled with a value of 1 to 5 depending on the number of workouts per week, but is not limited to this and may be labeled based on other criteria, and smoking status, eating habits, etc. may be labeled in a similar manner. Also, for example, health goals are (Brain Health: 0, Sleep: 1, Vitality/Fatigue: 2, Immunity: 3, Stress: 4, Blood Circulation: 5, Hair & Nails: 6, Skin: 7, Eye Health :8, joint/bone health: 9, liver health: 10, intestinal health: 11, weight control: 12, blood sugar control: 13, blood pressure management: 14, blood lipid improvement: 15, menopausal symptoms: 16, .... ), but is not limited to this and may be labeled based on other criteria, and family diseases, etc. may be labeled in a similar manner.

As illustrated in Table 3, the survey table is linked to the user ID in Table 1 and can store information such as number of exercises, smoking status, eating habits, family diseases, health goals, etc. as survey information for the user.

survey table number One User ID (ID) abcde Number of exercises (1~5) 3 Smoking status (1~5) 4 Eating habits (1~5) 2 family disease 0 health goals 2 ......

As shown in Table 4, the drug table can store information such as drug series, drug product name, and nutrient content by type as drug information.

medicine table number One Drug ID One product name Gentamicin line Aminoglycosides Vitamin A *** Vitamin B1 *** Vitamin B2 *** Vitamin B3 *** Vitamin B5 *** Vitamin B6 *** Vitamin B7 *** Vitamin B9 *** Vitamin B12 *** Vitamin C *** Vitamin D *** Vitamin E *** Vitamin K *** boron *** calcium *** chloride (choloride) *** copper *** Fluoride (fluorine) *** iodine *** zinc *** magnesium *** Manganese (Manganese) *** molybdenum *** nickel *** person *** .....

As shown in Table 5, the health functional food table can store information such as health functional food series, health functional food product name, health functional food manufacturer, and content of each type of nutrient.

Health functional food table number One Health functional food ID One product name Centrum Multi Gummy manufacturing company centrum line ****** Vitamin A *** Vitamin B1 *** Vitamin B2 *** Vitamin B3 *** Vitamin B5 *** Vitamin B6 *** Vitamin B7 *** Vitamin B9 *** Vitamin B12 *** Vitamin C *** Vitamin D *** Vitamin E *** Vitamin K *** boron *** calcium *** chloride (choloride) *** copper *** Fluoride (fluorine) *** iodine *** zinc *** magnesium *** Manganese (Manganese) *** Molybdenum *** nickel *** person *** .....

As shown in Table 6, the nutrient recommended intake table is linked to the user ID in Table 1 and can store information such as recommended intake information for each type of nutrient for the user.

Recommended intake table number One User ID (ID) abcde Vitamin A *** Vitamin B1 *** Vitamin B2 *** Vitamin B3 *** Vitamin B5 *** Vitamin B6 *** Vitamin B7 *** Vitamin B9 *** Vitamin B12 *** Vitamin C *** Vitamin D *** Vitamin E *** Vitamin K *** boron *** calcium *** chloride (choloride) *** copper *** Fluoride (fluorine) *** iodine *** zinc *** magnesium *** Manganese (Manganese) *** Molybdenum *** nickel *** person *** .....

The user-dose table can store drug information and health functional food information for each user, and for this purpose, it can be linked to the user ID, drug ID, and health functional food ID as shown in Table 7.

User-taking table number One 2 User ID (ID) abcde fghlj Health functional food ID 3 7 Health functional food ID 5 - Drug ID One 2 .....

The drug interaction analysis unit 520 is configured to analyze the drug interaction between the target health functional food for which information is obtained using the user terminal 200 and the drug or health functional food taken by the user and predict the possibility of side effects. You can. To this end, the drug interaction analysis unit 520 may construct an undirected weighted graph. Specifically, the drug interaction analysis unit 520 uses the target health functional food, the drug taken by the user, and the health functional food taken by the user as nodes, and the interaction between the two nodes as an edge. Drug interactions for users can be analyzed by constructing an undirected weighted graph with interaction risks set on the edges. Additionally, the drug interaction analysis unit 520 may be configured to determine that there is a possibility of side effects between two nodes connected by an edge whose interaction risk is greater than or equal to a preset threshold in the constructed undirected weighted graph.

In one embodiment, the interaction risk is the risk of interaction between nutrients contained in medicines/health functional foods and can be set based on knowledge of pharmaceutical science. Additionally, the interaction risk established based on pharmaceutical knowledge can be updated using statistical techniques such as moment correlation coefficient to further increase risk accuracy. For example, the risk of interaction between nutrients can be set within a risk value range from 0 to 5, and the threshold can be set to 3.0. In this case, if the risk of interaction between two nutrients is 3.0 or more, side effects between the two nutrients can occur. It can be determined that it is a possibility. In addition, for example, a drug/health functional food (first node) contains two or more nutrients, and for each of these nutrients, drug interactions with nutrients contained in other drugs/health functional foods (second node) When present, the undirected weighted graph may be configured to have two or more edges between the first node and the second node.

Regarding the undirected weight graph illustrated in FIG. 4, the user is taking contraceptive pill A (node A), a medicine, and multivitamin B (node C), a health functional food, and St. Petersburg as a target health functional food. Information on Johns Wort (Node B) can be obtained through the scan function of the user terminal 200. In this case, the interaction 'Heartburn' (with an interaction risk of 0.6) and the interaction 'Dizziness' (with an interaction risk of 0.37) between the nutrients contained in the two products being taken (node A and node C). If ' exists, as shown in FIG. 4, two edges can be formed on the graph between node A and node C to express the above interactions. In addition, the oral medication, Contraceptive Pill A (Node A), and the target health functional food, St. Interactions between nutrients contained in Johns Wort (Node B) include the interaction 'Dizziness' (with an interaction risk of 2.89) and the interaction 'Reduced Contraceptive Pill A Efficacy' (with an interaction risk of 4.83). In this case, as shown in FIG. 4, two or more edges including the above interactions may be formed between node A and node B on the graph. In this undirected weighted graph of FIG. 4, the drug interaction analysis unit 520 does not determine that interactions below the interaction risk threshold (3.0 in this example) are likely to have side effects, but nodes exceeding the threshold It may be determined that there is a possibility of side effects regarding the interaction 'decreased efficacy of contraceptive A' expressed in one of the trunk lines between A and node B, and the possibility of such side effects may be notified to the user terminal 200. As described above, the drug interaction analysis unit 520 can analyze drug interactions that exist between medicines and health functional foods or between health functional foods by constructing an undirected weighted graph.

Additionally, in one embodiment, the drug interaction analysis unit 520 may also determine the possibility of side effects due to overdose of a specific nutrient. For example, if the user's recommended intake of vitamin A is 3000 mcg and the vitamin A content of the health functional food the user is taking is 750 mcg, and the vitamin A content of the target health functional food selected by the user is 2500 mcg, drug interaction The analysis unit 520 may determine that there is a possibility of side effects due to overdose of vitamin A when taking the target health functional food together and notify the user terminal 200 of this.

In addition, in one embodiment, the drug interaction analysis unit 520 analyzes drug interactions between medicines/health functional foods currently being taken registered by the user, verifies whether there is a possibility of side effects among them, and displays the drug interaction analysis unit 520 in the user terminal 200. ) can be configured to provide.

The health functional food recommendation unit 530 determines the user's health condition type corresponding to the user's health condition obtained from the user's health checkup information and survey information, and recommends candidate health functional foods for the nutrients required for the health condition type. It may be configured to determine recommended health functional foods that reflect preferences for the corresponding health condition type among the candidate health functional foods. For this purpose, the health functional food recommendation unit 530 provides health status information (e.g., age, gender, blood pressure, bone density, etc.) from the user's health checkup information and health status information (e.g., For example, all or part of the number of exercises, smoking status, health goals, etc.) can be obtained as the user's health status, and Health condition types (e.g., eye health need type, skin protection need type, bone reinforcement need type, etc.) to classify health condition can be defined.

In one implementation, the health functional food recommendation unit 530 may be configured to apply a K-Nearest Neighbor (KNN) algorithm to determine which health condition type the user's health condition can be classified into. there is. To this end, the health functional food recommendation unit 530 may define health state types that match nutrient information in which normalized values are assigned to each nutrient according to the degree of need for each type of health state. For example, the degree to which a particular nutrient is needed for a particular type of health condition can be predicted based on pharmacological knowledge and normalized by nutrient to give a value between 0 and 1 through min-max normalization as represented in Table 8. It can be set to the specified number.

Here, _X is a variable representing the level of a specific nutrient, and

For example, in this way, for three types of health conditions (eye health needs, skin protection needs, and bone strength needs), three nutrients (vitamin A, calcium, and manganese) would have the A normalized value may be set.

category nutrient Vitamin A calcium manganese eye health 0.87 0.34 0.09 skin protection 0.15 0.21 0.92 bone reinforcement 0.19 0.94 0.25

The health functional food recommendation unit 530 determines the user's health state type corresponding to the user's health state through the KNN algorithm (i.e., classifies the user's health state into a specific health state type) and responds to the determined health state type. It can be configured to determine the necessary nutrients. In one implementation, the health functional food recommendation unit 530 may determine a specific nutrient whose normalized value of the nutrient is more than a preset threshold (e.g., 0.8) for a specific health condition type as a necessary nutrient for the specific health type. there is. For example, in the example in Table 9, when the user's health condition type is determined to be the eye health need type, vitamin A with a normalized value (0.87) greater than the preset threshold (0.8) is the nutrient required for the eye health type. can be decided. The health functional food recommendation unit 530 may search the database 550 to determine candidate health functional foods containing nutrients necessary for the determined health condition type.

The health functional food recommendation unit 530 recommends among the searched candidate health functional foods in the order of highest predicted preference in the user group belonging to the user's health condition type according to the user-based collaborative filtering (UCF: User-based Collaborative Filtering) algorithm. It can be configured to determine health functional foods. For example, the health functional food recommendation unit 530 recommends a predetermined number (e.g., 3, 5) in order of preference evaluated in a user group of the same health condition type (e.g., eye health need type) Information on recommended health functional foods (10, 10, etc.) can be sorted in descending order and provided to the user terminal 200, but is not limited to this and other display methods (e.g., ranking display, color display by ranking, etc.) Information on recommended health functional foods can be provided.

For example, preference prediction by user-based collaborative filtering can be done as expressed in Table 10.

Here, S _i,N is the similarity vector of the top-N items with the highest similarity to item i, and R _u,N is the actual rating vector for the top-N items with the highest similarity to user u and item i. , represents the personalized predicted rating value of user u and item i. Additionally, similarity measurement can be done by applying cosine similarity, Pearson similarity, or Euclidean distance. In preference prediction through collaborative filtering, recommended health functional foods and preference values can be updated as the amount of preference information of users belonging to the corresponding health condition type increases, and recommendation reliability can be improved accordingly.

In one implementation, the health functional food recommendation unit 530 may be configured to provide recommended health functional food information in connection with the drug interaction analysis unit 520. Specifically, the health functional food recommendation unit 530 excludes the medicine or health functional food taken by the user and the health functional food determined to have a possibility of side effects by the drug interaction analysis unit 520 from among the recommended health functional foods. Alternatively, it may be configured to provide information on the recommended health functional food along with information on the possibility of such side effects to the user terminal 200.

The nutritional information provision unit 540 provides dosage information for each nutrient compared to the recommended intake amount of the nutrients contained in the health functional foods being taken in the user's health functional food information, and determines whether each health functional food being taken is necessary. It can be configured to display. For example, the nutrition information provider 540 can provide information comparing the dosage of each nutrient contained in the health functional food being taken with the recommended intake amount as a user nutrition graph, as illustrated in FIG. 6, through which It can provide a clear visual indication of whether a specific nutrient is insufficient or exceeds the recommended intake amount. In addition, the nutritional information provision unit 540 may be configured to list health functional foods being taken, display health functional foods that are judged unnecessary to be taken because the dosage exceeds the recommended intake, and provide the information to the user terminal 200. there is.

In one embodiment, the nutritional information providing unit 540 may be configured to recommend additional health functional food information that can supplement specific nutrients whose nutrient dosage is less than the recommended intake, and the above-described collaborative filtering algorithm may be used to recommend additional health functional foods. It can also be applied to food recommendations. Additionally, in one embodiment, the nutritional information providing unit 540 is configured to recommend other health functional foods that meet the user's health goals or contain nutrients that the user needs or have not taken, based on the user's health condition. This can be done, and the above-described collaborative filtering algorithm can also be applied to these other health functional food recommendations.

Figure 3 is an exemplary flowchart showing a personalized health functional food recommendation method according to an embodiment of the present disclosure.

As shown in FIG. 3, the health functional food recommendation service system 400 builds user data including health checkup information, survey information, medication information, and health functional food information of the user of the user terminal 200. Can (110). As described above, health checkup information can be obtained from the health checkup information provider 600. The service system 400 may use the user terminal 200 to obtain target health functional food information (for example, through a scan function provided by an application or through a health functional food search function) (120). . The service system 400 can analyze the drug interaction between the target health functional food information obtained (for example, by constructing an undirected weight graph) and the user's medicine or health functional food and predict the possibility of side effects ( 130). The service system 400 may determine recommended health functional foods personalized to the user (140). To this end, the service system 400 determines the user's health status type corresponding to the user's health status obtained from the user's health checkup information and survey information (e.g., by applying the K-least closest algorithm) (e.g., by applying the K-least closest algorithm) or classification) and determine the nutrients required for the relevant health condition type, and determine candidate health functional foods for the determined required nutrients. Additionally, the service system 400 may determine recommended health functional foods that reflect preferences in the user's health status type from among the candidate health functional foods determined (eg, by applying a user-based collaborative filtering algorithm). The service system 400 may provide the user terminal 200 with a user nutrition graph based on the drug interaction results analyzed in step 130, recommended health functional food information determined in step 140, and health functional foods taken by the user. In addition, when additional target health functional food information is obtained through the user terminal 200, the service system 400 may provide the result of step 130 to the user terminal 200 for the additional target health functional food.

Figure 5 shows an example service screen of a health functional food recommendation application to which a user is logged in according to an embodiment of the present disclosure.

As described above, the service system 400 may provide an application for personalized drug interaction analysis and health functional food recommendation according to the present disclosure to the user terminal 200. When a user logs in to the application, the application provides an exemplary service screen 700 (e.g., a display screen when the user terminal 200 is a smartphone) as shown in FIG. 5 to the user terminal 200. You can. By way of example, the application may display a user image (or photo or icon) and basic user information (eg, name, age, height, weight, etc.) on the service screen (710). Additionally, the application may provide a menu 720 for obtaining the user's health checkup information. For example, when user authentication (e.g., public authentication, mobile authentication, etc.) is performed after clicking the menu 720, the service system 400 receives the user terminal 200 from the health checkup information provider 600. Health checkup information can be obtained and the obtained information can be stored in the database 550.

Additionally, the application may provide a scan menu 730 for obtaining target health functional food information. For example, when the scan menu 730 is selected, the user of the user terminal 200 can point the camera at the target health functional food 210 for which information is desired (e.g., service screen 731 in FIG. 7). , When the optically recognized information is matched with health functional food information in the database 550, the service system 400 can obtain the matched health functional food information as target health functional food information. Additionally, the application may provide a menu 740 for providing nutritional information to the user based on information on health functional foods taken by the user, which is described later in FIG. 6 .

6 illustrates an example service screen providing a user nutritional graph according to an embodiment of the present disclosure.

For example, when the user nutritional information provision menu 740 shown in FIG. 5 is selected, the application may provide an example service screen 700 as shown in FIG. 6 to the user terminal 200. The service screen 700 of FIG. 6 may provide a user nutrition graph 750 and a list 760 of health functional foods taken by the user. The user nutrition graph 750 is a graph of dose ('current nutritional content' in Figure 6) information compared to the recommended intake amount ('required nutrient content' in Figure 6) for each nutrient included in the health functional foods that the user is taking. It can be provided as . In the example of FIG. 6, for the nutrient vitamin D 751, the user nutrition graph 750 may display that the recommended intake is 1000 IU, but the dosage is 1400 IU. In addition, the user nutrition graph 750 can display the maximum intake amount of the corresponding nutrient. In the example of FIG. 6, for the nutrient vitamin A 752, the recommended intake amount is 700 mcg and the maximum intake amount 753 is 3000 mg, but the user's intake amount 753 is 3000 mg. The dosage can be displayed as 9477mcg. As such, when the dose of vitamin A 752 exceeds the maximum intake amount 753, the user nutrition graph 750 may display a warning 754 of possible side effects due to overdosage of vitamin A 752.

The health functional food list 760 can list the health functional foods that the user is taking and can indicate whether each health functional food is necessary. For example, in the example of Figure 6, the nutrient vitamin A (752) is an overdose of 9477mcg, so if you do not take 'vvvvitamin A 7500mcg' in the list (760), the user's vitamin A dose is (9477- 7500 mcg) < maximum intake (3000 mcg). In this case, 'vvv vitamin A 7500mcg' may be displayed (761) in the health functional food list (760) as an unnecessary health functional food.

FIG. 7 is an exemplary diagram illustrating scanning a target health functional food through the scan function of an application according to an embodiment of the present disclosure. As shown in the service screen 731 of FIG. 7, the target health functional food 210 is displayed. When illuminated by the camera of the user terminal 200, information displayed on the food may be optically recognized.

FIG. 8 shows an exemplary service screen for obtaining and providing information on health functional foods scanned in FIG. 7 according to an embodiment of the present disclosure.

In FIG. 7 , when the optically recognized information about the target health functional food 210 matches the health functional food information in the database 550, the application displays an example service screen 700 as shown in FIG. 8 on the user terminal. It can be provided as (200). The service screen 700 of FIG. 8 may provide information 770 on the target health functional food, a nutrient graph 771 of the target health functional food, and user evaluation information 772. Information 77 of the target health functional food may provide information including the product's image, product name, manufacturer, price, capacity, country of origin, etc. The nutrient graph 771 of the target health functional food can provide information on the content (dose) compared to the recommended intake amount of the corresponding nutrient in a graph form. User evaluation information 772 may provide information including other users' ratings and evaluation opinions regarding the health functional food. Additionally, the service screen 700 of FIG. 8 may additionally provide a scan menu 730 so that other target health functional foods can be additionally scanned.

FIG. 9 shows an example service screen that provides information on the possibility of a user's side effects with respect to the health functional food scanned in FIG. 7 according to an embodiment of the present disclosure.

In FIG. 7 , when the optically recognized information about the target health functional food 210 matches the health functional food information in the database 550, the application displays an example service screen 700 as shown in FIG. 9 on the user terminal. It can be provided as (200). The target health functional food information 770, the nutrient graph of the target health functional food, and the user evaluation information 772 provided by the service screen 700 of FIG. 9 are the same as described above with reference to FIG. 8. In addition, as described above, the service system 400 can analyze drug interactions between the target health functional food 210 for which information was obtained and the medicine/health functional food taken by the user, and if there is a possibility of side effects between them. As shown in FIG. 9, such information 773 and 774 can be provided to the user terminal 200. For example, in the example of Figure 9, when the target health functional food 210 is 'St Johns Wort 1000mg', the application consumes the target health functional food 210 in excess of the daily maximum (i.e., maximum intake amount). In this case, information on side effects that may occur (773) can be provided, and information on side effects (774) that reduce the efficacy of the contraceptive between the contraceptive pill, which is a medicine that the user is taking, and the target health functional food (210) can be provided.

Figure 10 shows an example service screen that provides information on nutrients and recommended health functional foods necessary for the user's health condition type according to an embodiment of the present disclosure.

As described above, the service system 400 may determine (or classify) the user's health state type (e.g., by applying a K-least closest algorithm) and determine nutrients required for the health state type. For example, in the example of FIG. 10, the user's determined health condition type is '50s/female/menopause' and the nutrients required for this health condition type may be vitamins C, D, K, and calcium (810). The service system 400 can search the database 550 for candidate health functional foods for each required nutrient (e.g., by applying a user-based collaborative filtering algorithm) and select the health condition type among the candidate health functional foods. The preference prediction can be reflected to determine recommended health functional foods and provide information on the corresponding health functional foods to the user terminal 200. For example, in the example of FIG. 10, the application sorts the information (812, 813, 814) of recommended health functional foods for vitamin D (811), a nutrient necessary for the user, in order of preference, and provides the information to the user terminal (200). It can be provided as . In addition, although not separately shown in FIG. 10, the application may provide recommended health functional food information to the user terminal 200 in the same manner for vitamins C, D, and calcium, which are other nutrients necessary for the user.

It is to be understood that any particular order or hierarchy of steps in any presented processes is an example of exemplary approaches. It is to be understood that the specific order or hierarchy of steps in processes may be rearranged within the scope of the present invention, based on design priorities. The appended method claims present elements of the various steps in an exemplary order but are not meant to be limited to the particular order or hierarchy presented.

As used herein, the terms "component", "unit (or part)", "module", "system", etc. refer to a computer-related entity, hardware, firmware, software, a combination of software and hardware, or an implementation of software. You can. For example, a component may be, but is not limited to, a process running on a processor, a processor, an object, a thread of execution, a program, and/or a computer. For example, both an application running on a computing device and the computing device can be a component. One or more components may reside within a processor and/or thread of execution, and a component may be localized within one computer, or distributed between two or more computers. Additionally, these components can execute from various computer-readable media having various data structures stored thereon.

The description of the presented embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments without departing from the scope of the invention. Thus, the present invention is not limited to the embodiments presented herein, but is to be construed in the broadest scope consistent with the principles and novel features presented herein.

200: user terminal
300: communication network
400: Service system
500: Server
510: User data collection unit
520: Drug interaction analysis unit
530: Health functional food recommendation section
540: Nutrition information section
550: database
600: Health checkup information provider

Claims (15)

As a personalized health functional food recommendation system,
A server configured to provide drug interaction analysis information and recommended health functional food information customized to the user to the user terminal based on user data; and
It includes a database configured to store and manage the user data, drug information, and health functional food information,
The server is,
a user data collection unit configured to construct the user data including the user's health checkup information, survey information, information on medications taken, and information on health functional foods taken;
a drug interaction analysis unit configured to analyze drug interactions between the target health functional food, information about which is obtained using the user terminal, and the taking medicine or the taking health functional food and predict the possibility of side effects; and
Determining a health state type of the user corresponding to the health state of the user obtained from the health checkup information and the survey information, determining candidate health functional foods for nutrients necessary for the health state type, and determining the candidate health condition. Comprising a health functional food recommendation unit configured to determine recommended health functional foods that reflect preferences in the health status type among health functional foods,
Personalized health functional food recommendation system. According to claim 1,
The drug interaction analysis unit,
The target health functional food, the medicine to be taken, and the health food to be taken are each taken as a node, the interaction between the two nodes is taken as an edge, and an undirected weighted graph is formed with an interaction risk set on each edge. configured to analyze drug interactions for the user and determine that there is a possibility of side effects between two nodes connected by a trunk line where the interaction risk is greater than or equal to a preset threshold,
Personalized health functional food recommendation system. According to claim 1,
The health functional food recommendation section,
According to the degree of need for each health state type, health state types are defined that match nutrient information in which normalized values are assigned to each nutrient, and the user's health state is determined through the KNN (K-Nearest Neighbor) algorithm. configured to determine a corresponding health state type of the user and determine the necessary nutrients corresponding to the determined health state type,
Personalized health functional food recommendation system. According to claim 3,
The health functional food recommendation section,
Among the candidate health functional foods, the recommended health functional foods are determined in the order of highest preference predicted in the user group belonging to the user's health status type according to a user-based collaborative filtering (UCF) algorithm. felled,
Personalized health functional food recommendation system. According to claim 4,
The health functional food recommendation section,
Among the recommended health functional foods determined above, health functional foods determined by the drug interaction analysis unit to have a possibility of side effects with the medicine or health functional food being taken are excluded, or information on the recommended health functional food along with the possibility of side effects is excluded. configured to provide,
Personalized health functional food recommendation system. According to claim 1,
The server provides an application for providing the drug interaction analysis information and the recommended health functional food information to the user terminal, and the application includes a health functional food scan function,
The user data collection unit,
Configured to acquire health functional food information in the database that matches information optically recognized through the camera of the user terminal by the health functional food scan function as target health food information,
Personalized health functional food recommendation system. According to claim 1,
A nutritional information provision section is provided to provide dosage information for each nutrient compared to the recommended intake amount of the nutrients contained in the health functional foods being taken in the health functional food information being taken, and to indicate whether each health functional food being taken is necessary. containing,
Personalized health functional food recommendation system. A personalized health functional food recommendation method performed by a health functional food recommendation system,
Building user data including the user's health checkup information, survey information, information on medications taken, and information on health functional foods taken;
Obtaining target health functional food information using a user terminal;
Analyzing drug interactions between the target health functional food and the taking medicine or the taking health functional food and predicting the possibility of side effects;
determining a health state type of the user corresponding to the health state of the user obtained from the health checkup information and the survey information and determining candidate health functional foods for nutrients required for the health state type; and
Comprising the step of determining recommended health functional foods that reflect preferences in the health status type among the candidate health functional foods,
How to recommend personalized health functional foods. According to claim 8,
The step of analyzing drug interactions and predicting the possibility of side effects is,
Constructing an undirected weighted graph in which each of the target health functional food, the medicine to be taken, and the health functional food to be taken are used as nodes, the interaction between the two nodes is used as an edge, and an interaction risk is set for each edge; and
Comprising the step of determining that there is a side effect between two nodes connected by an edge in the undirected weighted graph where the interaction risk is greater than or equal to a preset threshold,
How to recommend personalized health functional foods. According to claim 8,
The step of determining the candidate health functional foods is,
Defining health state types that match nutrient information in which normalized values are assigned to each nutrient according to the degree of need for each health state type; and
Determining a health state type of the user corresponding to the health state of the user through a K-least closest (KNN) algorithm and determining the necessary nutrients corresponding to the determined health state type,
How to recommend personalized health functional foods. According to claim 10,
The step of determining the recommended health functional foods is,
Among the candidate health functional foods, determining the recommended health functional foods in order of highest preference predicted in a user group belonging to the user's health status type according to a user-based collaborative filtering (UCF) algorithm,
How to recommend personalized health functional foods. According to claim 11,
The step of determining the recommended health functional foods is,
Further comprising the step of excluding from the determined recommended health functional foods a health functional food determined to have a possibility of side effects with the drug or the health functional food taken, or providing information on the recommended health functional food together with the possibility of side effects,
How to recommend personalized health functional foods. According to claim 8,
The health functional food recommendation system provides a health functional food recommendation application to the user terminal, and the application includes a health functional food scan function,
The step of acquiring the target health functional food information is,
Comprising the step of acquiring health functional food information in the database of the system that matches information optically recognized through the camera of the user terminal by the health functional food scan function as target health food information,
How to recommend personalized health functional foods. According to claim 8,
The above method is,
Providing dosage information for each nutrient compared to the recommended intake amount of the nutrients contained in the health functional foods being taken in the health functional food information being taken, and further comprising the step of indicating whether each health functional food being taken is necessary,
How to recommend personalized health functional foods. A computer program stored on a computer-readable medium, comprising computer-executable instructions for carrying out the method according to any one of claims 8 to 14.