Dietary survey method and evaluation system based on time, food and environment dimensions

Abstract

The invention provides a dietary survey method and evaluation system based on time, food and environment dimensions. The method comprises the following steps: collecting data of time, food and environment dimensions for evaluating an adult diet mode; according to attributes of time, food and environment information, parameters required by the diet mode evaluation model are generated respectively, data are converged, analyzed and cleaned, and scores of all composition factors are calculated; performing unified management on the analyzed and cleaned data, establishing an adult diet mode database, and updating an expert database; integrating the merged data introduced into the expert database with the user information of the mechanism to generate a historical attribute set, matching the historical attribute set with the diet mode model, and storing the matching result in a cloud end; and calling the diet mode and expert opinions of the current to-be-evaluated object to perform visual display, outputting pre-judgment information, and displaying the pre-judgment information to a user interface to prompt the to-be-evaluated object to perform improvement. The method is used for rapidly analyzing and presenting the dietary behavior of the investigated object in a short time.

Description

Technical Field

[0001] The present invention relates to the field of combining basic research in nutrition with computer technology, and more specifically, to a dietary survey method and evaluation system based on time, food and environment dimensions. Background Art

[0002] In traditional nutrition research, nutrients, specific foods and food groups, and more macro dietary patterns have always been hot topics that scholars have been concerned about, and the research methods and results have been relatively mature. However, with the continuous development of society, the differences in modern people's diets are no longer limited to the level of dietary food, but are gradually reflected in a wider dimension, and there are eating habits and methods that are not bound by the traditional three meals a day. From the perspective of "meal (Meal / Eating occasion / Eating episode)", people's diets have gradually shown a diversified pattern, such as skipping breakfast, eating late at night, irregular meals, eating more and less meals, often eating takeout or dining out, and other bad eating habits and emerging eating methods such as intermittent fasting. If we still only approach it from the perspective of dietary food, trying to improve the health hazards brought about by these bad eating habits is incomplete and challenging. Therefore, the nutrition community has begun to look at diet from a multidimensional perspective rather than being limited to traditional nutrition research.

[0003] In view of this, the foreign nutrition community has started to study the eating / meal pattern: In 1999, in a comparative study conducted in Nordic countries, In order to investigate whether regular eating will be replaced by irregular eating patterns, and to avoid the differences in subjective cognition of people in different social and cultural backgrounds about eating, et al. proposed a three-dimensional eating system model: eating pattern, meal format and social organization of eating. Since 2015, the multidimensionality of diet has begun to receive much attention, especially chrono-nutrition, which has become a hot topic of concern for many scholars. Leech et al. pointed out at the time that nutrition research was still limited to nutrients or dietary patterns, and rarely explored nutrient intake and dietary quality from the level of "meals". Therefore, based on the predecessors, patterning, format and context were used to further adjust the content of the three dimensions to make it more suitable for the use environment of nutrition research. Until 2021, the dietary pattern was further optimized by Cathal O'Hara and Eileen R Gibney in the field of statistical data mining technology into three dimensions: temporal, content and context, namely time dimension, food dimension and environment dimension.

[0004] Research on dietary patterns is gradually progressing. Most studies are still based on traditional dietary survey methods to collect dietary information. As a survey of the intake of various foods and nutrients for individuals, families or groups over a certain period of time, they can indeed accurately collect information on certain food dimensions, but it is difficult to analyze at the "meal" level. Furthermore, investigators still need to add questions about the time and environmental dimensions on their own. If the retrospective method or record method is used at the same time, it will bring more survey burdens to respondents on the basis of frequent reporting of food information, reducing the reliability of the research results. In addition, since there is no consensus on the selection of survey methods and the setting of supplementary questions, the reliability and validity have not been tested, resulting in a lack of comparability between research results, and its own credibility is greatly reduced.

[0005] The development of standardized survey methods will enable future dietary pattern studies to overcome the above-mentioned difficulties, improve the comprehensiveness of the survey, and the reliability, comparability, and credibility of the research results, thereby providing more accurate dietary information and recommendations for the formulation of guidelines to promote health. Summary of the invention

[0006] In view of the technical problems existing in the prior art, the present invention provides a dietary survey method and an evaluation system based on time, food and environment dimensions, so as to quickly analyze and present the dietary behavior of the survey subjects in a short time.

[0007] According to a first aspect of the present invention, a dietary survey method based on time, food, and environment dimensions is provided, comprising:

[0008] Step 1: Collect dietary behavior data for evaluating adult dietary patterns, including data on time, food, and environment, and provide timely reminders and precise guidance to monitor and record users’ diet in real time;

[0009] Step 2: Generate the parameters required for the dietary pattern assessment model based on the attributes of time, food, and environmental information, merge the data, parse and clean them, and calculate the scores of each component factor;

[0010] Step 3: Unify the management of the analyzed and cleaned data, establish an adult dietary pattern database, and update the expert database;

[0011] Step 4: Integrate the merged data after the introduction of the expert database with the user information of the institution to generate a historical attribute set, match the historical attribute set with the dietary pattern model, and store it in the cloud;

[0012] Step 5: retrieve the dietary pattern and expert opinions of the current subject to be evaluated for visual display, output prediction information, and display it on the user interface to prompt the subject to be evaluated to make improvements.

[0013] Based on the above technical solution, the present invention can also make the following improvements.

[0014] Optionally, the time dimension includes: the number of meals, snacks and snacks, the time of three meals, nighttime eating, eating time window, eating-sleep interval, eating differences between weekdays and weekends, monthly, quarterly and annual dietary patterns, long-term dietary preferences, binge eating, types and duration of restrictive eating behaviors, and specific dietary patterns;

[0015] The food dimensions include: energy distribution based on meal times, food group intake based on meal times, food group intake, and food cooking methods;

[0016] The environmental dimension includes items such as eating at home or outside, eating alone or with others, activities during mealtime, influence of the dining environment, and food accessibility.

[0017] Optionally, the dietary behavior data that can be used to evaluate the dietary patterns of adults are collected using smart wearable devices and smart terminals, and the smart wearable devices and smart terminals include interfaces developed with smart watches and health monitoring bracelet devices, and the smart wearable devices and smart terminals are provided with modules that can automatically remind and guide the wearer to respond to meal times and record food intake in real time.

[0018] Optionally, the parameters required for generating the dietary pattern assessment model according to the attributes of time, food and environmental information include:

[0019] The acquired entry data is parsed using a preset rule engine and rule algorithm, and the parsed data is cleaned, and the parameters required for the dietary pattern assessment model are generated according to the attributes of time, food and environmental information; the parameters include: continuous variables composed of time parameters and intake parameters, and categorical variables composed of frequency and time period information groups.

[0020] Optionally, the parameters required for the dietary pattern assessment model are represented by the user's answers in three dimensions: time, food, and environment, at least one of these three parameters or a combination of several of them. When the parameters of the three dimensions are all included in the assessment, the model is optimal.

[0021] Optionally, the converging data and performing parsing and cleaning to calculate the score of each component factor includes:

[0022] Generate corresponding parameters according to the attributes of feedback information and the variable type of parameters, and determine the corresponding parameter values. Synchronize the cleaned indicator data and their component factors into the database. According to the characteristics of different component factors, all component factors are dimensionlessly processed, and the reverse indicators are positively converted to calculate the scores of each component factor.

[0023] Optionally, the unified management of the analyzed and cleaned data, establishing an adult dietary pattern database, and updating the expert database includes:

[0024] The scores of each factor are added together to obtain the dietary pattern model score. Based on the dietary pattern model score, an adult dietary pattern model is formed. The parsed and cleaned data are unified according to standard specifications to obtain unified standardized data. The standardized data is stored to establish an adult dietary pattern database, and the expert database is continuously updated based on this.

[0025] Optionally, matching the historical attribute set with the dietary pattern model includes:

[0026] A trial collection of samples from several adults is conducted, and the dietary pattern model collected by the dietary pattern system is directly matched with the historical attribute set input by the user or obtained from hospitals and institutions. The collected samples are classified, and the historical attribute set and the dietary pattern model are probabilistically matched.

[0027] Optionally, the step of retrieving the current diet pattern and expert opinions of the subject to be evaluated for visual display, outputting the prediction information corresponding to the time, food, and environment dimensions of the subject to be evaluated, and displaying it on the user interface to prompt the subject to be evaluated to make improvements includes:

[0028] After obtaining the information of the subject to be evaluated, information is exchanged with the information systems of hospitals, welfare homes, and training camps to retrieve the current dietary patterns and expert opinions of the subject to be evaluated for visual display. The preprocessing module preprocesses the current status of the time, food, and environmental dimensions of the subject to be evaluated, outputs the prediction information corresponding to the time, food, and environmental dimensions of the subject to be evaluated, and displays it on the user interface to prompt the subject to be evaluated to make corresponding improvements.

[0029] According to a second aspect of the present invention, the present invention provides a dietary assessment system based on time, food, and environment dimensions, comprising:

[0030] The user diet information collection module collects dietary behavior data for evaluating adult dietary patterns, including data in three dimensions: time, food, and environment, and provides timely reminders and precise guidance to monitor and record users' diets in real time;

[0031] The data processing module is connected to the input module through wireless and / or wired communication; the data processing module includes a time dimension calculation unit, a food dimension calculation unit and an environment dimension calculation unit; wherein the time dimension calculation unit calculates time data through a preset time calculation model to generate time dimension feature data; the food dimension calculation unit calculates food data through a preset food calculation model to generate food dimension feature data; the environment dimension calculation unit calculates environment data through a preset environment calculation model to generate environment dimension feature data;

[0032] The cloud integration terminal is connected to the data processing module through wireless communication to realize the transmission and interaction of information in three dimensions. Based on its built-in integration logic, it integrates the corresponding data processing modules to form an overall evaluation and reads expert suggestions from the expert database;

[0033] Information terminal, which further processes the data of the cloud integrated terminal and transmits it to the local area through wireless communication connection and completes visualization;

[0034] Feedback terminal, used to display data processing results and expert advice for easy reference by adults who accept the system survey;

[0035] and expert pool.

[0036] Technical effects and advantages of the present invention:

[0037] The present invention provides a dietary survey method and evaluation system based on the dimensions of time, food, and environment. Starting from the dimensions of time, food, environment, etc., the present invention systematically explores a series of dietary pattern indicators. The system covers rich content and conducts a multi-dimensional survey on the diet of the population, which is consistent with the requirements of the dietary guidelines. Compared with the review method and the record method, it is less burdensome, and compared with the food frequency questionnaire, it supplements the questions of the time dimension and the environmental dimension, and improves on the basis of the traditional dietary survey. The system has a complete set of collection, cleaning, merging, and storage processes, and combines the expert database and user interaction to give instant feedback to the evaluation object. From the perspective of traditional nutrition and chrono-nutrition, it improves the user's eating habits in many aspects, prevents diseases that may be caused by dietary factors, and helps achieve the goal of a healthy China. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A flow chart of a dietary survey method based on time, food and environment dimensions provided by an embodiment of the present invention;

[0039] Figure 2 A schematic diagram of a dietary pattern system provided in an embodiment of the present invention.

[0040] In the figure, 101, 102, 103-input modules; 201, 202, 203-data processing modules; 300, cloud integration terminal; 400, information terminal; 500, feedback terminal; 600, expert database. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] Today's eating patterns can be roughly summarized as follows: ① Time dimension: number of meals, meal times, eating windows and regularity; ② Food dimension: food combinations in each meal and the order of food intake during the day; ③ Environmental dimension: dining locations, social groups and accompanying activities.

[0043] The embodiment of the present invention is based on the existing standardized dietary survey method, and explores a more comprehensive standardized survey method with high reliability and validity in the innovative research of multiple dietary pattern questionnaires. The computer system is used to quickly calculate and obtain relevant information on individual dietary behavior, namely, the Eating Pattern System (EPS).

[0044] It is understandable that based on the defects in the prior art, the embodiment of the present invention proposes a dietary survey method based on the dimensions of time, food, and environment, as shown in the following example: Figure 1 As shown, the method is as follows:

[0045] Step 1: Collect dietary behavior data for evaluating adult dietary patterns, including data on time, food, and environment; and provide timely reminders and precise guidance to monitor and record users’ diets in real time;

[0046] In this embodiment, smart wearable devices and smart terminals are used to collect the dietary behavior data that can be used to evaluate adult dietary patterns. The smart wearable devices and smart terminals include: developing interfaces with smart watches and health monitoring bracelets to achieve automatic synchronization of dietary data and reduce the burden of manual input by users. Since the system mainly collects information on meal frequency and time, food types and consumption, and meal location, unlike sleep or exercise, which can be automatically monitored by smart bracelets, we add a module to traditional smart wearable devices (such as bracelets) that can automatically remind and guide the wearer to respond to meal times and record food intake in real time, and then determine the wearer's dining environment through the positioning system.

[0047] The time dimension includes items such as: the number of meals, snacks and snacks (a meal is defined as the largest meal in the three time periods of 6:00-10:00, 11:00-15:00 and 17:00-21:00, i.e. breakfast, lunch and dinner. Please note that fruits can be included; various small amounts of food and drinks consumed outside of meals, but excluding water. Please note that fruits can be included; the time interval between each independent eating event is 15 minutes), three meal times (24-hour system), nighttime eating, eating time windows, eating-sleep intervals, eating differences between weekdays and weekends, monthly, quarterly and annual eating patterns, long-term eating preferences, binge eating, types and duration of restrictive eating behaviors, and specific dietary patterns;

[0048] The food dimension includes items such as: energy distribution based on meals, food group intake based on meals (the intake is based on the estimated portion size of the atlas, solid food is measured in grams, and liquid food is measured in milliliters; different foods in each food group must be considered separately. If the intake varies greatly, the most consumed representative food will be used as the substitute; if the difference is small, the average value will be calculated), intake of specific food groups (snacks, beverages, and seasonal fruits), and food cooking methods;

[0049] The environmental dimension includes items such as: eating at home or outside, eating alone or with others, activities during meals, the impact of the dining environment, and food accessibility. According to the needs of the investigators and users, determine the indicators that need to be collected under the three dimensions of adult dietary patterns. According to the actual situation, a number of items suitable for the surveyed population should be selected. For example, the collection of "differences in diet between weekdays and weekends, monthly, quarterly and annual dietary patterns, long-term dietary preferences, binge eating, types and duration of restrictive dietary behaviors, and specific dietary patterns" requires appropriate extension of the collection time and frequency.

[0050] This embodiment uses smart wearable devices and smart terminals to collect data, provide timely reminders and precise guidance, and thus monitor and record the user's diet in real time.

[0051] Step 2: Generate the parameters required for the dietary pattern assessment model based on the attributes of time, food, and environmental information, merge the data, parse and clean them, and calculate the scores of each component factor;

[0052] This embodiment uses a preset rule engine and rule algorithm to parse the acquired entry data, and cleans the parsed data; according to the attributes of time, food and environmental information, the parameters required for the dietary pattern assessment model are generated respectively. Except for the time parameter and the intake parameter which are continuous variables, the other parameters are categorical variables, including information such as frequency and time period. Therefore, based on the attributes of the feedback information and the variable type of the parameter, the data can be merged and more abundant parameters can be calculated from each other, so as to achieve the effect of efficient data cleaning.

[0053] It should be noted that the "parameters required for the dietary pattern assessment model" refers to the answers given by users in the three dimensions of time, food, and environment. It is at least one or a combination of several of these three parameters. When the parameters of the three dimensions are all included in the evaluation, the model is optimal.

[0054] In order to realize the parameters of the model for dietary pattern assessment, the data is merged and analyzed and cleaned, and the scores of each component factor are calculated, including:

[0055] According to the attributes of the above feedback information and the variable type of the parameters, the corresponding parameters are generated and the corresponding parameter values ​​are determined. The cleaned indicator data and their component factors are synchronized to the database. According to the characteristics of different component factors, all component factors are dimensionlessly processed, the reverse indicators are positively converted, and the scores of each component factor are calculated.

[0056] In the above-mentioned implementation, the reverse indicators are converted into positive ones, and the scores of each component factor are calculated, specifically including: in the dietary pattern system, the collected indicators have positive and reverse problems. For example, in the indicator of meal frequency, the frequency of three meals is positive, and the frequency of eating late at night is negative. Because the frequency of three meals makes a positive contribution to regular and healthy eating habits, while eating late at night is just the opposite, it is necessary to convert the reverse indicators into positive ones, such as directly taking the opposite number.

[0057] Step 3: Unify the management of the analyzed and cleaned data, establish an adult dietary pattern database, and update the expert database;

[0058] This example uses statistical software to perform exploratory factor analysis and confirmatory factor analysis, including:

[0059] The scores of each factor are added together to obtain the dietary pattern model score, and the dietary pattern model of adults is formed based on the dietary pattern model score. The parsed and cleaned data is processed uniformly through standard specifications to obtain unified standardized data; the standardized data is stored to establish an adult dietary pattern database, and the expert database is continuously updated based on this.

[0060] It should be noted that the unified processing here refers to the use of statistical analysis methods commonly used in nutritional epidemiology, such as multiple linear regression, logistic regression or Cox proportional hazard regression model, to explore the correlation between exposure factors (here refers to the indicators collected by the dietary pattern system) and diseases or physical indicators of a large sample population, forming a result library after data analysis.

[0061] Experts obtain a continuous stream of dietary pattern data from the merged data, conduct more research while providing feedback to experts, and achieve efficient utilization and visualization of the data.

[0062] Step 4: Integrate the merged data after the introduction of the expert database with the basic user information of the institution to generate a historical attribute set, match the historical attribute set with the dietary pattern model, and store it in the cloud;

[0063] The merged data after the introduction of the expert database is integrated with the user information in hospitals, welfare homes, training camps and other institutions, and the types of diseases that users may have and the basic information of these users are further included. Then this part of basic information is used to generate a historical attribute set, and the historical attribute set is matched with the dietary pattern model and stored in the cloud. Basic information includes age, education level, medical history, gender, living environment, geographical location, and whether living alone. Basic information can be entered by the user himself or matched with the hospital, welfare home, training camp and other institutions where the user is located.

[0064] The matching of the historical attribute set and the dietary pattern model comprises:

[0065] First, a trial collection of samples from several adults is conducted. Then, the dietary pattern model collected by the dietary pattern system is directly matched with the historical attribute set input by the user or obtained from hospitals and other institutions. The collected samples are classified, and the historical attribute set and the dietary pattern model are probabilistically matched.

[0066] Step 5: retrieve the dietary pattern and expert opinions of the current subject to be evaluated for visual display, output the prediction information corresponding to the time, food, and environmental dimensions of the subject to be evaluated, and display it on the user interface to prompt the subject to be evaluated to make improvements.

[0067] After obtaining the information of the current subject to be evaluated, the information is exchanged with the information systems of hospitals, welfare institutions, training camps and other institutions to retrieve the current subject's dietary patterns and expert opinions for visualization. The preprocessing module outputs the prediction information corresponding to the time, food, and environmental dimensions of the subject to be evaluated, and displays it on the user interface to prompt the subject to be evaluated to make corresponding improvements. The visual content is formed by preprocessing, which is convenient for giving the evaluation object an intuitive and easy-to-understand presentation. It should be noted that the preprocessing module converts data into visual content for the convenience of users to consult.

[0068] In summary, the dietary assessment method based on time, food, and environment dimensions described in the embodiment of the present invention starts from the dimensions of time, food, environment, etc., and systematically explores a series of dietary pattern indicators. The system covers rich content and conducts a multi-dimensional survey on the diet of the population, which is consistent with the requirements of the dietary guidelines. It is lighter than the review method and the record method, and compared with the food frequency questionnaire, it supplements the questions of the time dimension and the environmental dimension, and improves on the basis of the traditional dietary survey. The system has a complete set of collection, cleaning, merging, and storage processes, and combines the expert database and user interaction to give instant feedback to the evaluation object. From the perspective of traditional nutrition and chrono-nutrition, it improves the user's eating habits in many ways to prevent diseases that may be caused by dietary factors.

[0069] On the other hand, the embodiment of the present invention provides a dietary assessment system based on time, food, and environment dimensions, which evaluates the dietary patterns of adults and their long-term relationship with health based on their dietary behaviors in daily life and gives feedback to the subjects, specifically, Figure 2 As shown, it includes:

[0070] The input module collects dietary behavior data for evaluating adults’ dietary patterns, including content in three dimensions: time, food, and environment;

[0071] The data processing module 201 is connected to the input module via wireless and / or wired communication;

[0072] The cloud integration terminal 300 is connected to the data processing module through wireless communication to realize the transmission and interaction of the three-dimensional information. The cloud integration terminal can integrate the corresponding data processing modules to form an overall evaluation based on its built-in integration logic and read expert suggestions from the expert database;

[0073] The information terminal 400 further processes the data of the cloud integrated terminal and transmits it to the local through a wireless communication connection and completes the visualization;

[0074] Feedback terminal 500, used to display data processing results and expert advice, so that adults who accept the survey of the system can consult;

[0075] and an expert database of 600.

[0076] Specifically, the input module includes input module 101, input module 102 and input module 103; wherein, the input module 101 includes a meal frequency measurement index, a meal time measurement index and a regularity measurement index in the time dimension. The meal frequency measurement index is used to detect information about the frequency of adults' meals, including meals, snacks and snacks (ordered categorical variables 1-5), special meals (ordered categorical variables 1-5), nighttime food intake (unordered categorical variables 1-3), and total meals per day (ordered categorical variables 1-7). The meal time measurement index is used to detect information about adults' meal times, including eating windows (continuous variables), three meal times (unordered categorical variables 0&1, ordered categorical variables 1-4), and food intake distribution (unordered categorical variables 0&1). The regularity measurement index is used to detect information about adults' regularity, including the difference in diet between weekdays and weekends (unordered categorical variables 0&1), and the proportion of time spent on regular diet each month (ordered categorical variables 1-6). Regularity was scored using the difference in diet between weekdays and weekends, whether the three meals were fixed (additional points), and the monthly regular diet percentage (multiplied ratio). Differences in diet between weekdays and weekends were scored 1 point, irregular meal times were scored 1 point, and the monthly regular diet percentage was multiplied by the total score of 0-48, with higher scores representing more irregular diets.

[0077] The input module 102 includes food intake indicators for three meals and food intake indicators for snacks in the food dimension. The three-meal intake measurement indicators are used to detect and calculate the food and nutrient intake of 25 food groups for breakfast, lunch and dinner for adults (based on the Chinese Food Composition Table), including 1. Rice; 2. Porridge: porridge or soaked rice; 3. Flour foods: steamed buns, bread, noodles, cakes, etc.; 4. Sweets: snacks, cakes, sweet bread, etc.; 5. Fried foods: fried dough sticks, fried cakes, French fries, etc.; 6. Stuffed foods: steamed buns, wontons, dumplings; 7. Coarse grains: including brown rice, millet, corn, coix seed, oats, red beans, mung beans, etc.; 8. Potatoes: including sweet potatoes, potatoes, taro, yam, konjac, etc.; 9. Dairy products: milk, yogurt or milk powder; 10. Eggs: chicken eggs or duck eggs; 11. Red meat: such as pork, beef, and mutton; 12. Poultry: such as chicken, duck, and goose; 13. Processed meat products: Such as sausage, bacon, luncheon meat, ham, etc.; 14. Freshwater fish: such as black carp, sea bass, silver carp, river shrimp, etc.; 15. Seafood: such as hairtail, pomfret, yellow croaker, shrimp, etc.; 16. Soy products: such as tofu, dried tofu, vegetarian chicken, tripe, etc.; 17. Nuts: watermelon seeds, sunflower seeds, walnuts, peanuts, pistachios, etc.; 18. Dark vegetables: such as green vegetables, spinach, water spinach, tomatoes, green peppers, carrots, etc.; 19. Light vegetables: such as cabbage, radish, cucumber, etc.; 20. Mushrooms: such as mushrooms, shiitake mushrooms, straw mushrooms, oyster mushrooms, etc.; 21. Fruits: such as apples, pears, peaches, bananas, grapes, oranges, tangerines, watermelons, etc.; 22. Sweet drinks: such as cola, Sprite, iced black tea, sports drinks, etc.; 23. Beer; 24. Liquor; 25. Yellow wine (ordered categorical variables 1-5, continuous variables). The snack intake measurement index is used to detect and calculate the food and nutrient intake of 9 food groups of snacks and snacks in combination with the meal measurement index of the input module 101 (based on the Chinese Food Composition Table), including 1. sweet snacks such as sweets, desserts, and cakes; 2. salty snacks such as salty soda crackers; 3. fried snacks such as potato chips and French fries; 4. nut snacks such as peanuts and melon seeds; 5. marinated snacks such as duck necks and chicken wings; 6. chocolate; 7. sweet drinks; 8. dairy products; 9. fruits (ordered categorical variables 1-5, continuous variables).

[0078] The input module 103 includes a dining location measurement index, a social group measurement index, and an accompanying activity measurement index in the environmental dimension. The dining location measurement index is used to detect information about the adult's dining location, including home and take-out exposure (ordered categorical variables 1-5). The social group measurement index is used to detect information about the social group when the adult eats, including alone, family, friends, and work social activities (ordered categorical variables 1-5). The accompanying activity measurement index is used to detect information about accompanying activities when the adult eats, including entertainment activities (such as watching TV, playing with mobile phones, etc.), social activities (such as face-to-face chatting, etc.), and work (such as computer office, etc.) (ordered categorical variables 1-5).

[0079] Specifically, the data processing module includes a time dimension calculation unit, a food dimension calculation unit and an environment dimension calculation unit;

[0080] The time dimension calculation unit calculates time data by using a preset time operation model to generate time dimension feature data;

[0081] The food dimension calculation unit calculates food data through a preset food calculation model to generate food dimension feature data;

[0082] The environment dimension calculation unit calculates the environment data through a preset environment calculation model to generate environment dimension feature data;

[0083] The data processing modules 201, 202, and 203 are connected to the input modules 101, 102, and 103 respectively through wireless and / or wired communication, so that the data processing modules 201, 202, and 203 can generate characteristic data that can be used to evaluate the three dimensions of adult dietary patterns. The data processing module uses the characteristics of ordered categorical variables, uses the original classification or performs autonomous division to form a new classification, and performs correlation analysis with the outcome indicators; using the characteristics of continuous variables, the original data is used to directly calculate new variables (for example, eating window = the time to end eating in a day - the time to start eating in a day, eating midpoint = the time to start eating in a day + eating window / 2, early eating-sleep interval time = the time to start eating in a day - the time to wake up, late eating-sleep interval time = the time to fall asleep - the time to end eating in a day) or find nodes to convert into categorical variables, and perform correlation analysis with the outcome indicators. Use the multiple linear regression algorithm to establish a prediction model for the input data. Select appropriate independent variables and dependent variables, and estimate the parameters of the regression model by methods such as the least squares method. Use the logistic regression algorithm to establish a classification model for the input data. Select appropriate independent variables and classification labels, and estimate the parameters of the regression model by methods such as maximum likelihood estimation. Use multivariate statistical analysis methods such as principal component analysis, latent class analysis, cluster analysis, etc. to evaluate the pattern of standardized feature data, and reduce many possible feature patterns to several new interpretable patterns that can represent the sample population. Implement the PCA algorithm to reduce the dimension of the data and find the main variance in the data. This can be achieved by calculating eigenvalues ​​and eigenvectors and selecting the most important principal components. Use the latent class analysis algorithm to identify potential unlabeled categories in the data. The algorithm can identify categories based on the probability model and maximum likelihood estimation of the data, and classify them according to the potential association between the observed variables. Use the latent profile analysis algorithm to identify potential (hidden) profiles or groups in the data. The algorithm can determine the groups in the data based on the similarity measure between the variables by clustering the data. Use the cluster analysis algorithm to divide the data into different clusters or groups. This can be achieved by calculating the distance or similarity between data points and then assigning them to the corresponding clusters. Integrate the above steps into the data processing module.

[0084] The cloud integration terminal 300 is connected to the data processing modules 201, 202, and 203 through wireless communication to realize the transmission and interaction of three-dimensional information. The cloud integration terminal can integrate the corresponding data processing modules into an overall evaluation based on its built-in integration logic and read expert suggestions from the expert library 600.

[0085] The information terminal 400 further processes the data of the cloud integrated terminal and transmits it to the local through wireless communication connection and completes visualization.

[0086] The feedback terminal 500 is used to display the data processing results and expert suggestions for easy reference by adults who accept the survey of the system. For example, the feedback terminal 500 can be one of the smart devices such as a smart phone, an IPAD, a smart bracelet, etc. that is connected to the information terminal 400 for communication.

[0087] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A dietary survey method based on the dimensions of time, food, and environment, characterized in that: The following steps are involved: Step 1: Collect dietary behavior data for evaluating adult dietary patterns, including data on time, food, and environment, and provide timely reminders and precise guidance to monitor and record users’ diet in real time; Step 2: Generate the parameters required for the dietary pattern assessment model based on the attributes of time, food, and environmental information, merge the data, parse and clean them, and calculate the scores of each component factor; Step 3: Unify the data after analysis and cleaning, establish an adult dietary pattern database, and update the expert database; Step 4: Integrate the merged data after the introduction of the expert database with the user information of the institution to generate a historical attribute set, match the historical attribute set with the dietary pattern model, and store it in the cloud; Step 5: retrieve the dietary pattern and expert opinions of the current subject to be evaluated for visual display, output the prediction information corresponding to the time, food, and environmental dimensions of the subject to be evaluated, and display it on the user interface to prompt the subject to be evaluated to make improvements.

2. The dietary survey method based on time, food and environment dimensions according to claim 1 is characterized in that: The time dimension includes items such as: the number of meals, snacks and snacks, the time of three meals, nighttime eating, eating time window, eating-sleep interval, eating differences between weekdays and weekends, monthly, quarterly and annual eating patterns, long-term eating preferences, binge eating, type and duration of restrictive eating behavior, and specific eating patterns; The food dimension includes items such as: energy distribution based on meal times, food group intake based on meal times, food group intake, and food cooking methods; The environmental dimension includes items such as eating at home or outside, eating alone or with others, activities during meals, influence of the dining environment, and food accessibility.

3. The dietary survey method based on time, food and environment dimensions according to claim 1, characterized in that: The dietary behavior data that can be used to evaluate the dietary patterns of adults are collected using smart wearable devices and smart terminals. The smart wearable devices and smart terminals include interfaces developed with smart watches and health monitoring bracelet devices. The smart wearable devices and smart terminals are provided with modules that can automatically remind and guide the wearer to respond to meal times and record food intake in real time.

4. The dietary survey method based on time, food and environment dimensions according to claim 1, characterized in that: The parameters required for generating the dietary pattern assessment model according to the attributes of time, food and environmental information include: The acquired entry data is parsed using a preset rule engine and rule algorithm, and the parsed data is cleaned to generate the parameters required for the dietary pattern assessment model according to the attributes of time, food and environmental information; the parameters include: continuous variables composed of time parameters and intake parameters, and categorical variables composed of frequency and time period information groups.

5. The dietary survey method based on time, food and environment dimensions according to claim 4 is characterized in that: The parameters required for the dietary pattern assessment model are represented by the answers given by the user in the three dimensions of time, food, and environment, at least one of these three parameters or a combination of several of them. When the parameters of the three dimensions are all included in the assessment, the model is optimal.

6. The dietary survey method based on time, food and environment dimensions according to claim 1, characterized in that: The confluence data is analyzed and cleaned, and the scores of each component factor are calculated including: Generate corresponding parameters according to the attributes of feedback information and the variable type of parameters, and determine the corresponding parameter values. Synchronize the cleaned indicator data and their component factors into the database. According to the characteristics of different component factors, all component factors are dimensionlessly processed, and the reverse indicators are positively converted to calculate the scores of each component factor.

7. The dietary survey method based on time, food and environment dimensions according to claim 1, characterized in that: The unified management of the analyzed and cleaned data, establishment of an adult dietary pattern database, and updating of the expert database include: The scores of each factor are added together to obtain the dietary pattern model score. Based on the dietary pattern model score, an adult dietary pattern model is formed. The parsed and cleaned data are unified according to standard specifications to obtain unified standardized data. The standardized data is stored to establish an adult dietary pattern database, and the expert database is continuously updated based on this.

8. The dietary survey method based on time, food and environment dimensions according to claim 1, characterized in that: The matching of the historical attribute set and the dietary pattern model comprises: First, a trial sample collection is conducted on several adults. Then, the dietary pattern model collected by the dietary pattern system is directly matched with the historical attribute set input by the user or obtained from hospitals and institutions. The collected samples are classified, and the historical attribute set and the dietary pattern model are probabilistically matched.

9. The dietary survey method based on time, food and environment dimensions according to claim 1, characterized in that: The method of retrieving the current diet pattern and expert opinions of the subject to be evaluated for visual display, outputting the prediction information corresponding to the time, food, and environment dimensions of the subject to be evaluated, and displaying it on the user interface to prompt the subject to be evaluated to make improvements includes: After obtaining the information of the subject to be evaluated, information is exchanged with the information systems of hospitals, welfare homes, and training camp institutions to retrieve the current dietary patterns and expert opinions of the subject to be evaluated for visual display, pre-process the current status of the subject's time, food, and environmental dimensions, output the prediction information corresponding to the time, food, and environmental dimensions of the subject to be evaluated, and display it on the user interface to prompt the subject to be evaluated to make corresponding improvements.

10. A dietary assessment system based on time, food, and environment dimensions, characterized in that: include: The user diet information collection module collects dietary behavior data for evaluating adult dietary patterns, including data in three dimensions: time, food, and environment, and provides timely reminders and precise guidance to monitor and record users' diets in real time; The data processing module is connected to the input module through wireless and / or wired communication; the data processing module includes a time dimension calculation unit, a food dimension calculation unit and an environment dimension calculation unit; wherein the time dimension calculation unit calculates time data through a preset time calculation model to generate time dimension feature data; the food dimension calculation unit calculates food data through a preset food calculation model to generate food dimension feature data; the environment dimension calculation unit calculates environment data through a preset environment calculation model to generate environment dimension feature data; The cloud integration terminal is connected to the data processing module through wireless communication to realize the transmission and interaction of information in three dimensions. Based on its built-in integration logic, it integrates the corresponding data processing modules to form an overall evaluation and reads expert suggestions from the expert database; Information terminal, which further processes the data of the cloud integrated terminal and transmits it to the local area through wireless communication connection and completes visualization; Feedback terminal, used to display data processing results and expert advice for easy reference by adults who accept the system survey; and expert pool.

Images