calculating a scoring coefficient k of the daily energy requirement EN based on the human basal metabolic rate BMR and the physical activity level coefficient PAL:

k＝EN/1000kcal，

EN＝BMR×PAL，

wherein, the BMR (Harris nodal Equation) is as follows:

female: BMR =655+ (9.6X weight in kg) + (1.8X height in cm) - (4.7X age in years)

Male: BMR =66+ (13.7 x weight in kg) + (5 x height in cm) - (6.8 x age in years);

physical activity level coefficient PAL was taken from the PAL control table of the chinese society for nutrition as follows table 2:

TABLE 2PAL comparison Table

Calculating daily average intake w of each food dimension of the user based on the user meal information, wherein Tw is daily sum intake of each food dimension, and is equal to the product of single intake of the user in the food dimension and intake frequency or the value of total intake of the user in a meal counting period divided by the meal counting period in days according to a general calculation mode;

equally dividing the scoring interval (from 0 to full-scale interval) of the dietary scoring dimensionality of each food type based on a nutrition and dietary scoring rule, evaluating the interval of daily average intake data, acquiring the score s of each food dimensionality, and generally taking 2 decimal places;

the user overall meal present condition score S is the sum of all food dimension scores, the maximum score is 100 points, the minimum score is 0 point, 2 decimal points are reserved in background calculation logic, and the score is presented to the user in an integer form through front end feedback.

(S3) extracting key fields from the acquired user basic information, the user diet information and the user nutrition and diet indexes obtained through calculation, and introducing the key fields into a preset data processing program for analysis and processing to generate a user personalized intervention scheme presented in a daily task form, wherein the key fields comprise daily energy requirement, disease type, nutrition risk level and user overall diet present condition score;

specifically, extracting key fields from the user basic information includes: the type of disease, the nutritional risk level,

and on the basis of a nutritional intervention database, matching key fields according to the dietary behavior rules to generate a user personalized intervention scheme presented in a daily task mode, wherein each task of a single food category only appears once in each matching, and the number of the tasks obtained by matching is at least 1 and at most 4.

The preset data processing program carries out statistical analysis and sorting on the big data of healthy people through medical and nutriologists to form a nutrition intervention database and a diet action rule, wherein the nutrition intervention database comprises a food intake database, a diet mode database, a food risk grade database and an intervention task database based on disease types.

The food intake database based on the disease types can be based on the grading rules of HEI-2015 grading rules, chinese dietary guidelines and the like, and the recommended intake is used as the daily recommended intake standard of various foods of the user.

The meal mode database is formed by induction and sorting according to big data of the occurrence probability of each meal based on the design of nutriologists and the food types and is used as the occurrence meal standard of the food specifically recommended by the user; for example, the aforementioned obese patient has a criterion for recommended intake of the daily diet based on the HEI-2015 scoring criteria, with the full score criterion;

taking the marine fish food shown in table 3 below as an example, by performing expert evaluation setting on the number of meals of various foods, the number of meals of recommended food can be set from high to low according to the number of meals of food in the table when the user is matched. Meanwhile, the database can train data based on an artificial intelligence technology, a supervised machine learning algorithm, a boosting algorithm and the like, and a recommended meal probability weight relation model corresponding to the state indexes such as the occupation type and the disease type of the user is constructed so as to continuously optimize the recommended meal probability weight value corresponding to each state index.

TABLE 3 evaluation table of diet pattern of fish and marine products

A food risk grade database, which is designed by nutriologists and medical experts, and divides a large number of food categories into a recommendation category (G), a general category (Y) and a contraindication category (R) according to nutritional requirements and contraindications of different disease types, so as to serve as sources of food names specifically recommended by users;

the following table 4 shows examples of grades of marine fish food for obesity, hypertension and gout, and the food types of some disease type patients are graded by expert judgment, that is, the food names are matched and recommended to the user, and then the food names are sorted according to the order of food G grade preference, food Y grade suboptimum selection and food R grade forbidden selection in the table.

TABLE 4 Risk rating assessment table for fish and marine products

An intervention task database, which integrates knowledge and experience of experts in the field of nutrition to form a large number of nutrition requirement suggestions of users according with different disease states and is embodied as daily nutrition intervention tasks;

the user nutrition intervention task can be set based on the recommended intake of various foods of the user, and meanwhile, a restriction food intake task and a special food intake task are considered to be added. Specific task configurations may include meal intake, priority food names, alternative food names, task difficulty, intake, and other indicators.

Taking the fruit food tasks shown in the following table 5 as an example, when a user performs a fruit food pushing task, the recommended task number and priority can be judged and ranked according to the daily recommended intake of the user and the single task intake and the task difficulty set manually in advance in the following table. Meanwhile, the database can train data based on an artificial intelligence technology, a supervised machine learning algorithm and the like, and a task difficulty weight relation model corresponding to the state indexes such as the occupation type, the disease type and the like of the user is constructed so as to continuously optimize the task difficulty weight value corresponding to each state index.

Number of meals	Preferentially recommending food	Alternative recommended food	Difficulty of making	Intake amount (g)
					Breakfast	Eat one { front }	Or { other _ fruit } and other fruits (1 apple portion)	3	200
Lunch	Eat one { fruit }	Or { other _ fruit } and other fruits (1 apple portion)	3	200
					Dinner	Eat one { front }	Or { other _ fruit } and other fruits (1 apple portion)	3	200
Food with meal	Eat one { front }	Or { other _ fruit } and other fruits (1 apple portion)	1	200

Table 5 fruit food pushing task evaluation table

Where { fruit } represents one fruit name and { other _ fruit } represents another fruit name.

The meal action rules may be embodied in two forms:

firstly, matching a recommended food intake database according to user basic information mainly comprising disease types to form an ideal diet scheme; secondly, matching a meal mode database according to key fields including daily energy demand and the overall meal present condition score of the user to form a practical meal scheme of the user; and finally, performing cross matching on the ideal diet scheme and the actual diet scheme, combining a food risk level database and an intervention task database, obtaining the diet tasks which accord with the diet habits of the users and contain key fields related to food types, corresponding meals, single intake and the like, forming a task set for the personalized nutritional intervention of the users, and extracting daily nutritional intervention tasks from the task set to push and present the daily nutritional intervention tasks to the users.

Secondly, on the basis of a decision tree prediction model, nutrition requirements and nutrition intervention suggestion schemes based on the big data of healthy people are represented by non-leaf nodes and leaf nodes respectively, wherein each non-leaf node of the decision tree represents a judgment condition of the nutrition requirements of a user, branches of the decision tree represent objects meeting the conditions of the nodes, the leaf nodes of the decision tree represent nutrition intervention suggestion schemes, the nutrition intervention suggestion schemes can be pre-configured on the basis of nutrition intervention rules, and an associated intervention task database of the nutrition intervention suggestion schemes is pushed and presented to the user in a daily nutrition intervention task form. For example, whether a user suffers from a disease is judged based on disease type data, if not, a first ideal diet scheme is adopted and a corresponding daily nutritional intervention task is pushed, if yes, whether complications occur is judged continuously, if not, a second ideal diet scheme is adopted and a corresponding daily nutritional intervention task is pushed, if yes, whether the user is in an acute attack stage is judged continuously, if not, a third ideal diet scheme is adopted and a corresponding daily nutritional intervention task is pushed, and if yes, a preset gradual-evolution diet scheme is adopted and a corresponding daily nutritional intervention task is pushed.

(S4) receiving feedback of a user on the daily task completion condition, gradually evolving and adjusting the user personalized intervention scheme, and specifically comprising the following steps:

if the user is judged to finish the task on the current day, continuing to push the task on the second day, and if the unfinished task on the current day exists, continuing to push the task on the second day; if the same task is not completed in 3 consecutive days, stopping pushing the task and triggering background reminding, and auditing and adjusting the nutrition intervention scheme by professionals such as doctors and dieticians at the monitoring terminal;

in addition, professionals such as doctors and dieticians at the monitoring terminal can perform active monitoring through the background, adjust the intervention scheme according to the periodically updated evaluation report of the user, the periodic task completion report and the like, and display the intervention scheme to the user;

and according to the information fed back by the user card punching, the corresponding food dimension scores are counted into the overall diet present condition score of the user and are deducted after 7 days.

the information acquisition module is used for acquiring basic information of the user including sex, age, height, weight, disease type and nutritional risk level, and acquiring meal information of the user including food intake type, intake frequency, total food intake and outside dining behaviors in a built-in questionnaire mode;

the meal scoring module is used for calculating user nutrition and meal indexes including food dimension scores and user overall meal current condition scores according to the collected user basic information, the user meal information and a preset nutrition and meal scoring rule;

the core processing module is used for processing the acquired information, extracting key fields, incorporating the key fields into a preset data processing program for analysis and processing, and generating a user personalized intervention scheme containing daily tasks;

the input module is used for carrying out interactive operation by a user;

the storage module is used for storing related data information; and

Furthermore, the terminal also comprises a health habit module connected with the core processing module and used for evaluating the health habits of the user according to the collected user basic information and the user meal information.

The terminal can be presented in an APP or application software mode based on existing intelligent equipment such as a mobile phone, a tablet, a notebook and the like, so that a user can use the terminal conveniently.

Through the process, the method can analyze and evaluate the diet and health data of the user, calculate and adjust the diet quality current condition score matched with the demand, generate a daily intervention plan and task and food image recognition feedback intervention completion degree according with the diet habits of the user, and realize the function of dynamically adjusting the plan according to the task completion condition. Users of different nutritional needs may therefore obtain a complete and feasible set of evolutionary nutritional intervention programs based on mobile devices, rather than the traditional approach of providing only broad nutritional recommendations, without involving specific and effective touchdown actions.

The above-described embodiments are only preferred embodiments of the present invention, and are not intended to limit the scope of the present invention, but all changes that can be made by applying the principles of the present invention and performing non-inventive work on the basis of the principles shall fall within the scope of the present invention.

Claims

1. A method of intelligent nutritional intervention, comprising:

collecting basic information of a user including gender, age, height, weight, disease type and nutritional risk level, and collecting diet information of the user including food intake type, intake frequency, total food intake amount and outside dining behavior in a questionnaire form;

calculating user nutrition and diet indexes according to the collected user basic information, the user diet information and a preset nutrition and diet grading rule, wherein the user nutrition and diet indexes comprise food dimension grades and overall diet current condition grades of the user; the process of calculating the overall diet present score of the user is as follows:

calculating daily average intake w of each food dimension of the user based on the user meal information, wherein w = Tw/k, and Tw is the daily sum intake of each food dimension;

equally dividing the grading interval of the dietary grading dimensionality based on a nutrition and dietary grading rule, evaluating the interval of daily average intake data, and acquiring the grading s of each food dimensionality;

the user overall meal present condition score S is the sum of all food dimension scores;

receiving feedback of a user on the completion condition of each day of tasks, and gradually evolving and adjusting the personalized intervention scheme of the user;

the process of generating the user personalized intervention scheme comprises the following steps:

matching key fields according to the dietary behavior rules on the basis of a nutritional intervention database to generate a user personalized intervention scheme presented in a daily task form, wherein each time of matching, each task of a single food category only appears once, and the number of the tasks obtained by matching is at least 1 and at most 4;

the preset data processing program carries out statistical analysis and arrangement on the big data of healthy people through medical and nutriologists to form a nutrition intervention database and a diet action rule,

wherein the nutritional intervention database comprises a disease type-based food intake database, a meal pattern database, a food risk rating database, and an intervention task database;

the meal action rules include:

firstly, matching a recommended food intake database according to basic information of a user mainly based on disease types to form an ideal diet scheme; secondly, matching a diet mode database according to key fields including daily energy demand and the overall diet present score of the user to form a practical diet scheme of the user; then, the ideal diet scheme and the actual diet scheme are subjected to cross matching, a diet task conforming to the diet habit of the user is obtained by combining a food risk level database and an intervention task database, key fields contained in the diet task relate to food types, corresponding meals and single intake, so that a task set for personalized nutrition intervention of the user is formed, and daily tasks are extracted from the task set and pushed to the user;

2. The method of intelligent nutritional intervention of claim 1, wherein feedback is presented to the user after calculating the user's overall meal presence score and each food dimension score is incorporated into the user's overall meal presence score after the user completes a daily task.

3. The method of intelligent nutritional intervention of claim 2, wherein a meal scoring dimension comprising fruits, vegetables, beans, whole grains, dairy, seafood, refined grains, added sugars, total protein intake, total fatty acid intake, and saturated fat intake is pre-set in the nutrition and meal scoring rules, and the user's food intake status is scored based on the meal scoring dimension.

4. The method of intelligent nutritional intervention of claim 3, wherein receiving user feedback on daily task completion, wherein evolutionarily adjusting the user-customized intervention program specifically comprises:

5. A terminal for implementing the method of intelligent nutritional intervention according to any one of claims 1 to 4, comprising:

the display module is used for presenting intuitive user overall meal present score and daily tasks based on a user personalized intervention scheme for the user;

the input module is used for carrying out interactive operation by a user;

the storage module is used for storing related data information; and

6. The terminal of claim 5, further comprising a health habit module connected to the core processing module, for evaluating the health habits of the user according to the collected user basic information and the user meal information.