CN116525067A - Nutrient recipe recommendation system and method - Google Patents

Abstract

The invention is applicable to the technical fields of nutrition and artificial intelligence, and provides a nutrition recipe recommendation system and a method, wherein the system comprises the following steps: the system comprises a basic information acquisition module, a coding library index module, a nutrient algorithm module and a diet doctor order analysis and output module; the basic information acquisition module is used for acquiring basic information of a patient, wherein the basic information comprises disease diagnosis results, height, weight, age, laboratory examination and medical record summary information; the coding library index module is coupled with a predetermined nutrition knowledge coding library and is used for searching nutrition knowledge corresponding to the disease in the nutrition knowledge coding library according to the basic information; the nutrient algorithm module is used for calculating the theoretical nutrient range required by the patient according to the basic information; the invention uses computer intelligent technology, and uses various disease nutrition guidelines as calculation basis and reasoning basis, to ensure accurate nutrient recommendation scheme to be calculated according to the input information.

Description

Nutrient recipe recommendation system and method

Technical Field

The invention belongs to the technical field of nutrition and artificial intelligence, and provides a nutrition recipe recommendation system and method.

Background

In recent years, information technology is continuously developed, artificial intelligence (artificial intelligence, AI) is applied to a plurality of fields, and is increasingly widely applied to the nutrition science, and intelligent nutrition service work is performed by simulating thinking process, data grabbing and distinguishing screening capability. However, because the clinical nutrition doctors in all hospitals have uneven professional ability, the nutrition prescriptions of patients are different, and in order to give the optimal nutrient recommendation to the patients, the optimal nutrient recommendation is given according to the current information of the patients by means of artificial intelligence. Has the functions of warning, reminding and assisting the clinical nutrition department doctors to work.

Clinically, the types of diseases of patients are different, the occurrence and the development of the diseases of the patients are different, and information provided by the patients after admission is different, so that individual decision making is always the most main scheme of clinical treatment for disease treatment. The nutrition has important value for disease occurrence, development and prognosis, and has important significance for individual nutrient recommendation of patients. According to the data information of clinical patients and the characteristics of personalized diseases, the personalized nutrients are recommended to the patients, so that the disease treatment effect is facilitated, meanwhile, the workload of clinical nutrition doctors is reduced, and the treatment scheme is optimized. Firstly, combing the clinical nutrition department work content, mainly facing to clinical nutrition service work of all departments of the whole court of patients, and dividing the clinical nutrition service work into three categories of low protein diet, diabetes diet and other diets according to the disease category, nutrition department work development and nutrient recommending direction. And classifying according to the patient information, and finally determining a nutrient calculating method, wherein if the patient has only height and no weight, the patient can implement a standard weight method so as to calculate recommended amounts of calories, proteins, fat, vitamins and minerals.

Along with the continuous development of information technology, the identification and reasoning data are more and more accurate, and the knowledge of the clinical nutrition data is combined with artificial intelligence in order to alleviate the difficulty of recommending the existing nutrients in the clinical nutrition department. A set of intelligent recommended schemes aiming at clinical nutrition doctors work have been designed, and intelligent services are provided for further clinical nutrition work development.

However, the intelligent nutrient recommendation has certain problems and difficulties, mainly the defects of low recommendation accuracy and practicality, and the invention designs a nutrient recipe recommendation system and method of a nutrient knowledge coding library based on the problems.

Disclosure of Invention

The embodiment of the invention aims to provide a nutrition recipe recommendation system and method, which are used for solving the problems of low recommendation accuracy and low practicability of the existing nutrition recommendation mode in the background technology.

The embodiment of the invention is realized in such a way that a nutrition recipe recommendation system comprises: the system comprises a basic information acquisition module, a coding library index module, a nutrient algorithm module and a diet doctor order analysis and output module;

the basic information acquisition module is used for acquiring basic information of a patient, wherein the basic information comprises disease diagnosis results, height, weight, age, laboratory examination and medical record summary information;

The coding library index module is coupled with a predetermined nutrition knowledge coding library and is used for searching nutrition knowledge corresponding to the disease in the nutrition knowledge coding library according to the basic information;

the nutrient algorithm module is used for calculating the theoretical nutrient range required by the patient according to the basic information;

the diet doctor advice analysis and output module is used for analyzing according to the nutrition knowledge retrieved by the encoding library indexing module and the theoretical nutrient range required by the patient calculated by the nutrient algorithm module, selecting foods matched with the patient, and outputting at least one nutrition recipe meeting the requirement of the patient.

Preferably, the basic information acquisition module comprises a hospital information system, a test information system and an electronic medical record system;

the basic information can be acquired through at least one of the hospital information system, the inspection information system and the electronic medical record system.

Preferably, the nutrition knowledge coding library comprises a hospital meal coding library, a disease diagnosis coding library and a self-built meal coding library, wherein the hospital meal coding library and the disease diagnosis coding library are in communication with each other, and the self-built meal coding library and the disease diagnosis coding library are in communication with each other;

The hospital meal coding library and the disease diagnosis coding library are connected in a first common application range, the self-built meal coding library and the disease diagnosis coding library are connected in a second common application range, and the first common application range and the second common application range represent the matching degree of diseases, nutrients and foods.

Preferably, the self-built meal code library is built according to nutrient category classification, and the nutrient category is divided into macro nutrient and micro nutrient.

Preferably, the diet doctor advice analysis and output module is internally provided with a machine learning model, the machine learning model takes basic information and corresponding nutrition knowledge of a patient as a sample training set, the machine learning model learns according to the sample training set, and an optimized machine learning model is generated.

Preferably, the nutrient algorithm module comprises a standard weight method, a basal metabolic method and a DRIS method, and the theoretical nutrient range is calculated and determined by at least one algorithm of the standard weight method, the basal metabolic method and the DRIS method.

Preferably, the system further comprises: the allergen database is used for storing basic information and allergen information of the patient and the family members of the patient; the allergen database can assist the diet doctor order analysis and output module to analyze the theoretical nutrient range of the patient so as to formulate a nutrient recipe conforming to the patient.

Another object of the embodiments of the present invention is to provide a method for recommending a nutrition recipe, which is used for the nutrition recipe recommending system, and the method includes the following steps:

acquiring basic information of a patient;

retrieving nutritional knowledge corresponding to the disease from the nutritional knowledge code base according to the basic information;

calculating the theoretical nutrient range required by the patient according to the basic information;

and analyzing according to the nutrition knowledge and the theoretical nutrient range, selecting foods matched with the patient, and outputting at least one nutrition recipe which meets the needs of the patient.

Preferably, the step of retrieving, in the nutrition knowledge code base, nutrition knowledge corresponding to the disease according to the basic information specifically includes:

inputting the acquired basic information of the patient into an encoding library index module;

judging whether the disease diagnosis result in the basic information is recorded in a hospital meal coding library and a disease diagnosis coding library or not through a coding library index module;

if the nutrition knowledge is recorded in the hospital meal coding library and the disease diagnosis coding library, directly outputting the nutrition knowledge corresponding to the disease;

if the nutrient is not recorded in the hospital meal coding library and the disease diagnosis coding library, the nutrient categories are classified into macro nutrients and micro nutrients according to the nutrient categories;

And searching the divided results in the self-built meal coding library and the disease diagnosis coding library, and outputting nutrition knowledge corresponding to the disease according to the search results.

Preferably, the step of calculating the theoretical nutrient range required by the patient according to the basic information specifically includes:

selecting a calculation mode of theoretical nutrients required by a patient according to the basic information;

according to the selected calculation mode, basic information of a patient is input based on a preset guideline, and calculation of heat required by the patient is performed; alternatively, the patient's required calories are calculated by a nutrient algorithm module to output the patient's required theoretical nutrient range.

Compared with the prior art, the nutrition recipe recommendation system provided by the embodiment of the invention has the following beneficial effects: the defect that the professional capacities of clinical nutrition doctors in all hospitals are uneven in level and different in nutrition prescriptions of patients can be overcome by utilizing a computer intelligent technology; basic information of a patient is input, and each module is used as a calculation basis and an reasoning basis according to the existing various disease nutrition guidelines, so that an accurate nutrient recommendation scheme is calculated according to the input information, and an effective method is provided for supporting treatment of clinical nutrition departments.

Drawings

Fig. 1 is a block diagram of a nutrition recipe recommendation system according to the present embodiment;

fig. 2 is a diagram illustrating the concept of a nutrition recipe recommendation system according to the present embodiment;

FIG. 3 is a flowchart of a method for recommending a nutrition recipe according to the present embodiment;

fig. 4 is an internal structural diagram of a computer device in one embodiment.

In the figure: the system comprises a basic information acquisition module 10, a coding library index module 20, a nutrition knowledge coding library 30, a nutrient algorithm module 40 and a diet doctor order analysis and output module 50.

Detailed Description

The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.

In one embodiment, as shown in fig. 1, a block diagram of a nutrition recipe recommendation system provided in this embodiment is shown, and fig. 2 is a mind map of a nutrition recipe recommendation system provided in this embodiment; the system comprises: the system comprises a basic information acquisition module 10, a coding library index module 20, a nutrient algorithm module 40 and a diet doctor order analysis and output module 50;

The basic information acquisition module 10 is configured to acquire basic information of a patient, where the basic information includes disease diagnosis results, height, weight, age, laboratory examination, and medical record summary information;

the coding library index module 20 is coupled with a predetermined nutrition knowledge coding library 30, and is used for searching nutrition knowledge corresponding to the disease in the nutrition knowledge coding library 30 according to the basic information;

the nutrient algorithm module 40 is configured to calculate a theoretical nutrient range required by the patient according to the basic information;

the diet doctor's advice analysis and output module 50 is configured to analyze according to the nutrition knowledge retrieved by the code library indexing module 20 and the theoretical nutrient range required by the patient calculated by the nutrient algorithm module 40, select foods matching the patient, and output at least one nutrition recipe matching the patient's requirement.

The embodiment utilizes the computer intelligent technology, and can overcome the defects that the professional capacities of clinical nutrition doctors in all hospitals are uneven in level and different in nutrition prescriptions are issued to patients; basic information of a patient is input, and each module is used as a calculation basis and an reasoning basis according to the existing various disease nutrition guidelines, so that an accurate nutrient recommendation scheme is calculated according to the input information, and an effective method is provided for supporting treatment of clinical nutrition departments.

In one embodiment, the basic information acquisition module 10 includes a hospital information system, a test information system, an electronic medical record system;

the basic information can be acquired through at least one of the hospital information system, the inspection information system and the electronic medical record system.

The hospital information system, the inspection information system and the electronic medical record system are all existing hospital information systems, inspection information systems and electronic medical record systems, and generally, disease diagnosis, height, weight, age, laboratory examination and medical record summary information related to patients can be obtained through inquiry by any one of the hospital information systems, the inspection information systems and the electronic medical record systems.

Of course, prior to the query operation described above, authorization of the patient or patient's family has been obtained; or the inquiring end is authorized by a hospital manager, and meets the requirements and specifications of the industry on related privacy information or ethics. The authorization mode is to obtain authorization by reading a patient's visit card or medical insurance card, but is not limited to this authorization mode, and may be identity IC card information, face biological information, or the like.

For example: the hospital information system can obtain the information of disease diagnosis, name, height, weight, age and the like of the patient; by checking information about the name, height, weight, age, laboratory examination, etc. of the information system patient.

The nutrition knowledge coding base can be determined in advance because the data magnitude of the nutrition knowledge coding base is large;

in one embodiment, the nutritional knowledge coding library comprises a hospital meal coding library, a disease diagnosis coding library;

the reference basis established by the hospital meal coding library and the disease diagnosis coding library can be a Chinese disease knowledge base (China Disease Knowledge Total Database), which is called a disease library for short; is common knowledge to the person skilled in the art and is not described in detail herein.

In an example of an embodiment, the nutritional knowledge coding library comprises a hospital meal coding library, a disease diagnosis coding library, and a self-built meal coding library, wherein the hospital meal coding library and the disease diagnosis coding library are in communication with each other, and the self-built meal coding library and the disease diagnosis coding library are in communication with each other;

the hospital meal coding library and the disease diagnosis coding library are connected in a first common application range, the self-built meal coding library and the disease diagnosis coding library are connected in a second common application range, and the first common application range and the second common application range represent the matching degree of diseases, nutrients and foods.

In an example of an embodiment, the self-built meal code library is established according to nutrient class classification, and the nutrient classes are classified into macro-nutrients and micro-nutrients.

In one embodiment, the nutrient algorithm module includes a standard weight method, a basal metabolic method, a DRIS method, and the theoretical nutrient range is determined by at least one algorithm calculation of the standard weight method, the basal metabolic method, the DRIS method.

In an example of an embodiment, the hospital meal code library, in particular:

the hospital meal is divided into basic meal, treatment meal and test meal, wherein the test meal is removed in the embodiment, and the test meal is removed as shown in the table 1 because the test meal is used for assisting a clinician in diagnosing diseases, and the embodiment is only aimed at the aspect of nutrition recommendation and does not relate to disease diagnosis;

the number of the hospital meal codes is 13 at present, the later period can be increased or decreased according to the need, and the codes are ordered according to the same coding principle and sequence;

in one example, the basic diet may be analyzed from the list of five aspects, namely category code, application scope, diet principle, usage, and optional food, according to the diet order, as shown in table 2:

In one example, the treatment diet may be analyzed in a list from three aspects of category coding, application scope, and diet principle and usage, respectively, according to the diet orders, as shown in table 3:

in one embodiment, a disease diagnosis database is selected: the disease diagnosis database of the embodiment is derived from a Chinese disease knowledge base (China Disease Knowledge Total Database), which is called a disease base for short;

according to the hospital meal coding library and the disease diagnosis database, carrying out corresponding coding on the hospital meal coding library and the disease diagnosis coding library through a first common application range, and establishing a relational database;

in one example of this embodiment, due to the huge database content, two typical diseases corresponding to each diet order are now taken as references, as shown in table 4:

in one embodiment, the self-built meal code library may be built according to nutrient class classification, and the nutrient classes are classified into macro-nutrients and micro-nutrients;

for example: macronutrients (energy producing nutrients) and micronutrients (non-energy producing nutrients). Wherein: macronutrients mainly include carbohydrates, proteins, fats, dietary fibers, etc.; the micronutrients mainly comprise calcium, iron, zinc, iodine, copper, potassium, sodium and the like;

Carrying out diet doctor order naming according to the content of the nutrients, constructing a diet coding library, such as high-protein low-sodium diet, low-protein low-iodine diet and the like, finding the relationship between the high-protein low-sodium diet and the low-protein low-iodine diet by combining a second common application range of diseases in the disease diagnosis coding library, and finally, incorporating a nutrition knowledge coding library, wherein the steps are similar to the construction of the nutrition knowledge coding library according to the hospital diet coding library and the disease diagnosis coding library;

in one example of this embodiment, an own nutrition knowledge coding base may be built: and establishing a corresponding coding library according to the association between the hospital meal and the disease diagnosis, and finding out corresponding diseases according to the applicable range of the hospital meal.

In one embodiment, the diet doctor order analysis and output module is internally provided with a machine learning model, the machine learning model takes basic information and corresponding nutrition knowledge of a patient as a sample training set, the machine learning model learns according to the sample training set, and an optimized machine learning model is generated.

In one embodiment, the system further comprises: a basic information updating module;

the basic information updating module is used for periodically updating the basic information of the patient; and sharing the patient's nutritional recipe on different terminals.

In this embodiment, the basic information updating module includes an information marking unit and an information updating unit;

marking certain categories in the basic information through an information marking unit, and then replacing or rewriting the marked categories of the basic information according to the acquired latest basic information so as to realize updating;

generally, the name, sex and height of the patient are not changed in a longer period of time, so that the patient does not need to be updated; the disease diagnosis, laboratory examination and medical record abstract information of the patient need to be updated in time; updating this information can result in a more accurate nutritional recipe that matches the patient.

In one embodiment, the system further comprises: the allergen database is used for storing basic information and allergen information of the patient and the family members of the patient; the allergen database can assist the diet doctor order analysis and output module to analyze the theoretical nutrient range of the patient so as to formulate a nutrient recipe conforming to the patient.

In the embodiment, the allergen of the patient and the family members of the patient is recorded, so that the manufacturing efficiency of the nutrition recipe can be accelerated, and the risk of the patient and the family members of the patient can be ensured by verifying the examination information of the laboratory; meanwhile, data support can be provided for the confirmation of the allergen of the patient and the family members of the patient;

For example: in practice, some patients with physique are allergic to eggs and some patients are allergic to soybeans, and eggs and soybeans are commonly used as common foods in a recommended list of the nutrition recipes, so that the established allergen database can provide reliable and safe auxiliary support for outputting the nutrition recipes.

Referring to fig. 3, an embodiment of the present invention provides a method for recommending a nutrition recipe, which is used in the nutrition recipe recommendation system, and the specific method includes the following steps:

s10: basic information, namely disease diagnosis, height, weight, age, laboratory examination and medical record summary information of a patient in hospital is acquired through at least one system of a Hospital Information System (HIS), a test information system (LIS) and an electronic medical record system (EMR);

s20: and (5) nutrition knowledge retrieval: according to the disease diagnosis, laboratory examination and medical record abstract information of the patient obtained in the step S10, retrieving the nutritional knowledge required by the patient from a nutritional knowledge coding library through a coding library index module;

wherein, the step of retrieving the nutritional knowledge through the basic information is as follows:

firstly, inputting the obtained disease diagnosis, laboratory examination and medical record summary information of a patient into a coding library index module;

The coding library index module judges whether the information recorded in the hospital meal coding library and the disease diagnosis coding library is recorded;

if the information recorded in the hospital meal coding library and the disease diagnosis coding library is recorded, nutrition knowledge can be directly output;

if the information is not in the hospital meal coding library and the disease diagnosis coding library, the nutrients are divided into macro nutrients and micro nutrients according to the nutrient category, and the information recorded in the self-built meal coding library and the disease diagnosis coding library outputs nutrition knowledge;

s30: nutrient calculation: according to the height, weight and age information of the patient obtained in the step S10, calculating a theoretical nutrient range required by the patient in a nutrient algorithm module by at least one algorithm of a standard weight method, a basic metabolism method and a DRIS method;

the theoretical nutrient range required by the patient is calculated by the basic information as follows:

firstly, according to age classification, if a patient is more than or equal to 18 years old, the patient is included in a group of databases of adults to calculate related nutrients; patients < 18 years old, the related nutrients are calculated by inclusion in the underage group of databases;

for example: patient < 18 years old, since calculated according to calories described in the national institutes of diet, the following table 5 is specific:

If the patient is more than or equal to 18 years old, the method is divided into the following steps according to the height and weight information of the patient: three kinds of height, weight, height, weight and weight;

when the height and weight information exists, calculating by adopting a standard weight method;

the standard weight calculation method is based on the world health organization calculation method;

male: (height cm-80) ×70% =standard body weight;

female: (height cm-70) ×60% = standard weight;

for the calculation of calories corresponding to standard weight, the labor intensity of the patient needs to be evaluated, and the patient can be judged according to a recognized physical labor comparison table (table 6):

for example: the patient's occupation is a teacher, and then the labor intensity can be judged as a light manual worker.

Finally, the daily required heat of the patient is calculated, and then the daily required heat calculation table of different crowds can be consulted for judgment, and the specific table is shown as the following table 7:

for the final heat required by the patient, the calculation is required according to the standard weight calculation and the manual labor comparison table.

Examples: patient Li Mou, male, 178cm, teacher;

firstly, calculating the standard weight of the weight-reducing agent: (178-80) ×70% =68.6 kg;

and judging the physical activity of the teacher as a light physical worker according to the working property of the teacher.

Finally, calculating according to normal weight, calculating the daily required calories according to 25-30kcal per kilogram per day, and calculating Li Mou daily required calories interval ranges as follows:

(25×68.6-30×68.6) = (1715-2058) kcal/day.

Three major energy-producing nutrients, such as protein, carbohydrate, fat, should be calculated after daily caloric value is calculated for the patient.

Protein = standard body weight x recommended protein amount (g/kg/day);

remarks: the recommended amount of protein is in a specific recommended range according to the Chinese resident nutrient guidelines, for example, the daily required amount of the protein for adults is (1.0-1.2) g/kg/day;

fat= (calories x percentage recommended)/9 (g/kg/day);

remarks: fat is recommended to have an interval range according to the Chinese resident nutrition guidelines, for example, the daily proportion of fat of adults is 20-30% of total calories;

carbohydrate= (calories-protein x 4-fat x 9)/4 (g/kg/day);

the standard weight of patient Li Mou calculated above was 68.6kg and the daily required caloric range was (1715-2058) kcal/day, and the daily required protein, fat and carbohydrate range for patient Li Mou was obtained according to the protein, fat and carbohydrate calculation formula, as follows:

Protein= (1.0×68.6-1.2×68.6) = (68.6-82.3) g/day;

fat= [ (1715-2058) × (20-30%) ]/9= (38-68.6) g/day;

carbonized compound= [ (1715-2058) - (68.6-82.3) ×4- (38-68.6) ×9]/4

= (274.7-277.85) g/day;

in one example, when there is height and weight information, a basal metabolic method is adopted for calculation;

the basic metabolic method calculates the formula:

male: (13.75×body weight+5×height-6.76×age+ 66.47) ×1.16 (activity coefficient);

female: (9.56×weight+1.85×height-4.68×age+ 655.10) ×1.19 (activity coefficient);

remarks: this formula is derived from the international basic energy formula, from washington, usa in 1919: harris-Benedict finds that it is suitable for most people, and can customize a relatively reasonable nutrition scheme for individuals, and is widely used.

Remarks: the reason for adopting the basic metabolism method is that the information of the patient is collected, the patient has a height and a weight, and the method is suitable for calculating the height, the weight and the age of the patient by adopting the basic metabolism method.

Examples: patient Li Mou, male, 178cm,65kg,35 years old, teacher;

the daily required calories are calculated according to the basal metabolic method:

(13.75x65+5 x 178-6.76 x 35+66.47) ×1.16= 1871.8 kcal/day;

The caloric calculation interval needs to acquire the final daily required caloric range of the patient according to international unified standard according to + -5% fluctuation, so that (1871.8 ×0.95-1871.8 ×1.05) = (1778.2-1965.4) kcal/day; likewise, the three nutritional products protein, fat, carbohydrate can be calculated according to the calculation formula in step III, as follows:

protein= (1.0×65-1.2×65) = (65-78) g/day;

fat= [ (1778.2-1965.4) × (20-30%) ]/9= (39.5-65.5) g/day;

carbohydrate = [ (1778.2-1965.4) - (65-78) ×4- (39.5-65.5) ×9]/4

= (266-290.7) g/day;

in one example, the DRIS method is used to calculate when there is no height or weight;

dietary nutrient reference intake (DRIS, dietary Reference Intakes) refers to a set of reference values for daily dietary nutrient intake for different populations that are generated to ensure proper nutrient intake by the human body, avoiding the occurrence of excessive or deficient amounts of certain nutrients. The data sources are all "Chinese resident dietary nutrient reference intake (2013)".

Examples: patient Li Mou, male, 35 years old;

the average daily required caloric value of adult males is found to be 1875kcal according to the reference intake (2013) of dietary nutrients of Chinese residents, and the final daily required caloric range of patients is obtained according to the international unified standard and the fluctuation of +/-5% in a caloric calculation interval, so that (1875×0.95-1875×1.05) = (1781.2-1968.7) kcal/day, for three-capacity nutrients, the average daily required caloric value of adult males is recommended to be 65 g/day, the average daily required caloric value of females is recommended to be 55 g/day, the final daily required caloric value of protein is calculated according to the international unified standard +/-10%, and fat and carbide can be calculated according to the calculation formula when weight information exists in height, specifically, protein= (65×0.9-65×1.1) = (58.5-71.7) g/day, fat= (39.6-65.6) g/day, and carbohydrate= (272.9-297.7) g/day.

It should be noted that: for the nutrient algorithm module, the heat, the carbohydrate, the protein and the fat are calculated according to basic information of a patient, such as height, weight, height and weight. It can be seen that the results calculated by each method have certain difference, because the basic information of the patient is different, according to different algorithms, the smaller the basic information of the patient is, the larger the estimated difference is, and the final difference of the results is within +/-20%. This also alerts the healthcare worker that the patient basic information is as detailed as possible, such as the patient height and weight, so that the calculation is more accurate than the estimation of the weight without height.

In one embodiment, the method further comprises:

s40: comprehensive nutrition comparison analysis: and comprehensively comparing the nutritional knowledge required by the patient retrieved in the step S20 with the theoretical nutrient range required by the patient calculated in the step S30 in a diet doctor advice analysis and output module, and outputting a nutrient recipe meeting the requirements of the patient according to the comparison result.

According to the method, a standardized nutrient recommended quantity is obtained, and a corresponding nutrient recipe is provided;

in one example of this embodiment, if none of the patient's personal disease conditions are within the established nutrient model, i.e., the hospital meal code library; the nutrient recommendation can be empirically added or modified and can be set as a classical nutritional formula template for the next patient of the same type.

In the embodiment, by utilizing a computer intelligent technology, only basic information of the existing patient is needed to be input, and each set module can be used as a calculation basis and an inference basis according to various disease nutrition guidelines, so that an accurate nutrient recommendation scheme is ensured to be calculated according to information input, and an effective method is provided for supporting treatment of clinical nutrition departments.

One specific application of this embodiment is:

step 101, acquiring basic information of a patient through a Hospital Information System (HIS), which is specifically shown in the following table 8:

table 8 shows basic information of patient (Li Mou wind)

Name of name

Sex (sex)

Age of

Height of body

Weight of body

Occupation of

Diagnostic information

Li Mou wind

Man's body

28

170cm

Without any means for

Sports teacher

Thyroiditis (thyroiditis)

Step 102, inputting diagnostic information "thyroiditis" in the obtained basic information into a coding library index module, and searching a thyroiditis disease code DIS001 through the coding library index module, wherein the corresponding diet code is 001: the general diet, based on table 2 above, can give the patient the desired nutritional knowledge, as shown in table 9 below:

table 9 shows the nutritional knowledge required by the patient (Li Mou wind)

/>

Step 103, inputting age, sex, height, weight and occupation information in the obtained basic information into a system, and calculating a theoretical nutrient range required by a patient, wherein the patient lacks weight information, so that the patient needs to calculate according to the standard weight method, and the method is specifically as follows:

Standard body weight= (170-80) ×70% = 63 (kg);

according to the physical labor comparison table, the patient can know that the patient is in middle physical labor, so that the daily required heat of the patient is (30-35) kcal/kg/day, and the daily required heat range of the patient can be (1890-2205) kcal/day;

further, the range of protein, fat, and carbide, which are required daily, can be calculated, as follows:

protein=63× (1.0-1.2) = (63-75.6) g/day;

fat= [ (1890-2205) × (20-30%) ]/9= (42-73.5) g/day;

carbohydrate = [ (1890-2205) - (63-75.6) ×4- (42-73.5) ×9]/4

= (310.3-315) g/day;

step 104, according to the heat, protein, fat and carbohydrate data calculated in step 103 and the nutrition knowledge queried in step 102, a nutrition recipe of the patient can be output, specifically:

breakfast: about 200ml of milk, about 50g of eggs, about 100g of corns and about 60g of cucumbers;

lunch: about 200g of oat meal, about 120g of metapenaeus ensis, about 100g of lean meat and about 200g of green vegetables;

dinner: about 150g of noodles, about 50g of mushrooms, 100g of green vegetables and about 80g of drumstick meat.

The second specific application of this embodiment is:

step 201, acquiring basic information of a patient through a Hospital Information System (HIS), which is specifically shown in the following table 10:

Table 10 shows basic information of patient (Zhao Moulan)

Name of name

Sex (sex)

Age of

Height of body

Weight of body

Occupation of

Diagnostic information

Zhao Moulan

Female

45

158cm

50kg

Office staff

High heat

Step 202, inputting the diagnosis information 'high fever' in the obtained basic information into a coding library index module, and retrieving a disease code DIS004 of 'high fever' through the coding library index module, wherein the corresponding diet code is 004: the fluid diet, based on the above table 2, can give the patient the desired nutritional knowledge, as shown in table 11 below:

table 11 is the nutritional knowledge required by the patient (Zhao Moulan)

/>

Step 203, inputting the obtained age, height, weight and occupation information into a system, and calculating the theoretical nutrient range required by the patient, wherein the calculation is required according to the basic metabolic method because the height and weight information of the patient are complete, and the method is specifically as follows:

patient daily required calories = (9.56×50+1.85×158-4.68×45+655.10) ×1.19

= 1445.6 kcal/day;

the patient's daily caloric requirement ranges from 1445.6 × (0.95-1.05) = (1373.3-1517.9) kcal/day;

further, the range of protein, fat, and carbide, which are required daily, can be calculated, as follows:

protein=50× (1.0-1.2) = (50-60) g/day;

Fat= [ (1373.3-1517.9) × (20-30%) ]/9= (30.5-50.6) g/day;

carbohydrate = [ (1373.3-1517.9) - (50-60) ×4- (30.5-50.6) ×9]/4

= (205.6-224.7) g/day;

step 204, according to the heat, protein, fat and carbohydrate data calculated in step 203 and the nutrition knowledge queried in step 202, a nutrition recipe can be prescribed to the patient, specifically:

breakfast: about 300ml millet soup;

early noon point: about 200ml of milk;

lunch: about 250ml of vegetable juice;

late noon: about 200ml of soybean milk;

dinner: about 250ml of fruit juice;

late: the thin lotus root starch is about 200 ml.

The third specific application of this embodiment is:

step 301, obtaining basic information of a patient through a test information system (LIS), which is shown in the following table 12:

table 12 basic information of patient (Ma Mou country)

Name of name

Sex (sex)

Age of

Height of body

Weight of body

Occupation of

Diagnostic information

Ma Mou nation

Man's body

68

Without any means for

Without any means for

Retirement staff

Hypercholesterolemia

Step 302, inputting the diagnosis information "hypercholesterolemia" in the obtained basic information into the coding library index module, and retrieving the disease code DIS009 of "hypercholesterolemia" by the coding library index module, wherein the corresponding diet code is 009: the low cholesterol diet, based on table 3 above, can give the patient the desired nutritional knowledge, as shown in table 13 below:

TABLE 13 knowledge of nutrition required by patients (Ma Mou nation)

Step 303, inputting the obtained age, height, weight and occupation information into a system, and calculating the theoretical nutrient range required by the patient, wherein the calculation is required according to the DRIS method because the information of the height and the weight of the patient is not available, and the method specifically comprises the following steps:

firstly, finding out that the daily average value of the calories required by the aged men is 2050kcal according to the reference intake (2013) of dietary nutrients of Chinese residents. For three major-capacity nutrients, the average protein recommended by elderly men is recommended to be 65 g/day for the chinese resident dietary nutrient reference intake (2013);

the patient's daily required calories range is (2050 x 0.95-2050 x 1.05) kcal/day= (1947.5-2152.5) kcal/day;

further, the range of protein, fat, and carbide, which are required daily, can be calculated, as follows:

protein (g) = (58.5-71.7) g/day;

fat (g) = (43.3-71.8) g/day;

carbohydrate (g) = (243.4-349.8) g/day;

step 304, according to the heat, protein, fat and carbohydrate data calculated in step 303 and the nutrition knowledge queried in step 302, a nutrition recipe can be prescribed to the patient, specifically:

Breakfast: about 250ml of milk, about 30g of oatmeal, about 40g of baked whole wheat steamed bread, about 50g of nuts and about 250g of fruits;

lunch: about 150g of sweet potato rice (wherein the sweet potato is about 100 g), about 100g of pure lean meat, about 200g of green vegetables and about 100g of bean curd.

Dinner: about 200ml of eight-treasure porridge, about 200g of fried green vegetables and about 50g of peeled chicken leg meat.

A fourth specific application of this embodiment is:

step 401, obtaining basic information of a patient through an electronic medical record system (EMR), which is specifically shown in the following table 14:

table 14 shows basic information of patient (plum)

Name of name

Sex (sex)

Age of

Height of body

Weight of body

Occupation of

Diagnostic information

Some plums

Female

55

161

Without any means for

At home

Chemotherapy stage of breast cancer

Step 402, inputting the diagnosis information "breast cancer chemotherapy period" in the obtained basic information into a coding library index module, searching the coding information of the disease code of the "breast cancer chemotherapy period" in the existing nutrition coding library through the coding library index module, at this time, dividing according to nutrient categories and diet of patients in the breast cancer tumor chemotherapy period, and establishing a new code 014 according to the Chinese tumor nutrition therapy guideline that diet medical advice should be given as high-calorie high-protein diet, and combining macro-nutrition high-calorie and high-protein nutrition: high calorie high protein diet, corresponding to disease codes: DIS014: patients with tumor chemotherapy stage and incorporating new disease diagnosis and diet codes into the existing hospital nutrition knowledge code library for the next use, as shown in table 15 below;

Table 15 shows new codes 014 in the nutrition knowledge code base

Step 403, inputting the obtained age, height, weight and occupation information into a system, and calculating the theoretical nutrient range required by the patient, wherein the calculation is required according to the standard weight method due to the lack of the weight information of the patient, and the specific steps are as follows:

standard body weight= (161-70) ×60% = 54.6 (kg);

according to the physical labor comparison table, the patient can be known to be lightly labor-saving, so that the daily required heat of the patient is (25-30) kcal/kg/day, and the daily required heat range of the patient is (1365-1638);

further, the range of protein, fat, and carbide, which are required daily, can be calculated, as follows:

protein=54.6x6 (1.2-1.5) = (65.5-81.9) g/day;

fat= [ (1365-1638) × (20-30%) ]/9= (30.3-54.6) g/day;

carbohydrate= [ (1365-1638) - (65.5-81.9) ×4- (30.3-54.6) ×9]/4

= (204.8-207.6) g/day;

step 404, according to the heat, protein, fat, carbohydrate data calculated in step 403 and the nutrition knowledge of step 402, a nutrition recipe can be prescribed to the patient, specifically:

breakfast: about 250ml of coarse cereal porridge, about 50g of steamed bread, about 60g of eggs and about 40g of nuts;

Breakfast: about 250ml of milk and about 250g of fruit;

lunch: about 200g of rice, about 100g of lean meat, about 60g of fish meat, about 80g of bean products and about 200g of green vegetables;

noon point: about 200ml of soybean milk and about 30g of instant oat;

dinner: about 60g of noodles, about 40g of lean meat and about 100g of green vegetables, and vegetable oil is selected as much as possible for cooking.

In another embodiment, a method for recommending a nutrient recipe is used for the nutrient recipe recommendation system, and the method comprises the following steps:

acquiring basic information of a patient;

retrieving nutritional knowledge corresponding to the disease from the nutritional knowledge code base according to the basic information;

calculating the theoretical nutrient range required by the patient according to the basic information;

and analyzing according to the nutrition knowledge and the theoretical nutrient range, selecting foods matched with the patient, and outputting at least one nutrition recipe which meets the needs of the patient.

In this embodiment, the analysis is performed according to the nutrition knowledge and the theoretical nutrient range, and the food matched with the patient is selected, so that two nutrition recipes meeting the needs of the patient can be output;

the two nutrition recipes which meet the needs of the patient comprise a first nutrition recipe and a second nutrition recipe;

The first nutrient recipe is a recipe that is more matched to the range of nutrients required by the patient, and the second nutrient recipe is a recipe that is more matched to the range of nutrients required by the patient; however, the first nutrient recipe contains allergens of the family members of the patient, so that in this scenario the second nutrient recipe can be replaced by the first nutrient recipe and output.

In one example of this embodiment, the analysis is based on the nutritional knowledge and the theoretical nutrient range, and selecting the food matching the patient is not limited to outputting a nutrient recipe that meets the patient's needs.

In one embodiment, the step of retrieving the nutritional knowledge corresponding to the disease in the nutritional knowledge code library according to the basic information specifically includes:

inputting the acquired basic information of the patient into an encoding library index module;

judging whether the disease diagnosis result in the basic information is recorded in a hospital meal coding library and a disease diagnosis coding library or not through a coding library index module;

if the nutrition knowledge is recorded in the hospital meal coding library and the disease diagnosis coding library, directly outputting the nutrition knowledge corresponding to the disease;

if the nutrient is not recorded in the hospital meal coding library and the disease diagnosis coding library, the nutrient categories are classified into macro nutrients and micro nutrients according to the nutrient categories;

And searching the divided results in the self-built meal coding library and the disease diagnosis coding library, and outputting nutrition knowledge corresponding to the disease according to the search results.

In one embodiment, the step of calculating the theoretical nutrient range required by the patient according to the basic information specifically includes:

selecting a calculation mode of theoretical nutrients required by a patient according to the basic information;

according to the selected calculation mode, basic information of a patient is input based on a preset guideline, and calculation of heat required by the patient is performed; alternatively, the patient's required calories are calculated by a nutrient algorithm module to output the patient's required theoretical nutrient range.

The selected calculation method can be selected from standard body weight method, basal metabolic method, and DRIS method.

In an example of this embodiment, the basic information obtaining module may use a keyboard, a card reader, an input interface of a touch display screen, or a protocol; the input interface or protocol is connected with the input/output interface of the nutrition knowledge coding base and is used for inputting/outputting data; and the related APIs can be called based on Python.

In one example, the base information update module has associated with it one or more Applications (APP) that may be installed in one or more device terminals for use by associated staff, such as a patient's family doctor, child, companion, nutritional technician, etc.

The nutrition recipe recommendation system provided by the above can be realized by a computer program installed in a computer device; as shown in fig. 4, the computer device includes a processor, a memory, a network interface, an input device, and a display screen connected by a system bus. The memory includes a nonvolatile storage medium and an internal memory. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, the input device of the computer equipment can be a touch layer covered on the display screen, can also be keys, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.

The patient's nutritional recipe can be viewed on different terminals using an associated Application (APP); different terminals include, but are not limited to, cell phones, computers, displays, etc.

The nutrient recipe recommendation system provided by the embodiment and the nutrient recipe recommendation method provided based on the nutrient recipe recommendation system can overcome the defects that the professional ability of clinical nutrition doctors in all hospitals is uneven and the nutrient prescriptions of patients are different by utilizing a computer intelligent technology; basic information of a patient is input, and each module is used as a calculation basis and an reasoning basis according to the existing various disease nutrition guidelines, so that an accurate nutrient recommendation scheme is calculated according to the input information, and an effective method is provided for supporting treatment of clinical nutrition departments.

It should be noted that, fig. 4 shows a computer device; the computer device includes a processor, a memory, a network interface, an input device, and a display screen connected by a system bus. The memory includes a nonvolatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program which, when executed by the processor, causes the processor to implement a method of nutritional recipe recommendation. The internal memory may also have stored therein a computer program which, when executed by the processor, causes the processor to perform a method of nutritional recipe recommendation. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, the input device of the computer equipment can be a touch layer covered on the display screen, can also be keys, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.

Those skilled in the art will appreciate that all or part of the processes in the methods of the above embodiments may be implemented by a computer program for instructing relevant hardware, where the program may be stored in a non-volatile computer readable storage medium, and where the program, when executed, may include processes in the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the various embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.

The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, and alternatives falling within the spirit and principles of the invention.

Claims (10)

1. A nutritional recipe recommendation system, the system comprising: the system comprises a basic information acquisition module, a coding library index module, a nutrient algorithm module and a diet doctor order analysis and output module;

the basic information acquisition module is used for acquiring basic information of a patient, wherein the basic information comprises disease diagnosis results, height, weight, age, laboratory examination and medical record summary information;

the coding library index module is coupled with a predetermined nutrition knowledge coding library and is used for searching nutrition knowledge corresponding to the disease in the nutrition knowledge coding library according to the basic information;

the nutrient algorithm module is used for calculating the theoretical nutrient range required by the patient according to the basic information;

the diet doctor advice analysis and output module is used for analyzing according to the nutrition knowledge retrieved by the encoding library indexing module and the theoretical nutrient range required by the patient calculated by the nutrient algorithm module, selecting foods matched with the patient, and outputting at least one nutrition recipe meeting the requirement of the patient.

2. The nutritional recipe recommendation system of claim 1, wherein the basic information acquisition module comprises a hospital information system, a test information system, an electronic medical record system;

the basic information can be acquired through at least one of the hospital information system, the inspection information system and the electronic medical record system.

3. The nutritional recipe recommendation system of claim 1, wherein the nutritional knowledge coding library comprises a hospital meal coding library, a disease diagnosis coding library, and a self-built meal coding library, wherein the hospital meal coding library and the disease diagnosis coding library are in communication with each other, and wherein the self-built meal coding library and the disease diagnosis coding library are in communication with each other;

the hospital meal coding library and the disease diagnosis coding library are connected in a first common application range, the self-built meal coding library and the disease diagnosis coding library are connected in a second common application range, and the first common application range and the second common application range represent the matching degree of diseases, nutrients and foods.

4. A nutrient recipe recommendation system as claimed in claim 3 wherein the self-built meal code library is built according to nutrient class division and nutrient classes are divided into macro-and micro-nutrients.

5. The nutritional recipe recommendation system according to claim 1, wherein the diet physician order analysis and output module is built with a machine learning model that uses basic information of a patient and corresponding nutritional knowledge as a sample training set, the machine learning model learns from the sample training set, and generates an optimized machine learning model.

6. The nutritional recipe recommendation system of claim 1, wherein the nutrient algorithm module comprises a standard weight method, a basal metabolic method, a DRIS method, and the theoretical nutrient range is determined by at least one algorithm calculation of the standard weight method, basal metabolic method, and DRIS method.

7. The nutritional recipe recommendation system according to claim 1, wherein the system further comprises: the allergen database is used for storing basic information and allergen information of the patient and the family members of the patient; the allergen database can assist the diet doctor order analysis and output module to analyze the theoretical nutrient range of the patient so as to formulate a nutrient recipe conforming to the patient.

8. A method of nutritional recipe recommendation, for use in a nutritional recipe recommendation system as claimed in any one of claims 1 to 7, the method comprising the steps of:

Acquiring basic information of a patient;

retrieving nutritional knowledge corresponding to the disease from the nutritional knowledge code base according to the basic information;

calculating the theoretical nutrient range required by the patient according to the basic information;

and analyzing according to the nutrition knowledge and the theoretical nutrient range, selecting foods matched with the patient, and outputting at least one nutrition recipe which meets the needs of the patient.

9. The method according to claim 8, wherein the step of retrieving the nutritional knowledge corresponding to the disease in the nutritional knowledge code base according to the basic information comprises:

inputting the acquired basic information of the patient into an encoding library index module;

judging whether the disease diagnosis result in the basic information is recorded in a hospital meal coding library and a disease diagnosis coding library or not through a coding library index module;

if the nutrition knowledge is recorded in the hospital meal coding library and the disease diagnosis coding library, directly outputting the nutrition knowledge corresponding to the disease;

if the nutrient is not recorded in the hospital meal coding library and the disease diagnosis coding library, the nutrient categories are classified into macro nutrients and micro nutrients according to the nutrient categories;

And searching the divided results in the self-built meal coding library and the disease diagnosis coding library, and outputting nutrition knowledge corresponding to the disease according to the search results.

10. The method of claim 8, wherein the step of calculating a theoretical nutrient range required by the patient based on the basic information, comprises:

selecting a calculation mode of theoretical nutrients required by a patient according to the basic information;

according to the selected calculation mode, basic information of a patient is input based on a preset guideline, and calculation of heat required by the patient is performed; alternatively, the patient's required calories are calculated by a nutrient algorithm module to output the patient's required theoretical nutrient range.