Intelligent hospital catering platform and method integrating dining features and health management

Abstract

The invention provides an intelligent hospital catering platform and method fusing dining features and health management, and an architecture diagram of the platform, and the platform comprises a patient ordering system which is used for recognizing the dining features of a patient according to the illness state, the dining habits and the nutritional requirements of the patient, and recommending a first personalized menu to the patient based on the dining features of the patient; the employee ordering payment system is used for identifying employee dining features according to the health data, the dietary habits and the nutritional requirements of the employees and recommending a second personalized menu to the employees based on the employee dining features; and the recommendation system is used for recommending the menu to the corresponding user side. By combining an existing payment ordering system on the market and through AI upgrading, the system is more humanized, and scientific diet is achieved. The AI technology is integrated into the patient ordering system and the staff ordering payment system of the hospital, so that the dining experience can be remarkably improved, the nutrition management is optimized, and the patient and the staff are helped to establish healthy dietary habits.

Description

Technical Field

[0001] The present invention relates to the field of catering technology, and in particular to an intelligent hospital catering platform that integrates dining features and health management, and its application method, electronic equipment and computer-readable storage medium. Background Art

[0002] The operating steps of the existing payment ordering system on the market:

[0003] Users download and install the food ordering app or visit the food ordering website.

[0004] The user registers an account and logs in.

[0005] Users browse the menu and select the food and drinks they want to order.

[0006] The user adds the selected food items to the shopping cart.

[0007] The user confirms the order details on the shopping cart page, including food selection, quantity, price, etc.

[0008] The user selects a payment method such as credit card, debit card, e-wallet, bank transfer or cash on delivery.

[0009] The user enters payment information such as card number, expiration date, security code or performs e-wallet authorization.

[0010] The user submits the order and waits for payment confirmation.

[0011] The payment system processes the payment request and confirms that the funds have been successfully transferred.

[0012] The user receives payment confirmation and order status updates.

[0013] The restaurant starts preparing the food after receiving the order.

[0014] The delivery person delivers food according to the user's address.

[0015] The user receives the food and confirms that the order is complete.

[0016] Existing payment and meal ordering systems on the market primarily select meals based on merchant offerings, and are unable to provide personalized, nutritious meal orders based on in-depth user information. This is particularly true for hospital patients and outpatients, who require a balanced diet to aid recovery. Existing payment and meal ordering systems on the market are unable to recommend personalized menus based on a patient's condition, eating habits, or nutritional needs. They are unable to provide personalized dietary advice to aid recovery, and lack AI analysis support.

[0017] Similarly, hospital staff face high-risk tasks daily, interacting with a wide range of patients and requiring a balanced, healthy diet. Existing meal ordering systems are unable to integrate these patient recovery meal options to provide personalized, nutritious meal ordering services for hospital staff. Summary of the Invention

[0018] In order to solve the technical problems existing in the prior art, the present invention provides the following technical solutions:

[0019] On the one hand, an intelligent hospital catering platform integrating dining features and health management is provided, the platform comprising:

[0020] a patient meal ordering system for identifying patient meal characteristics according to the patient's condition, eating habits, and nutritional needs, and recommending a first personalized meal menu to the patient based on the patient meal characteristics;

[0021] An employee meal ordering and payment system, which is used to identify employee dining characteristics based on their health data, eating habits, and nutritional needs, and recommend a second personalized menu to the employee based on the employee dining characteristics;

[0022] Recommendation system, used to recommend menus to corresponding user terminals;

[0023] The patient meal ordering system and the employee meal ordering and payment system are respectively connected to the recommendation system for communication.

[0024] Preferably, the patient meal ordering system includes:

[0025] (1) A quick ordering subsystem, which is used to provide ordering services based on voice input or image recognition technology, and to input a list of dishes to be ordered;

[0026] (2) A nutritional component analysis subsystem, which is used to: obtain the list of dishes ordered by the patient; traverse and parse the name and description information of each dish in the dish list based on NLP technology; input the name and description information of each dish into a preset nutritional database, and the nutritional database calculates the nutritional components contained in each dish in real time; count the nutritional components contained in each dish in the dish list, generate the nutritional components of the patient's order and send it to the recommendation system, and the recommendation system recommends the nutritional components of the order to the corresponding patient end.

[0027] Preferably, the patient meal ordering system further includes:

[0028] (3) Healthy diet suggestion subsystem, used for:

[0029] Based on the patient's health data and diet records, a deep learning model is used to predict the patient's health risks, and healthy diet suggestions that match the patient's health risks are recommended and sent to the personalized recommendation subsystem.

[0030] Preferably, the patient meal ordering system further includes:

[0031] (4) Personalized recommendation subsystem, used for:

[0032] Analyzing the healthy eating recommendations to generate the patient's nutritional needs;

[0033] Using a machine learning algorithm, identifying the patient's dining characteristics from the patient's health data, historical dietary data, and nutritional needs, and recommending a menu that matches the patient's dining characteristics;

[0034] Combined with the preset nutritional data of dishes, each dish whose nutritional components meet the requirements of the menu is recommended, and each dish is written into the preset patient meal list to obtain the first personalized menu and send it to the recommendation system, which then recommends the first personalized menu to the corresponding patient end.

[0035] Preferably, the employee meal ordering and payment system includes:

[0036] (1) Healthy diet suggestion subsystem, used for:

[0037] Based on employees' health data and dietary records, a deep learning model is used to predict their health risks, and daily nutritional intake reports and healthy diet suggestions that match their health risks are recommended and sent to the personalized recommendation subsystem;

[0038] (2) Personalized recommendation subsystem, used for:

[0039] Analyzing the daily nutritional intake report and the healthy diet recommendations to generate health goals and nutritional needs of employees, respectively;

[0040] using a machine learning algorithm to identify the employee's dining characteristics from the employee's health data, historical dietary data, the health goals, and the nutritional needs, and recommending a menu that matches the employee's dining characteristics;

[0041] Based on the preset nutritional data of the dishes, recommend dishes whose nutritional components meet the requirements of the menu, write the dishes into the preset employee meal list, obtain the second personalized menu, and send it to the recommendation system. The recommendation system recommends the second personalized menu to the corresponding employee terminal;

[0042] (3) Fast payment subsystem, used for:

[0043] According to the preset dish pricing rules, each dish in the second personalized menu is charged and counted, and a meal bill for the employee's meal is generated and sent to the recommendation system. The recommendation system recommends the nutritional components of the employee's meal to the corresponding employee end.

[0044] Preferably, the employee meal ordering and payment system further includes:

[0045] (4) A nutritional component analysis subsystem, which is used to: obtain the second personalized menu of the employee; traverse and parse the name and description information of each dish in the second personalized menu based on NLP technology; input the name and description information of each dish into a preset nutrition database, and the nutrition database calculates the nutritional components contained in each dish in real time; count the nutritional components contained in each dish in the second personalized menu, generate the nutritional components of the employee's meal this time and send it to the recommendation system, and the recommendation system recommends the nutritional components of the employee's meal this time to the corresponding employee end.

[0046] On the other hand, an application method of an intelligent hospital catering platform integrating dining characteristics and health management is provided, the application method comprising:

[0047] The user logs into the platform through the terminal and enters the corresponding ordering / dining information;

[0048] The platform's patient meal ordering system identifies the patient's dining characteristics based on the patient's condition, eating habits, and nutritional needs, recommends a first personalized menu to the patient based on the patient's dining characteristics, and recommends the first personalized menu to the patient via the recommendation system;

[0049] The platform's employee meal ordering and payment system identifies employee dining characteristics based on their health data, eating habits, and nutritional needs, and recommends a second personalized menu to the employees based on the employee dining characteristics, and then recommends the second personalized menu to the employee through the recommendation system.

[0050] On the other hand, an electronic device is provided, comprising: a processor; and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the above method is implemented.

[0051] On the other hand, a computer-readable storage medium is provided, wherein at least one instruction is stored in the storage medium, and the at least one instruction is loaded and executed by a processor to implement the above method.

[0052] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0053] The platform provided by the present invention includes: a patient meal ordering system, which is used to identify the patient's dining characteristics based on the patient's condition, eating habits and nutritional needs, and recommend a first personalized menu to the patient based on the patient's dining characteristics; an employee meal ordering and payment system, which is used to identify the employee's dining characteristics based on the employee's health data, eating habits, and nutritional needs, and recommend a second personalized menu to the employee based on the employee's dining characteristics; a recommendation system, which is used to recommend the menu to the corresponding user end. By combining the existing payment and ordering systems on the market and upgrading through AI, it will be more humane and achieve a scientific diet. Integrating AI technology into the hospital's patient meal ordering system and employee meal ordering and payment system can significantly improve the dining experience, optimize nutritional management, and help patients and employees establish healthy eating habits.

[0054] The following expected effects can be produced:

[0055] Patients: Receive personalized diet plans to promote recovery. Monitor daily nutritional intake in real time and improve eating habits.

[0056] Employees: Get healthy eating advice to prevent chronic diseases. Improve ordering and payment efficiency, and enhance the dining experience.

[0057] Hospitals: Smart logistics will enhance canteen management efficiency and reduce waste. This will improve the hospital's information technology and service quality. By integrating AI technology, the hospital's meal ordering and ordering systems will become more intelligent and personalized, providing better services for patients and staff while also contributing to the achievement of health management goals. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0059] Figure 1 This is a block diagram of an intelligent hospital catering platform that integrates dining features and health management, provided by an embodiment of the present invention;

[0060] Figure 2 This is a schematic diagram of the application system composition of a patient meal ordering system provided by an embodiment of the present invention;

[0061] Figure 3 This is a schematic diagram of the application system composition of an employee meal ordering and payment system provided by an embodiment of the present invention;

[0062] Figure 4 It is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0063] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0064] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0065] In the embodiments of the present invention, the terms "image" and "picture" may be used interchangeably. It should be noted that, when the distinction between them is not emphasized, their intended meanings are the same. The terms "of," "corresponding," and "corresponding" may be used interchangeably. It should be noted that, when the distinction between them is not emphasized, their intended meanings are the same.

[0066] In the embodiments of the present invention, sometimes a subscript such as W1 may be mistakenly written as a non-subscript form such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0067] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0068] The embodiment of the present invention provides an intelligent hospital catering platform that integrates dining features and health management. Figure 1 The architecture diagram of the intelligent hospital catering platform that integrates dining features and health management is shown, and the platform includes:

[0069] a patient meal ordering system for identifying patient meal characteristics according to the patient's condition, eating habits, and nutritional needs, and recommending a first personalized meal menu to the patient based on the patient meal characteristics;

[0070] An employee meal ordering and payment system, which is used to identify employee dining characteristics based on their health data, eating habits, and nutritional needs, and recommend a second personalized menu to the employee based on the employee dining characteristics;

[0071] Recommendation system, used to recommend menus to corresponding user terminals;

[0072] The patient meal ordering system and the employee meal ordering and payment system are respectively connected to the recommendation system for communication.

[0073] This platform can be based on the existing payment and ordering system and upgraded through AI, making it more humane and achieving scientific diet. AI technology will be integrated into the hospital's patient ordering system and employee ordering and payment system.

[0074] The platform provides a user-side login API interface. The user side includes the patient side for patients and the employee side for employees. The user side can be an app, mini-program, or webpage, etc., which can log in to the platform to order meals and manage meals.

[0075] Specific: system architecture design

[0076] 1. Patient meal ordering system

[0077] Target users: inpatients, outpatients.

[0078] Core Functions: Recommend personalized menus based on the patient's condition, eating habits, and nutritional needs. Analyze daily nutritional intake in real time. Provide dietary advice to help patients recover.

[0079] 2. Employee meal ordering and payment system

[0080] Target users: Hospital staff.

[0081] Core Functions: Recommends menus based on employees' health data, eating habits, and nutritional needs. Analyzes daily nutritional intake to help employees develop healthy eating habits. Supports fast ordering, payment, and pickup.

[0082] 3. Specific Applications of AI

[0083] 3.1 Personalized Recommendation System

[0084] Technical Implementation: Machine learning algorithms (such as collaborative filtering and content recommendation) are used to analyze users' historical dietary data, health data (such as weight, blood sugar, blood pressure, etc.), and nutritional needs. Combined with the nutritional composition data of dishes provided by the hospital's dietetic department, menus that meet the user's needs are recommended.

[0085] Application scenarios:

[0086] Patients: Recommend low-sugar, low-salt, and low-fat menus based on their condition (such as diabetes, hypertension).

[0087] Employees: Recommend high-protein, low-calorie menus based on health goals (such as weight loss and muscle gain).

[0088] 3.2 Nutritional analysis

[0089] Technical Implementation: Natural Language Processing (NLP) technology is used to parse dish names and descriptions and extract nutritional information. Combined with nutritional databases (such as the Chinese Food Composition Table), the nutritional composition of each meal (such as calories, protein, fat, carbohydrates, vitamins, etc.) is calculated in real time.

[0090] Application scenarios:

[0091] Patients: Generate daily nutritional intake reports to help doctors adjust diet plans.

[0092] Employees: Provide daily nutritional intake analysis to help employees understand their dietary structure.

[0093] 3.3 Healthy Diet Recommendations

[0094] Technical Implementation: Based on the user's health data (such as the Chinese Dietary Guidelines, physical examination reports, medical history) and dietary records, a deep learning model is used to predict health risks (such as obesity and malnutrition). Personalized dietary recommendations are provided (such as increasing dietary fiber and reducing sugar intake).

[0095] Application scenarios:

[0096] Patients: Provide dietary advice during recovery.

[0097] Employees: Help employees prevent chronic diseases and improve their health.

[0098] 3.4 Smart Ordering and Payment

[0099] Technical implementation:

[0100] Use computer vision technology (such as OCR) to identify dish images and automatically generate orders.

[0101] Combined with face recognition or fingerprint recognition technology, fast payment can be achieved.

[0102] Application scenarios:

[0103] Patients: Order food through the tablet device or mobile phone APP in the ward, which supports voice input.

[0104] Employees: Order and pay quickly through the cafeteria’s self-service ordering machine or mobile app.

[0105] The platform can be integrated with hospital management systems such as HIS systems and employee management systems to achieve data sharing and interoperability. The platform can obtain health data, medical records, living and dining habits, dietary records and other data of patients or employees, and make personalized dining recommendations based on AI technology. The specific data sources and integration can be referred to as follows:

[0106] 1. Data Source

[0107] Patient data: Electronic medical records (EMR): medical conditions, allergies, nutritional needs. Dietary records: historical meal ordering data, eating habits.

[0108] Employee data: Health records: physical examination reports, health goals. Dietary records: historical meal ordering data, eating habits.

[0109] Dish data:

[0110] Nutritional information provided by the Dietary Department. Information including images, descriptions, and prices of dishes.

[0111] 2. Data integration: Establish a unified database to integrate patient, employee, and food data. Use data cleaning and standardization techniques to ensure data accuracy and consistency.

[0112] The functions of the platform system will be further described below.

[0113] like Figure 2 As shown, preferably, the patient meal ordering system includes:

[0114] (1) A quick ordering subsystem, which provides an ordering service based on voice input or image recognition technology, and inputs a list of dishes for this order. For example: Voice recognition: The iFlytek Medical ASR engine is used to convert the patient's voice into text (e.g., "a portion of braised pork" → dish ID: A203), with an accuracy rate of ≥95%. Image recognition: The ResNet50 model is used to train 3,000 dish images, and a lightweight MobileNet deployment is used to realize code scanning and dish recognition (e.g., identifying "stir-fried seasonal vegetables" and mapping them to a nutrition database). This embodiment can also use computer vision technology (e.g., OCR) to recognize dish images and automatically generate orders.

[0115] (2) A nutritional component analysis subsystem, which is used to: obtain the list of dishes ordered by the patient; traverse and parse the name and description information of each dish in the dish list based on NLP technology; input the name and description information of each dish into a preset nutritional database, and the nutritional database calculates the nutritional components contained in each dish in real time; count the nutritional components contained in each dish in the dish list, generate the nutritional components of the patient's order and send it to the recommendation system, and the recommendation system recommends the nutritional components of the order to the corresponding patient end.

[0116] System architecture design of nutrient composition analysis subsystem:

[0117] 1. Data input layer

[0118] Obtain patient meal order data through the hospital HIS system interface, including dish ID, name and description text.

[0119] RabbitMQ message queue is used to realize asynchronous data transmission, with peak processing capacity ≥1000TPS.

[0120] 2.NLP parsing engine

[0121] Use the BERT-Base model (Chinese version) for dish entity recognition:

[0122] #Example: Identify the main ingredients and cooking methods of a dish

[0123] nlp_pipeline=BertForSequenceClassification.from_pretrained('bert-base-chinese')

[0124] ingredients = nlp_pipeline.predict("Steamed Sea Bass") # Output: ['Sea Bass', 'Steamed'].

[0125] Build a synonym database to handle dialect expressions (such as "tomato" vs. "tomato").

[0126] 3. Nutrition Database

[0127] Contains nutritional information for over 3000 standard dishes (content per 100g);

[0128] Supports fuzzy matching and ingredient superposition calculation (such as "Fish-flavored Shredded Pork = Pork + Fungus + Carrot").

[0129] Real-time calculation process:

[0130] A[order]-->B{NLP analysis};

[0131] B-->C[Nutrient database query];

[0132] C-->D[ingredient summary];

[0133] D-->E[recommendation system].

[0134] Key algorithms:

[0135] Nutritional weight calculation:

[0136] Total protein = ∑((dish weight_i)×(protein content_i)), where i is the dish type.

[0137] Patient suitability score: Dynamically adjusted in combination with the disease contraindication database (such as potassium restriction in patients with kidney disease).

[0138] 3. Output and Feedback

[0139] Data Output

[0140] The JSON format includes 12 indicators such as total calories, protein, and fat.

[0141] The information is pushed to the mobile app in real time, showing a progress bar on reaching the nutritional standard (e.g. "Protein intake: 65% / day requirement").

[0142] Exception handling

[0143] Unrecognized dishes automatically trigger the manual review process;

[0144] When nutrition exceeds the standard, a graded warning will be triggered (yellow / orange / red color prompts).

[0145] Preferably, the patient meal ordering system further includes:

[0146] (3) Healthy diet suggestion subsystem, used for:

[0147] Based on the patient's health data and diet records, a deep learning model is used to predict the patient's health risks, and healthy diet suggestions that match the patient's health risks are recommended and sent to the personalized recommendation subsystem.

[0148] The specific system architecture of the healthy diet recommendation subsystem can be described as follows:

[0149] Data collection layer

[0150] Obtain patient health indicators (BMI / blood sugar, etc.) through the hospital HIS system connection;

[0151] The mobile APP records daily diet (supports photo recognition and manual input).

[0152] Core Model

[0153] Adopting a dual-channel neural network architecture:

[0154] Channel 1: LSTM processes time-series health data (blood sugar fluctuations, etc.);

[0155] Channel 2: ResNet-18 processes food image recognition;

[0156] Recommendation Engine

[0157] Output three-dimensional suggestions: taboo ingredients / recommended nutrients / menu combinations.

[0158] 2. Model Training Process

[0159] Data preprocessing

[0160] Missing value filling: KNN algorithm is used to fill in the physical examination data;

[0161] Dietary text standardization: BERT-CRF model identifies dish ingredients.

[0162] Feature Engineering

[0163] Key feature dimensions:

[0164]

[0165]

[0166] Feature labeling: Label the corresponding health risks and corresponding healthy eating recommendations.

[0167] Model training

[0168] Loss function: weighted cross entropy (increasing weights for high-risk patients);

[0169] Optimizer: AdamW(lr=3e-5, warmup_steps=1000);

[0170] Early stopping mechanism: If the AUC of the validation set decreases for three consecutive rounds, the algorithm is terminated.

[0171] Model validation performance metrics

[0172] Indicators training set validation set risk prediction AUC recommendation acceptance rate 82.6% 78.3%.

[0173] Preferably, the patient meal ordering system further includes:

[0174] (4) Personalized recommendation subsystem, used for:

[0175] Analyzing the healthy eating recommendations to generate the patient's nutritional needs;

[0176] Using a machine learning algorithm, identifying the patient's dining characteristics from the patient's health data, historical dietary data, and nutritional needs, and recommending a menu that matches the patient's dining characteristics;

[0177] Combined with the preset nutritional data of dishes, each dish whose nutritional components meet the requirements of the menu is recommended, and each dish is written into the preset patient meal list to obtain the first personalized menu and send it to the recommendation system, which then recommends the first personalized menu to the corresponding patient end.

[0178] The architecture and application of the personalized recommendation subsystem can be referred to as follows:

[0179] 1. The architectural design of the personalized recommendation subsystem is as follows

[0180] Module division

[0181] Data collection layer: connects to health monitoring equipment, hospital information systems, and user-side apps to obtain real-time patient health data (blood sugar, blood pressure, BMI, etc.), historical dietary records, allergen information, medical advice and contraindications, and other structured and unstructured data.

[0182] Data processing layer: Deploy the ETL (Extract-Transform-Load) tool chain, including Apache Kafka real-time data stream processing and Spark batch data cleaning, to complete data normalization (such as unifying units to kilocalories and milligrams), fill missing values (mean or interpolation based on medical records), and filter outliers (such as marking daily intake exceeding 3,000 kilocalories).

[0183] Recommendation engine layer: Integrates the TensorFlow / PyTorch machine learning framework to build a collaborative filtering model based on user portraits, a constraint satisfaction model based on the nutritional composition of dishes, and a real-time feedback reinforcement learning model.

[0184] Service interface layer: Provides menu generation services through RESTful API, defining the request body (including user ID, time range, special scenario tags) and response body (including menu ID, dish list, nutritional analysis report) in JSON format.

[0185] Technology stack selection

[0186] Database: PostgreSQL (storing user portrait feature vectors), MongoDB (storing unstructured diet logs), and Redis (caching frequently accessed nutritional data of dishes).

[0187] Computing framework: Spark MLlib is used for offline model training, and Flink processes real-time diet feedback data streams.

[0188] Deployment environment: Docker containerization is deployed in a Kubernetes cluster, and service monitoring is achieved through Prometheus+Grafana.

[0189] 2. Health Data and Nutritional Needs Analysis Module

[0190] Multi-source data integration

[0191] The system is connected to the hospital's EMR system through the HL7 protocol to parse nutrition-related instructions in the medical order text (such as "low-sodium diet" and "high protein after surgery"), and the BERT-NER model is used to extract entities (restricted substances, target intake).

[0192] Wearable device data (such as smart bracelet step counts and body fat scale data) is obtained through OAuth2.0 authorization and converted into resource objects according to the FHIR standard.

[0193] Nutrient requirement quantification model

[0194] Calculation of basal metabolic rate (BMR): Use the Mifflin-St Jeor formula to calculate static energy expenditure based on age, gender, weight, and height.

[0195] Dynamic demand adjustment:

[0196] Disease factors: The carbohydrate intake of diabetic patients is set at 45%-60% of total calories according to the "Chinese Diabetes Dietary Guidelines", and the protein intake of kidney disease patients is controlled in segments according to the eGFR value.

[0197] Exercise compensation: Based on the exercise energy consumption data provided by Huawei HealthKit, 50 kcal of dietary supply is added for every 100 kcal of exercise.

[0198] Nutrient element distribution: Through a linear programming model, the optimal ratio of protein (1.2-1.5g / kg), fat (20-30% of total calories), carbohydrates, and vitamins (such as vitamin C ≥ 100mg / day) is solved under the total calorie constraint.

[0199] 3. Patient Dining Feature Modeling

[0200] Feature Engineering

[0201] Static features:

[0202] Demographics: age groups (children / adults / elderly), gender-derived characteristics (e.g., women have a 20% increase in iron requirements).

[0203] Pathological features: ICD-10 codes are mapped to nutritional intervention categories (e.g., E11.9 → diabetic diet).

[0204] Dynamic features:

[0205] Dietary preferences: Construct dish embedding vectors through historical ordering data, and use t-SNE to reduce the dimension and cluster them (e.g., Cluster 1: preference for high-sodium Sichuan dishes, Cluster 2: preference for Mediterranean diet).

[0206] Compliance assessment: Calculate the historical violation rate of contraindications in medical advice (e.g., the number of times kidney patients ordered high-potassium dishes / total number of meals ordered).

[0207] Machine learning model training

[0208] Collaborative filtering improvements:

[0209] Pathological constraints are introduced into the matrix decomposition model, and L2 regularization terms are added to the loss function (according to the importance of each term).

[0210] The SGD optimizer was used with a learning rate of 0.005 and 500 epochs.

[0211] Deep Interest Network:

[0212] A dual-tower DNN model was constructed. The user tower input included health indicators (200 dimensions) and diet log sequences (encoded into 128-dimensional vectors through LSTM); the dish tower input included nutritional ingredients (20 dimensions) and cooking methods (one-hot encoding).

[0213] Cosine similarity is calculated as the recommendation score, and Triplet Loss optimizes the feature space distance.

[0214] 4. Menu Generation and Dish Matching

[0215] Candidate set screening

[0216] Construct a dish map based on a nutritional database (such as the Chinese Food Composition Table). The node attributes include:

[0217] Macronutrients: calories, protein, fat, carbohydrates

[0218] Trace elements: sodium (accurate to mg), dietary fiber, vitamins A / B / C

[0219] Allergen labeling: 8 common allergens including peanuts, gluten, etc.

[0220] Use Dijkstra's algorithm to find the optimal path under nutrition constraints: def generate_menu(nutrition_constraints):

[0221] candidate_dishes=filter_dishes(constraints),

[0222] graph=build_nutrition_graph(candidate_dishes),

[0223] return find_optimal_path(graph,constraints).

[0224] Multi-objective optimization

[0225] Establish the objective function: MaximizeΣ(user_preference_score)

[0226] Subject to:

[0227] Σ(calories)∈[BMR×0.9,BMR×1.1]

[0228] Σ(sodium)≤disease_limit

[0229] diversity_index ≥ 0.7.

[0230] The NSGA-II genetic algorithm is used to solve the Pareto frontier with a population size of 100, a crossover probability of 0.8, and a mutation probability of 0.2.

[0231] 5. System Integration and Deployment

[0232] Service interface design

[0233] Menu generation API: POST / api / v1 / menu / generate

[0234] {

[0235] "user_id":"P202311001",

[0236] "date_range":["2023-11-01","2023-11-07"],

[0237] "context":{"is_postoperative":true,"allergy_alert":["peanut"]}

[0238] }

[0239] Example response: {

[0240] "menu_id":"M20231101-001",

[0241] "dishes":[

[0242] {

[0243] "dish_id":"D02345",

[0244] "name":"Steamed Sea Bass",

[0245] "nutrition":{"protein":28.3,"fat":5.7,"carbs":2.1},

[0246] "allergens":[]

[0247] }

[0248] ],

[0249] "nutrition_summary":{

[0250] "total_calories":1850,

[0251] "protein_ratio":22%

[0252] }

[0253] }.

[0254] Real-time feedback mechanism

[0255] Deploy Kafka consumer groups to process user ratings (1-5 stars) and food waste rate data (automatically collected through cafeteria RFID plates).

[0256] Online learning module: Update user embedding vectors every hour and use the FTRL optimizer to achieve real-time update of model parameters.

[0257] VI. Quality Control and Verification

[0258] AB Testing Framework

[0259] Traffic grouping: Bucketing is done by user ID hash value. The control group uses rule-based recommendations (such as a standard diabetic menu), while the experimental group uses a machine learning model.

[0260] Core indicators:

[0261] Compliance improvement rate = (number of times the experimental group complies with the doctor's orders - control group) / control group × 100%

[0262] Customer satisfaction = NPS score difference

[0263] Significance test: Two-sample T test was used, and the strategy was considered effective when p-value < 0.05.

[0264] Nutritionist review process

[0265] Development Review Workbench:

[0266] Visualize the radar chart of the NRV (nutrient reference value) ratio of the menu

[0267] Provide mandatory coverage function (such as specifying that a meal must contain dishes with a calcium content of ≥200mg)

[0268] Establish an audit knowledge base: Encode 200 correction suggestions from the nutrition department of a tertiary hospital into business rules (for example, prohibit the use of high-fiber vegetables in the first week after surgery).

[0269] VII. Security and Compliance

[0270] Data privacy protection

[0271] Health data storage complies with HIPAA standards, uses AES-256 encryption for storage, and data transmission uses TLS1.3.

[0272] Implement GDPR compliance plan: users can export all dietary records with one click through the management background, and support complete data erasure.

[0273] Medical compliance

[0274] The logic of dish recommendations has been certified by the "Clinical Nutrition Risk Screening Guidelines" and has 47 built-in clinical rules (such as prohibiting raw food during cancer chemotherapy).

[0275] Automatically synchronize the latest dietary guidelines released by the National Health Commission every week, and trigger model retraining through the Jenkins pipeline.

[0276] like Figure 3 As shown, preferably, the employee meal ordering and payment system includes:

[0277] (1) Healthy diet suggestion subsystem, used for:

[0278] Based on employees' health data and dietary records, a deep learning model is used to predict their health risks, and daily nutritional intake reports and healthy diet suggestions that match their health risks are recommended and sent to the personalized recommendation subsystem;

[0279] (2) Personalized recommendation subsystem, used for:

[0280] Analyzing the daily nutritional intake report and the healthy diet recommendations to generate health goals and nutritional needs of employees, respectively;

[0281] using a machine learning algorithm to identify the employee's dining characteristics from the employee's health data, historical dietary data, the health goals, and the nutritional needs, and recommending a menu that matches the employee's dining characteristics;

[0282] Based on the preset nutritional data of the dishes, recommend dishes whose nutritional components meet the requirements of the menu, write the dishes into the preset employee meal list, obtain the second personalized menu, and send it to the recommendation system. The recommendation system recommends the second personalized menu to the corresponding employee terminal;

[0283] (3) Fast payment subsystem, used for:

[0284] According to the preset dish pricing rules, each dish in the second personalized menu is charged and counted, and a meal bill for the employee's meal is generated and sent to the recommendation system. The recommendation system recommends the nutritional components of the employee's meal to the corresponding employee end.

[0285] Preferably, the employee meal ordering and payment system further includes:

[0286] (4) A nutritional component analysis subsystem, which is used to: obtain the second personalized menu of the employee; traverse and parse the name and description information of each dish in the second personalized menu based on NLP technology; input the name and description information of each dish into a preset nutrition database, and the nutrition database calculates the nutritional components contained in each dish in real time; count the nutritional components contained in each dish in the second personalized menu, generate the nutritional components of the employee's meal this time and send it to the recommendation system, and the recommendation system recommends the nutritional components of the employee's meal this time to the corresponding employee end.

[0287] The following example describes the application process of the healthy diet suggestion subsystem:

[0288] Employee health data is collected through the company's HR system, wearable devices (such as smart bracelets and body fat scales), and physical examination reports submitted by employees. It includes information such as height, weight, BMI, blood pressure, blood sugar, allergens, and chronic disease history. Daily dietary records are manually entered by employees in the ordering system or automatically generated by using OCR technology to recognize post-meal photos. The data cleaning module standardizes the raw data, for example, converting "hypertension stage 2" into a numerical label and eliminating invalid data (such as blood sugar values exceeding the medically reasonable range).

[0289] The health risk prediction model uses an LSTM neural network. The input layer contains time-series health data (such as 30-day blood sugar fluctuations) and static data (such as genetic disease history). The hidden layer is designed as a three-layer neuron structure with a Reluctant Unit (ReLU) activation function. The output layer uses Softmax to generate a health risk probability distribution (such as an 85% risk of diabetes and a 12% risk of cardiovascular disease). The model is trained using the TensorFlow framework with a dataset containing 100,000 anonymized employee health records, and parameters are updated quarterly through incremental learning.

[0290] The daily nutrient intake report generation module calls upon the WHO nutrition standard library based on risk prediction results. For example, for employees at high risk of diabetes, it automatically matches the "daily carbohydrate intake ≤ 130g" rule and calculates protein requirements (1.2g / kg → 84g / d) based on the employee's weight (70kg). The healthy diet recommendation generator utilizes a rules engine. If it detects that an employee has insufficient dietary fiber intake for three consecutive days, it triggers a prompt to "increase oatmeal and broccoli intake" and outputs multilingual recommendation text using natural language generation (NLG) technology. The data interface uses a RESTful API, which can push 500 reports per second to the personalized recommendation subsystem.

[0291] Personalized recommendation subsystem application process

[0292] The health goal analysis module converts the "84g protein" requirement in the nutrition report into specific constraints, combining it with time to generate dynamic goals. For example, breakfast should provide 30% of the daily protein intake (25.2g). The nutritional needs calculator incorporates a metabolic equivalent (MET) compensation mechanism, automatically increasing calorie intake by 10% for employees who clock in at the gym.

[0293] User feature extraction uses the XGBoost algorithm, with input features including historical ordering frequency (e.g., beef three times a week), dietary preferences (halal / vegetarian), and meal affordability (15-25 yuan / meal). Feature importance analysis revealed a correlation of 0.78 between "dinner carbohydrate intake" and "post-meal blood sugar changes," and the feature vector space was constructed based on this correlation.

[0294] The menu recommendation engine utilizes a hybrid model: a content-based filtering module calculates dish similarity (for example, the cosine similarity between Kung Pao Chicken and Stir-fried Chicken with Soy Sauce is 0.92), while a collaborative filtering module taps into group preferences (80% of male employees of the same age group choose high-protein meal options). The multi-objective optimization algorithm, using NSGA-II, finds a Pareto optimal solution within a database of 3,200 dishes, balancing maximizing protein (objective function f1) and minimizing price (objective function f2).

[0295] The dish nutrition matcher connects to a MySQL database to verify recommendations in real time. When the system recommends braised pork, an alert is triggered: The fat content (22g) in a single serving exceeds 50% of the recommended daily value, and the dish is immediately replaced with steamed fish of equal protein. The resulting second personalized menu includes alternative options, such as offering both chicken and fish as a main course. A / B testing is used to determine the optimal presentation order.

[0296] Fast payment subsystem application process:

[0297] The pricing rules engine loads corporate subsidy policies (e.g., an 8 yuan dinner allowance) and dynamically calculates the actual payment amount. When an employee chooses a 28 yuan steak set meal, the system automatically splits the payment into 8 yuan for the corporate account and 20 yuan for the personal account. The discount calculation module supports multi-dimensional stacking, for example, allowing membership levels (82% off for VIP) to be applied simultaneously with holiday promotions (3 yuan off for purchases over 25 yuan).

[0298] The real-time billing module uses distributed transaction processing and a TCC (Try-Confirm-Cancel) model to ensure data consistency. When an employee uses their meal allowance balance and WeChat Pay simultaneously, their meal allowance account is frozen for 8 yuan and a WeChat payment order is created. If no WeChat callback is received within 20 seconds, a compensation transaction is automatically triggered, releasing the frozen amount. The bill generation service processes 1,200 transactions per second, with peak latency kept within 200ms.

[0299] The payment risk control system monitors for unusual behavior patterns and automatically initiates facial recognition verification if it detects three or more payment requests from the same employee account within five minutes. Transaction data is synchronized to a blockchain-based evidence storage platform, where each block includes a timestamp, transaction hash, and digital signature to ensure it cannot be tampered with.

[0300] Application of the Nutritional Analysis Subsystem: The dish parsing module uses the BERT model for multimodal processing: extracting text from menu images (with an OCR accuracy of 98.7%) and performing semantic disambiguation on ambiguous terms such as "Fish-flavored Shredded Pork" (distinguishing between Sichuan and Beijing styles). The ingredient identifier uses a named entity recognition (NER) model to extract key ingredients from the description text, for example, parsing "tomato beef soup" into beef (150g), tomato (200g), and onion (50g).

[0301] The nutrition database was built using knowledge graph technology. The entities include 12,000 food ingredients and 300 nutrients, with edge definitions such as "100g of beef contains 26g of protein and 2.6mg of iron." The real-time computing engine, using the Apache Flink stream processing framework, automatically associates 150g of tofu, 30g of ground beef, and 10ml of chili oil when parsing "Mapo Tofu," and accumulates the nutrients.

[0302] The report generation module uses dynamic visualization technology to break down the 1200kcal intake into a pie chart (45% carbohydrates, 30% protein, 25% fat), and plots a bar chart comparing it to the recommended values. Warning messages are highlighted through color coding. For example, if a gluten-allergic employee mistakenly selects a dish containing wheat, a pop-up window will immediately prompt and recommend an alternative. The data push service supports the WebSocket protocol, ensuring that the employee's end-user displays the latest nutritional analysis results in real time.

[0303] The description of the employee meal ordering and payment system can be understood in conjunction with the functions of the patient meal ordering system described above.

[0304] On the other hand, an application method of an intelligent hospital catering platform integrating dining characteristics and health management is provided, the application method comprising:

[0305] The user logs into the platform through the terminal and enters the corresponding ordering / dining information;

[0306] The platform's patient meal ordering system identifies the patient's dining characteristics based on the patient's condition, eating habits, and nutritional needs, recommends a first personalized menu to the patient based on the patient's dining characteristics, and recommends the first personalized menu to the patient via the recommendation system;

[0307] The platform's employee meal ordering and payment system identifies employee dining characteristics based on their health data, eating habits, and nutritional needs, and recommends a second personalized menu to the employees based on the employee dining characteristics, and then recommends the second personalized menu to the employee through the recommendation system.

[0308] For details on the method and steps, please refer to the previous platform description.

[0309] Figure 4 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention, such as Figure 4 As shown, the electronic device 410 may include a first processor 2001 .

[0310] Optionally, the electronic device 410 may further include a memory 2002 and a transceiver 2003 .

[0311] The first processor 2001, the memory 2002 and the transceiver 2003 may be connected via a communication bus.

[0312] The following combination Figure 4 The components of the electronic device 410 are described in detail.

[0313] The first processor 2001 is the control center of the electronic device 410 and can be a single processor or a collective term for multiple processing elements. For example, the first processor 2001 can be one or more central processing units (CPUs), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement an embodiment of the present invention, such as one or more digital signal processors (DSPs) or one or more field programmable gate arrays (FPGAs).

[0314] Optionally, the first processor 2001 can execute various functions of the electronic device 410 by running or executing software programs stored in the memory 2002 and calling data stored in the memory 2002.

[0315] In a specific implementation, as an embodiment, the first processor 2001 may include one or more CPUs, such as Figure 4 CPU0 and CPU1 are shown in FIG.

[0316] In a specific implementation, as an embodiment, the electronic device 410 may also include multiple processors, such as Figure 41 and 2. The first processor 2001 and the second processor 2004 are shown in FIG. Each of these processors can be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). A processor herein can refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0317] The memory 2002 is used to store the software program for executing the solution of the present invention, and is controlled by the first processor 2001 for execution. The specific implementation method can refer to the above method embodiment and will not be repeated here.

[0318] Alternatively, the memory 2002 may be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM) or other type of dynamic storage device that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compact disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 2002 may be integrated with the first processor 2001, or may exist independently and be accessed through the interface circuit ( Figure 4 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0319] The transceiver 2003 is used to communicate with a network device or a terminal device.

[0320] Optionally, the transceiver 2003 may include a receiver and a transmitter ( Figure 4 (not shown separately in the figure). The receiver is used to implement a receiving function, and the transmitter is used to implement a sending function.

[0321] Optionally, the transceiver 2003 may be integrated with the first processor 2001, or may exist independently and communicate with the first processor 2001 through the interface circuit ( Figure 4 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0322] It should be noted that Figure 4 The structure of the electronic device 410 shown in the figure does not constitute a limitation on the router. The actual knowledge structure recognition device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0323] In addition, the technical effects of the electronic device 410 can refer to the technical effects of the intelligent hospital catering platform and method that integrates dining features and health management as described in the above method embodiment, and will not be repeated here.

[0324] It should be understood that the first processor 2001 in the embodiment of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0325] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM).

[0326] The above embodiments can be implemented in whole or in part through software, hardware (such as circuits), firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the processes or functions described in accordance with the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired method (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains a collection of one or more available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, or magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0327] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0328] In this disclosure, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, "at least one of a, b, or c" can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or plural.

[0329] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0330] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0331] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0332] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interface, indirect coupling or communication connection of the device or unit, which can be electrical, mechanical or other forms.

[0333] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0334] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0335] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0336] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. An intelligent hospital catering platform that integrates dining features and health management, characterized by: The platform includes: a patient meal ordering system for identifying patient meal characteristics according to the patient's condition, eating habits, and nutritional needs, and recommending a first personalized meal menu to the patient based on the patient meal characteristics; An employee meal ordering and payment system, which is used to identify employee dining characteristics based on their health data, eating habits, and nutritional needs, and recommend a second personalized menu to the employee based on the employee dining characteristics; Recommendation system, used to recommend menus to corresponding user terminals; The patient meal ordering system and the employee meal ordering and payment system are respectively connected to the recommendation system for communication.

2. The intelligent hospital catering platform integrating dining features and health management according to claim 1 is characterized in that: The patient meal ordering system includes: (1) A quick ordering subsystem, which is used to provide ordering services based on voice input or image recognition technology, and to input a list of dishes to be ordered; (2) A nutritional component analysis subsystem, which is used to: obtain the list of dishes ordered by the patient; traverse and parse the name and description information of each dish in the dish list based on NLP technology; input the name and description information of each dish into a preset nutritional database, and the nutritional database calculates the nutritional components contained in each dish in real time; count the nutritional components contained in each dish in the dish list, generate the nutritional components of the patient's order and send it to the recommendation system, and the recommendation system recommends the nutritional components of the order to the corresponding patient end.

3. The intelligent hospital catering platform integrating dining features and health management according to claim 2 is characterized in that: The patient meal ordering system further includes: (3) Healthy diet suggestion subsystem, used for: Based on the patient's health data and diet records, a deep learning model is used to predict the patient's health risks, and healthy diet suggestions that match the patient's health risks are recommended and sent to the personalized recommendation subsystem.

4. The intelligent hospital catering platform integrating dining features and health management according to claim 3 is characterized in that: The patient meal ordering system further includes: (4) Personalized recommendation subsystem, used for: Analyzing the healthy eating recommendations to generate the patient's nutritional needs; Using a machine learning algorithm, identifying the patient's dining characteristics from the patient's health data, historical dietary data, and nutritional needs, and recommending a menu that matches the patient's dining characteristics; Combined with the preset nutritional data of dishes, each dish whose nutritional components meet the requirements of the menu is recommended, and each dish is written into the preset patient meal list to obtain the first personalized menu and send it to the recommendation system, which then recommends the first personalized menu to the corresponding patient end.

5. The intelligent hospital catering platform integrating dining features and health management according to claim 1 is characterized in that: The employee meal ordering and payment system includes: (1) Healthy diet suggestion subsystem, used for: Based on employees' health data and dietary records, a deep learning model is used to predict their health risks, and daily nutritional intake reports and healthy diet suggestions that match their health risks are recommended and sent to the personalized recommendation subsystem. (2) Personalized recommendation subsystem, used for: Analyzing the daily nutritional intake report and the healthy diet recommendations to generate health goals and nutritional needs of employees, respectively; using a machine learning algorithm to identify the employee's dining characteristics from the employee's health data, historical dietary data, the health goals, and the nutritional needs, and recommending a menu that matches the employee's dining characteristics; Based on the preset nutritional data of the dishes, recommend dishes whose nutritional components meet the requirements of the menu, write the dishes into the preset employee meal list, obtain the second personalized menu, and send it to the recommendation system. The recommendation system recommends the second personalized menu to the corresponding employee terminal; (3) Fast payment subsystem, used for: According to the preset dish pricing rules, each dish in the second personalized menu is charged and counted, and a meal bill for the employee's meal is generated and sent to the recommendation system. The recommendation system recommends the nutritional components of the employee's meal to the corresponding employee end.

6. The intelligent hospital catering platform integrating dining features and health management according to claim 5 is characterized in that: The employee meal ordering and payment system further includes: (4) A nutritional component analysis subsystem, which is used to: obtain the second personalized menu of the employee; traverse and parse the name and description information of each dish in the second personalized menu based on NLP technology; input the name and description information of each dish into a preset nutrition database, and the nutrition database calculates the nutritional components contained in each dish in real time; count the nutritional components contained in each dish in the second personalized menu, generate the nutritional components of the employee's meal this time and send it to the recommendation system, and the recommendation system recommends the nutritional components of the employee's meal this time to the corresponding employee end.

7. An application method of the intelligent hospital catering platform integrating dining features and health management according to any one of claims 1 to 6, characterized in that: The application method comprises: The user logs into the platform through the terminal and enters the corresponding ordering / dining information; The platform's patient meal ordering system identifies the patient's dining characteristics based on the patient's condition, eating habits, and nutritional needs, recommends a first personalized menu to the patient based on the patient's dining characteristics, and recommends the first personalized menu to the patient via the recommendation system; The platform's employee meal ordering and payment system identifies employee dining characteristics based on their health data, eating habits, and nutritional needs, and recommends a second personalized menu to the employees based on the employee dining characteristics, and then recommends the second personalized menu to the employee through the recommendation system.

8. An electronic device, characterized in that: The electronic device comprises: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the method according to claim 7 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program code, which can be called by a processor to execute the method according to claim 7.

Images