System and method for providing personalized nutrition recommendation for old surgical patients

Abstract

The invention provides a system and method for providing individualized nutrition recommendation for old surgical patients, and relates to the technical field of nutrition recommendation, and the method comprises the steps: obtaining a basic information set of a target patient; constructing a nutrition recommendation space; inputting information into the nutrition recommendation space for retrieval, and determining initial nutrition recommendation information; acquiring a food intake type sequence and a food intake quantity sequence of the target patient; according to the food intake type sequence, the food intake quantity sequence and the K pieces of nutrition intake information, recipe configuration is carried out, and K-stage nutrition recipe schemes are determined. The technical problems that in the prior art, personalized nutrition recommendation cannot be provided for old surgical patients, and a mechanism for analyzing historical food intake of the patients and continuously monitoring and correcting in the nutrition scheme implementation process is lacked are solved. Through multi-module cooperation, the technical effect of accurate, scientific and dynamically adjustable personalized nutrition recommendation for the elderly surgery patients is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of nutrition recommendation, and in particular to a system and method for providing personalized nutrition recommendation to elderly surgical patients. Background Art

[0002] With the aging of the population, the number of elderly surgical patients is increasing. Reasonable nutritional support is crucial in the surgical rehabilitation process of elderly patients. However, the existing technology has many deficiencies in providing nutritional recommendations for elderly surgical patients. On the one hand, the existing nutrition recommendation methods often lack personalized considerations, and are usually based on general nutritional standards, without fully considering the unique nutritional needs, dietary preferences and personal physical information of each elderly surgical patient. Different types of surgery, physical conditions and rehabilitation stages have huge differences in nutritional needs, and a unified nutrition recommendation plan is difficult to meet the personalized needs of patients. On the other hand, the existing technology pays insufficient attention to the patient's historical food intake. Without understanding the patient's past food intake type, intake and frequency, it is difficult to develop a nutritional recipe plan that meets nutritional needs and is acceptable to patients. At the same time, in the implementation of the nutrition plan, there is a lack of effective monitoring and correction mechanism, and the nutrition plan cannot be adjusted in time according to the patient's actual implementation and physical changes, resulting in unsatisfactory results of nutritional support.

[0003] Therefore, existing technologies have technical problems such as being unable to provide personalized nutrition recommendations for elderly surgical patients, and lacking analysis of patients' historical food intake and a continuous monitoring and correction mechanism during the implementation of nutrition plans. Summary of the invention

[0004] The present application provides a system and method for providing personalized nutritional recommendations for elderly surgical patients, which is used to solve the technical problems that the existing technology cannot provide personalized nutritional recommendations for elderly surgical patients, and lacks a mechanism for analyzing the patient's historical food intake and continuously monitoring and correcting the implementation of the nutritional plan.

[0005] In view of the above problems, the present application provides a system and method for providing personalized nutritional recommendations for elderly surgical patients.

[0006] In a first aspect, the present application provides a system for providing personalized nutrition recommendations for elderly surgical patients, the system comprising:

[0007] A basic information set acquisition module, the basic information set acquisition module is used to acquire the basic information set of the target patient, wherein the basic information set includes nutritional requirement information, dietary preference information, and personal physical information; a nutritional recommendation space acquisition module, the nutritional recommendation space acquisition module is used to construct a nutritional recommendation space; an initial nutritional recommendation information determination module, the initial nutritional recommendation information determination module is used to input the nutritional requirement information, dietary preference information and personal physical information into the nutritional recommendation space for retrieval, and determine the initial nutritional recommendation information, wherein the initial nutritional recommendation information includes K nutritional intake information of K stages, K being an integer greater than or equal to 1; a patient intake information acquisition module, the patient intake information acquisition module is used to acquire the food intake type sequence and food intake amount sequence of the target patient in a preset historical window, wherein the food intake type and the food intake amount correspond one to one; a nutritional recipe plan determination module, the nutritional recipe plan determination module is used to use the nutritional intake analysis network layer to configure recipes according to the food intake type sequence, the food intake amount sequence and the K nutritional intake information, and determine the nutritional recipe plans for K stages.

[0008] In a second aspect, the present application provides a method for providing personalized nutrition recommendations for elderly surgical patients, the method comprising:

[0009] Obtain a basic information set of the target patient, wherein the basic information set includes nutritional requirement information, dietary preference information, and personal physical information; construct a nutritional recommendation space; input the nutritional requirement information, dietary preference information, and personal physical information into the nutritional recommendation space for retrieval to determine initial nutritional recommendation information, wherein the initial nutritional recommendation information includes K nutritional intake information of K stages, K being an integer greater than or equal to 1; obtain a food intake type sequence and a food intake amount sequence of the target patient within a preset historical window, wherein the food intake type and the food intake amount correspond one to one; use a nutritional intake analysis network layer to configure a recipe according to the food intake type sequence, the food intake amount sequence, and the K nutritional intake information, and determine a nutritional recipe plan for K stages.

[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0011] The embodiment of the present application provides a system and method for providing personalized nutrition recommendations for elderly surgical patients, which includes obtaining a basic information set of the target patient; constructing a nutrition recommendation space; inputting the nutrition requirement information, dietary preference information, and personal physical information into the nutrition recommendation space for retrieval to determine the initial nutrition recommendation information; obtaining the food intake type sequence and food intake sequence of the target patient in a preset historical window; using the nutrition intake analysis network layer to configure recipes according to the food intake type sequence, the food intake sequence, and the K nutrition intake information, and determining K stage nutrition recipe plans. The technical effect of achieving accurate, scientific, and dynamically adjustable personalized nutrition recommendations for elderly surgical patients through multi-module collaboration is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A schematic diagram of a system structure for providing personalized nutrition recommendations to elderly surgical patients is provided for this application;

[0013] Figure 2 A schematic flow chart of a method for providing personalized nutritional recommendations to elderly surgical patients is provided for this application.

[0014] Explanation of the reference numerals: basic information set acquisition module 10 , nutrition recommendation space acquisition module 20 , initial nutrition recommendation information determination module 30 , patient intake information acquisition module 40 , nutrition recipe plan determination module 50 . DETAILED DESCRIPTION

[0015] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0016] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0017] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0018] Embodiment 1: Figure 1 As shown, the present application provides a system for providing personalized nutrition recommendations for elderly surgical patients, the system comprising:

[0019] The basic information set acquisition module 10 is used to acquire the basic information set of the target patient, wherein the basic information set includes nutritional requirement information, dietary preference information, and personal physical information.

[0020] Specifically, the basic information set acquisition module 10 is responsible for acquiring the basic information set of the target patient, which covers important information in multiple aspects. The first is the nutritional demand information, which is determined based on the operation time and operation method. The length of the operation time has different consumption levels on the patient's body. Longer operations may cause a large consumption of the patient's body energy reserves, and increase the demand for nutrients such as protein and carbohydrates to promote wound healing and body recovery. The surgical method will also affect nutritional needs. For example, different gastrointestinal surgeries affect the patient's ability to digest and absorb food, and thus the intake of nutrients needs to be adjusted. For example, it may be necessary to increase the supplementation of certain vitamins and minerals to make up for the insufficient absorption caused by the surgical method.

[0021] Dietary preference information is also an important part of the basic information set. It is very important to understand the patient's likes and dislikes of food, which directly affects the patient's compliance with the nutrition intake plan. Different patients have different preferences for various foods such as meat, vegetables, fruits, grains, etc. At the same time, it is also necessary to consider whether the patient has allergies or intolerance to certain specific foods. This information will provide an important basis for the subsequent formulation of personalized nutrition recipes.

[0022] Personal physical information is also indispensable, including the patient's basic physical characteristics data, such as age, gender, height, weight, body mass index (BMI), etc. These data reflect the patient's physical condition and metabolic level, and are of great significance for accurately assessing the patient's nutritional needs and formulating a reasonable nutrition plan. For example, elderly patients have decreased physical function and slowed metabolism, and their ability to absorb and utilize nutrients is not as good as that of young patients, so these factors need to be considered comprehensively when determining nutritional needs. At the same time, gender differences may also affect nutritional needs, and there are differences in the needs of certain nutrients between men and women. Height and weight are used to calculate the body mass index, which further assesses the patient's body obesity and health status, so as to more accurately determine nutritional needs and formulate appropriate nutrition plans.

[0023] The nutrition recommendation space acquisition module 20 is used to construct the nutrition recommendation space.

[0024] Specifically, the nutrition recommendation space acquisition module 20 needs to determine the dimensions and coordinate axes of the construction space, and use several key factors closely related to the patient's basic information as the coordinate axes, and use nutrition demand information, dietary preference information, and personal physical information as the coordinate axes to construct a three-dimensional coordinate system. This choice is based on the important influence of these factors on nutrition recommendations. Next, the module needs to collect a large amount of data to fill this space, and obtain relevant information of multiple sample patients, including sample basic information sets and corresponding sample nutrition recommendation information. The data of these sample patients covers various situations such as different ages, genders, types of surgery, dietary preferences, etc., and is widely representative. Then, the sample basic information sets and sample nutrition recommendation information of these sample patients are input into the constructed three-dimensional coordinate system, and the data of each sample patient will form a sample point in this coordinate system. In this way, multiple sample points will appear in the three-dimensional space. Finally, based on these large number of sample points and the structure of the three-dimensional coordinate system, the nutrition recommendation space acquisition module 20 constructs a complete nutrition recommendation space. This space can reflect the complex relationship between different combinations of nutritional needs information, dietary preference information, and personal body information and the corresponding nutritional recommendation information, providing an effective data environment and theoretical basis for the subsequent retrieval of nutritional recommendation information suitable for target patients in this space.

[0025] The initial nutrition recommendation information determination module 30 is used to input the nutrition requirement information, dietary preference information and personal body information into the nutrition recommendation space for retrieval to determine the initial nutrition recommendation information, wherein the initial nutrition recommendation information includes K nutrition intake information of K stages, K being an integer greater than or equal to 1.

[0026] Specifically, the main task of the initial nutrition recommendation information determination module is to input the target patient's nutrition demand information, dietary preference information and personal physical information into the nutrition recommendation space for retrieval to determine the initial nutrition recommendation information. First of all, the nutrition demand information is determined according to the patient's surgical situation and physical condition. At different stages, such as preoperative, early postoperative and postoperative rehabilitation, the patient's nutrition needs will be different. In the preoperative stage, it is necessary to prepare the body for surgery and adjust the nutritional status to improve surgical tolerance; in the early postoperative period, the body is in a critical period of recovery and requires specific nutrition to promote wound healing and restore body function; in the postoperative rehabilitation period, more attention is paid to comprehensive nutritional support to help patients return to their preoperative health as soon as possible. Dietary preference information reflects the patient's preference for different foods, taking into account the patient's eating habits and preferences to improve compliance with the nutrition recommendation plan. If the patient does not like a certain food, but the nutrition recommendation plan contains a large amount of this food, it will be difficult for the patient to adhere to the plan. Personal physical information includes the patient's age, gender, body mass index, etc., which will affect the patient's nutritional needs and metabolic capacity. For example, older patients need more protein to maintain muscle mass, while patients with a higher body mass index need to control calorie intake to avoid weight gain. After this information is entered into the nutrition recommendation space, the module will search to determine the initial nutrition recommendation information. This information includes K stages (preoperative, early postoperative, and postoperative recovery stages), that is, determining the food and intake that patients should consume at each stage. By analyzing and matching the data in the nutrition recommendation space, find the sample points closest to the target patient's situation, and determine the initial recommendation plan based on the nutrition recommendation information of these sample points. Such a plan can better meet the actual needs of patients and provide more personalized and effective nutritional support.

[0027] The patient intake information acquisition module 40 is used to obtain the food intake type sequence and food intake amount sequence of the target patient within a preset historical window, wherein the food intake type and the food intake amount correspond one to one.

[0028] Specifically, the patient intake information acquisition module 40 is mainly responsible for acquiring the food intake related information of the target patient in the preset history window. The setting of the preset history window can be adjusted according to the actual situation, such as the past week, month and other time periods. By analyzing the food intake of the patient in this specific time period, the patient's eating habits and nutritional intake status can be better understood. The information acquired by the module includes a food intake type sequence and a food intake sequence. The food intake type covers various types of food consumed by the patient, such as vegetables, fruits, meat, grains, etc. Each food intake type corresponds to a specific food intake. For example, if the food intake type is apples, the corresponding food intake may be a specific value such as one apple per day or three apples per week. This one-to-one correspondence ensures an accurate description of the patient's food intake. By collecting this information, the patient's past dietary preferences and nutritional intake levels can be understood, providing an important reference for the subsequent formulation of personalized nutritional recipes. At the same time, the patient's nutritional imbalance can also be analyzed based on the patient's historical food intake, so as to make targeted adjustments and optimizations in the new nutritional recommendation plan.

[0029] The nutritional recipe plan determination module 50 is used to utilize the nutritional intake analysis network layer to configure recipes according to the food intake type sequence, the food intake amount sequence and the K nutritional intake information, and determine the nutritional recipe plans for K stages.

[0030] Specifically, the nutritional recipe determination module 50 is responsible for formulating personalized nutritional recipes for elderly surgical patients. This module uses the nutritional intake analysis network layer to configure recipes. First, it receives the food intake type sequence and food intake sequence from the patient intake information acquisition module 40. These two sequences reflect the actual diet of the target patient within the preset historical window. The food intake type sequence shows the various types of food consumed by the patient, while the food intake sequence corresponds to the specific intake of each food. By analyzing this information, the module understands the patient's dietary preferences and current nutritional intake level. At the same time, the module also receives K nutritional intake information from the initial nutritional recommendation information determination module 30. This information includes the specific nutritional needs of K stages (such as preoperative, early postoperative, postoperative rehabilitation, etc.), and determines the amount of various nutrients that the patient should consume at each stage. Then, the nutritional intake analysis network layer combines these input information to configure recipes. It will comprehensively consider the patient's historical dietary situation, current nutritional needs, and recovery goals at different stages, and tailor nutritional recipes for patients in K stages. In the process of formulating the plan, the network layer will optimize according to factors such as the nutritional composition of the food, the patient's dietary preferences, and physical condition. For example, for patients in the early postoperative period, we recommend some foods that are easy to digest, rich in protein and vitamins, while avoiding foods that may cause gastrointestinal discomfort. For patients with specific dietary preferences, we try to choose foods that suit their tastes while meeting their nutritional needs. The final K-stage nutritional recipe plan can provide patients with detailed dietary guidance, helping them to consume appropriate foods at different recovery stages, meet the body's nutritional needs, and promote surgical recovery.

[0031] In a possible implementation, the nutrition recommendation space acquisition module 20 further includes:

[0032] A three-dimensional coordinate system is constructed with nutritional requirement information, dietary preference information, and personal physical information as coordinate axes respectively; multiple sample basic information sets and multiple sample nutritional recommendation information of multiple sample patients are obtained; the multiple sample basic information sets and the multiple sample nutritional recommendation information are input into the three-dimensional coordinate system to obtain multiple sample points; and the nutritional recommendation space is constructed based on the multiple sample points and the three-dimensional coordinate system.

[0033] Specifically, in the process of constructing the nutrition recommendation space, a three-dimensional coordinate system is first constructed with nutrition requirement information, dietary preference information, and personal physical information as coordinate axes. These three coordinate axes are selected because they cover the key factors that affect the patient's nutrition recommendation. Nutritional requirement information reflects the specific nutrients and intake required by the patient based on his or her own physical condition and surgical situation. Different types of surgery and physical conditions will lead to huge differences in nutritional needs. Dietary preference information reflects the patient's preference for different foods, which is crucial for formulating practical nutrition plans. If the recommended food does not meet the patient's taste preferences, it will be difficult for the patient to adhere to the nutrition plan. Personal physical information includes age, gender, body mass index, etc. These factors will affect the patient's metabolic level and demand for different nutrients.

[0034] Next, multiple sample basic information sets and multiple sample nutrition recommendation information of multiple sample patients are obtained. The data of these sample patients come from a large number of clinical practices and studies, covering patients in various situations and having broad representativeness. The sample basic information set includes the nutrition requirement information, dietary preference information and personal physical information of each sample patient, while the sample nutrition recommendation information is a nutrition plan suitable for each sample patient formulated based on this basic information.

[0035] Then, these multiple sample basic information sets and multiple sample nutritional recommendation information are input into the constructed three-dimensional coordinate system, in which the basic information of each sample patient corresponds to a point, thereby obtaining multiple sample points. The distribution of these sample points in three-dimensional space reflects the relationship between different nutritional needs, dietary preferences, and personal physical information combinations and corresponding nutritional recommendations.

[0036] Finally, a nutrition recommendation space is constructed based on these multiple sample points and a three-dimensional coordinate system. This space is a model that integrates a large amount of sample data and can provide a reference for the subsequent determination of nutrition recommendations for target patients. In this space, the most suitable nutrition recommendation plan for the target patient can be found by analyzing the relationship between the target patient's basic information and the sample points.

[0037] In a possible implementation, the initial nutrition recommendation information determination module 30 further includes:

[0038] The nutritional requirement information, dietary preference information and personal physical information are input into the nutritional recommendation space to obtain a target point; sample points whose distance to the target point is a preset distance are extracted from the nutritional recommendation space to obtain multiple matching sample points; and balanced analysis is performed on the nutritional recommendation information of multiple matching samples of the multiple matching sample points to determine the initial nutritional recommendation information.

[0039] Specifically, in the process of personalized nutrition recommendation, the target patient's nutrition requirement information, dietary preference information and personal physical information are input into the constructed nutrition recommendation space. Nutrition requirement information reflects the specific nutrients and intake required by the patient based on the surgical situation and physical condition. For example, if the patient has just undergone a major gastrointestinal surgery, the demand for nutrients such as protein and vitamin C will increase greatly during the postoperative recovery phase to promote wound healing and repair of body tissues. Entering this information into the nutrition recommendation space helps to screen out those options that can meet the patient's specific nutritional needs from many possible nutrition plans. Dietary preference information reflects the patient's preference for different foods. If the patient prefers vegetarianism, or has a special preference or aversion to certain foods, then these factors need to be fully considered when determining the nutrition recommendation plan. After entering the dietary preference information into the nutrition recommendation space, the system can try to choose food that suits the patient's taste while meeting the nutritional needs, thereby improving the patient's compliance with the nutrition plan. Personal physical information includes age, gender, body mass index, etc., which affect the patient's metabolic level and demand for different nutrients. For example, older patients may have a relatively slow metabolism and need to control their calorie intake. They may also need more nutrients such as calcium to prevent osteoporosis. After entering personal physical information into the nutrition recommendation space, the system can make more accurate nutrition recommendations based on the patient's specific physical condition. When these three aspects of information are entered into the nutrition recommendation space, a specific point, namely the target point, will be determined in the space. This target point represents the comprehensive characteristics of the target patient in terms of nutritional needs, dietary preferences, and personal physical condition, and provides a clear starting point and reference for the subsequent determination of personalized nutrition recommendation plans.

[0040] Extracting sample points in the nutrition recommendation space is one of the key steps to determine the initial nutrition recommendation information. When the target patient's nutrition requirement information, dietary preference information, and personal physical information determine the target point in the nutrition recommendation space, the system begins to look for sample points with a preset distance from the target point in this space. The preset distance is carefully considered. It must ensure that the extracted sample points have a certain similarity with the target patient so as to provide a valuable reference for the initial nutrition recommendation, but it should not be too broad, so as not to introduce too many irrelevant sample points and affect the accuracy of the recommendation. The multiple sample points obtained in this way are called matching sample points, which represent other patients with similar characteristics to the target patient in some aspects. For example, in terms of nutritional needs, there are specific nutrient requirements caused by similar surgical types; in terms of dietary preferences, there may be common favorite food types; in terms of personal physical information, there are similar ages, genders, or body mass indexes. The existence of these matching sample points provides rich reference data for determining the initial nutrition recommendation information of the target patient. By analyzing the nutrition recommendation information corresponding to these sample points similar to the target patient, we can better understand the nutrition plans of other patients in similar situations, so as to formulate a more reasonable and personalized nutrition recommendation plan for the target patient. The process of extracting sample points with a preset distance from the target point in the nutrition recommendation space and obtaining multiple matching sample points is an important step in combining a large amount of sample data with the specific conditions of the target patients, laying the foundation for subsequent nutrition recommendation analysis.

[0041] Balanced analysis is performed on multiple matched sample nutrition recommendation information of multiple matched sample points. First, these matched sample points represent other patients with certain similarities to the target patient, and each matched sample point corresponds to specific nutrition recommendation information, which differs in food types, nutrient intake, dietary frequency, etc. When performing balanced analysis, multiple factors need to be considered comprehensively. On the one hand, the commonality between the nutrition recommendation information of different matched samples should be considered. For example, if multiple matched sample points recommend a certain food rich in specific nutrients, then this may mean that this food has important nutritional value for people with similar characteristics to the target patient, and a higher weight is given when determining the initial nutrition recommendation information. On the other hand, differences should also be considered. Different matched sample points may have some differences in nutrition recommendations due to individual differences or other factors. At this time, it is necessary to carefully analyze the reasons for these differences, which are caused by different surgical recovery stages, different physical reactions, and other factors. In the process of balanced analysis, the nutrition recommendation information of multiple matched samples is averaged to determine a rough average recommendation plan. Then, according to the distance between the nutrition recommendation information of each matched sample and the average recommendation plan, their weights are determined. Recommended information that is closer is considered more reliable and is given a higher weight; recommended information that is farther away may need further analysis of its rationality or be given a lower weight. Through such a balanced analysis, a relatively reasonable initial nutrition recommendation information is determined. This information not only takes into account the experience of other patients similar to the target patient, but also fully combines the unique situation of the target patient, providing the target patient with a nutrition recommendation starting point that is scientific and feasible. In the subsequent process, this initial nutrition recommendation information is further adjusted and optimized based on the actual situation and feedback of the target patient to ensure that the final nutrition recipe plan can meet the needs of the target patient to the greatest extent.

[0042] In a possible implementation, the initial nutrition recommendation information determination module 30 further includes:

[0043] Perform mean processing on the multiple matching sample nutrition recommendation information to determine the mean of the matching sample nutrition recommendation information; determine multiple matching weights according to the distance from the multiple matching sample nutrition recommendation information to the mean of the matching sample nutrition recommendation information; perform weighted calculation on the multiple matching sample nutrition recommendation information based on the multiple matching weights to obtain the initial nutrition recommendation information.

[0044] Specifically, first, the nutrition recommendation information of multiple matching samples is averaged. This step aims to find an intermediate value representing the nutrition recommendation of multiple similar patients. By calculating the average value of these recommendation information in various aspects (such as the intake of different nutrients, the distribution of food types, etc.), a comprehensive mean of the nutrition recommendation information of matching samples is obtained. This mean reflects the more common nutrition recommendation situation in the group with similar characteristics to the target patient.

[0045] Next, multiple matching weights are determined according to the distance from multiple matching sample nutrition recommendation information to the mean of the matching sample nutrition recommendation information. The distance calculation measures the degree of difference between each matching sample nutrition recommendation information and the average. Recommendation information with a closer distance is considered to be closer to the average, representing a more universal and reliable recommendation, and therefore will be assigned a higher weight. On the contrary, recommendation information with a farther distance has certain particularities and its rationality needs to be considered more carefully, so it will be assigned a lower weight. This way of weight distribution ensures that more reliable recommendation information can play a greater role in subsequent calculations.

[0046] Multiple matching weights are determined based on the distance between the nutrition recommendation information of the matching samples and the mean of the nutrition recommendation information of the matching samples. These weights reflect the relative reliability and importance of each nutrition recommendation information of the matching samples. Recommendation information with a higher weight means that it is closer to the average recommendation of most similar patients, or has special value in some aspects. When performing weighted calculations, each nutrition recommendation information of the matching samples is assigned a corresponding weight, and the recommendation information with a higher weight will be given a greater influence in the calculation. For example, if a nutrition recommendation information of a matching sample is more consistent with the target patient's situation in terms of the intake of a specific nutrient, and its weight is higher, then in the weighted calculation, the nutrient intake corresponding to this recommendation information will have a greater impact on the nutrient intake in the initial nutrition recommendation information. By performing weighted calculations on the nutrition recommendation information of multiple matching samples and comprehensively considering the characteristics and reliability of different matching samples, the initial nutrition recommendation information obtained in this way is no longer a simple average of multiple recommendation information, but fully considers the differences between each recommendation information and their relative importance. This initial nutrition recommendation information can better reflect the actual situation of the target patient, and also combines the experience of other patients with similar characteristics to the target patient. Based on multiple matching weights, the nutritional recommendation information of multiple matching samples is weightedly calculated, providing the target patients with a more scientific, reasonable and personalized initial nutritional recommendation information, which provides an important basis for the subsequent determination and optimization of nutritional recipe plans.

[0047] In a possible implementation, the patient intake information acquisition module 40 further includes:

[0048] Acquire multiple historical food intake types, multiple historical food intake amount sets and multiple historical food intake frequencies of the target patient within a preset historical window; serialize the multiple historical food intake types based on the multiple historical food intake frequencies to obtain the food intake type sequence; perform mean processing on the multiple historical food intake amount sets, and serialize the processing results according to the mapping relationship between food intake type and food intake amount in combination with the food intake type sequence to obtain the food intake amount sequence.

[0049] Specifically, in the process of determining personalized nutrition recommendations for target patients, it is a crucial step to obtain multiple historical food intake types, multiple historical food intake sets, and multiple historical food intake frequencies of target patients within a preset historical window. The preset historical window is determined to collect the patient's dietary information over a period of time. This time range can be adjusted according to specific circumstances, such as the past month, three months, or half a year. By analyzing the patient's diet during this period, their eating habits and nutritional intake can be better understood. Multiple historical food intake types include various types of food consumed by the target patient within the preset historical window, which covers many food categories such as vegetables, fruits, grains, meat, fish, beans, etc. Understanding the patient's food intake type determines the diversity and preference of his diet. For example, if it is found that the patient often eats a certain vegetable or fruit, then when formulating a nutrition recommendation plan, the recommended amount of the relevant food is increased or adjusted to meet the patient's taste needs while ensuring nutritional balance. Multiple historical food intake sets record the specific amount of each food consumed by the target patient each time, which is a specific unit of weight (such as grams, kilograms) or volume (such as milliliters, liters). By analyzing the food intake set, we can understand the patient's intake level of different foods and determine whether there is excessive or insufficient intake. For example, if it is found that the patient's intake of a certain nutrient is insufficient due to the low intake of certain foods, then the recommended amount of foods rich in this nutrient can be specifically increased in the nutrition recommendation plan. Multiple historical food intake frequencies reflect the frequency of the target patient's consumption of various foods within the preset historical window, helping to understand whether the patient's eating habits are regular and the degree of preference for different foods. For example, if it is found that the patient eats a certain specific food several times a week, then it can be considered that this food occupies a relatively important position in the patient's diet, and its role needs to be fully considered when formulating a nutrition recommendation plan.

[0050] Multiple historical food intake frequencies reflect the frequency of consumption of different foods by the target patient in the past period of time. These frequency data provide important clues to understand the patient's eating habits. By analyzing these frequencies, it is determined which foods the patient prefers and which foods are less consumed. Multiple historical food intake types are serialized, and this process is to sort the food types according to the frequency of food intake. For example, food types frequently consumed by patients will occupy a higher position in the sequence, while food types that are less consumed will be relatively late. Such serialization can intuitively show the patient's dietary preference order. The food intake type sequence obtained through this serialization process can not only help to better understand the patient's eating habits, but also provide a basis for subsequent nutritional analysis and recipe configuration. When formulating a nutritional recommendation plan, priority should be given to the food types that patients often consume to improve the patient's acceptance and compliance with the recommended plan. At the same time, for those food types that are at the end of the sequence, appropriate recommendations should be made based on their nutritional characteristics and the needs of the patient to enrich the patient's dietary diversity and ensure balanced nutrition. Based on multiple historical food intake frequencies, multiple historical food intake types are serialized to obtain a food intake type sequence, which helps to improve the pertinence and effectiveness of the recommendation plan.

[0051] The mean processing of multiple historical food intake sets means summarizing the patient's intake of the same food at different time points and calculating the average intake. Through mean processing, a relatively stable reference value of food intake is obtained to avoid the influence of individual outliers on the overall analysis. For example, if the patient's intake of a certain food is particularly high or low on certain days, the mean processing makes this data smoother and better reflects the patient's general diet. Then, according to the mapping relationship between food intake type and food intake, the previously obtained food intake type sequence is combined for serialization processing. The food intake type sequence has been sorted according to the patient's food intake frequency. Now the processed average food intake is matched with the corresponding food type. This is done to establish an ordered food intake sequence that matches the food intake type sequence. Through this serialization processing, the average intake of different food types of patients can be clearly understood. For example, the corresponding average food intake of the food type that ranks first in the food intake type sequence can find the corresponding position in the food intake sequence, which helps to analyze whether the patient's diet structure is reasonable and determine which food intake may need to be adjusted. The final food intake sequence provides an important basis for personalized nutrition recommendations. By comparing this sequence with nutritional requirements, we can find out possible deficiencies or excesses in the patient's diet, thereby developing a more scientific and reasonable nutritional recipe plan to meet the nutritional needs of patients at different stages.

[0052] In a possible implementation, the nutritional recipe determination module 50 further includes:

[0053] Acquire multiple sample food intake type sequences, multiple sample food intake amount sequences, multiple sample nutrient intake information sets, and corresponding multiple sample stage nutrient recipe plans as training data, wherein each sample nutrient intake information set includes K sample nutrient intake information, and each sample stage nutrient recipe plan includes K sample stage nutrient recipe plans; utilize the training data to perform supervised training on a framework constructed based on a feedforward neural network until the training converges, and obtain the nutrient intake analysis network layer after training.

[0054] Specifically, multiple sample data are collected to build a rich database to cover a variety of different dietary situations and nutritional needs. The sample food intake type sequence shows the order of food types consumed by different individuals over a period of time. For example, the order of appearance of different food types, including vegetables, fruits, meat, and grains, corresponds to the specific intake of each food type, reflecting the individual's consumption of different foods. The sample nutritional intake information set includes K sample nutritional intake information, where the K information corresponds to the nutritional needs of different stages (such as preoperative, early postoperative, postoperative recovery, etc.). The nutritional intake information of each stage includes specific requirements for nutrients such as protein, carbohydrates, fat, vitamins, and minerals. Each sample stage nutritional recipe plan includes K sample stage nutritional recipe plans, which are formulated based on the corresponding nutritional intake information and food intake situations, and provide specific food combinations and intake recommendations to meet the nutritional needs of different stages.

[0055] Then, the training data is used to supervise the training of the framework built on the feedforward neural network, which is a powerful machine learning model that can learn the complex relationship between input data (sample food intake type sequence, sample food intake sequence and sample nutrient intake information set) and output data (sample stage nutrient recipe plan). During the training process, the network continuously adjusts its weights and parameters to minimize the error between the predicted nutrient recipe plan and the actual sample nutrient recipe plan. The training continues until the network converges, which means that the network can stably produce output results close to the actual situation when processing new input data. At this time, the trained nutrient intake analysis network layer is obtained. This network layer accurately determines the nutrient recipe plan suitable for the patient at different stages based on the new patient's food intake type sequence, food intake sequence and nutrient intake information. By obtaining a large amount of sample data and conducting supervised training, the constructed nutrient intake analysis network layer can provide reliable support for the formulation of personalized nutrient recipe plans, help elderly surgical patients better meet their nutritional needs and promote rehabilitation.

[0056] In a possible implementation, the nutritional recipe determination module 50 further includes:

[0057] Obtain a preset monitoring cycle; monitor the execution records of the K-stage nutritional recipe plans of the target patient within the preset monitoring cycle to obtain monitoring feedback results; and revise the K-stage nutritional recipe plans based on the monitoring feedback results.

[0058] Specifically, first, obtain the preset monitoring cycle, which is determined based on the specific situation and needs of the target patient, such as one week, one month or other appropriate time period. The preset monitoring cycle is set to regularly evaluate and provide feedback on the implementation of the patient's nutritional diet plan.

[0059] During the preset monitoring period, the execution records of the target patient's K-stage nutritional diet plan are monitored, including recording the type of food, amount of food intake, and time of intake, etc. The execution records are monitored to understand whether the patient follows the nutritional diet plan, as well as the problems and difficulties encountered during the implementation process. For example, it is found that the patient has a resistance to certain foods, or is unable to eat at the time specified in the plan due to actual conditions.

[0060] The monitoring feedback results obtained are a comprehensive assessment of the patient's implementation of the nutritional diet plan. This result includes information on the patient's acceptance of food, whether the nutritional intake reaches the expected goal, changes in physical condition, etc. Based on the monitoring feedback results, the nutritional diet plans for K stages are revised. If it is found that the patient does not like or has difficulty obtaining certain foods, the types and combinations of food are adjusted while ensuring nutritional needs. If the patient's physical condition changes, such as accelerated recovery after surgery or complications, the nutritional diet plan needs to be adjusted accordingly to meet the new nutritional needs. By obtaining the preset monitoring cycle, the execution records of the target patient's nutritional diet plan are monitored and feedback results are obtained, and then the plan is revised, and the personalized nutritional diet plan is continuously optimized to make it more in line with the patient's actual situation and needs, and promote the patient's recovery process.

[0061] Embodiment 2: Based on the same inventive concept as the system for providing personalized nutrition recommendations to elderly surgical patients in the aforementioned embodiment, Figure 2 As shown, the present application provides a method for providing personalized nutrition recommendations for elderly surgical patients, the method comprising:

[0062] Step S100: Obtain a basic information set of a target patient, wherein the basic information set includes nutritional requirement information, dietary preference information, and personal physical information.

[0063] Step S200: constructing a nutrition recommendation space.

[0064] Step S300: input the nutritional requirement information, dietary preference information and personal physical information into the nutritional recommendation space for retrieval to determine initial nutritional recommendation information, wherein the initial nutritional recommendation information includes K nutritional intake information of K stages, K being an integer greater than or equal to 1.

[0065] Step S400: Acquire the food intake type sequence and food intake amount sequence of the target patient in a preset history window, wherein the food intake type and the food intake amount correspond one to one.

[0066] Step S500: Utilizing the nutrition intake analysis network layer to configure a recipe according to the food intake type sequence, the food intake amount sequence and the K nutrition intake information, and determining nutrition recipe plans for K stages.

[0067] Furthermore, step S200 further includes:

[0068] Step S210: constructing a three-dimensional coordinate system using the nutritional requirement information, the dietary preference information, and the personal body information as coordinate axes.

[0069] Step S220: Acquire multiple sample basic information sets and multiple sample nutritional recommendation information of multiple sample patients.

[0070] Step S230: input the multiple sample basic information sets and the multiple sample nutrition recommendation information into the three-dimensional coordinate system to obtain multiple sample points.

[0071] Step S240: constructing the nutrition recommendation space based on the multiple sample points and the three-dimensional coordinate system.

[0072] Furthermore, step S300 also includes:

[0073] Step S310: inputting the nutrition requirement information, dietary preference information and personal physical information into the nutrition recommendation space to obtain a target point.

[0074] Step S320: extracting sample points whose distance to the target point is a preset distance in the nutrition recommendation space, and obtaining a plurality of matching sample points.

[0075] Step S330: performing a balanced analysis on the nutrition recommendation information of the multiple matching samples of the multiple matching sample points to determine the initial nutrition recommendation information.

[0076] Furthermore, step S330 also includes:

[0077] Step S331: performing mean processing on the plurality of matching sample nutrition recommendation information to determine the mean of the matching sample nutrition recommendation information.

[0078] Step S332: Determine a plurality of matching weights according to the distances from the plurality of matching sample nutrition recommendation information to the mean of the matching sample nutrition recommendation information.

[0079] Step S333: performing weighted calculation on the plurality of matching sample nutrition recommendation information based on the plurality of matching weights to obtain the initial nutrition recommendation information.

[0080] Furthermore, step S400 further includes:

[0081] Step S410: Acquire multiple historical food intake types, multiple historical food intake amount sets, and multiple historical food intake frequencies of the target patient within a preset historical window.

[0082] Step S420: Serializing the multiple historical food intake types based on the multiple historical food intake frequencies to obtain the food intake type sequence.

[0083] Step S430: performing mean processing on the multiple historical food intake sets, and performing serialization processing on the processing results according to the mapping relationship between food intake type and food intake in combination with the food intake type sequence to obtain the food intake sequence.

[0084] Furthermore, step S500 further includes:

[0085] Step S510: Obtain multiple sample food intake type sequences, multiple sample food intake amount sequences, and multiple sample nutrition intake information sets, as well as corresponding multiple sample stage nutrition recipe plans as training data, wherein each sample nutrition intake information set includes K sample nutrition intake information, and each sample stage nutrition recipe plan includes K sample stage nutrition recipe plans.

[0086] Step S520: using the training data to perform supervised training on a framework constructed based on a feedforward neural network until the training reaches convergence, thereby obtaining the nutrition intake analysis network layer after training.

[0087] Furthermore, step S500 further includes:

[0088] Step S530: Obtain a preset monitoring period.

[0089] Step S540: Monitor the execution records of the K-stage nutritional recipe plans of the target patient within the preset monitoring period to obtain monitoring feedback results.

[0090] Step S550: Modify the K-stage nutritional recipe plans based on the monitoring feedback results.

[0091] Through the above detailed description of a system for providing personalized nutritional recommendations to elderly surgical patients, those skilled in the art can clearly understand the system and method for providing personalized nutritional recommendations to elderly surgical patients in this embodiment. As for the method disclosed in the embodiment, since it corresponds to the system disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0092] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A system for providing personalized nutrition recommendations for elderly surgical patients, characterized in that: The system comprises: A basic information set acquisition module, the basic information set acquisition module is used to acquire a basic information set of a target patient, wherein the basic information set includes nutritional requirement information, dietary preference information, and personal physical information; A nutrition recommendation space acquisition module, wherein the nutrition recommendation space acquisition module is used to construct a nutrition recommendation space; An initial nutrition recommendation information determination module, the initial nutrition recommendation information determination module is used to input the nutrition requirement information, dietary preference information and personal physical information into the nutrition recommendation space for retrieval, and determine the initial nutrition recommendation information, wherein the initial nutrition recommendation information includes K nutrition intake information of K stages, K being an integer greater than or equal to 1; A patient intake information acquisition module, the patient intake information acquisition module is used to acquire a food intake type sequence and a food intake amount sequence of the target patient within a preset historical window, wherein the food intake type and the food intake amount correspond one to one; A nutrition recipe plan determination module is used to utilize a nutrition intake analysis network layer to configure a recipe according to the food intake type sequence, the food intake amount sequence and the K nutrition intake information, and determine nutrition recipe plans for K stages.

2. A system for providing personalized nutrition recommendations for elderly surgical patients as claimed in claim 1, characterized in that: The nutrition recommendation space acquisition module also includes: Construct a three-dimensional coordinate system using nutritional requirement information, dietary preference information, and personal physical information as coordinate axes; Obtain multiple sample basic information sets and multiple sample nutritional recommendation information of multiple sample patients; Inputting the plurality of sample basic information sets and the plurality of sample nutritional recommendation information into the three-dimensional coordinate system to obtain a plurality of sample points; The nutrition recommendation space is constructed based on the multiple sample points and the three-dimensional coordinate system.

3. A system for providing personalized nutrition recommendations for elderly surgical patients as claimed in claim 2, characterized in that: The initial nutrition recommendation information determination module also includes: Inputting the nutritional requirement information, dietary preference information and personal physical information into the nutritional recommendation space to obtain a target point; Extracting sample points whose distance to the target point is a preset distance in the nutrition recommendation space to obtain a plurality of matching sample points; A balanced analysis is performed on the nutrition recommendation information of the multiple matching samples of the multiple matching sample points to determine the initial nutrition recommendation information.

4. A system for providing personalized nutrition recommendations for elderly surgical patients as claimed in claim 3, characterized in that: The initial nutrition recommendation information determination module also includes: Performing mean processing on the plurality of matching sample nutrition recommendation information to determine a mean of the matching sample nutrition recommendation information; Determine a plurality of matching weights according to the distances between the plurality of matching sample nutrition recommendation information and the mean of the matching sample nutrition recommendation information; The nutrition recommendation information of the multiple matching samples is weightedly calculated based on the multiple matching weights to obtain the initial nutrition recommendation information.

5. A system for providing personalized nutrition recommendations for elderly surgical patients as claimed in claim 1, characterized in that: The patient intake information acquisition module also includes: Acquire multiple historical food intake types, multiple historical food intake amount sets, and multiple historical food intake frequencies of the target patient within a preset historical window; Serializing the multiple historical food intake types based on the multiple historical food intake frequencies to obtain the food intake type sequence; The plurality of historical food intake sets are averaged, and the processing results are serialized according to a mapping relationship between food intake type and food intake in combination with the food intake type sequence to obtain the food intake sequence.

6. A system for providing personalized nutrition recommendations for elderly surgical patients as claimed in claim 1, characterized in that: The nutritional recipe program determination module also includes: Acquire multiple sample food intake type sequences, multiple sample food intake amount sequences, multiple sample nutrition intake information sets, and corresponding multiple sample stage nutrition recipe plans as training data, wherein each sample nutrition intake information set includes K sample nutrition intake information, and each sample stage nutrition recipe plan includes K sample stage nutrition recipe plans; The training data is used to perform supervised training on a framework constructed based on a feedforward neural network until the training converges, thereby obtaining the training-completed nutrient intake analysis network layer.

7. A system for providing personalized nutrition recommendations for elderly surgical patients as claimed in claim 1, characterized in that: The nutritional recipe program determination module also includes: Get the preset monitoring period; Monitoring the execution records of the K-stage nutritional recipe plans of the target patient within the preset monitoring period to obtain monitoring feedback results; The K-stage nutritional recipe plans are modified based on the monitoring feedback results.

8. A method for providing personalized nutritional recommendations for elderly surgical patients, characterized in that: The method comprises: Acquire a basic information set of the target patient, wherein the basic information set includes nutritional requirement information, dietary preference information, and personal physical information; Building a nutrition recommendation space; Inputting the nutrition requirement information, dietary preference information and personal physical information into the nutrition recommendation space for retrieval to determine initial nutrition recommendation information, wherein the initial nutrition recommendation information includes K nutrition intake information of K stages, where K is an integer greater than or equal to 1; Acquire a food intake type sequence and a food intake amount sequence of the target patient within a preset history window, wherein the food intake type and the food intake amount correspond one to one; The nutrition intake analysis network layer is used to configure a recipe according to the food intake type sequence, the food intake amount sequence and the K nutrition intake information to determine K stage nutrition recipe plans.

Images