Knowledge graph database construction method based on chronic disease diet management

Abstract

The invention provides a knowledge graph database construction method based on chronic disease diet management, and relates to the field of health management, and the method comprises the steps: determining a plurality of chronic disease types; according to the kth type of chronic disease type, acquiring historical patient information of the kth type of historical patients; according to the historical patient information of the kth type of historical patients, determining a diet health relation function of the kth type of chronic disease type; determining names of various types of food; according to the name of the ith type of food, acquiring nutritional ingredients of the ith type of food; determining a food health score according to the nutritional ingredients of the ith food and the diet health relation function of the kth chronic disease type; and generating a knowledge graph database according to the food health score, the food name and the chronic disease type. According to the method and the device, the knowledge graph database generation accuracy can be improved.

Description

Technical Field

[0001] The present invention relates to the field of health management, and particularly to a method for constructing a knowledge graph database based on chronic disease diet management. Background Art

[0002] In the related art, CN109145119A discloses a knowledge graph construction device and a construction method in the field of health management. The knowledge graph construction device includes: a data processing unit, which is connected to each other and is used to obtain a plurality of entities and generate relationships between the entities. Among them, the plurality of entities include any multiple of sports entities, diet entities, sleep entities, and psychological entities; a display unit, which is used to construct and display a knowledge graph according to the entities and the relationships between the entities. The knowledge graph includes a plurality of nodes, and the knowledge graph is used to display the current node among the plurality of nodes, the remaining nodes related to the current node, and the relationships between the current node and the remaining nodes. The plurality of nodes respectively represent the displayed entities, and the connection lines between the plurality of nodes represent the relationships between the displayed entities. The above technical solution of this solution has the beneficial effects of rich content, wide coverage of the knowledge system, and high scalability.

[0003] CN114925216A discloses a health management method based on a knowledge graph. The method includes: obtaining a target keyword indicating the health management needs of a user; querying a target entity corresponding to the target keyword and other entities having a relationship with the target entity in the knowledge graph; and outputting and displaying the queried target entity, the other entities, and the relationships between the target entity and the other entities as search results to the user.

[0004] Based on the above related technologies, the scalability of the knowledge graph can be improved. However, the related technologies do not analyze the specific relationships between entities, that is, the specific relationships between entities cannot be analyzed, and a knowledge graph database cannot be established according to the specific relationships between entities and the entities.

[0005] The information disclosed in the background art part of the present application is only intended to deepen the understanding of the general background art of the present application, and should not be regarded as an admission or any form of implication that this information constitutes the prior art known to those skilled in the art. Summary of the Invention

[0006] The present invention provides a method for constructing a knowledge graph database based on chronic disease diet management, which can solve the technical problems that the related technologies cannot analyze the specific relationships between entities and cannot establish a knowledge graph database according to the specific relationships between entities and the entities.

[0007] According to a first aspect of the present invention, there is provided a method for constructing a knowledge graph database based on chronic disease diet management, including:

[0008] Determine multiple types of chronic disease types;

[0009] According to the k-th type of chronic disease type, obtain the historical patient information of the k-th type of historical patients, where the historical patient information includes: patient diet information, patient health data, and the number of disease occurrences of the patient;

[0010] According to the historical patient information of the k-th type of historical patients, determine the diet-health relationship function of the k-th type of chronic disease type;

[0011] Determine multiple types of food names;

[0012] According to the i-th type of food name, obtain the nutritional components of the i-th food;

[0013] According to the nutritional components of the i-th food and the diet-health relationship function of the k-th type of chronic disease type, determine the food health score;

[0014] According to the food health score, the food name, and the chronic disease type, generate a knowledge graph database.

[0015] According to the second aspect of the present invention, there is provided a knowledge graph database construction system for chronic disease diet management, including:

[0016] A chronic disease type module for determining multiple types of chronic disease types;

[0017] A patient information module for obtaining the historical patient information of the k-th type of historical patients according to the k-th type of chronic disease type, where the historical patient information includes: patient diet information, patient health data, and the number of disease occurrences of the patient;

[0018] A relationship function module for determining the diet-health relationship function of the k-th type of chronic disease type according to the historical patient information of the k-th type of historical patients;

[0019] A food name module for determining multiple types of food names;

[0020] A food information module for obtaining the nutritional components of the i-th food according to the i-th type of food name;

[0021] A food health module for determining the food health score according to the nutritional components of the i-th food and the diet-health relationship function of the k-th type of chronic disease type;

[0022] A database generation module for generating a knowledge graph database according to the food health score, the food name, and the chronic disease type.

[0023] Technical effects: According to the present invention, the relationship between the patient's diet information and the patient's health data can be accurately analyzed, and based on this relationship, the impact of various foods on the patient's health can be determined. Further, based on the impact of various foods on the patient's health, the food names, and the types of chronic diseases, a knowledge graph database is generated, improving the accuracy of generating the knowledge graph database. When determining the patient's health score, the patient's health score can be determined according to the patient's blood glucose data, the standard deviation of the systolic blood pressure change rate, the standard deviation of the diastolic blood pressure change rate, the preset standard systolic blood pressure data, the preset standard blood glucose data, the patient's systolic blood pressure data, the patient's systolic blood pressure change rate, the patient's diastolic blood pressure change rate, and the number of patient disease occurrences. During the calculation process, the patient's blood pressure health status within the historical period can be determined according to the patient's systolic blood pressure status, the patient's systolic blood pressure change status, and the patient's diastolic blood pressure change status. Further, the patient's overall health status can be determined according to the patient's blood pressure health status, the patient's blood glucose health status, and the number of patient disease occurrences, improving the comprehensiveness and accuracy of the patient's health score. When determining the diet-health relationship function, the diet-health relationship function for the k-th type of chronic disease is determined according to the amount of food consumed by the patient, the nutritional components of the food consumed by the patient, and the patient's health score, accurately describing the impact relationship of the amount of food consumed by the patient and the nutritional components of the food consumed by the patient on the optimization degree of the patient's blood glucose health status and the optimization degree of the patient's blood pressure health status, improving the accuracy of the diet-health relationship function. When determining the food health score, the food health score of the i-th food for the k-th type of chronic disease can be determined according to the fat content, protein content, carbohydrate content, and vitamin content of the i-th food, and the diet-health relationship function for the k-th type of chronic disease, improving the accuracy of the food health score.

[0024] It should be understood that the above general description and the following detailed description are merely exemplary and explanatory, and do not limit the present invention. According to the following detailed description of exemplary embodiments with reference to the accompanying drawings, other features and aspects of the present invention will become clearer. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or in the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other embodiments can be obtained based on these drawings without creative efforts;

[0026] Figure 1 Exemplarily shows a schematic flowchart of a method for constructing a knowledge graph database based on chronic disease diet management according to an embodiment of the present invention;

[0027] Figure 2A block diagram of a knowledge graph database construction system for chronic disease diet management according to an embodiment of the present invention is exemplarily shown. Detailed implementation manners

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are only a part rather than all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0029] The technical solutions of the present invention will be described in detail below with specific embodiments. These specific embodiments may be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0030] Figure 1 A flowchart of a method for constructing a knowledge graph database for chronic disease diet management according to an embodiment of the present invention is exemplarily shown. The method includes:

[0031] Step S101, determining multiple types of chronic disease types;

[0032] Step S102, according to the kth type of chronic disease type, obtaining historical patient information of the kth type of historical patients, where the historical patient information includes: patient diet information, patient health data, and the number of patient disease occurrences;

[0033] Step S103, according to the historical patient information of the kth type of historical patients, determining a diet-health relationship function for the kth type of chronic disease type;

[0034] Step S104, determining multiple types of food names;

[0035] Step S105, according to the ith type of food name, obtaining the nutritional components of the ith food;

[0036] Step S106, according to the nutritional components of the ith food and the diet-health relationship function of the kth type of chronic disease type, determining a food health score;

[0037] Step S107, generating a knowledge graph database according to the food health score, the food name, and the chronic disease type.

[0038] The method for constructing a knowledge graph database based on chronic disease diet management according to an embodiment of the present invention can accurately analyze the relationship between patients' diet information and patients' health data, and determine the impact of various foods on patients' health based on this relationship. Further, based on the impact of various foods on patients' health, food names, and chronic disease types, a knowledge graph database is generated, improving the accuracy of generating the knowledge graph database.

[0039] According to an embodiment of the present invention, in step S101, multiple types of chronic disease types are determined.

[0040] For example, according to a professional chronic disease management database and knowledge base, multiple types of chronic disease types are determined, such as hypertension and diabetes.

[0041] According to an embodiment of the present invention, in step S102, according to the k-th type of chronic disease type, historical patient information of the k-th type of historical patients is obtained, where the historical patient information includes: patients' diet information, patients' health data, and the number of times of patients' disease onset.

[0042] For example, the k-th type of historical patients are patients suffering from the k-th type of chronic disease type, and through authoritative health institutions and medical institutions, the patients' diet information, patients' health data, and the number of times of patients' disease onset (the number of times of complication of chronic disease complications) of patients suffering from the k-th type of chronic disease type in the historical time period are obtained.

[0043] According to an embodiment of the present invention, in step S103, according to the historical patient information of the k-th type of historical patients, a diet-health relationship function of the k-th type of chronic disease type is determined.

[0044] According to an embodiment of the present invention, step S103 includes:

[0045] According to the patients' diet information, determine the amount of food consumed by the patients and the nutritional components of the food consumed by the patients, where the nutritional components of the food consumed by the patients include: protein content, carbohydrate content, vitamin content, and fat content;

[0046] According to the patients' health data, determine the patients' blood pressure data and the patients' blood glucose data;

[0047] According to the patients' blood pressure data, the patients' blood glucose data, and the number of times of patients' disease onset, determine the patients' health score;

[0048] According to the amount of food consumed by the patients, the nutritional components of the food consumed by the patients, and the patients' health score, determine the diet-health relationship function of the k-th type of chronic disease type.

[0049] For example, based on the patient's dietary information, determine the amounts of various foods consumed by the chronic disease patient within a historical time period, that is, the amounts of foods consumed by the patient, and determine the nutritional components of various foods consumed by the chronic disease patient; obtain the patient's blood pressure data and the patient's blood glucose data within the historical time period; evaluate the patient's health status within the historical period based on the patient's blood pressure data, the patient's blood glucose data, and the number of times the patient has fallen ill, and determine the patient's health score; the patient's health status is related to the patient's dietary status to a certain extent. For example, when the vitamin content of the consumed foods is relatively high, the patient's blood pressure and blood glucose conditions are also better. Based on the correlation of the above data, determine the dietary health relationship function between the patient's health score, the amounts of foods consumed by the patient, and the nutritional components of the foods consumed by the patient.

[0050] According to an embodiment of the present invention, determining the patient's health score based on the patient's blood pressure data, the patient's blood glucose data, and the number of times the patient has fallen ill includes:

[0051] Determine the preset standard high blood pressure data and the preset standard blood glucose data according to the kth type of chronic disease type;

[0052] Determine the patient's high blood pressure data and the patient's low blood pressure data based on the patient's blood pressure data;

[0053] Perform fitting based on the patient's high blood pressure data and the moments in the historical time period to obtain the high blood pressure data function of the patient's high blood pressure data in the historical time period;

[0054] Determine the derivative function of the high blood pressure data according to the high blood pressure data function;

[0055] Determine the patient's high blood pressure change rate at multiple moments in the historical time period according to the derivative function of the high blood pressure data;

[0056] Determine the standard deviation of the high blood pressure change rate according to the patient's high blood pressure change rate at multiple moments in the historical time period;

[0057] Perform fitting based on the patient's low blood pressure data and the moments in the historical time period to obtain the low blood pressure data function of the patient's low blood pressure data in the historical time period;

[0058] Determine the derivative function of the low blood pressure data according to the low blood pressure data function;

[0059] Determine the patient's low blood pressure change rate at multiple moments in the historical time period according to the derivative function of the low blood pressure data;

[0060] Determine the standard deviation of the low blood pressure change rate according to the patient's low blood pressure change rate at multiple moments in the historical time period;

[0061] Determine the patient's health score based on the patient's blood glucose data, the standard deviation of the high blood pressure change rate, the standard deviation of the low blood pressure change rate, the preset standard high blood pressure data, the preset standard blood glucose data, the patient's high blood pressure data, the patient's high blood pressure change rate, the patient's low blood pressure change rate, and the number of times the patient has had an illness.

[0062] For example, according to the k-th type of chronic disease type, determine the preset standard high blood pressure data and the preset standard blood glucose data. For example, when the k-th type of chronic disease type is diabetes, the preset standard blood glucose data is 7 mmol / L; when the k-th type of chronic disease type is not diabetes, the preset standard blood glucose data is 4 mmol / L; when the k-th type of chronic disease type is not hypertension, the preset standard high blood pressure data is 120 mmHg; when the k-th type of chronic disease type is hypertension, the preset standard high blood pressure data is 140 mmHg. Determine the patient's high blood pressure data and the patient's low blood pressure data during the historical time period. Fit the patient's high blood pressure data and the time points during the historical time period to obtain a high blood pressure data function that describes the law of change of the patient's high blood pressure data over time. Take the derivative of the high blood pressure data function to determine the high blood pressure derivative function. Substitute the time points during the historical time period into the high blood pressure derivative function to determine the patient's high blood pressure change rate at multiple time points during the historical time period. Calculate the standard deviation of the high blood pressure change rate based on the patient's high blood pressure change rate at multiple time points during the historical time period. Fit the patient's low blood pressure data and the time points during the historical time period to obtain a low blood pressure data function that describes the law of change of the patient's low blood pressure data over time. Take the derivative of the low blood pressure data function to determine the low blood pressure derivative function. Substitute the time points during the historical time period into the low blood pressure derivative function to determine the patient's low blood pressure change rate at multiple time points during the historical time period. Calculate the standard deviation of the low blood pressure change rate based on the patient's low blood pressure change rate at multiple time points during the historical time period. Evaluate the patient's health status based on the patient's blood glucose data, the standard deviation of the high blood pressure change rate, the standard deviation of the low blood pressure change rate, the preset standard high blood pressure data, the preset standard blood glucose data, the patient's high blood pressure data, the patient's high blood pressure change rate, the patient's low blood pressure change rate, and the number of times the patient has had an illness, and determine the patient's health score.

[0063] According to an embodiment of the present invention, determining the patient's health score based on the patient's blood glucose data, the standard deviation of the high blood pressure change rate, the standard deviation of the low blood pressure change rate, the preset standard high blood pressure data, the preset standard blood glucose data, the patient's high blood pressure data, the patient's high blood pressure change rate, the patient's low blood pressure change rate, and the number of times the patient has had an illness includes: determining the patient's health score of the e-th historical patient of the k-th type of chronic disease type in the j-th historical time period according to formula (1) , (1) where and is a preset weight value, is a preset multiple, if is a conditional function, and max is a function for taking the maximum value. is a preset standard high - pressure data, is the high - pressure data of the e - th historical patient at the r - th moment in the j - th historical time period, is the blood - glucose data of the e - th historical patient at the r - th moment in the j - th historical time period, is a preset standard blood - glucose data, is the r - th moment in the historical time period, is the high - pressure change rate of the e - th historical patient at the r - th moment in the j - th historical time period, is the standard deviation of the high - pressure change rate of the e - th historical patient in the j - th historical time period, is the low - pressure change rate of the e - th historical patient at the r - th moment in the j - th historical time period, is the standard deviation of the low - pressure change rate of the e - th historical patient in the j - th historical time period, is the number of disease occurrences of the e - th historical patient in the j - th historical time period. R is the number of moments in the historical time period, r ≤ R, and both r and R are positive integers.

[0064] According to an embodiment of the present invention, is to take the maximum value of the high - pressure data of the e - th historical patient at R moments in the j - th historical time period. The above - mentioned process of taking the maximum value can be used to determine whether the e - th historical patient has a condition of excessive blood pressure in the j - th historical time period. is the relative difference between the preset standard high - pressure data and the maximum value of the high - pressure data of the e - th historical patient in the j - th historical time period. The larger this ratio is, the smaller the maximum value of the high - pressure data of the e - th historical patient in the j - th historical time period is relatively, the smaller the possibility that the patient has a condition of excessive blood pressure, and the better the blood - pressure health condition of the patient.

[0065] According to an embodiment of the present invention, in formula (1), the following two situations can be represented in the form of a conditional function. When the condition of is satisfied, it means that the high - pressure change rate of the e - th historical patient at the r - th moment in the j - th historical time period is centered on the average high - pressure change rate of patients and is within two times the preset multiple (for example, the preset multiple Within the interval where the standard deviation of (3) is the interval length, the value of the conditional function is 1, indicating that the deviation between the patient's high blood pressure change rate and the mean at the r-th moment in the j-th historical time period of the e-th historical patient is small, so the patient's high blood pressure change is relatively stable at the r-th moment. Otherwise, the deviation between the patient's high blood pressure change rate and the mean at the r-th moment in the j-th historical time period of the e-th historical patient is large, so the patient's high blood pressure change is relatively abnormal. is the ratio of the number of moments when the patient's high blood pressure change is relatively stable to the number of moments in the historical time period. The larger this ratio, the relatively more stable the high blood pressure change of the e-th historical patient in the j-th historical time period, and the better the patient's blood pressure health condition.

[0066] According to an embodiment of the present invention, in formula (1), the following two situations can be represented in the form of a conditional function. When the condition is satisfied, it indicates that the patient's low blood pressure change rate at the r-th moment in the j-th historical time period of the e-th historical patient is within the interval centered on the average patient's low blood pressure change rate and with a standard deviation of twice the preset multiple (e.g., the preset multiple is 3). The value of the conditional function is 1, indicating that the deviation between the patient's low blood pressure change rate and the mean at the r-th moment in the j-th historical time period of the e-th historical patient is small, so the patient's low blood pressure change is relatively stable at the r-th moment. Otherwise, the deviation between the patient's low blood pressure change rate and the mean at the r-th moment in the j-th historical time period of the e-th historical patient is large, so the patient's low blood pressure change is relatively abnormal. is the ratio of the number of moments when the patient's low blood pressure change is relatively stable to the number of moments in the historical time period. The larger this ratio, the relatively more stable the low blood pressure change of the e-th historical patient in the j-th historical time period, and the better the patient's blood pressure health condition.

[0067] According to an embodiment of the present invention, represents determining the patient's blood pressure health condition in the historical period based on the patient's high blood pressure condition, the patient's high blood pressure change condition, and the patient's low blood pressure change condition.

[0068] According to an embodiment of the present invention, is the relative difference between the patient's blood glucose data and the preset standard blood glucose data at the r-th moment in the j-th historical time period of the e-th historical patient. The smaller this ratio, the closer the patient's blood glucose data at the r-th moment is to the preset standard blood glucose data, and the better the patient's blood glucose health condition. is to calculate the average value based on the number of moments in the historical time period. The larger this average value, the better the patient's blood glucose health condition in the historical time period.

[0069] According to an embodiment of the present invention, Determine the overall health status of a patient based on the patient's blood pressure health status, blood sugar health status, and the number of times the patient has fallen ill, where, and are preset weights determined according to the type of chronic disease of the patient. For example, when the type of chronic disease is hypertension, , such as, , , indicating that the blood pressure health status has a greater impact on the overall health status of the patient. When the type of chronic disease is diabetes, , such as, , , indicating that the blood sugar health status has a greater impact on the overall health status of the patient.

[0070] In this way, based on the patient's blood sugar data, standard deviation of systolic blood pressure change rate, standard deviation of diastolic blood pressure change rate, preset standard systolic blood pressure data, preset standard blood sugar data, patient's systolic blood pressure data, patient's systolic blood pressure change rate, patient's diastolic blood pressure change rate, and the number of times the patient has fallen ill, the patient's health score can be determined. During the calculation process, based on the patient's systolic blood pressure status, systolic blood pressure change status, and diastolic blood pressure change status of the patient, the blood pressure health status of the patient within the historical period can be determined. Further, based on the patient's blood pressure health status, blood sugar health status, and the number of times the patient has fallen ill, the overall health status of the patient can be determined, improving the comprehensiveness and accuracy of the patient's health score.

[0071] According to an embodiment of the present invention, determine the dietary health relationship function of the k-th type of chronic disease based on the amount of food consumed by the patient, the nutritional components of the food consumed by the patient, and the patient's health score, including: determining the first undetermined coefficient equation of the dietary health relationship function of the k-th type of chronic disease according to formula (2), (2) where, and are preset weights, is the patient's health score of the e-th historical patient of the k-th type of chronic disease in the j-th historical time period, is the patient's health score of the e-th historical patient of the k-th type of chronic disease in the (j - 1)-th historical time period, is the amount of food consumed by the patient of the q-th type of food in the j-th historical time period, is the protein content of the q-th type of food consumed by the patient, is the carbohydrate content of the q-th type of food consumed by the patient, is the vitamin content of the q-th type of food consumed by the patient, is the fat content of the q-th type of food consumed by the patient, , , , , , , , , , , and are the first undetermined coefficients of the first undetermined coefficient equation, Q is the number of types of food consumed by the patient, q ≤ Q, and both q and Q are positive integers;

[0072] Solve the first undetermined coefficient according to the amount of food consumed by the patient, the patient's diet composition, and the patient's health score to obtain the solution value of the first undetermined coefficient;

[0073] Determine the diet-health relationship function of the kth type of chronic disease according to the solution value of the first undetermined coefficient and the first undetermined coefficient equation.

[0074] According to an embodiment of the present invention, is the difference between the health scores of the e-th historical patient of the kth type of chronic disease in the j-th historical time period and the (j - 1)-th historical time period, indicating the degree of optimization of the health status of the e-th historical patient.

[0075] According to an embodiment of the present invention, represents the part that shows the influence of the carbohydrate content, vitamin content, and fat content ingested by the patient on the patient's blood pressure health status among the patient's health status, represents the total amount of vitamins ingested by the patient in the j-th historical time period, represents the average vitamin content of all the foods ingested by the patient in the j-th historical time period, represents the total amount of carbohydrates ingested by the patient in the j-th historical time period, represents the average carbohydrate content of all the foods ingested by the patient in the j-th historical time period, represents the total amount of fat ingested by the patient in the j-th historical time period, represents the average fat content of all the foods ingested by the patient in the j-th historical time period, represents that the average vitamin content of all the foods ingested by the patient in the j-th historical time period has a positive correlation with the degree of optimization of the patient's blood pressure health status. For example, antioxidants such as vitamin C can protect the blood vessel wall, reduce vascular inflammation, help lower blood pressure, and maintain blood pressure stability. represents that the average carbohydrate content of all the foods ingested by the patient in the j-th historical time period has a negative correlation with the degree of optimization of the patient's blood pressure health status. For example, a long-term high-carbohydrate diet may lead to weight gain, which in turn increases the risk of hypertension. It indicates that there is a negative correlation between the average fat content of all the foods consumed by the patient in the j-th historical time period and the degree of optimization of the patient's blood pressure health condition. For example, saturated fatty acids and trans fatty acids in dietary fat can cause blood pressure to increase.

[0076] According to an embodiment of the present invention, The part representing the influence of the carbohydrate content, vitamin content, and protein content consumed by the patient on the patient's blood glucose health condition among the patient's health conditions, Represents the protein content consumed by the patient in the j-th historical time period, Represents the average protein content of all the foods consumed by the patient in the j-th historical time period, It indicates that there is a positive correlation between the average vitamin content of all the foods consumed by the patient in the j-th historical time period and the degree of optimization of the patient's blood glucose health condition. For example, vitamin C can promote sugar metabolism and help maintain blood glucose stability. It indicates that there is a negative correlation between the average protein content of all the foods consumed by the patient in the j-th historical time period and the degree of optimization of the patient's blood glucose health condition. For example, when too much protein is consumed, it may cause blood glucose to rise. It indicates that there is a negative correlation between the average carbohydrate content of all the foods consumed by the patient in the j-th historical time period and the degree of optimization of the patient's blood glucose health condition. For example, excessive carbohydrate intake can cause blood glucose to rise rapidly. Based on the above relationships, the first undetermined coefficient equation of the dietary health relationship function of the k-th type of chronic disease type can be obtained.

[0077] According to an embodiment of the present invention, fitting can be performed based on multiple parameters involved in the above first undetermined coefficient equation, that is, based on the amount of food consumed by the patient, the nutritional components of the food consumed by the patient, and the patient's health score for fitting, and the above multiple first undetermined coefficients are solved. There are 12 first undetermined coefficients, namely, , , , , , , , , , , and , the above 12 first undetermined coefficients are solved according to the amount of food consumed by the patient, the nutritional components of the food consumed by the patient, and the patient's health score of at least 12 patients, the solution values of the above 12 first undetermined coefficients are obtained, and the solution values of the above 12 first undetermined coefficients are substituted into the first undetermined coefficient equation to determine the dietary health relationship function of the k-th type of chronic disease type.

[0078] In this way, based on the amount of food consumed by the patient, the nutritional components of the food consumed by the patient, and the patient's health score, the dietary health relationship function of the k-th type of chronic disease is determined, which accurately describes the influence relationship between the amount of food consumed by the patient and the nutritional components of the food consumed by the patient on the optimization degree of the patient's blood glucose health status and the optimization degree of the patient's blood pressure health status, and improves the accuracy of the dietary health relationship function.

[0079] According to an embodiment of the present invention, in step S104, multiple types of food names are determined.

[0080] For example, multiple common types of food names are obtained, such as potatoes and tomatoes.

[0081] According to an embodiment of the present invention, in step S105, according to the i-th type of food name, the nutritional components of the i-th food are obtained.

[0082] For example, according to the food name, the nutritional components of the food are queried on a professional nutrition data website.

[0083] According to an embodiment of the present invention, in step S106, according to the nutritional components of the i-th food and the dietary health relationship function of the k-th type of chronic disease, the food health score is determined.

[0084] According to an embodiment of the present invention, step S106 includes:

[0085] According to the nutritional components of the i-th food, the fat content, protein content, carbohydrate content, and vitamin content of the i-th food are determined;

[0086] According to the fat content, protein content, carbohydrate content, and vitamin content of the i-th food, and the dietary health relationship function of the k-th type of chronic disease, the food health score of the i-th food with respect to the k-th type of chronic disease is determined.

[0087] For example, according to a professional nutrition data website, the fat content, protein content, carbohydrate content, and vitamin content of the i-th food are queried; the fat content, protein content, carbohydrate content, and vitamin content of the i-th food are substituted into the dietary health relationship function of the k-th type of chronic disease to evaluate the impact of consuming the i-th food on the health status of patients with the k-th type of chronic disease, and the food health score of the i-th food with respect to the k-th type of chronic disease is determined.

[0088] According to an embodiment of the present invention, the food health score of the i-th food with respect to the k-th type of chronic disease is determined according to the fat content, protein content, carbohydrate content, and vitamin content of the i-th food, and the diet-health relationship function of the k-th type of chronic disease, including: determining the food health score of the i-th food with respect to the k-th type of chronic disease according to formula (3). ,

[0089] (3)

[0090] wherein, and are preset weights, is the protein content of the i-th food, is the carbohydrate content of the i-th food, is the vitamin content of the i-th food, is the fat content of the i-th food, is the solution value of, is the solution value of, is the solution value of, is the solution value of, is the solution value of, is the solution value of, is the solution value of, is the solution value of, is the solution value of, is the solution value of, is the solution value of, is the solution value of.

[0091] According to an embodiment of the present invention, the fat content, protein content, carbohydrate content, and vitamin content of the i-th food are substituted into the diet-health relationship function of the k-th type of chronic disease to obtain formula (3), thereby obtaining the food health score of the i-th food with respect to the k-th type of chronic disease.

[0092] In this way, the food health score of the i-th food with respect to the k-th type of chronic disease can be determined according to the fat content, protein content, carbohydrate content, and vitamin content of the i-th food, and the diet-health relationship function of the k-th type of chronic disease, which can improve the accuracy of the food health score.

[0093] According to an embodiment of the present invention, in step S107, a knowledge graph database is generated based on the food health score, the food name, and the type of chronic disease.

[0094] According to an embodiment of the present invention, step S107 includes:

[0095] Determine the relationship between the food and the type of chronic disease according to the food health score and the set food health score threshold;

[0096] Generate knowledge graph database entities based on the food name and the type of chronic disease;

[0097] Generate a knowledge graph database according to the knowledge graph database entities and the relationship between the food and the type of chronic disease.

[0098] For example, set the food health score threshold to , if the food health score of the i-th food for the k-th type of chronic disease is greater than , it means that consuming the i-th food is beneficial to the recovery of the k-th type of chronic disease, and the relationship between the i-th food and the k-th type of chronic disease is "beneficial". If the food health score of the i-th food for the k-th type of chronic disease is less than or equal to , it means that consuming the i-th food is not beneficial to the recovery of the k-th type of chronic disease, and the relationship between the i-th food and the k-th type of chronic disease is "not beneficial"; identify the food name and the type of chronic disease as entities through named entity recognition technology; select the storage method of the knowledge graph (such as, a graph database), import the relationship between the entities into the graph database, optimize the data in the knowledge graph, and regularly update the data in the knowledge graph to maintain its accuracy and integrity.

[0099] The method for constructing a knowledge graph database based on chronic disease diet management according to an embodiment of the present invention can accurately analyze the relationship between the patient's diet information and the patient's health data, and determine the impact of various foods on the patient's health based on this relationship. Further, based on the impact of various foods on the patient's health, the food names, and the types of chronic diseases, a knowledge graph database is generated, improving the accuracy of generating the knowledge graph database. When determining the patient's health score, the patient's health score can be determined according to the patient's blood glucose data, the standard deviation of the systolic blood pressure change rate, the standard deviation of the diastolic blood pressure change rate, the preset standard systolic blood pressure data, the preset standard blood glucose data, the patient's systolic blood pressure data, the patient's systolic blood pressure change rate, the patient's diastolic blood pressure change rate, and the number of patient disease occurrences. During the calculation process, the patient's blood pressure health status within the historical period can be determined according to the patient's systolic blood pressure status, the patient's systolic blood pressure change status, and the patient's diastolic blood pressure change status. Further, according to the patient's blood pressure health status, the patient's blood glucose health status, and the number of patient disease occurrences, the patient's overall health status is determined, improving the comprehensiveness and accuracy of the patient's health score. When determining the diet-health relationship function, the diet-health relationship function of the k-th type of chronic disease is determined according to the amount of food consumed by the patient, the nutritional components of the food consumed by the patient, and the patient's health score, accurately describing the impact relationship of the amount of food consumed by the patient and the nutritional components of the food consumed by the patient on the optimization degree of the patient's blood glucose health status and the optimization degree of the patient's blood pressure health status, improving the accuracy of the diet-health relationship function. When determining the food health score, the food health score of the i-th food regarding the k-th type of chronic disease can be determined according to the fat content, protein content, carbohydrate content, and vitamin content of the i-th food, and the diet-health relationship function of the k-th type of chronic disease, improving the accuracy of the food health score.

[0100] Figure 2 Exemplarily shown is a block diagram of a knowledge graph database construction system based on chronic disease diet management according to an embodiment of the present invention. The system includes:

[0101] A chronic disease type module for determining multiple types of chronic disease types;

[0102] A patient information module for obtaining the historical patient information of the k-th type of historical patients according to the k-th type of chronic disease type, where the historical patient information includes: patient diet information, patient health data, and the number of patient disease occurrences;

[0103] A relationship function module for determining the diet-health relationship function of the k-th type of chronic disease type according to the historical patient information of the k-th type of historical patients;

[0104] A food name module for determining multiple types of food names;

[0105] A food information module for obtaining the nutritional components of the i-th type of food according to the name of the i-th type of food;

[0106] A food health module for determining a food health score according to the nutritional components of the i-th type of food and the dietary health relationship function of the k-th type of chronic disease;

[0107] A database generation module for generating a knowledge graph database according to the food health score, the food name, and the chronic disease type.

[0108] The present invention may be a method, an apparatus, a system, and / or a computer program product. The computer program product may include a computer-readable storage medium having thereon computer-readable program instructions for performing various aspects of the present invention.

[0109] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the drawings are only examples and do not limit the present invention. The object of the present invention has been fully and effectively achieved. The functions and structural principles of the present invention have been shown and described in the embodiments, and the embodiments of the present invention may have any deformation or modification without departing from the principle.

Claims

1. A method for constructing a knowledge graph database based on chronic disease dietary management, characterized in that: include: Identify various types of chronic diseases; According to the k-th type of chronic disease, historical patient information of the k-th historical patient is obtained, wherein the historical patient information includes: patient diet information, patient health data and patient disease attack frequency; Determining a diet-health relationship function of a k-th type of chronic disease type according to the historical patient information of the k-th type of historical patients; Identify the names of various types of food; According to the name of the i-th type of food, obtain the nutritional components of the i-th food; Determine a food health score according to the nutritional components of the i-th food and the dietary health relationship function of the k-th type of chronic disease; A knowledge graph database is generated based on the food health score, the food name and the chronic disease type.

2. The method for constructing a knowledge graph database based on chronic disease dietary management according to claim 1 is characterized in that: According to the historical patient information of the k-th type of historical patients, the diet-health relationship function of the k-th type of chronic disease is determined, including: Determine the amount of food consumed by the patient and the nutritional components of the food consumed by the patient according to the patient's dietary information, wherein the nutritional components of the food consumed by the patient include: protein content, carbohydrate content, vitamin content and fat content; Determine the patient's blood pressure data and the patient's blood sugar data according to the patient's health data; Determine the patient's health score according to the patient's blood pressure data, the patient's blood sugar data and the number of illness attacks of the patient; A diet-health relationship function of the kth type of chronic disease is determined according to the amount of food consumed by the patient, the nutritional components of the food consumed by the patient and the health score of the patient.

3. The method for constructing a knowledge graph database based on chronic disease diet management according to claim 2 is characterized in that: Determining the patient's health score according to the patient's blood pressure data, the patient's blood sugar data, and the number of illnesses of the patient includes: Determining preset standard high blood pressure data and preset standard blood sugar data according to the kth type of chronic disease; Determine the patient's high pressure data and the patient's low pressure data according to the patient's blood pressure data; Perform fitting based on the patient's high pressure data and the moments in the historical time period to obtain a high pressure data function of the patient's high pressure data in the historical time period; Determining a high-voltage data derivative function according to the high-voltage data function; Determine the patient's high pressure change rate at multiple moments in a historical time period according to the high pressure data derivative function; Determine a standard deviation of the high pressure change rate according to the high pressure change rate of the patient at multiple moments in the historical time period; Perform fitting based on the patient's low pressure data and the moments in the historical time period to obtain a low pressure data function of the patient's low pressure data in the historical time period; Determining a low-pressure data derivative function according to the low-pressure data function; Determine the patient's low pressure change rate at multiple moments in a historical time period according to the low pressure data derivative function; Determine a standard deviation of the low pressure change rate according to the low pressure change rate of the patient at multiple moments in the historical time period; The patient health score is determined based on the patient's blood sugar data, the standard deviation of the high pressure change rate, the standard deviation of the low pressure change rate, the preset standard high pressure data, the preset standard blood sugar data, the patient's high pressure data, the patient's high pressure change rate, the patient's low pressure change rate and the number of times the patient has an illness.

4. The method for constructing a knowledge graph database based on chronic disease diet management according to claim 3 is characterized in that: Determining the patient health score according to the patient blood sugar data, the high pressure change rate standard deviation, the low pressure change rate standard deviation, the preset standard high pressure data, the preset standard blood sugar data, the patient high pressure data, the patient high pressure change rate, the patient low pressure change rate and the number of illnesses of the patient, including: According to the formula Determine the patient health score of the e-th historical patient of the k-th type of chronic disease in the j-th historical time period ,in, and is the preset weight, is the preset multiple, if is the conditional function, and max is the maximum value function. To preset standard high voltage data, is the high pressure data of the e-th historical patient at the r-th moment in the j-th historical time period, is the blood glucose data of the e-th historical patient at the r-th moment in the j-th historical time period, To preset standard blood sugar data, is the rth moment of the historical time period, is the high pressure change rate of the e-th historical patient at the r-th moment in the j-th historical time period, is the standard deviation of the high pressure change rate of the e-th historical patient in the j-th historical time period, is the low pressure change rate of the e-th historical patient at the r-th moment in the j-th historical time period, is the standard deviation of the low pressure change rate of the e-th historical patient in the j-th historical time period, is the number of attacks of the e-th historical patient in the j-th historical time period, R is the number of moments in the historical time period, r≤R, r and R are both positive integers.

5. The method for constructing a knowledge graph database based on chronic disease diet management according to claim 2, characterized in that: According to the amount of food consumed by the patient, the nutritional components of the food consumed by the patient and the health score of the patient, a diet-health relationship function of the kth type of chronic disease is determined, including: According to the formula Determine the first undetermined coefficient equation of the diet-health relationship function of the kth type of chronic disease, where: and is the preset weight, The health score of the e-th historical patient of the k-th chronic disease type in the j-th historical time period, The health score of the e-th historical patient of the k-th chronic disease type in the j-1-th historical time period, The amount of food consumed by the qth patient in the jth historical period, is the protein content of the food consumed by the qth patient, is the carbohydrate content of the food consumed by the qth patient, is the vitamin content of the food consumed by the qth patient, is the fat content of the food consumed by the qth patient, , , , , , , , , , , and is the first undetermined coefficient of the first undetermined coefficient equation, Q is the number of types of food consumed by the patient, q≤Q, and q and Q are both positive integers; Solving the first undetermined coefficient according to the amount of food consumed by the patient, the dietary components of the patient and the health score of the patient to obtain a solution value of the first undetermined coefficient; According to the solved value of the first undetermined coefficient and the first undetermined coefficient equation, the diet-health relationship function of the kth type of chronic disease is determined.

6. The method for constructing a knowledge graph database based on chronic disease diet management according to claim 5, characterized in that: Determining a food health score according to the nutritional components of the i-th food and the dietary health relationship function of the k-th type of chronic disease includes: Determine the fat content, protein content, carbohydrate content and vitamin content of the i-th food according to the nutritional components of the i-th food; Based on the fat content, protein content, carbohydrate content and vitamin content of the i-th food and the dietary health relationship function of the k-th type of chronic disease, the food health score of the i-th food with respect to the k-th type of chronic disease is determined.

7. The method for constructing a knowledge graph database based on chronic disease diet management according to claim 6, characterized in that: According to the fat content, protein content, carbohydrate content and vitamin content of the i-th food and the dietary health relationship function of the k-th chronic disease type, the food health score of the i-th food with respect to the k-th chronic disease type is determined, including: According to the formula Determine the food health score of the i-th food with respect to the k-th type of chronic disease ,in, and is the preset weight, is the protein content of the ith food, is the carbon content of the ith food, is the vitamin content of the ith food, is the fat content of the ith food, for The solution value of for The solution value of for The solution value of for The solution value of for The solution value of for The solution value of for The solution value of for The solution value of for The solution value of for The solution value of for The solution value of for The solution value of .

8. The method for constructing a knowledge graph database based on chronic disease diet management according to claim 1, characterized in that: According to the food health score, the food name and the chronic disease type, a knowledge graph database is generated, including: Determining the relationship between food and chronic disease type according to the food health score and a set food health score threshold; Generate a knowledge graph database entity according to the food name and the chronic disease type; A knowledge graph database is generated based on the relationship between the knowledge graph database entities and the food and chronic disease types.

9. A knowledge graph database construction system based on chronic disease diet management, characterized in that: include: The chronic disease type module is used to determine the types of various chronic diseases; A patient information module is used to obtain historical patient information of a k-th historical patient according to the k-th type of chronic disease, wherein the historical patient information includes: patient diet information, patient health data and patient disease attack frequency; A relationship function module, used for determining a diet-health relationship function of a k-th type of chronic disease type according to the historical patient information of the k-th type of historical patients; Food name module, used to determine the names of various types of food; The food information module is used to obtain the nutritional components of the i-th type of food according to the name of the i-th type of food; A food health module, used to determine a food health score according to the nutritional components of the i-th food and the dietary health relationship function of the k-th type of chronic disease; The database generation module is used to generate a knowledge graph database based on the food health score, the food name and the chronic disease type.

Images