Analysis and detection method and system for cardiovascular and cerebrovascular diseases

Abstract

The invention discloses a cardiovascular and cerebrovascular disease analysis and detection method and system. Analyzing the relationship between the influence factors of the sex, age, region, heart rate, sleep time, exercise condition, height, weight, blood pressure, blood sugar, total cholesterol, smoking condition, drinking condition, coronary heart disease / cerebral apoplexy past history and family history condition of the user and the risk of the cerebral cardiovascular disease; single risk factor searching is carried out based on a Logistic regression algorithm; related interaction risk factors are searched on the basis of a Pearson algorithm; a binary logic method is adopted to distinguish the relationship between the traditional Chinese medicine constitution and the cerebral cardiovascular disease risk, transverse data statistical analysis is finally included, and a cerebral cardiovascular disease risk assessment model based on an ensemble learning algorithm is constructed; the cardiovascular and cerebrovascular disease risk assessment model is used for performing cardiovascular and cerebrovascular disease assessment, so that disease reasons of patients can be conveniently searched, and reasonable cardiovascular and cerebrovascular disease risk assessment is given.

Description

Technical Field

[0001] The present invention relates to the field of medical information processing and control technology, and in particular to a cardiovascular and cerebrovascular disease analysis and detection method and system. Background Art

[0002] Cardiovascular and cerebrovascular diseases, as one of the most important health problems worldwide, pose a huge threat to human life and health. The diagnosis process of this type of disease is complex and requires comprehensive consideration of various physiological signals and clinical indicators. However, existing cardiovascular and cerebrovascular disease analysis and testing is difficult to find the relationship between influencing factors and the risk of cerebrovascular and cardiovascular diseases, conduct cardiovascular and cerebrovascular disease assessments, and cannot propose good health promotion plans. Summary of the invention

[0003] In order to solve the defects of the prior art, the present invention provides a cardiovascular and cerebrovascular disease analysis and detection method and system.

[0004] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0005] The present invention provides a cardiovascular and cerebrovascular disease analysis and detection method and system, comprising the following steps:

[0006] Step 1: Collect influencing data on patients with cerebrovascular diseases, including but not limited to gender, age, region, heart rate, sleep time, exercise status, height, weight, blood pressure, blood sugar, total cholesterol, smoking status, drinking status, history of coronary heart disease / stroke and family history;

[0007] Step 2: Preprocess the influencing data, and then analyze the relationship between the influencing factors of the user's gender, age, region, heart rate, sleep time, exercise status, height, weight, blood pressure, blood sugar, total cholesterol, smoking status, drinking status, history of coronary heart disease / stroke and family history and the risk of cerebrovascular disease;

[0008] Step 3: Search for single risk factors based on the Logistic regression algorithm; search for relevant interactive risk factors based on the Pearson algorithm;

[0009] Step 4: Use binary logic method to identify the relationship between TCM constitution and cerebrovascular disease risk, and finally incorporate it into horizontal data statistical analysis to construct a cerebrovascular disease risk assessment model based on ensemble learning algorithm.

[0010] Step 5: Use the cerebrovascular disease risk assessment model to assess cardiovascular and cerebrovascular diseases.

[0011] As a preferred technical solution of the present invention, after the cerebrovascular disease risk assessment model assesses the risk of cerebrovascular disease, the dietary intake, exercise status, sleep schedule, tobacco and alcohol taboos, health care methods and other intervention precautions of patients with cerebrovascular disease are imported into the knowledge graph after formatting and normalizing the data. Traditional Chinese medicine constitution, cerebrovascular disease risk factors, diet, exercise, self-care techniques, etc. are used as nodes of the knowledge graph, and the corresponding intervention methods are used as edges. The Neo4j graph database is used to establish a knowledge graph, and the Cypher statement of the Neo4j graph database is used to recommend effective health promotion plans for individuals according to the cerebrovascular disease risk assessment model.

[0012] As a preferred technical solution of the present invention, the method for preprocessing the influencing data is to use a convolution kernel with a size of K*1 to perform a temporal convolution operation, expand the original single variable data set into an m-dimensional feature data set, and then use the expanded m-dimensional feature data set as the input set of the LSTM neural network, and then construct a cerebrovascular disease risk assessment model based on an integrated learning algorithm.

[0013] As a preferred technical solution of the present invention, a cardiovascular disease analysis and detection system includes a user layer, an intermediate layer, a data receiving layer, a data processing layer, a cerebrovascular disease risk assessment module and a health promotion program recommendation module;

[0014] The user layer includes a doctor login module and a patient login module, wherein the doctor login module is used for medical personnel to log in to the cardiovascular and cerebrovascular disease analysis system, and the patient login module is used for patients to log in to the cardiovascular and cerebrovascular disease analysis system; the middle layer is used to realize information exchange between the doctor login module and the patient login module, and the middle layer includes a video service module, a voice service module and a short message service module;

[0015] The data receiving layer is used to receive vital signs data, which includes text and chart types; the data receiving layer can also extract vital signs data of the chart type. The data processing layer is used to preprocess the vital signs data collected by the data receiving layer, and the preprocessed vital signs data is used as the input layer and input into the cerebrovascular disease risk assessment module to assess the patient's cardiovascular and cerebrovascular diseases, and through the health promotion program recommendation module, effective health promotion programs are recommended to individuals based on the assessment results of the cerebrovascular disease risk assessment module.

[0016] As a preferred technical solution of the present invention, it also includes a Chinese medicine recommendation module, which is used to recommend Chinese medicine treatment to patients. The recommendation method of the Chinese medicine recommendation module includes the following steps:

[0017] Step A: Collect historical diagnosis and treatment data of patients with cardiovascular and cerebrovascular diseases; obtain a training data set based on the historical diagnosis and treatment data and a generative adversarial network model;

[0018] Step B, inputting historical diagnosis and treatment data into the generative adversarial network model for training to obtain generated diagnosis and treatment data;

[0019] Step C: training a prediction model based on the training data set and the deep neural network model;

[0020] Step D: The prediction model is optimized using a parameter differential evolution algorithm to obtain a target prediction model;

[0021] Step E: obtaining target diagnosis and treatment data of target cardiovascular and cerebrovascular disease patients; based on the target prediction model and the target diagnosis and treatment data, obtaining a target prognosis Chinese medicine treatment recommendation for the target cardiovascular and cerebrovascular disease patients.

[0022] As a preferred technical solution of the present invention, the target prediction model is used to predict the prognosis of traditional Chinese medicine treatment recommendations for patients with cardiovascular and cerebrovascular diseases, including: obtaining a test diagnosis and treatment data group, the test diagnosis and treatment data group including multiple test diagnosis and treatment data sequences; performing a mutation operation on one or more test diagnosis and treatment data in the test diagnosis and treatment data sequence to obtain a variant diagnosis and treatment data sequence; performing a crossover operation on any two or more of the variant diagnosis and treatment data sequences to generate a child diagnosis and treatment data sequence; based on the test diagnosis and treatment data sequence and the child diagnosis and treatment data sequence, optimizing the prediction model to obtain a target prediction model for predicting the prognosis of traditional Chinese medicine treatment recommendations for patients with cardiovascular and cerebrovascular diseases.

[0023] The beneficial effects of the present invention are:

[0024] 1. The cardiovascular disease analysis and detection method and system collects influencing data of patients with cerebrovascular diseases, wherein the influencing data include but are not limited to gender, age, region, heart rate, sleep time, exercise status, height, weight, blood pressure, blood sugar, total cholesterol, smoking status, drinking status, history of coronary heart disease / stroke and family history; pre-processes the influencing data, and then analyzes the relationship between the influencing factors of the user's gender, age, region, heart rate, sleep time, exercise status, height, weight, blood pressure, blood sugar, total cholesterol, smoking status, drinking status, history of coronary heart disease / stroke and family history and the risk of cerebrovascular disease; Single risk factors are searched based on the Logistic regression algorithm; related interactive risk factors are searched based on the Pearson algorithm; binary logic method is used to distinguish the relationship between TCM constitution and cerebrovascular disease risk, and finally horizontal data statistical analysis is included to build a cerebrovascular disease risk assessment model based on an integrated learning algorithm; cardiovascular disease risk assessment model is used to assess cardiovascular and cerebrovascular diseases; the present invention searches for the relationship between influencing factors and the risk of cerebrovascular diseases, and obtains single risk factors and related interactive risk factors, so as to facilitate the search for the cause of the patient's illness and provide a reasonable cerebrovascular disease risk assessment.

[0025] 2. In this cardiovascular disease analysis and detection method, after evaluating the risk of cerebrovascular disease in the cerebrovascular disease risk assessment model, the dietary intake, exercise status, sleep schedule, tobacco and alcohol taboos, health care methods and other intervention precautions of patients with cerebrovascular disease are formatted and normalized and then imported into the knowledge graph. Traditional Chinese medicine constitution, cerebrovascular disease risk factors, diet, exercise, self-care techniques, etc. are used as nodes of the knowledge graph, and the corresponding intervention methods are used as edges. The knowledge graph is established using the Neo4j graph database, and the Cypher statement of the Neo4j graph database is used to recommend effective health promotion plans for individuals according to the cerebrovascular disease risk assessment model, so that reasonable treatment plan suggestions can be given.

[0026] 3. The method for preprocessing the influencing data described in the present invention is to use a convolution kernel with a size of K*1 to perform a temporal convolution operation, expand the original single variable data set into an m-dimensional feature data set, and then use the expanded m-dimensional feature data set as the input set of the LSTM neural network, and then construct a cerebrovascular disease risk assessment model based on an integrated learning algorithm. In this way, the single variable data set is expanded to an m-dimensional feature data set, and then the expanded m-dimensional feature data set is used as the input set of the LSTM neural network, so that the obtained model is more accurate.

[0027] 4. The present invention also includes a traditional Chinese medicine recommendation module, which is used to recommend traditional Chinese medicine treatment to patients. The recommendation method of the traditional Chinese medicine recommendation module is to collect historical diagnosis and treatment data of historical cardiovascular and cerebrovascular disease patients; based on the historical diagnosis and treatment data and the generative adversarial network model, a training data set is obtained; the historical diagnosis and treatment data is input into the generative adversarial network model for training to obtain generated diagnosis and treatment data; based on the training data set and the deep neural network model, a prediction model is trained; the prediction model is optimized using a parameter differential evolution algorithm to obtain a target prediction model; the target diagnosis and treatment data of the target cardiovascular and cerebrovascular disease patient is obtained; based on the target prediction model and the target diagnosis and treatment data, a target prognosis traditional Chinese medicine treatment recommendation plan for the target cardiovascular and cerebrovascular disease patient is obtained, so that the use of drugs can be recommended. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0029] In the attached picture:

[0030] Figure 1 It is a schematic diagram of a process of a cardiovascular and cerebrovascular disease analysis and detection method of the present invention;

[0031] Figure 2 It is a structural schematic diagram of a cardiovascular and cerebrovascular disease analysis and detection system of the present invention;

[0032] Figure 3 It is a workflow diagram of a drug recommendation module in a cardiovascular and cerebrovascular disease analysis and detection system of the present invention;

[0033] Figure 4 It is a schematic diagram of a cerebrovascular disease risk assessment module in a cardiovascular disease analysis and detection system of the present invention.

[0034] In the figure: 1. User layer; 101. Doctor login module; 102. Patient login module; Middle layer; 2. Middle layer; 201. Video service module; 202. Voice service module; 203. SMS service module; 3. Data receiving layer; 4. Data processing layer; 5. Cerebrovascular disease risk assessment module; 6. Health promotion program recommendation module; 7. Traditional Chinese medicine recommendation module. DETAILED DESCRIPTION

[0035] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0036] Example: Figure 1As shown, the present invention provides a cardiovascular and cerebrovascular disease analysis and detection method, comprising the following steps:

[0037] Step 1: Collect influencing data on patients with cerebrovascular diseases, including but not limited to gender, age, region, heart rate, sleep time, exercise status, height, weight, blood pressure, blood sugar, total cholesterol, smoking status, drinking status, history of coronary heart disease / stroke and family history;

[0038] Step 2: Preprocess the influencing data, and then analyze the relationship between the influencing factors of the user's gender, age, region, heart rate, sleep time, exercise status, height, weight, blood pressure, blood sugar, total cholesterol, smoking status, drinking status, history of coronary heart disease / stroke and family history and the risk of cerebrovascular disease;

[0039] Step 3: Search for single risk factors based on the Logistic regression algorithm; search for relevant interactive risk factors based on the Pearson algorithm;

[0040] Step 4: Use binary logic method to identify the relationship between TCM constitution and cerebrovascular disease risk, and finally incorporate it into horizontal data statistical analysis to construct a cerebrovascular disease risk assessment model based on ensemble learning algorithm.

[0041] Step 5: Use the cerebrovascular disease risk assessment model to assess cerebrovascular disease. The present invention seeks the relationship between influencing factors and the risk of cerebrovascular disease, and obtains single risk factors and related interactive risk factors, so as to facilitate the search for the cause of the patient's illness and provide a reasonable cerebrovascular disease risk assessment.

[0042] After the cerebrovascular disease risk assessment model assesses the risk of cerebrovascular disease, the dietary intake, exercise status, sleep schedule, tobacco and alcohol taboos, health care methods and other intervention precautions of patients with cerebrovascular disease are imported into the knowledge graph after formatting and normalizing the data. The traditional Chinese medicine constitution, cerebrovascular disease risk factors, diet, exercise, self-care techniques, etc. are used as nodes of the knowledge graph, and the corresponding intervention methods are used as edges. The knowledge graph is established using the Neo4j graph database, and the Cypher statement of the Neo4j graph database is used to recommend effective health promotion plans for individuals based on the cerebrovascular disease risk assessment model. In this way, reasonable treatment plan recommendations can be given.

[0043] The method for preprocessing the influencing data is to use a convolution kernel with a size of K*1 to perform a time convolution operation, expand the original single variable data set into an m-dimensional feature data set, and then use the expanded m-dimensional feature data set as the input set of the LSTM neural network, and then build a cerebrovascular disease risk assessment model based on an integrated learning algorithm; in this way, the single variable data set is expanded into an m-dimensional feature data set, and then the expanded m-dimensional feature data set is used as the input set of the LSTM neural network, so that the obtained model is more accurate.

[0044] A cardiovascular disease analysis and detection system comprises a user layer 1, an intermediate layer 2, a data receiving layer 3, a data processing layer 4, a cerebrovascular disease risk assessment module 5 and a health promotion program recommendation module 6;

[0045] The user layer 1 includes a doctor login module 101 and a patient login module 102, wherein the doctor login module 101 is used by medical personnel to log in to the cardiovascular and cerebrovascular disease analysis system, and the patient login module 102 is used by patients to log in to the cardiovascular and cerebrovascular disease analysis system; the middle layer 2 is used to realize information exchange between the doctor login module 3 and the patient login module 4, and the middle layer 2 includes a video service module 201, a voice service module 202 and a text message service module 203;

[0046] The data receiving layer is used to receive vital sign data, which includes text and chart data; the data receiving layer can also extract chart-type vital sign data, and the data processing layer is used to pre-process the vital sign data collected by the data receiving layer, and the pre-processed vital sign data is used as the input layer and input into the cerebrovascular disease risk assessment module 5 to assess the patient's cardiovascular disease, and the health promotion program recommendation module 6 is used to recommend effective health promotion programs to individuals based on the assessment results of the cerebrovascular disease risk assessment module 5. It is convenient for remote treatment.

[0047] Among them, it also includes a traditional Chinese medicine recommendation module, which is used to recommend traditional Chinese medicine treatment to patients. The recommendation method of the traditional Chinese medicine recommendation module includes the following steps:

[0048] Step A: Collect historical diagnosis and treatment data of patients with cardiovascular and cerebrovascular diseases; obtain a training data set based on the historical diagnosis and treatment data and a generative adversarial network model;

[0049] Step B, inputting historical diagnosis and treatment data into the generative adversarial network model for training to obtain generated diagnosis and treatment data;

[0050] Step C: training a prediction model based on the training data set and the deep neural network model;

[0051] Step D: The prediction model is optimized using a parameter differential evolution algorithm to obtain a target prediction model;

[0052] Step E: obtaining target diagnosis and treatment data of target cardiovascular and cerebrovascular disease patients; based on the target prediction model and the target diagnosis and treatment data, obtaining a target prognosis Chinese medicine treatment recommendation for the target cardiovascular and cerebrovascular disease patients.

[0053] Among them, the historical diagnosis and treatment data include: gender, age, medical history data, genetic history data, examination results, test results, physical examination results, drug use records and health monitoring data of historical cardiovascular and cerebrovascular disease patients, among which the drug use records include: type of drugs used and dosage of drugs used.

[0054] Among them, the target prediction model is used to predict the prognosis of patients with cardiovascular and cerebrovascular diseases and the recommended Chinese medicine treatment plan, including: obtaining a test diagnosis and treatment data group, the test diagnosis and treatment data group includes multiple test diagnosis and treatment data sequences; performing a mutation operation on one or more test diagnosis and treatment data in the test diagnosis and treatment data sequence to obtain a variant diagnosis and treatment data sequence; performing a cross operation on any two or more of the variant diagnosis and treatment data sequences to generate a child diagnosis and treatment data sequence; based on the test diagnosis and treatment data sequence and the child diagnosis and treatment data sequence, optimizing the prediction model to obtain a target prediction model for predicting the prognosis of patients with cardiovascular and cerebrovascular diseases and the recommended Chinese medicine treatment plan.

[0055] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for analyzing and detecting cardiovascular and cerebrovascular diseases, characterized in that: The following steps are involved: Step 1: Collect influencing data on patients with cerebrovascular diseases, including but not limited to gender, age, region, heart rate, sleep time, exercise status, height, weight, blood pressure, blood sugar, total cholesterol, smoking status, drinking status, history of coronary heart disease / stroke and family history; Step 2: Preprocess the influencing data, and then analyze the relationship between the influencing factors of the user's gender, age, region, heart rate, sleep time, exercise status, height, weight, blood pressure, blood sugar, total cholesterol, smoking status, drinking status, history of coronary heart disease / stroke and family history and the risk of cerebrovascular disease; Step 3: Search for single risk factors based on the Logistic regression algorithm; Search for relevant interaction risk factors based on Pearson algorithm; Step 4: Use binary logic method to identify the relationship between TCM constitution and cerebrovascular disease risk, and finally incorporate it into horizontal data statistical analysis to construct a cerebrovascular disease risk assessment model based on ensemble learning algorithm. Step 5: Use the cerebrovascular disease risk assessment model to assess cardiovascular and cerebrovascular diseases.

2. A cardiovascular and cerebrovascular disease analysis and detection method according to claim 1, characterized in that: After the cerebrovascular disease risk assessment model assesses the risk of cerebrovascular disease, the dietary intake, exercise status, sleep schedule, tobacco and alcohol taboos, health care methods and other intervention precautions of patients with cerebrovascular disease are formatted and normalized before being imported into the knowledge graph. Traditional Chinese medicine constitution, cerebrovascular disease risk factors, diet, exercise, self-care techniques, etc. are used as nodes of the knowledge graph, and the corresponding intervention methods are used as edges. The Neo4j graph database is used to establish a knowledge graph, and the Cypher statement of the Neo4j graph database is used to recommend effective health promotion plans for individuals based on the cerebrovascular disease risk assessment model.

3. A cardiovascular and cerebrovascular disease analysis and detection method according to claim 1, characterized in that: The method for preprocessing the influencing data is to use a convolution kernel with a size of K*1 to perform a temporal convolution operation, expand the original single variable data set into an m-dimensional feature data set, and then use the expanded m-dimensional feature data set as the input set of the LSTM neural network, and then construct a cerebrovascular disease risk assessment model based on an integrated learning algorithm.

4. A cardiovascular and cerebrovascular disease analysis system according to claim 1, characterized in that: It includes a user layer (1), an intermediate layer (2), a data receiving layer (3), a data processing layer (4), a cerebrovascular disease risk assessment module (5) and a health promotion program recommendation module (6); The user layer (1) comprises a doctor login module (101) and a patient login module (102), wherein the doctor login module (101) is used for medical personnel to log in to the cardiovascular and cerebrovascular disease analysis system, and the patient login module (102) is used for patients to log in to the cardiovascular and cerebrovascular disease analysis system; the middle layer (2) is used to realize information exchange between the doctor login module (3) and the patient login module (4), and the middle layer (2) comprises a video service module (201), a voice service module (202) and a short message service module (203); The data receiving layer is used to receive vital sign data, which includes text and chart data; the data receiving layer can also extract chart-type vital sign data. The data processing layer is used to pre-process the vital sign data collected by the data receiving layer, and the pre-processed vital sign data is used as an input layer and input into a cerebrovascular disease risk assessment module (5) to assess the patient's cardiovascular disease, and through a health promotion program recommendation module (6), an effective health promotion program is recommended to the individual based on the assessment result of the cerebrovascular disease risk assessment module (5).

5. A cardiovascular and cerebrovascular disease analysis and detection system according to claim 4, characterized in that: It also includes a Chinese medicine recommendation module (7), which is used to recommend Chinese medicine treatment to patients. The recommendation method of the Chinese medicine recommendation module includes the following steps: Step A: Collect historical diagnosis and treatment data of patients with cardiovascular and cerebrovascular diseases; obtain a training data set based on the historical diagnosis and treatment data and a generative adversarial network model; Step B, inputting historical diagnosis and treatment data into the generative adversarial network model for training to obtain generated diagnosis and treatment data; Step C: training a prediction model based on the training data set and the deep neural network model; Step D: The prediction model is optimized using a parameter differential evolution algorithm to obtain a target prediction model; Step E: obtaining target diagnosis and treatment data of target cardiovascular and cerebrovascular disease patients; based on the target prediction model and the target diagnosis and treatment data, obtaining a target prognosis Chinese medicine treatment recommendation for the target cardiovascular and cerebrovascular disease patients.

6. A cardiovascular and cerebrovascular disease analysis and detection method and system according to claim 5, characterized in that: The historical diagnosis and treatment data include: gender, age, medical history data, genetic history data, examination results, test results, physical examination results, drug use records and health monitoring data of historical cardiovascular and cerebrovascular disease patients, among which the drug use records include: type of drugs used and dosage of drugs used.

7. A cardiovascular and cerebrovascular disease analysis and detection method and system according to claim 5, characterized in that: The target prediction model is used to predict the prognosis of patients with cardiovascular and cerebrovascular diseases and the recommended Chinese medicine treatment plan, including: obtaining a test diagnosis and treatment data group, the test diagnosis and treatment data group including multiple test diagnosis and treatment data sequences; performing a mutation operation on one or more test diagnosis and treatment data in the test diagnosis and treatment data sequence to obtain a variant diagnosis and treatment data sequence; performing a cross operation on any two or more of the variant diagnosis and treatment data sequences to generate a child diagnosis and treatment data sequence; based on the test diagnosis and treatment data sequence and the child diagnosis and treatment data sequence, optimizing the prediction model to obtain a target prediction model for predicting the prognosis of patients with cardiovascular and cerebrovascular diseases and the recommended Chinese medicine treatment plan.

Images