Acute myocardial infarction patient self-management system based on knowledge graph

Abstract

The invention discloses an acute myocardial infarction patient self-management system based on a knowledge graph. The system comprises a patient condition acquisition module, a patient condition analysis module, a knowledge graph analysis module and a patient self-management module. According to online and offline acute myocardial infarction literatures, the acute myocardial infarction literatures are corrected, enough knowledge reserve is obtained, then a knowledge graph is constructed according to keywords of various acute myocardial infarction literatures and association between the keywords, knowledge accumulation of acute myocardial infarction is obtained, and then the knowledge of acute myocardial infarction is determined according to the current condition of a patient. According to the system provided by the invention, the patient can obtain the corresponding taste management guidance and monitoring only by inputting the actual condition of the patient, so that medical help is provided for the patient in real time, and the patient can obtain the corresponding taste management guidance and monitoring. The cardiovascular disease prevention and rehabilitation efficiency of patients is improved, and the workload of doctors is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of patient self - management systems, and more particularly to a self - management system for acute myocardial infarction patients based on a knowledge graph. Background Art

[0002] Acute Myocardial Infarction (AMI) is a cardiovascular disease with high incidence and high fatality rate. Self - management is crucial for improving the survival rate and quality of life of AMI patients. However, currently, patients need to carry out self - management under the guidance of doctors, and the minimum frequency for patients to go to the hospital is once a week. That is to say, the self - management methods for patients are updated only once a week. Obviously, this is not conducive to the prevention and rehabilitation efficiency of patients' cardiovascular diseases and will also increase the workload of doctors. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a self - management system for acute myocardial infarction patients based on a knowledge graph in view of the above - mentioned deficiencies in the prior art. By combining the constructed knowledge graph with the current situation of patients, it provides guidance and monitoring for patients' self - management, thereby improving the prevention and rehabilitation efficiency of patients for cardiovascular diseases and reducing the workload of doctors.

[0004] To solve the above - mentioned technical problem, the technical solution adopted by the present invention is: A self - management system for acute myocardial infarction patients based on a knowledge graph, comprising:

[0005] A patient condition acquisition module, configured to acquire the recent physical condition and current physical condition of the patient;

[0006] A patient condition analysis module, configured to determine the current physical state of the patient according to the recent physical condition and current physical condition of the patient;

[0007] A knowledge graph analysis module, configured to obtain the self - management needs of the patient through a pre - established knowledge graph according to the current physical state of the patient;

[0008] A patient self - management module, configured to provide corresponding services to complete the self - management of the patient according to the self - management needs of the patient;

[0009] The knowledge graph is established by obtaining online and offline acute myocardial infarction literatures and analyzing the acute myocardial infarction literatures to construct the knowledge graph.

[0010] Further, the establishment of the knowledge graph includes the following steps:

[0011] S1: Obtain acute myocardial infarction literature through online and offline methods, and correct the acute myocardial infarction literature obtained online except for the set channels to obtain a number of acute myocardial infarction literature;

[0012] S2: Extract the keywords and the logical relationships between the keywords from each acute myocardial infarction literature respectively;

[0013] S3: Logically connect each acute myocardial infarction literature according to the keywords and the logical relationships between the keywords in each acute myocardial infarction literature;

[0014] S4: Correspond the keywords with the acute myocardial infarction literature respectively, form a knowledge graph according to the logical relationships between the keywords, and call the corresponding acute myocardial infarction literature according to the keywords.

[0015] Furthermore, in step S1, the correction of the acute myocardial infarction literature obtained online except for the set channels includes:

[0016] Screen the acute myocardial infarction literature obtained from each online channel according to the channel, and obtain the acute myocardial infarction literature from each online channel except for the set channels;

[0017] Extract the keywords and the logical relationships between the keywords in the acute myocardial infarction literature from each online channel except for the set channels, and obtain modification suggestions through a pre-constructed neural network model for the keywords and the logical relationships between the keywords;

[0018] Modify the acute myocardial infarction literature from each online channel except for the set channels according to the modification suggestions;

[0019] The neural network model outputs modification suggestions by inputting the entered keywords and the logical relationships between the keywords.

[0020] Furthermore, the modification of the acute myocardial infarction literature from each online channel except for the set channels according to the modification suggestions includes:

[0021] Decompose the modification suggestions to obtain multiple modification contents;

[0022] Determine the paragraphs in the corresponding acute myocardial infarction literature for the modification contents according to the keywords in the modification contents;

[0023] Expand the modification contents according to the keywords in the modification contents to obtain corrected contents;

[0024] Replace the paragraphs in the corresponding acute myocardial infarction literature for the modification contents with the corrected contents to obtain the modified acute myocardial infarction literature.

[0025] Further, expanding the modified content according to the keywords of the modified content to obtain the corrected content, including:

[0026] Searching for the paragraph with the highest similarity in the set channels online and the acute myocardial infarction literature offline according to the order of the keywords to obtain the similar paragraphs;

[0027] Rewriting the similar paragraphs using AI rewriting technology to obtain the corrected content.

[0028] Further, in the knowledge graph analysis module, obtaining the patient's self-management needs according to the patient's current physical condition through the pre-established knowledge graph, including:

[0029] Converting the patient's current physical condition into keywords;

[0030] Searching in the knowledge graph according to the keywords of the physical condition, and finding the acute myocardial infarction literature corresponding to the keywords of the physical condition;

[0031] Extracting the associations between the keywords in the acute myocardial infarction literature corresponding to the physical condition keywords;

[0032] Determining the self-management needs according to the logical relationship between the keywords in the acute myocardial infarction literature.

[0033] Further, determining the self-management needs according to the logical relationship between the keywords in the acute myocardial infarction literature, including:

[0034] Determining the keywords included in the self-management needs according to the logical relationship between the keywords in the acute myocardial infarction literature;

[0035] Determining the keywords of the self-management needs according to the labels of the keywords included in the self-management needs;

[0036] Obtaining the self-management needs using the AI general writing method according to the keywords of the self-management needs.

[0037] Further, the patient self-management module provides corresponding services according to the patient's self-management needs to complete the patient's self-management, including:

[0038] Decomposing the needs of the patient's self-management needs to obtain each self-management content;

[0039] Decomposing each self-management content respectively to obtain the management time and management matters;

[0040] Generating corresponding services according to the management practice and management matters;

[0041] According to the services corresponding to each self - management content, start the corresponding service resources to provide services for patients.

[0042] The present invention has the following advantages compared with the prior art:

[0043] The present invention provides a self - management system for acute myocardial infarction patients based on a knowledge graph. According to the online and offline acute myocardial infarction literatures, and correcting the acute myocardial infarction literatures, sufficient knowledge reserves are obtained. Then, according to the keywords of various acute myocardial infarction literatures and the associations between them, a knowledge graph is constructed to obtain the knowledge accumulation of acute myocardial infarction. After that, combined with the current situation of the patient, the self - management needs of the patient are analyzed, and guidance and monitoring for the patient's self - management are provided. For the system provided by the present invention, as long as the patient inputs their actual situation, corresponding guidance and monitoring for self - management can be obtained, so as to provide medical assistance for the patient in real - time, improve the prevention and rehabilitation efficiency of the patient for cardiovascular diseases, and reduce the workload of doctors.

[0044] The technical solutions of the present invention will be further described in detail below through the accompanying drawings and embodiments. Brief Description of the Drawings

[0045] Figure 1 It is a schematic diagram of the overall structure of a self - management system for acute myocardial infarction patients based on a knowledge graph provided by the present invention. Detailed Embodiments

[0046] As Figure 1 shown, a self - management system for acute myocardial infarction patients based on a knowledge graph provided by the present invention includes:

[0047] A patient condition acquisition module, which is used to acquire the recent physical condition and current physical condition of the patient;

[0048] A patient condition analysis module, which is used to determine the current physical state of the patient according to the recent physical condition and current physical condition of the patient;

[0049] A knowledge graph analysis module, which is used to obtain the self - management needs of the patient through the pre - established knowledge graph according to the current physical state of the patient;

[0050] A patient self - management module, which is used to provide corresponding services to complete the self - management of the patient according to the self - management needs of the patient;

[0051] The establishment of the knowledge graph is achieved by obtaining online and offline acute myocardial infarction literatures and analyzing the acute myocardial infarction literatures to construct the knowledge graph.

[0052] The above-mentioned modules interact with each other to perform their respective tasks, and ultimately obtain the self-management needs of the patient, assist the patient in self-management, and ensure their physical health.

[0053] The patient condition acquisition module is used to obtain the patient's recent physical condition and current physical condition. In the present invention, the recent physical condition and the current physical condition refer to the physical conditions at different times. Among them, the recent physical condition can be obtained from the physical examination reports at different past times, or can be obtained according to the data retained in the current physical condition each time the system of the present invention was used in the past. The current physical condition can be obtained not only by on-site collection and entry, but also by facial analysis, or a combination of both, to obtain the current physical condition.

[0054] The physical condition in the present invention refers to various physical index parameters of acute myocardial infarction patients.

[0055] The patient condition analysis module is used to determine the patient's current physical state according to the patient's recent physical condition and current physical condition. In the present invention, according to the recent physical condition and the current physical condition, the trend of the development of the physical condition is determined, that is, the patient's current physical state.

[0056] In the present invention, the patient's current physical state represents the development trend of the patient's physical indicators. In this development trend, the degree of the development trend is represented by the speed of development growth. That is to say, the current physical state represents the degree of development of various physical index parameters.

[0057] The knowledge graph analysis module is used to obtain the patient's self-management needs through a pre-established knowledge graph according to the patient's current physical state. Among them, the establishment of the knowledge graph is achieved by obtaining online and offline acute myocardial infarction literatures and analyzing the acute myocardial infarction literatures to construct the knowledge graph.

[0058] The present invention provides the most comprehensive help to acute myocardial infarction patients through the knowledge graph. The knowledge graph will be automatically improved with the progress of technology, and the most advanced knowledge and experience are obtained to help patients.

[0059] In the present invention, the knowledge graph is established through various online and offline literatures. These technical literatures record the relevant experiences of acute myocardial infarction. According to the keywords of each acute myocardial infarction literature and the logical relationship between the keywords, the content of the acute myocardial infarction literature can be obtained, which is similar to expressing the meaning of the whole article through a logical expression.

[0060] The establishment of the knowledge graph includes the following steps:

[0061] S1: Obtain acute myocardial infarction literature through online and offline methods, and correct the acute myocardial infarction literature obtained online except for the set channels to obtain a number of acute myocardial infarction literature;

[0062] S2: Extract the keywords and the logical relationships between the keywords from each acute myocardial infarction literature respectively;

[0063] S3: Logically connect each acute myocardial infarction literature according to the keywords and the logical relationships between the keywords in each acute myocardial infarction literature;

[0064] S4: Correlate the keywords with the acute myocardial infarction literature respectively, form a knowledge graph according to the logical relationships between the keywords, and call the corresponding acute myocardial infarction literature according to the keywords.

[0065] Among them, in step S1, online and offline refer to online and offline of the Internet, and the set channels refer to technical academic websites such as CNKI and journals. The acquisition method can be obtained in real time through a crawler process to ensure the currency of the knowledge graph. In step S2, through keyword extraction technology, the extracted keywords are generally nouns, and the logical relationships between them are generally logical words. The article structure of the literature is analyzed by artificial intelligence, and combined with the article structure, keywords and logical words, the logical relationships between the keywords are obtained. Finally, the overall meaning of the article can be expressed by an expression. Step S3 is to associate each acute myocardial infarction literature together, that is, fuse the expressions corresponding to each acute myocardial infarction literature together to form a grid graph, which is the knowledge graph in step S4 and also the knowledge graph of the present invention. When calling later, since there is a corresponding citation relationship between the keywords and the acute myocardial infarction literature, the acute myocardial infarction literature can be retrieved according to the keywords.

[0066] Specifically, since the content recorded in the acute myocardial infarction literature obtained from non-set channels may have some differences from the actual situation, in order to ensure the accuracy of the knowledge, in step S1 of the present invention, the correction of the acute myocardial infarction literature obtained online except for the set channels includes:

[0067] Screen the acute myocardial infarction literature obtained from each online channel according to the channel, and obtain the acute myocardial infarction literature from each online channel except for the set channels;

[0068] Extract the keywords and the logical relationships between the keywords in the acute myocardial infarction literature from each online channel except for the set channels, and obtain modification suggestions through a pre-constructed neural network model for the keywords and the logical relationships between the keywords;

[0069] Modify the acute myocardial infarction literature on each online channel except the set channel according to the modification suggestions;

[0070] The neural network model outputs modification suggestions by inputting the entered keywords and the logical relationships between the keywords.

[0071] Based on the acute myocardial infarction literature of the set channel, this invention corrects the acute myocardial infarction literature of other online channels. The correction method still uses keywords and their corresponding logical relationships. The neural network model used is pre-constructed and continuously corrected later. Of course, the correction process is calibrated again with artificial experience. The modification suggestions obtained by this model are the basis for modifying the acute myocardial infarction literature, and the modification suggestions can be decomposed into keywords and their corresponding logical relationships later.

[0072] Among them, modifying the acute myocardial infarction literature on each online channel except the set channel according to the modification suggestions includes:

[0073] Decompose the modification suggestions to obtain multiple modification contents;

[0074] Determine the paragraphs in the acute myocardial infarction literature corresponding to the modification contents according to the keywords in the modification contents;

[0075] Expand the modification contents according to the keywords in the modification contents to obtain the corrected contents;

[0076] Replace the paragraphs in the acute myocardial infarction literature corresponding to the modification contents with the corrected contents to obtain the modified acute myocardial infarction literature.

[0077] After sorting the modification opinions in this invention, determine the paragraphs corresponding to each modification content, and then modify each paragraph separately. When modifying, in order to ensure the correctness of the modification, the method of deleting and rewriting is used. This invention uses expanding according to the keywords, and uses vernacular to expand the keywords so that ordinary people can also easily understand.

[0078] In this invention, there are basically one or two keywords corresponding to each paragraph. The keywords can be not only nouns, but also verb-object phrases consisting of a verb and a noun.

[0079] In this invention, when expanding, the keywords of the paragraph are advanced by at least two.

[0080] Among them, expanding the modification contents according to the keywords in the modification contents to obtain the corrected contents includes:

[0081] Search for the paragraph with the highest similarity in the set online channel and the offline acute myocardial infarction literature according to the order of the keywords to obtain the similar paragraph;

[0082] Rewrite similar paragraphs using AI rewriting technology to obtain corrected content.

[0083] In order to accurately express the professional content of acute myocardial infarction, the present invention searches for paragraphs with high similarity on the Internet and then uses rewriting. When determining the similarity of paragraphs, it is based on the order of the keywords corresponding to the paragraphs and the order of the keywords in the modified content. After determining the similarity, AI rewriting technology is used for rewriting to ensure that the content of the corrected paragraph is correct knowledge of acute myocardial infarction.

[0084] Based on the obtained knowledge graph, in the knowledge graph analysis module, according to the current physical state of the patient, through the pre-established knowledge graph, the self-management needs of the patient are obtained, including:

[0085] Convert the current physical state of the patient into keywords;

[0086] Search in the knowledge graph according to the keywords of the physical state, and find the acute myocardial infarction literature corresponding to the keywords of the physical state;

[0087] Extract the associations between the keywords in the acute myocardial infarction literature corresponding to the physical state keywords;

[0088] Determine the self-management needs according to the logical relationship between the keywords in the acute myocardial infarction literature.

[0089] The present invention converts the physical state into keywords. Since the above physical state is the development degree of each physical index of the body, the keywords are the degrees of each physical index of the body, that is to say, the keywords here are multiple keywords. Then search in the knowledge graph according to the existing keywords of the physical state to obtain the literature. Searching directly for words has the effects of simplicity, high efficiency and rapidity, so as to obtain the corresponding acute myocardial infarction literature, and then determine the logical relationship between its keywords and the keywords, that is, the expression corresponding to the acute myocardial infarction literature. Finally, determine the self-management needs according to the acute myocardial infarction literature, which achieves the purpose of the present invention.

[0090] Among them, determining the self-management needs according to the logical relationship between the keywords in the acute myocardial infarction literature includes:

[0091] Determine the keywords included in the self-management needs according to the logical relationship between the keywords in the acute myocardial infarction literature;

[0092] Determine the keywords of the self-management needs according to the labels of the keywords included in the self-management needs;

[0093] According to the keywords of self - management needs, use AI to roughly write methods to obtain self - management needs.

[0094] In the present invention, in order to obtain self - management needs more quickly, keywords and their corresponding logical relationships are also used. At the same time, tags of keywords are used. The tags of keywords represent the nature of the keywords, that is, whether this keyword is principle - based, state - source - based, or result - based. In the present invention, the tag of self - management needs is result - based, the keywords of physical states are source - based, and the principle - based keywords are the intermediate process principles. According to the tags of keywords, the keywords of self - management needs can be quickly locked, so as to obtain self - management needs.

[0095] Acute myocardial infarction is a very urgent disease and can achieve an emergency effect within the golden time. Therefore, the present invention has extremely high requirements for processing timeliness. Therefore, the present invention uses the grid of the knowledge graph as a whole, combines keywords to search for knowledge, so as to improve the efficiency of the output result and strive for the fastest golden time for patients.

[0096] The patient self - management module is used to provide corresponding services according to the patient's self - management needs to complete the patient's self - management, and it includes:

[0097] Decompose the needs of the patient's self - management needs to obtain each self - management content;

[0098] Decompose each self - management content respectively to obtain management time and management matters;

[0099] Generate corresponding services according to management practice and management matters;

[0100] According to the services corresponding to each self - management content, start the corresponding service resources to provide services for patients.

[0101] In the present invention, when providing services to patients according to self - management, the corresponding service resources can be connections to various departments of the hospital, so that medical staff can arrive quickly to provide emergency assistance to patients. Through the system of the present invention, judge the urgency of the patient's need for services and adjust the order of service of medical staff, so as to provide good services for patients with acute myocardial infarction.

[0102] The above - mentioned are only the preferred embodiments of the present invention, and do not impose any limitations on the present invention. Any simple modifications, changes, and equivalent structural changes made to the above - mentioned embodiments according to the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A self-management system for patients with acute myocardial infarction based on knowledge graph, characterized in that: include: A patient condition acquisition module is used to obtain the patient's recent physical condition and current physical condition; A patient condition analysis module, used to determine the patient's current physical condition based on the patient's recent physical condition and current physical condition; The knowledge graph analysis module is used to obtain the patient's self-management needs based on the patient's current physical condition through a pre-established knowledge graph; Patient self-management module, used to provide corresponding services to complete patient self-management according to the patient's self-management needs; The knowledge graph is established by acquiring online and offline acute myocardial infarction literature and analyzing the acute myocardial infarction literature to construct a knowledge graph.

2. A self-management system for patients with acute myocardial infarction based on a knowledge graph according to claim 1, characterized in that: The establishment of the knowledge graph includes the following steps: S1: Acute myocardial infarction literature was obtained through online and offline methods, and the acute myocardial infarction literature obtained online except for the set channels was corrected to obtain several acute myocardial infarction literature; S2: Extract the keywords from each acute myocardial infarction literature and the logical relationship between the keywords; S3: Logically connect each acute myocardial infarction literature based on the keywords of each acute myocardial infarction literature and the logical relationship between the keywords; S4: Match the keywords with the acute myocardial infarction literature respectively, and form a knowledge graph based on the logical relationship between the keywords, and call the corresponding acute myocardial infarction literature based on the keywords.

3. A self-management system for patients with acute myocardial infarction based on a knowledge graph according to claim 2, characterized in that: In step S1, the correction of the acute myocardial infarction literature obtained online except for the set channel includes: The acute myocardial infarction literature obtained from various online channels is screened according to the channels, and the acute myocardial infarction literature from various online channels except the set channels is obtained; Extract keywords and logical relationships between keywords from acute myocardial infarction literature in various online channels except the set channels, and obtain modification suggestions for keywords and logical relationships between keywords through a pre-built neural network model; According to the modification suggestions, the acute myocardial infarction literature in various online channels except the set channels was modified; The neural network model outputs modification suggestions by inputting the entered keywords and the logical relationships between the keywords.

4. A self-management system for patients with acute myocardial infarction based on knowledge graph according to claim 3, characterized in that: According to the modification suggestions, the acute myocardial infarction literature in various online channels other than the set channels will be modified, including: Decomposing the modification suggestion to obtain multiple modification contents; Determine the paragraphs in the acute myocardial infarction literature corresponding to the modified content based on the keywords in the modified content; Expand the modified content according to the keywords of the modified content to obtain the revised content; The revised content replaces the paragraph in the acute myocardial infarction document corresponding to the revised content to obtain a revised acute myocardial infarction document.

5. A self-management system for patients with acute myocardial infarction based on a knowledge graph according to claim 4, characterized in that: The step of expanding the modified content according to the keywords of the modified content to obtain the revised content includes: According to the order of keywords, the most similar paragraphs are searched in the online set channels and offline acute myocardial infarction literature to obtain similar paragraphs; Similar paragraphs are rewritten using AI rewriting technology to obtain corrected content.

6. A self-management system for patients with acute myocardial infarction based on knowledge graph according to claim 5, characterized in that: In the knowledge graph analysis module, the patient's self-management needs are obtained based on the patient's current physical condition through a pre-established knowledge graph, including: Translate the patient's current physical condition into keywords; Search the knowledge graph according to the keywords of the physical condition to find the literature on acute myocardial infarction corresponding to the keywords of the physical condition; According to the physical condition keywords corresponding to the acute myocardial infarction literature, the association between the keywords of the acute myocardial infarction literature was extracted; Self-management needs were identified based on the logical relationships between key words in the literature on acute myocardial infarction.

7. A self-management system for patients with acute myocardial infarction based on knowledge graph according to claim 6, characterized in that: The self-management needs were determined based on the logical relationship between key words in the literature on acute myocardial infarction, including: According to the logical relationship between the keywords in the literature on acute myocardial infarction, the keywords included in the self-management needs were determined; Determine the keywords of self-management needs according to the tags of the keywords contained in the self-management needs; According to the keywords of self-management needs, use AI rough writing method to obtain self-management needs.

8. A self-management system for patients with acute myocardial infarction based on knowledge graph according to claim 1, characterized in that: The patient self-management module provides corresponding services to complete the patient's self-management according to the patient's self-management needs, including: Decomposing the patient's self-management needs to obtain various self-management contents; Decompose each self-management content separately to obtain management time and management matters; Generate corresponding services based on management practices and management matters; According to the services corresponding to each self-management content, the corresponding service resources are activated to provide services to patients.

Images