A data intelligent recommendation method, system and medium based on medical services

Abstract

The present invention discloses a method, system and medium for intelligent data recommendation based on medical services. A three-dimensional map model is constructed for the target medical area, and information is extracted based on the user's initial patient data, and search tags are generated to retrieve medical big data, recommend service content and medical treatment procedures. The user's movement route is tracked by a camera device, and the medical treatment route is recorded and recommended in the map model. Furthermore, medical feature extraction is performed in combination with the user's medical treatment results and initial patient information, and medical service data of similar users is extracted after cluster analysis, forming second recommendation data and sending it to the user. Through the present invention, more efficient and accurate medical service recommendations can be provided.

Description

Technical Field

[0001] The present invention relates to the field of medical data analysis, and more specifically, to a data intelligent recommendation method, system and medium based on medical services. Background Art

[0002] With the rapid development of medical technology and the continuous accumulation of medical data, how to provide more efficient and accurate medical services through data analysis has become an urgent need in the medical industry. The traditional medical service model is often based on the doctor's experience for diagnosis and treatment, and lacks in-depth analysis and application of a large amount of medical data. Although the existing medical data recommendation system can initially realize data-based recommendation services, it still needs to be improved in terms of dynamic analysis of patient needs, recommendation accuracy and personalization. The existing technology lacks similar patient analysis of the target user and lacks effective mining of historical data. Therefore, it is necessary to develop a data intelligent recommendation method based on medical services. Summary of the invention

[0003] The present invention overcomes the defects of the prior art and proposes a data intelligent recommendation method, system and medium based on medical services.

[0004] The first aspect of the present invention provides a data intelligent recommendation method based on medical services, comprising:

[0005] Based on the target medical area, build a three-dimensional map model;

[0006] Acquire initial patient data of a target user, extract patient information based on the initial patient data, and form initial patient information in a preset format;

[0007] Generate retrieval tags based on initial patient information, conduct preliminary retrieval from medical big data based on the retrieval tags, obtain recommended service content data, conduct medical process recommendation analysis based on initial patient information, and form medical process recommendation data;

[0008] The recommended service content data and the recommended medical consultation process data are sent to the user terminal as the first recommended medical service data, and the user's moving route is tracked in the target medical area by a camera device, and recorded and recommended in the map model;

[0009] Obtain the user's medical treatment result data, extract the user's medical features based on semantic analysis based on the user's medical treatment result data and the initial patient information, use the medical features as cluster sample data to perform cluster analysis from historical patient data and obtain cluster group patient data;

[0010] Medical service data of similar users is extracted based on the cluster group patient data, and second recommended data is formed, and the second recommended medical service data is sent to the user terminal.

[0011] In this solution, the three-dimensional map model is constructed based on the target medical area, specifically:

[0012] Obtain the regional outline, regional area, and regional layout information of the target medical area;

[0013] Construct a three-dimensional visual map model based on regional outline, regional area, and regional layout information;

[0014] Generate movable paths and route range information in the map model through regional layout information.

[0015] In this solution, the initial patient data of the target user is obtained, patient information is extracted based on the initial patient data, and initial patient information in a preset format is formed, specifically:

[0016] Obtaining initial patient data of a target user through a user terminal, wherein the initial patient data includes basic information, registration information, and medical service demand information of the target user;

[0017] Based on the initial patient data, information extraction and information organization in a preset format are performed to form initial patient information.

[0018] In this solution, the retrieval tags are generated based on the initial patient information, and a preliminary search is performed from the medical big data based on the retrieval tags to obtain the recommended service content data. The medical consultation process recommendation analysis is performed through the initial patient information to form the medical consultation recommendation process data, which is specifically:

[0019] Generate information retrieval tags based on the initial patient information, conduct preliminary retrieval of medical big data in the system database based on the information retrieval tags, and obtain recommended service content data;

[0020] The medical treatment process recommendation analysis is performed through the initial patient information, and the process is matched from multiple preset medical treatment processes to form the medical treatment process recommendation data.

[0021] In this solution, the recommended service content data and the recommended medical treatment process data are sent to the user terminal as the first recommended medical service data, and the user's movement route is tracked in the target medical area by a camera device, and recorded and recommended in the map model. Specifically:

[0022] Sending the recommended service content data and the recommended medical consultation process data as first recommended medical service data to the user terminal;

[0023] Generate multiple waypoints based on the recommended medical treatment process data, perform shortest route planning based on the multiple waypoints combined with the map model, and generate a recommended medical treatment route;

[0024] In the target medical area, real-time video data is collected by a camera device, key frames are extracted from the real-time video data, and a key image set is obtained;

[0025] Import the key image set into the CNN-based image recognition model for user identification and tracking, and combine it with the map model to generate the user's real-time movement route;

[0026] Compare and warn the user's movement route based on the recommended medical treatment route and the user's real-time movement route.

[0027] In this solution, the user's medical treatment result data is obtained, the user's medical treatment result data and the initial patient information are subjected to semantic analysis to extract the user's medical features, and the medical features are used as cluster sample data to perform cluster analysis from the historical patient data to obtain cluster group patient data, specifically:

[0028] After the consultation process, the user's consultation result data is obtained through the target medical terminal;

[0029] Clean the user's medical result data and the initial patient information and convert them into text format to form first text data and second text data respectively;

[0030] Build a BERT-based semantic model, import the first text data and the second text data into the semantic model, first perform word segmentation, convert each word into a vector representation through an embedding layer, and obtain first vector data and second vector data;

[0031] Using the first vector data and the second vector data as medical feature data of the current user;

[0032] Obtain the medical characteristic data of historical patients, use all the medical characteristic data of the current user and historical patients as clustering sample data, and perform cyclic clustering on the clustering sample data based on the Kmeans clustering algorithm until the clustering group no longer changes. During the clustering process, the distance analysis between the data is based on the standard Euclidean distance calculation, and finally the clustering result is obtained;

[0033] Based on the clustering results, the cluster group data corresponding to the current user is obtained and marked as cluster group patient data.

[0034] In this solution, the medical service data of similar users is extracted based on the cluster group patient data, and the second recommended data is formed, and the second recommended medical service data is sent to the user terminal, specifically:

[0035] Obtain similar patient information based on clustered group patient data;

[0036] Obtain the corresponding user's medical consultation result data and initial patient information through similar patient information, and mark them as second medical consultation result data and second initial patient information respectively;

[0037] Generate a search tag through the second consultation result data and the second initial patient information, perform medical service data retrieval through the medical big data, and generate second recommendation data;

[0038] The second recommended medical service data is sent to the user terminal.

[0039] The second aspect of the present invention further provides a data intelligent recommendation system based on medical services, the system comprising: a memory, a processor, the memory comprising a data intelligent recommendation program based on medical services, the data intelligent recommendation program based on medical services being executed by the processor to implement the following steps:

[0040] Based on the target medical area, build a three-dimensional map model;

[0041] Acquire initial patient data of a target user, extract patient information based on the initial patient data, and form initial patient information in a preset format;

[0042] Generate retrieval tags based on initial patient information, conduct preliminary retrieval from medical big data based on the retrieval tags, obtain recommended service content data, conduct medical process recommendation analysis based on initial patient information, and form medical process recommendation data;

[0043] The recommended service content data and the recommended medical consultation process data are sent to the user terminal as the first recommended medical service data, and the user's moving route is tracked in the target medical area by a camera device, and recorded and recommended in the map model;

[0044] Obtain the user's medical treatment result data, extract the user's medical features based on semantic analysis based on the user's medical treatment result data and the initial patient information, use the medical features as cluster sample data to perform cluster analysis from historical patient data and obtain cluster group patient data;

[0045] Medical service data of similar users is extracted based on the cluster group patient data, and second recommended data is formed, and the second recommended medical service data is sent to the user terminal.

[0046] The third aspect of the present invention also provides a computer-readable storage medium, which includes a data intelligent recommendation program based on medical services. When the data intelligent recommendation program based on medical services is executed by a processor, the steps of the data intelligent recommendation method based on medical services as described in any one of the above items are implemented.

[0047] The present invention discloses a method, system and medium for intelligent data recommendation based on medical services. A three-dimensional map model is constructed for the target medical area, and information is extracted based on the user's initial patient data, and search tags are generated to retrieve medical big data, recommend service content and medical treatment procedures. The user's movement route is tracked by a camera device, and the medical treatment route is recorded and recommended in the map model. Furthermore, medical feature extraction is performed in combination with the user's medical treatment results and initial patient information, and medical service data of similar users is extracted after cluster analysis, forming second recommendation data and sending it to the user. Through the present invention, more efficient and accurate medical service recommendations can be provided. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A flow chart of a data intelligent recommendation method based on medical services of the present invention is shown;

[0049] Figure 2 A block diagram of a data intelligent recommendation system based on medical services of the present invention is shown. DETAILED DESCRIPTION

[0050] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0051] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0052] Figure 1 A flow chart of a data intelligent recommendation method based on medical services of the present invention is shown.

[0053] like Figure 1 As shown, the first aspect of the present invention provides a data intelligent recommendation method based on medical services, comprising:

[0054] S102, constructing a three-dimensional map model based on the target medical area;

[0055] S104, acquiring initial patient data of the target user, extracting patient information based on the initial patient data, and forming initial patient information in a preset format;

[0056] S106, generating a search tag based on the initial patient information, performing a preliminary search from the medical big data based on the search tag, obtaining recommended service content data, performing a medical consultation process recommendation analysis based on the initial patient information, and forming medical consultation process recommendation data;

[0057] S108, sending the recommended service content data and the recommended medical consultation process data as first recommended medical service data to the user terminal, tracking the user's moving route in the target medical area through a camera device, and recording and recommending the medical consultation route in a map model;

[0058] S110, obtaining user medical treatment result data, extracting user medical features based on semantic analysis of the user medical treatment result data and initial patient information, performing cluster analysis on the medical features as cluster sample data from historical patient data and obtaining cluster group patient data;

[0059] S112, extracting medical service data of similar users based on the cluster group patient data, and forming second recommended data, and sending the second recommended medical service data to the user terminal.

[0060] According to an embodiment of the present invention, the three-dimensional map model is constructed based on the target medical area, specifically:

[0061] Obtain the regional outline, regional area, and regional layout information of the target medical area;

[0062] Construct a three-dimensional visual map model based on regional outline, regional area, and regional layout information;

[0063] Generate movable paths and route range information in the map model through regional layout information.

[0064] It should be noted that the target medical area is generally a hospital or clinic area, and the map model is constructed for use. The regional layout information includes the spatial layout information of each department and room in the medical area, which is used to analyze the corresponding personnel movement route information.

[0065] According to an embodiment of the present invention, the initial patient data of the target user is obtained, patient information is extracted based on the initial patient data, and initial patient information in a preset format is formed, specifically:

[0066] Obtaining initial patient data of a target user through a user terminal, wherein the initial patient data includes basic information, registration information, and medical service demand information of the target user;

[0067] Based on the initial patient data, information extraction and information organization in a preset format are performed to form initial patient information.

[0068] It should be noted that the information extraction in the preset format is to extract the corresponding patient information in text format and form the initial patient information based on the preset arrangement form, which is equivalent to a standardization process, which is conducive to subsequent feature extraction. The basic information includes the user's gender, age, medical history and other information. The user terminal includes a mobile terminal and a computer terminal, which is used to interact with the terminal of the medical institution through network interconnection.

[0069] According to an embodiment of the present invention, the retrieval tag is generated based on the initial patient information, and a preliminary search is performed from the medical big data based on the retrieval tag to obtain the recommended service content data, and the medical consultation process recommendation analysis is performed through the initial patient information to form the medical consultation recommendation process data, which is specifically:

[0070] Generate information retrieval tags based on the initial patient information, conduct preliminary retrieval of medical big data in the system database based on the information retrieval tags, and obtain recommended service content data;

[0071] The medical treatment process recommendation analysis is performed through the initial patient information, and the process is matched from multiple preset medical treatment processes to form the medical treatment process recommendation data.

[0072] It should be noted that the initial patient information includes key patient information of the target user, such as disease type, corresponding department, examination service, medical needs, living habits, etc., and is the data for recommendation analysis before the user sees the doctor. It is general data with certain universality (among patients of the same type). By generating recommended service content data and recommended medical process data, it is possible to make corresponding data recommendations and medical process recommendations for users before seeing the doctor (or before registering), improve the user's service experience, and at the same time, it can collect information data for users before seeing the doctor, so as to better facilitate the subsequent medical big data analysis, which has certain information application value.

[0073] According to an embodiment of the present invention, the recommended service content data and the recommended medical consultation process data are sent to the user terminal as the first recommended medical service data, the user's moving route is tracked in the target medical area by a camera device, and recorded and recommended in the map model, specifically:

[0074] Sending the recommended service content data and the recommended medical consultation process data as first recommended medical service data to the user terminal;

[0075] Generate multiple waypoints based on the recommended medical treatment process data, perform shortest route planning based on the multiple waypoints combined with the map model, and generate a recommended medical treatment route;

[0076] In the target medical area, real-time video data is collected by a camera device, key frames are extracted from the real-time video data, and a key image set is obtained;

[0077] Import the key image set into the CNN-based image recognition model for user identification and tracking, and combine it with the map model to generate the user's real-time movement route;

[0078] Compare and warn the user's movement route based on the recommended medical treatment route and the user's real-time movement route.

[0079] It should be noted that the image recognition model based on CNN (convolutional neural network) is an AI recognition model with learning function, which can be trained based on the face database before recognition. The first recommended medical service data is the basic recommendation data, and data recommendation analysis is performed before the user visits the doctor to meet general service recommendations. The recommended service content data includes the patient's popular science data, disease warning and diagnosis history data, disease risk assessment data, disease-related information, etc., which are used to enable users to understand the disease and medical conditions, better understand medical services, and improve the efficiency of medical service communication and service efficiency.

[0080] According to an embodiment of the present invention, the user's medical treatment result data is obtained, the user's medical treatment result data and the initial patient information are subjected to semantic analysis to extract the user's medical features, and the medical features are used as cluster sample data to perform cluster analysis from the historical patient data to obtain cluster group patient data, specifically:

[0081] After the consultation process, the user's consultation result data is obtained through the target medical terminal;

[0082] Clean the user's medical result data and the initial patient information and convert them into text format to form first text data and second text data respectively;

[0083] Build a BERT-based semantic model, import the first text data and the second text data into the semantic model, first perform word segmentation, convert each word into a vector representation through an embedding layer, and obtain first vector data and second vector data;

[0084] Using the first vector data and the second vector data as medical feature data of the current user;

[0085] Obtain the medical characteristic data of historical patients, use all the medical characteristic data of the current user and historical patients as clustering sample data, and perform cyclic clustering on the clustering sample data based on the Kmeans clustering algorithm until the clustering group no longer changes. During the clustering process, the distance analysis between the data is based on the standard Euclidean distance calculation, and finally the clustering result is obtained;

[0086] Based on the clustering results, the cluster group data corresponding to the current user is obtained and marked as cluster group patient data.

[0087] It should be noted that the process of the cluster group no longer changes, that is, the center point of each cluster group no longer moves. In the present invention, a text semantic analysis process is introduced to perform semantic analysis on the user's user medical treatment result data and the initial patient information, and the vector obtained based on the semantic analysis is used as the cluster sample data, which can effectively group similar patients for users. In the process based on semantic analysis, the potential data connection features can be effectively analyzed. The semantic analysis is based on natural language processing, which conforms to the semantic analysis of general medical terms. Semantic feature extraction is performed based on the two combined dimensions of diagnosis results and initial patient information, and the clustering of patient big data is realized in one step, which effectively extracts similar historical patients and further makes accurate and accurate recommendations for medical data. In traditional technologies, similar group analysis of target users is generally based on simple similarity analysis of diagnostic results, for example, whether they suffer from the same disease. In large-scale cluster analysis, it is difficult to perform medical feature association mining from the data, difficult to analyze potential similar patients for research, and difficult to achieve accurate medical data recommendation.

[0088] The first vector data and the second vector data correspond to the user's medical treatment result data and the initial patient information respectively, and the two are combined for cluster analysis.

[0089] According to an embodiment of the present invention, the medical service data of similar users is extracted based on the cluster group patient data, and the second recommended data is formed, and the second recommended medical service data is sent to the user terminal, specifically:

[0090] Obtain similar patient information based on clustered group patient data;

[0091] Obtain the corresponding user's medical consultation result data and initial patient information through similar patient information, and mark them as second medical consultation result data and second initial patient information respectively;

[0092] Generate a search tag through the second consultation result data and the second initial patient information, perform medical service data retrieval through the medical big data, and generate second recommendation data;

[0093] The second recommended medical service data is sent to the user terminal.

[0094] It should be noted that the second recommendation data includes medical recommendation data and recommended service content data. The recommendation data is the second recommendation data obtained by analyzing the users in the cluster group after the users have visited the doctor. It can effectively screen and mine historical big data, and retrieve recommended service data that is effective for the target users in the form of cluster group users, thereby greatly improving the user experience.

[0095] According to an embodiment of the present invention, the obtaining of medical characteristic data of historical patients, taking all medical characteristic data of the current user and historical patients as clustering sample data, further includes:

[0096] Determine whether the total amount of historical patient medical characteristic data is greater than the preset data amount. If so, obtain the initial patient information and user medical result data;

[0097] Import the initial patient information and user medical result data into the semantic model to perform semantic feature analysis based on medical services to form user diagnosis and treatment information;

[0098] Based on the user's diagnosis and treatment information, obtain the corresponding patient statistics from the system database;

[0099] Based on patient statistics, calculate and analyze the growth trend of patient personnel. The trend analysis is based on seasonal statistical calculations and obtains the patient growth rate in each time period;

[0100] Mark each time period in which the patient growth rate is greater than the preset growth rate, and select the most recent time period as the expected time period;

[0101] Extracting patient users from patient statistical data based on an expected time period to obtain selected patient users;

[0102] The medical characteristic data of the selected patient user and the medical characteristic data of the current user are used as clustering sample data.

[0103] It should be noted that the user diagnosis and treatment information includes disease type information, diagnosis result information, and treatment method information. In the process of large-scale patient data clustering, since the amount of data is often large, the actual analysis may result in excessive consumption of hardware resources. It is necessary to further streamline the data for clustering, and to ensure the integrity and rationality of the data as much as possible. The various time periods are generally seasonal periods. The patient statistical data includes statistical data such as the number of patients, patient types, and patient consultation time, which are used to count the growth trend of historical patients.

[0104] The present invention collects patient data and further analyzes the growth trend of patients similar to or consistent with the target patient, marks the statistical data of the recent growth trend, extracts the corresponding patients as cluster sample data, and reduces the amount of data to be analyzed as much as possible while ensuring the rationality of the cluster sample data, thereby improving the recommended data analysis of the target patient.

[0105] Through user diagnosis and treatment information, statistical data of patients similar or consistent with the target patient can be obtained from all databases, and the search is specifically based on the diagnosis and treatment dimensions.

[0106] It is worth mentioning that some diseases often have a certain seasonality, such as colds and fevers. Based on the patient trend, the present invention simplifies the corresponding statistical data and selects patient data with a certain growth trend period for feature analysis and clustering, which can effectively reduce the amount of original data and accurately analyze the clustering group of target users, improve recommendation efficiency and reduce system resource consumption.

[0107] Figure 2 A block diagram of a data intelligent recommendation system based on medical services of the present invention is shown.

[0108] The second aspect of the present invention further provides a data intelligent recommendation system 2 based on medical services, the system comprising: a memory 21, a processor 22, the memory comprising a data intelligent recommendation program based on medical services, the data intelligent recommendation program based on medical services when executed by the processor implements the following steps:

[0109] Based on the target medical area, build a three-dimensional map model;

[0110] Acquire initial patient data of a target user, extract patient information based on the initial patient data, and form initial patient information in a preset format;

[0111] Generate retrieval tags based on initial patient information, conduct preliminary retrieval from medical big data based on the retrieval tags, obtain recommended service content data, conduct medical process recommendation analysis based on initial patient information, and form medical process recommendation data;

[0112] The recommended service content data and the recommended medical consultation process data are sent to the user terminal as the first recommended medical service data, and the user's moving route is tracked in the target medical area by a camera device, and recorded and recommended in the map model;

[0113] Obtain the user's medical treatment result data, extract the user's medical features based on semantic analysis based on the user's medical treatment result data and the initial patient information, use the medical features as cluster sample data to perform cluster analysis from historical patient data and obtain cluster group patient data;

[0114] Medical service data of similar users is extracted based on the cluster group patient data, and second recommended data is formed, and the second recommended medical service data is sent to the user terminal.

[0115] According to an embodiment of the present invention, the three-dimensional map model is constructed based on the target medical area, specifically:

[0116] Obtain the regional outline, regional area, and regional layout information of the target medical area;

[0117] Construct a three-dimensional visual map model based on regional outline, regional area, and regional layout information;

[0118] Generate movable paths and route range information in the map model through regional layout information.

[0119] It should be noted that the target medical area is generally a hospital or clinic area, and the map model is constructed for use. The regional layout information includes the spatial layout information of each department and room in the medical area, which is used to analyze the corresponding personnel movement route information.

[0120] According to an embodiment of the present invention, the initial patient data of the target user is obtained, patient information is extracted based on the initial patient data, and initial patient information in a preset format is formed, specifically:

[0121] Obtaining initial patient data of a target user through a user terminal, wherein the initial patient data includes basic information, registration information, and medical service demand information of the target user;

[0122] Based on the initial patient data, information extraction and information organization in a preset format are performed to form initial patient information.

[0123] It should be noted that the information extraction in the preset format is to extract the corresponding patient information in text format and form the initial patient information based on the preset arrangement form, which is equivalent to a standardization process, which is conducive to subsequent feature extraction. The basic information includes the user's gender, age, medical history and other information. The user terminal includes a mobile terminal and a computer terminal, which is used to interact with the terminal of the medical institution through network interconnection.

[0124] According to an embodiment of the present invention, the retrieval tag is generated based on the initial patient information, and a preliminary search is performed from the medical big data based on the retrieval tag to obtain the recommended service content data, and the medical consultation process recommendation analysis is performed through the initial patient information to form the medical consultation recommendation process data, which is specifically:

[0125] Generate information retrieval tags based on the initial patient information, conduct preliminary retrieval of medical big data in the system database based on the information retrieval tags, and obtain recommended service content data;

[0126] The medical treatment process recommendation analysis is performed through the initial patient information, and the process is matched from multiple preset medical treatment processes to form the medical treatment process recommendation data.

[0127] It should be noted that the initial patient information includes key patient information of the target user, such as disease type, corresponding department, examination service, medical needs, living habits, etc., and is the data for recommendation analysis before the user sees the doctor. It is general data with certain universality (among patients of the same type). By generating recommended service content data and recommended medical process data, it is possible to make corresponding data recommendations and medical process recommendations for users before seeing the doctor (or before registering), improve the user's service experience, and at the same time, it can collect information data for users before seeing the doctor, so as to better facilitate the subsequent medical big data analysis, which has certain information application value.

[0128] According to an embodiment of the present invention, the recommended service content data and the recommended medical consultation process data are sent to the user terminal as the first recommended medical service data, the user's moving route is tracked in the target medical area by a camera device, and recorded and recommended in the map model, specifically:

[0129] Sending the recommended service content data and the recommended medical consultation process data as first recommended medical service data to the user terminal;

[0130] Generate multiple waypoints based on the recommended medical treatment process data, perform shortest route planning based on the multiple waypoints combined with the map model, and generate a recommended medical treatment route;

[0131] In the target medical area, real-time video data is collected by a camera device, key frames are extracted from the real-time video data, and a key image set is obtained;

[0132] Import the key image set into the CNN-based image recognition model for user identification and tracking, and combine it with the map model to generate the user's real-time movement route;

[0133] Compare and warn the user's movement route based on the recommended medical treatment route and the user's real-time movement route.

[0134] It should be noted that the image recognition model based on CNN (convolutional neural network) is an AI recognition model with learning function, which can be trained based on the face database before recognition. The first recommended medical service data is the basic recommendation data, and data recommendation analysis is performed before the user visits the doctor to meet general service recommendations. The recommended service content data includes the patient's popular science data, disease warning and diagnosis history data, disease risk assessment data, disease-related information, etc., which are used to enable users to understand the disease and medical conditions, better understand medical services, and improve the efficiency of medical service communication and service efficiency.

[0135] According to an embodiment of the present invention, the user's medical treatment result data is obtained, the user's medical treatment result data and the initial patient information are subjected to semantic analysis to extract the user's medical features, and the medical features are used as cluster sample data to perform cluster analysis from the historical patient data to obtain cluster group patient data, specifically:

[0136] After the consultation process, the user's consultation result data is obtained through the target medical terminal;

[0137] Clean the user's medical result data and the initial patient information and convert them into text format to form first text data and second text data respectively;

[0138] Build a BERT-based semantic model, import the first text data and the second text data into the semantic model, first perform word segmentation, convert each word into a vector representation through an embedding layer, and obtain first vector data and second vector data;

[0139] Using the first vector data and the second vector data as medical feature data of the current user;

[0140] Obtain the medical characteristic data of historical patients, use all the medical characteristic data of the current user and historical patients as clustering sample data, and perform cyclic clustering on the clustering sample data based on the Kmeans clustering algorithm until the clustering group no longer changes. During the clustering process, the distance analysis between the data is based on the standard Euclidean distance calculation, and finally the clustering result is obtained;

[0141] Based on the clustering results, the cluster group data corresponding to the current user is obtained and marked as cluster group patient data.

[0142] It should be noted that the process of the cluster group no longer changes, that is, the center point of each cluster group no longer moves. In the present invention, a text semantic analysis process is introduced to perform semantic analysis on the user's user medical treatment result data and the initial patient information, and the vector obtained based on the semantic analysis is used as the cluster sample data, which can effectively group similar patients for users. In the process based on semantic analysis, the potential data connection features can be effectively analyzed. The semantic analysis is based on natural language processing, which conforms to the semantic analysis of general medical terms. Semantic features are extracted based on the two combined dimensions of the diagnosis results and the initial patient information, and the clustering of patient big data is realized in one step, which effectively extracts similar historical patients and further makes accurate and accurate recommendations for medical data. In traditional technologies, similar group analysis of target users is generally based on simple similarity analysis of diagnostic results, for example, whether they suffer from the same disease. In large-scale cluster analysis, it is difficult to mine the association of medical features from the data, difficult to analyze potential similar patients for research, and difficult to achieve accurate medical data recommendation.

[0143] The first vector data and the second vector data correspond to the user's medical treatment result data and the initial patient information respectively, and the two are combined for cluster analysis.

[0144] According to an embodiment of the present invention, the medical service data of similar users is extracted based on the cluster group patient data, and the second recommended data is formed, and the second recommended medical service data is sent to the user terminal, specifically:

[0145] Obtain similar patient information based on clustered group patient data;

[0146] Obtain the corresponding user's medical consultation result data and initial patient information through similar patient information, and mark them as second medical consultation result data and second initial patient information respectively;

[0147] Generate a search tag through the second consultation result data and the second initial patient information, perform medical service data retrieval through the medical big data, and generate second recommendation data;

[0148] The second recommended medical service data is sent to the user terminal.

[0149] It should be noted that the second recommendation data includes medical recommendation data and recommended service content data. The recommendation data is the second recommendation data obtained by analyzing the users in the cluster group after the users have visited the doctor. It can effectively screen and mine historical big data, and retrieve recommended service data that is effective for the target users in the form of cluster group users, thereby greatly improving the user experience.

[0150] The third aspect of the present invention also provides a computer-readable storage medium, which includes a data intelligent recommendation program based on medical services. When the data intelligent recommendation program based on medical services is executed by a processor, the steps of the data intelligent recommendation method based on medical services as described in any one of the above items are implemented.

[0151] The present invention discloses a method, system and medium for intelligent data recommendation based on medical services. A three-dimensional map model is constructed for the target medical area, and information is extracted based on the user's initial patient data, and search tags are generated to retrieve medical big data, recommend service content and medical treatment procedures. The user's movement route is tracked by a camera device, and the medical treatment route is recorded and recommended in the map model. Furthermore, medical feature extraction is performed in combination with the user's medical treatment results and initial patient information, and medical service data of similar users is extracted after cluster analysis, forming second recommendation data and sending it to the user. Through the present invention, more efficient and accurate medical service recommendations can be provided.

[0152] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0153] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0154] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0155] A person of ordinary skill in the art can understand that: all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: a mobile storage device, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0156] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0157] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A data intelligent recommendation method based on medical services, characterized in that: include: Based on the target medical area, build a three-dimensional map model; Acquire initial patient data of a target user, extract patient information based on the initial patient data, and form initial patient information in a preset format; Generate retrieval tags based on initial patient information, conduct preliminary retrieval from medical big data based on the retrieval tags, obtain recommended service content data, conduct medical process recommendation analysis based on initial patient information, and form medical process recommendation data; The recommended service content data and the recommended medical consultation process data are sent to the user terminal as the first recommended medical service data, and the user's moving route is tracked in the target medical area by a camera device, and recorded and recommended in the map model; Obtain the user's medical treatment result data, extract the user's medical features based on semantic analysis based on the user's medical treatment result data and the initial patient information, use the medical features as cluster sample data to perform cluster analysis from historical patient data and obtain cluster group patient data; Extracting medical service data of similar users based on the cluster group patient data, forming second recommended data, and sending the second recommended medical service data to the user terminal; The step of obtaining the initial patient data of the target user, extracting patient information based on the initial patient data, and forming the initial patient information in a preset format is as follows: Obtaining initial patient data of a target user through a user terminal, wherein the initial patient data includes basic information, registration information, and medical service demand information of the target user; Extracting and arranging information in a preset format based on the initial patient data to form initial patient information; The method generates a search tag based on the initial patient information, performs a preliminary search from the medical big data based on the search tag, obtains the recommended service content data, and performs a medical consultation process recommendation analysis based on the initial patient information to form the medical consultation process recommendation data, specifically: Generate information retrieval tags based on the initial patient information, conduct preliminary retrieval of medical big data in the system database based on the information retrieval tags, and obtain recommended service content data; Perform a recommendation analysis of the medical process based on the initial patient information, match the process from multiple preset medical processes, and form the medical process recommendation data; The recommended service content data and the recommended medical treatment process data are sent to the user terminal as the first recommended medical service data, and the user's moving route is tracked in the target medical area by a camera device, and recorded and recommended in the map model, specifically: Sending the recommended service content data and the recommended medical consultation process data as first recommended medical service data to the user terminal; Generate multiple waypoints based on the recommended medical treatment process data, perform shortest route planning based on the multiple waypoints combined with the map model, and generate a recommended medical treatment route; In the target medical area, real-time video data is collected by a camera device, key frames are extracted from the real-time video data, and a key image set is obtained; Import the key image set into the CNN-based image recognition model for user identification and tracking, and combine it with the map model to generate the user's real-time movement route; Compare and warn users based on recommended medical treatment routes and real-time user movement routes; The method of obtaining the user's medical treatment result data, extracting the user's medical features based on semantic analysis from the user's medical treatment result data and the initial patient information, and performing cluster analysis on the medical features as cluster sample data from the historical patient data to obtain cluster group patient data is as follows: After the consultation process, the user's consultation result data is obtained through the target medical terminal; Clean the user's medical treatment result data and the initial patient information and convert them into text format to form first text data and second text data respectively; Build a BERT-based semantic model, import the first text data and the second text data into the semantic model, first perform word segmentation, convert each word into a vector representation through an embedding layer, and obtain first vector data and second vector data; Using the first vector data and the second vector data as medical feature data of the current user; Obtain the medical characteristic data of historical patients, use all the medical characteristic data of the current user and historical patients as clustering sample data, and perform cyclic clustering on the clustering sample data based on the Kmeans clustering algorithm until the clustering group no longer changes. During the clustering process, the distance analysis between the data is based on the standard Euclidean distance calculation, and finally the clustering result is obtained; Based on the clustering results, the cluster group data corresponding to the current user is obtained and marked as cluster group patient data; The method of extracting medical service data of similar users based on the cluster group patient data and forming second recommended data, and sending the second recommended medical service data to the user terminal, is specifically as follows: Obtain similar patient information based on clustered group patient data; Obtain the corresponding user's medical consultation result data and initial patient information through similar patient information, and mark them as second medical consultation result data and second initial patient information respectively; Generate a search tag through the second consultation result data and the second initial patient information, perform medical service data retrieval through the medical big data, and generate second recommendation data; sending the second recommended medical service data to the user terminal; The obtaining of medical characteristic data of historical patients, taking all medical characteristic data of the current user and historical patients as clustering sample data, includes: Determine whether the total amount of historical patient medical characteristic data is greater than the preset data amount. If so, obtain the initial patient information and user medical result data; Import the initial patient information and user medical result data into the semantic model to perform semantic feature analysis based on medical services to form user diagnosis and treatment information; Based on the user's diagnosis and treatment information, obtain the corresponding patient statistics from the system database; Based on patient statistics, calculate and analyze the growth trend of patient personnel. The trend analysis is based on seasonal statistical calculations and obtains the patient growth rate in each time period; Mark each time period in which the patient growth rate is greater than the preset growth rate, and select the most recent time period as the expected time period; Extracting patient users from patient statistical data based on an expected time period to obtain selected patient users; The medical characteristic data of the selected patient user and the medical characteristic data of the current user are used as clustering sample data; The step of obtaining medical characteristic data of historical patients and taking all medical characteristic data of the current user and historical patients as clustering sample data further includes: Determine whether the total amount of historical patient medical characteristic data is greater than the preset data amount. If so, obtain the initial patient information and user medical result data; Import the initial patient information and user medical result data into the semantic model to perform semantic feature analysis based on medical services to form user diagnosis and treatment information; Based on the user's diagnosis and treatment information, obtain the corresponding patient statistics from the system database; Based on patient statistics, calculate and analyze the growth trend of patient personnel. The trend analysis is based on seasonal statistical calculations and obtains the patient growth rate in each time period; Mark each time period in which the patient growth rate is greater than the preset growth rate, and select the most recent time period as the expected time period; Extracting patient users from patient statistical data based on an expected time period to obtain selected patient users; The medical characteristic data of the selected patient user and the medical characteristic data of the current user are used as clustering sample data.

2. According to claim 1, a data intelligent recommendation method based on medical services is characterized in that: The three-dimensional map model is constructed based on the target medical area, specifically: Obtain the regional outline, regional area, and regional layout information of the target medical area; Construct a three-dimensional visual map model based on regional outline, regional area, and regional layout information; Generate movable paths and route range information in the map model through regional layout information.

3. A data intelligent recommendation system based on medical services, characterized in that: The system includes: a memory and a processor, wherein the memory includes a data intelligent recommendation program based on medical services, and when the data intelligent recommendation program based on medical services is executed by the processor, the steps of the data intelligent recommendation method based on medical services as described in any one of claims 1 to 2 are implemented.

4. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a data intelligent recommendation program based on medical services. When the data intelligent recommendation program based on medical services is executed by a processor, the steps of the data intelligent recommendation method based on medical services as described in any one of claims 1 to 2 are implemented.

Images